Geologic studies in Alaska by the U.S. Geological Survey, 1991

1992 · Report

<p>This collection of twenty-one papers continues the annual series of U.S. Geological Survey reports on the geology of Alaska. These contributions, which…

Public-domain full text preserved in the Mountain Man Mining Library. Original source: pubs.usgs.gov.

Geologic Studies in Alaska by the U.S. Geological Survey, 1991

AVAILABILITY OF BOOKS AND MAPS OF THE U.S. GEOLOGICAL SURVEY Instructions on ordering publications of the U.S. Geological Survey, along with the last offerings, are given in the current-year issues of the monthly catalog "New Publications of the U.S. Geological Survey." Prices of available U.S. Geological Survey publications released prior to the current year are listed in the most recent annual "Price and Availability List" Publications that are listed in various U.S. Geological Survey catalogs (see back inside cover) but not listed in the most recent annual "Price and Availability List" are no longer available. Prices of reports released to the open files are given in the listing "U.S. Geological Survey Open-File Reports," updated monthly, which is for sale in microfiche from U.S. Geological Survey Book and Open-File Report Sales, Box 25425, Denver, CO 80225. Order U.S. Geological Survey publications by mail or over the counter from the offices given below. BY MAIL Books Professional Papers, Bulletins, Water-Supply Papers, Techniques of Water-Resources Investigat;ions, Circulars, publications of general interest (such as leaflets, pamphlets, booklets), single copies of periodicals (Earthquakes & Volcanoes, Preliminary Determination of Epicenters), and some miscellaneous reports, including some of the foregoing series that have gone out of print at the Superintendent of Documents, are obtainable by mail from U.S. Geological Survey, Book and Open-File Report Sales Box2S425 Denver, CO 80225 Subscriptions to periodicals (Earthquakes & Volcanoes and Preliminary Determination of Epicenters) can be obtained ONLY from Superintendent of Documents U.S. Government Printing Office Washington, DC 20402 (Check or money order must be payable to Superintendent of Documents.) Maps For maps , address mail orders to U.S. Geological Survey, Map Sales Box 25286 Denver, CO 80225 Residents of Alaska may order maps from U.S. Geological Survey, Map Sales 101 Twelfth Ave.- Box 12 Fairbanks, AK 99701 OVER THE COUNTER Books Books of the U.S. Geological Survey are available over the counter at the following U.S. Geological Survey offices, all of which are authorized agents of the Superintendent of Documents. ANCHORAGE, Alaska--4230 University Dr., Rm. 101 ANCHORAGE, Alaska--605 West 4th Ave., Rm G-84 DENVER, Colorado--Federal Bldg., Rm. 169, 1961 Stout St. LAKEWOOD, Colorado-- Federal Center, Bldg. 810 MENLO PARK, California--Bldg. 3,Rm. 3128, 345Middlefield Rd. RESTON, Virginia--National Center, Rm. 1C402, 12201 Sunrise Valley Dr. SALT LAKE CITY, Utah--Federal Bldg., Rm. 8105, 125 South State St. SAN FRANCISCO, California--Customhouse, Rm. 504, 555 Battery St. SPOKANE, Washington--U.S. Courthouse, Rm. 678, West 920 Riverside Ave. WASHINGTON, D.C.--U.S. Department of the Interior Bldg., Rm. 2650, 1849 C St., NW. Maps Maps may be purchased over the counter at the U.S. Geological Survey offices where books are sold (all addresses in above list) and at the following Geological Survey offices: ROLLA, Missouri--1400 Independence Rd. FAIRBANKS, Alaska--New Federal Building, 101 Twelfth Ave.

Geologic Studies in Alaska by the U.S. Geological Survey, 1991 DWIGHT C. BRADLEY and CYNTHIA DUSEL-BACON, Editors U.S. GEOLOGICAL SURVEY BULLETIN 2041

U.S. DEPARTMENT OF THE INTERIOR MANUEL LUJAN, JR., Secretary U.S. GEOLOGICAL SURVEY Dallas L. Peck, Director Any use of trade, product, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. Government UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1992 For sale by Book and Open-File Report Sales U.S. Geological Survey Federal Center, Box 25286 Denver, CO 80225 COVER: Aerial view looking southwest from Muldrow Glacier toward Mount McKinley (elevation 6,193 m), the summit of which is covered by an ice cloud. The combination of more than 4,000 m vertical relief and granitoid bedrock containing minerals suitable for fission-track dating is ideal for determination of uplift rates (see article by Plafker and others, page 202). Photograph by George Plafker, july 1976.

CONTENTS Introduction Dwight C. Bradley and Cynthia Dusel-Bacon MINERAL RESOURCE STUDIES- ARTICLES Fluid-inclusion study of the Rock Creek area, Nome Mining District, Seward Peninsula, Alaska Lori E. Apodaca Geochemistry of lode-gold deposits, Nuka Bay district, southern Kenai Peninsula J. Carter Borden, Richard J. Goldfarb, Carol A. Gent, Robert C. Burruss, and Bruce H. Roushey Placer gold of the Kenai lowland Barrett A. Cieutat, Richard J. Goldfarb, Dwight C. Bradley, and Bruce H. Roushey Summary of results of the mineral resource assessment of the Bethel and southeastern part of the Russian Mission 1 o by 3° quadrangles, Alaska Thomas P. Frost, Stephen E. Box, and Elizabeth J. Moll-Stalcup Comparison of the effectiveness of stream-sediment, heavy-mineral-concentrate, aquaticmoss, and stream-water geochemical sample media for the mineral assessment study of the Iditarvd quadrangle, Alaska John E. Gray, Philip L. Hageman, and Jean L. Ryder Geochemically anomalous areas in the west-central part of the Howard Pass quadrangle, National Petroleum Reserve, Alaska: Evidence for sediment-hosted Zn-Pb-Ag-Ba mineralization Karen D. Kelly, J. Carter Borden, Elizabeth A. Bailey, David L. Fey, Jerry M. Motooka, and Bruce H. Roushey A followup geochemical survey of base-metal anomalies in the Ward Creek/Windfall Harbor and Gambier Bay areas, Admiralty Island, Southeast Alaska Cliff D. Taylor, Barrett A. Cieutat, and Lance D. Miller Experimental abrasion of detrital gold in a tumbler Warren Y eend MINERAL RESOURCE STUDIES -GEOLOGIC NOTES Gold in the Usibelli Group coals, Nenana Coal Field, Alaska Gary D. Stricker, Richard B. Tripp, John B. McHugh, Ronald H. Affolter, and John B. Cathrall Rare earth minerals in "thunder eggs" from Zarembo Island, southeast Alaska John Philpotts and John R. Evans Contents

Contents GEOLOGIC FRAMEWORK STUDIES -ARTICLES Upper Devonian shallow-marine siliciclastic strata and associated fauna and flora, Lime Hills D-4 quadrangle, southwest Alaska 106 · Robert B. Blodgett and Wyatt G. Gilbert Petrography and provenance of sandstones from the Nation River Formation (Devonian) and the Step Conglomerate (Permian), Kandik region, east-central Alaska Thomas Brocculeri, Michael B. Underwood, and David G. Howell Magnetic susceptibilities and iron content of plutonic rocks across the coast plutonicmetamorphic complex near Juneau, Alaska James L. Drinkwater, Arthur B. Ford, and David A. Brew High-pressure amphibolite-facies metamorphism and deformation within the Yukon-Tanana and Taylor Mountain terranes, eastern Alaska Cynthia Dusel-Bacon and Vicki L. Hansen Some facies aspects of the upper part of the Kenai Group, southern Kenai Peninsula Romeo M. Flores and Gary D. Stricker Sedimentology of the Bay of Pillars and Point Augusta Formations, Alexander Archipelago, Alaska Susan M. Karl and Closey F. Giffen Depositional environments and some aspects of the fauna of middle Ordovician rocks of the Telsitna Formation, northern Kuskokwim Mountains, Alaska Elizabeth A. Measures, David M. Rohr, and Robert B. Blodgett Cenozoic uplift history of the Mount McKinley area in the central Alaska Range based on fission-track dating George Plafker, Charles W. Naeser, RobertA. Zimmerman, JohnS. Lull, and Travis Hudson Isotopic variations in calcite veins from the Kandik region of east-central Alaska Kevin L. Shelton, Michael B. Underwood, Deborah Bergfeld, and David G. Howell Statistical comparison between illite crystallinity and vitrinite reflectance, Kandik region of east-central Alaska Michael B. Underwood, Thomas Brocculeri, Deborah Bergfeld, David G. Howell, and Mark Pawlewicz GEOLOGIC FRAMEWORK STUDIES -GEOLOGIC NOTE The Arctic Alaska superterrane Thomas E. Moore BIBLIOGRAPHIES U.S. Geological Survey reports on Alaska released in 1991 Compiled by Ellen R. White Reports about Alaska in non-USGS publications released in 1991 that include USGS authors Compiled by Ellen R. White

CONTRIBUTORS TO THIS BULLETIN Anchorage U.S. Geological Survey 4200 University Drive Anchorage, Alaska 99508-4667 Bailey, Elizabeth A. Bradley, Dwight, C. Karl, Susan M. Denver U.S. Geological Survey MSBox 25046 Denver Federal Center Lakewood, Colorado 80225-0046 Affolter, Ronald H. MS 972 Apodaca, Lori E. MS 973 Borden,}. Carter MS 973 Burruss, Robert C. MS 973 Cathrall, john B. MS 973 Cieutat, Barrett A. MS 973 Fey, David L. MS 973 Flores, Romeo M. MS 972 Gent, Carol A. MS 973 Goldfarb, Richard}. MS 973 Gray, john E. MS 973 Hageman, Philip L. MS 973 Kelley, Karen D. MS 973 McHugh, john B. MS 973 Motooka, jerry M. MS 973 Naeser, Charles W. MS 963 Pawlewicz, Mark MS 940 Roushey, Bruce H. MS 973 Ryder, jean L. MS 973 Stricker, Gary D. MS 972 Taylor, Cliff D. MS 973 Tripp, Richard B. MS 973 Zimmerman, Robert A. MS 905 Menlo Park U.S. Geological Survey MS-904 345 Middlefield Road Menlo Park, California 94025 Brew, David A. Drinkwater, james L. Dusei-Bacon, Cynthia Ford, Arthur B. Howell, David G. Lull, johnS. Moore, Thomas E. Plafker, George White, Ellen R. (MS 955) Yeend, Warren Contents

Contents Reston U.S. Geological Survey National Center, MS12201 Sunrise Valley Drive Reston, Virginia 22092 Blodgett, Robert B. MS 970 Evans, john R. MS 957 Mo/1-Sta/cup, Elizabeth j. MS 959 Phi/potts, john MS 923 Spokane U.S. Geological Survey W 920 Riverside Ave, Rm. 656 Spokane, WA 99201 Box, Stephen E. Frost, Thomas P. Outside Bergfeld, Deborah Brocculeri, Thomas Shelton, Kevin L. Underwood, Michael B. Department of Geological Sciences University of Missouri (olumbia, Missouri 65211 Giffen, Closey F. North Pacific Mining Corporation 121 W. Fireweed Road, Suite 102 Anchorage, Alaska 99503 Gilbert, Wyatt G. Alaska Division of Geological and Geophysical Surveys 794 University Avenue Fairbanks, Alaska 99708 Hansen, Vicki L. Department of Geological Sciences Southern Methodist University Dallas, Texas 75275 Hudson, Travis ARCO Alaska Inc. P.O. Box 100360 Anchorage, Alaska 99510 Measures, Elizabeth A Department of Geology University of Idaho Moscow, Idaho 83843 Miller, Lance D. Department of Geosciences University of Arizona Tucson, Arizona 85721 Rohr, David M. Department of Geology Sui Ross State University Alpine, Texas 79832

Geologic Studies in Alaska by the U. S. Geological Survey, 1991 Dwight C. Bradley and Cynthia Dusei-Bacon, Editors INTRODUCTION This collection of twenty-one papers continues the annual series of U.S. Geological Survey reports on the geology of Alaska. These contributions, which include full-length Articles and shorter Geologic Notes, are grouped under two broad headings: Mineral Resource Studies (ten papers) and Geologic Framework Studies (eleven papers). Reports on mineral resources discuss exploration geochemistry in the Howard Pass quadrangle, lditarod quadrangle, and Admiralty Island, a mineral resource appraisal of the Bethel quadrangle, a fluid-inclusion study in the Nome Gold District, geochemistry of lode-gold deposits in the Seldovia quadrangle, a new occurrence of placer gold in the Seldovia quadrangle, gold in coal in the Healy quadrangle, experimental abrasion of detrital gold, and a new occurrence of rareearth minerals in southeastern Alaska. Under the heading of Geologic Framework Studies are reports on sedimentology and (or) stratigraphy in the Seldovia, Lime Hills, and Medfra quadrangles, the Kandik region, and the Alexander terrane in southeastern Alaska. Other papers report on the isotope geochemistry of veins and thermal maturity of the Kandik region, metamorphism and deformation of the Yukon-Tanana and Taylor Mountain terranes in eastcentral Alaska, magnetic susceptibilities of plutonic rocks in southeastern Alaska, terrane nomenclature in northern Alaska, and uplift of Mt. McKinley. These studies span nearly the entire State from the North Slope and Brooks Range to interior, western, southwestern, southcentral, and southeastern Alaska (fig. 1 ). Two bibliographies on Alaskan geology at the end of the volume list ( 1) reports about Alaska in USGS publications released in 1991, and (2) reports about Alaska by USGS authors in publications outside the USGS in 1991. Manuscript approved for publication July 9, 1992. Introduction

N "

:r r &

::r tD c

" tl)

"""

""" 66of 64° 60of 58°f 170° 174° 171° 168° 165° 162° 159° 153° 150° 147° 144° 141° 70° [0 '0

(l

168° 't>

166° r HI At' H 1 164° 162° 160° p... R 158° 156° 154° 66° STATEWIDE 8- Yeend 64° 22, 23 - White 62° Dusei-Bacon and Hansen 60° 58° 56° p A C I F I C 54° j1 KISK.A lw , 134° 132° 130° Figure 1. Index map of Alaska showing 1 :250,000-scale quadrangles and general locations of study areas discussed in this bulletin. Numbers are approximately located in center of study area and indicate order of papers in this volume; names of authors follow numbers.

MINERAL RESOURCE STUDIES-ARTICLES Fluid-Inclusion Study of the Rock Creek Area, Nome Mining District, Seward Peninsula, Alaska By Lori E. Apodaca Abstract Microthermometry and mass spectrometry were used to study fluid inclusions in quartz veins concordant to the metamorphic foliation and gold-bearing quartz veins in greenschist-facies metamorphic rocks of the Rock Creek area, Nome Mining District. Fluids from both vein types are characterized by a H20-C02-NaCI±N2±CH4 composition. The fluids that formed the auriferous quartz veins were emplaced postkinematically as indicated by the observation that the veins crosscut the metamorphic foliation of their host rocks. These veins are composed of quartz, with albite, chlorite, carbonate, and sulfides occurring along the vein selvages. Mineralized veins formed at minimum temperatures of 184°C to 272°C compared with 259°C to 312°C for concordant veins. Minimum pressures during vein formation are inferred to be 1 kbar and 1 .4 kbar for the mineralized and concordant quartz, respectively, and it was possible that immiscibility was involved in the vein-forming processes. INTRODUCTION Since the discovery of placer gold in 1898 on the Seward Peninsula, western Alaska, placer production has exceeded 6.2 million oz of gold, of which about 75 to 80 percent is from the Nome Mining District. Lode-gold deposits on the Seward Peninsula have remained relatively unexploited. The only recorded lode production is from the Big Hurrah Mine located 65 km east of Nome; with a reported production of 27,000 oz of gold, principally between 1903 and 1907 (Read and Meinert, 1986). Largely owing to the poor bedrock exposure, the geology and geochemistry of the lode-gold deposits of the Nome Mining District are not well understood. Characterization of the ore-bearing fluids, vein mineral assemblages, and alteration is important in understanding the genesis of gold-bearing, low-sulfide quartz veins. As part of the investigation of the gold-bearing veins, a fluid-inclusion study has been undertaken to place constraints on the pressure, temperature, and composition of ore deposition within the Rock Creek area of the Nome Mining District. Previous work in the Nome district has concentrated on geologic mapping, but little work· has been done on the geochemistry of the gold-bearing quartz veins. However, Read and Meinert (1986), in a study of the Big Hurrah Mine, discussed the genesis of the gold-bearing veins during a fluid-inclusion study of the different generations of quartz veins. Their fluid-inclusion data do not clearly define significant geochemical differences between vein types (that is, mineralized vs. premineralized conditions) owing to the limited measurements of primary inclusions in the different vein stages at the Big Hurrah Mine. The Rock Creek area is located in the N orne mining district approximately 14 km north of Nome, Alaska (fig. 1). Known lode-gold deposits occur as quartz-sulfide veins from Anvil Creek northward to Lindblom Creek within greenschist-facies metasedimentary rocks. Samples collected for fluid-inclusion studies from the Rock Creek area include vein material from Anvil Creek on the south end, north to Lindblom Creek; the highest concentration of samples was obtained from Sophie Gulch. These samples include mineralized quartz veins that cut the metamorphic foliation and earlier unmineralized quartz veins parallel to the metamorphic foliation. GEOLOGY Metamorphic rocks in the southwestern part of the Seward Peninsula are composed of the 4.5-km-thick Nome Group, which consists of (1) a basal quartz-rich pelitic schist unit, (2) a mixed unit consisting predominantly of quartz-graphite schist and marble with minor lenses of mafic schists, (3) mafic or chlorite-rich schist with calcareous components, and (4) a chloritic .marble (Till, 1984 ). These rocks were originally lower Paleozoic sediments that underwent regional blueschist metamorphism during the Jurassic, followed by Late Jurassic to Early Cretaceous overprinting under greenschist-facies metamorphic conditions during decompression of the metamorphic pile (Patrick and Evans, 1989). Fluid-Inclusion Study of the Rock Creek Area, Nome Mining District, Seward Peninsula

64°30' 165°30' KILOMETERS Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Cooper +Anvil Mtn. oDiscoverJ Gulch Kl LOMETERS 165°20' MILES NOME Figure 1. Rock Creek study area in Nome Mining District, Seward Peninsula, Alaska.

Amphibolite- to granulite-facies rocks of the eastwest-trending Kigluaik Mountains, approximately 40 km north of the Rock Creek area, form the northern boundary of the Nome Mining District. Intrusive rocks are not present in the Rock Creek area, but granitic intrusions are common in the Kigluaik Mountains to the north. A granitic stock of Cretaceous age is located at Cape Nome, 25 km southeast of the Rock Creek area. The Cape Nome stock has a weakly foliated texture parallel to the regional metamorphic fabric (Gillette, 1989). In the Nome Mining District, the gold-bearing quartz veins are preferentially hosted in the mixed unit of the Nome Group. Schists within the district are composed of quartz, muscovite, chlorite, and carbonate with some zones of albite porphyroblasts in which chlorite typically replaces the albite. Accessory minerals in the host rocks consist of pyrrhotite, pyrite, and carbon (Gillette, 1989). VEIN TYPES Three distinct generations of veins occur in the Rock Creek area (from oldest to youngest). First-generation veins consist of barren quartz emplaced as lenses and bands parallel to the metamorphic foliation generally not exceeding a few centimeters in width. The secondgeneration consists of low-sulfide gold-bearing quartz veins that crosscut the metamorphic foliation at high angles. The third-generation consists of barren carbonate-pyrite veins that fill fractures and cut both the firstand second-generation veins. Minor scheelite in drillcore samples appears to be younger than the second-generation veins, as it fills fractures in these veins. Veins containing gold mineralization are confined to the second-generation and typically range in size from a few millimeters to about 10 em in width. These lowsulfide gold-bearing quartz veins are inferred to be later than the barren concordant quartz lenses and bands, as they cut the lensoidal quartz. Mineralized veins tend to occur in swarms (fig. 2). These veins have a core composed of milky quartz, with vein margins consisting of quartz, carbonate (dolomite to ankerite), chlorite, albite, gold, and sulfides (arsenopyrite, pyrite, galena, sphalerite, and stibnite). Gold occurs in the second-generation quartz veins as free gold within fractures in the quartz and in sulfides, and near or attached to sulfide grains. Several silicified fault zones up to 90 em thick are seen in drill core, containing fragments of second-generation veins. These faults are often high grade owing to the fragments of second-generation veins in them. The silica that forms the matrix in these fault zones is not mineralized. FLUID-INCLUSION MICROTHERMOMETRY A study of fluid inclusions has been carried out to characterize the fluids associated with the mineralizing and earlier metamorphic event. Inclusions were measured in generation I and II quartz. A modified U.S. Geological Survey gas-flow fluid-inclusion stage was used. Temperature range for the stage is from -196°C to + 700°C at a temperature resolution of 0.1 oc . Accuracy for the stage is ±0.2°C between -60.0°C and 0°C, and ±5°C up to 400°C (Werre and others, 1979). The fluid inclusions studied in mineralized vein material were isolated inclusions trapped near sulfide grain embayments. Fluid inclusions found in sphalerite have gas/liquid ratios similar to inclusions found in quartz, but because of the opacity of the sphalerite, microthermometric measurements of these inclusions were difficult to obtain. In the mineralized veins, fluid inclusions may be secondary or pseudosecondary inclusions that were trapped along fractures and later healed. Figure 2. Surface exposure of gold-sulfide veins at Sophie Gulch. Fluid-Inclusion Study of the Rock Creek Area, Nome Mining District, Seward Peninsula

The inclusions in quartz are compositionally similar to those analyzed in sphalerite, which may also be secondary inclusions. Thus, they are presumed to be associated with sulfide deposition. In the unmineralized quartz first-generation veins, isolated inclusions and groups of inclusions that are believed to be primary were measured. Gold-Sulfide Vein Quartz Fluid inclusions were measured in quartz from second-generation veins and in fragments of second-generation veins from the silicified fault zones. Three types of inclusions have been identified in the mineralized vein quartz. Type I inclusions are single-phase, primary fluid inclusions of C02 liquid and (or) other gases (fig. 3A). Type II inclusions are two- and three-phase, primary inclusions. At room temperature, type II two-phase inclusions contain C02 liquid and other gases plus low-salinity water (fig. 3B), while type II three-phase inA elusions contain C02 liquid, C02 gas, and H20 liquid (fig. 3C). Vapor to liquid ratios are fairly consistent within groups of inclusions, with the vapor occupying approximately 75 to 90 volume percent for these inclusions. The density of the C02-rich phase in the inclusions determines whether these inclusions exhibit two or three phases at room temperature. Type III inclusions are aqueous two-phase inclusions of vapor and low-salinity water, with the vapor occupying approximately 10 volume percent (fig. 3D). These inclusions are believed to be secondary because they often occur in trails within the quartz. Measured fluid inclusions ranged in size from 5 to 13 J.1Ill. In addition to C02, type I and II inclusions contain minor CH4 and (or) N2 gases as indicated by the depression of the triple point for pure C02. Type I and II inclusions had measured TmC02 (melting temperature of C02) that ranged from -59.8°C to -56.8°C and ThC02 (homogenization temperature of C02) ranged from 0.3°C to 29.2°C with homogenization to a liquid. Type I inclusions had lower ThC02 than B D Figure 3. Fluid inclusions in vein and concordant quartz, Rock Creek area. A, Single-phase C02-rich fluid inclusions. 8, Two-phase C02-rich fluid inclusion. C Three-phase C02 (I iqu id, vapor)-H 20 fluid inclusions. 0, T rai I of H 20-rich secondary i ncl us ions. Field of view 0.16 mm. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

type II inclusions. For type II inclusions, T mclathrate (melting temperature of clathrate) values ranged from 7 .9°C to 11.4 oc. Depression of the T mclathrate values from that expected for pure C02 indicate a salinity of less than 5 equivalent weight percent NaCl for inclusions with relatively high TmC02 (Collins, 1979). Final Th (temperature of homogenization) for individual inclusions was often not reproducible, suggesting that microfracturing or leakage of the inclusions may have occurred during heating. For those inclusions that had reproducible values, Th ranged from 184 °C to 272°C. Type IT inclusions homogenized to a liquid. Observing the final homogenization to a liquid along with the small sizes of the inclusions made it difficult to determine final Th . Type ill secondary inclusions have T mice (melting temperature of ice) of -2.1 oc to -1.3°C, which yields salinity values of equivalent weight percent NaCl, based on the freezing-point depression in the H20-NaCl system (Roedder, 1984 ). Type ill inclusions homogenized to a liquid, with Th values ranging from 118°C to 179°C. Coeval C02-rich and H20-rich inclusions would indicate immiscibility. This evidence for immiscibility is not observed in the samples studied. If it is assumed that immiscibility did not occur, then homogenization temperatures are minimum trapping temperatures. Concordant Vein Quartz Fluid inclusions in the concordant quartz frrst-generation veins contain the same three types of inclusions as described in the gold-bearing quartz veins. Type I and IT inclusions had T mC02 values ranging from -59.3°C to -57.1°C, and ThC02 values of 0.3°C to 23.7°C. Type IT inclusions had Tmclathrate of 9.4°C to 10.5°C. Salinity values cannot be estimated from T mclathrate of type II inclusions owing to the presence of other gases (Collins, 1979). In the gold-sulfide quartz veins, many type IT inclusions did not provide reproducible Th values because of microfracturing or leakage. A few inclusions decrepitated before homogenization was reached. For those type II inclusions that did not decrepitate or leak, Th values to a liquid ranged from 259°C to 312°C. Type ill secondary inclusions consisted of water and 10 volume percent vapor, with Tmice ranging from -0.4°C to -1.8°C ( equivalent weight percent NaCl). Secondary inclusions homogenized to a liquid at Th ranging from 147°C to 228°C. Comparisons Between Different Generations of Quartz The data for T mC02, T mice, T mclathrate, ThC02, and Thin figure 4 have been compiled for fluid inclusions in the concordant quartz veins and in the gold-bearing quartz veins. The histograms include data f~r both primary and secondary inclusions. For type II inclusions, T mC02 and T mclathrate values are within the same range for both types of quartz (fig. 4A, C). ThC02 values are slightly lower for concordant quartz relative to goldbearing vein quartz (fig. 4D), which may indicate that a denser fluid formed the concordant quartz. Type ill inclusions (secondary inclusions) are the only inclusions in which ice melt could be determined; the range of T mice is the same for both types of quartz (fig. 4B). Th for concordant quartz occurs at temperatures >256°C, whereas values for gold-bearing vein quartz range from - 200°C to 235°C. Differences in trapping pressures or composition may explain the temperature differences between the concordant and gold-bearing vein quartz, as the Th values are minimum temperatures. FLUID-INCLUSION MASS SPECTROMETRY Microthermometry analyses provide semiquantitative information on the composition of the gaseous phase in inclusions. A quadrupole mass spectrometer was used to quantitatively determine the volatile composition (H20, C02, CH4, N2, and H2S) of the fluid inclusions (Landis and others, 1987). Fluid inclusions in quartz were thermally decrepitated during ramped beatings (30°C intervals) over a temperature range of 110°C to 410°C. Although thermal decrepitation releases gases from different generations of inclusions, monitoring of the gases released at decrepitation allows for a quantitative study of the ore fluid at a level comparable to results obtained from individual inclusions. The gas released is profiled as a function of temperature, allowing different fluidinclusion populations to be defined. Gold-Sulfide Vein Quartz Thermal decrepitation of gold-bearing quartz vein material has shown a bimodal population of inclusions having a mode of 68 mole percent C02, along with another less well-defined population having a mode of 5 mole percent C02 (fig. 5). There is also a faint indication of a third population having a mode of 40 mole percent C02. In measured inclusions within sphalerite, a single population shows an identical high C02 mode. Higher values of CH4 and N2 are found in C02-rich inclusions than in H20-rich inclusions, with both CH4 and N2 having a mode at 10 mole percent. Trace amounts of H2S were also detected in a few larger inclusions. In the H20-rich inclusions, CH4 and N2 are present in very low amounts, generally around 0 to 2 mole percent. No H2S was found in H20-rich inclusions. Fluid-Inclusion Study of the Rock Creek Area, Nome Mining District, Seward Peninsula

(J) z 0 u;

() z IJ.. A a: w CD

z (J) z 0 u;

()

IJ.. a: w CD

z (J) z 0 u;

() z IJ.. a: w CD

z C02 MELTING TEMPERATURE IN oc PRIMARY INCLUSIONS CLATHRATE MELTING TEMPERATURE IN °C PRIMARY INCLUSIONS E FINAL HOMOGENIZATION TEMPERATURE IN oc ALL INCLUSIONS ICE MELTING TEMPERATURE IN °C SECONDARY INCLUSIONS D C02 HOMOGENIZATION TEMPERATURE IN oc PRIMARY INCLUSIONS

D EXPLANATION Concordant quartz vein Gold-bearing quartz vein Secondary inclusions, concordant quartz vein Secondary inclusions, gold-bearing quartz vein Figure 4. Microthermometric data for (A) melting of C02, (8) melting of ice, (0 melting of clathrate, (0) homogenization of C02, and (f) final homogenization temperatures for concordant and gold-bearing quartz veins. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Concordant Vein Quartz Inclusions in the concordant quartz also exhibit a bimodal distribution, with two modes present in the histogram plot of C02-one at 68 mole percent C02 and the other at 2 mole percent C02 (fig. 5). H20 has modes at 15 and 97 mole percent, CH4 has a mode at 2.5 mole percent, and N2 exhibits modes at 1 and 15 mole percent, with the higher values corresponding to C02-rich inclusions. H2S was not present in any of the inclusions in the concordant quartz. Comparisons Between Different Generations of Quartz and Sphalerite The fairly consistent COiH20 ratios between the inclusions measured in the gold-bearing vein quartz, sphalerite, and concordant quartz suggest that these inclusions might have been trapped above the C02-H20NaCl solvus (fig. 6). However, the bimodal C02 and H20 populations in the quadrupole mass spectrometer en z U5 J () z u. a: w co

A H20 MOLE PERCENT CH4 MOLE PERCENT Figure 5. Mass-spectrometry data from fluid inclusions in concordant and gold-bearing vein quartz, and in sphalerite. A, H 20 mole percent. 8, C02 mole percent. C, CH4 mole percent. D, N 2 mole percent. data hint that the inclusions may represent end members of an unmixed parent C02-H20 fluid. PRESSURE AND TEMPERATURE CONDITIONS Limits on pressure (P) and temperature (T) conditions of ore deposition can be estimated from the density of the fluids in the fluid inclusions and known experimental data on the C02-H20 system. If the fluid inclusions in the one-phase field were trapped above the C02-H20-NaCl solvus (for example, because of the consistent gas/liquid ratios for groups of inclusions), then isobaric densities of 0.84 g/cm3 for the concordant quartz and 0.74 g/cm3 for the gold-bearing vein quartz (using calculations in Shepherd and others, 1985) can be used to put constraints on the P-T conditions of formation. Using homogenization temperatures and densities of the inclusions, minimum trapping pressures of about 1 kbar and 1.4 kbar are estimated for the gold-bearing vein quartz and concordant quartz, respectively (fig. 7). Actual trapping P-T conditions would lie to the right of the D C02 MOLE PERCENT N2 MOLE PERCENT EXPLANATION

Concordant quartz vein D Gold-bearing quartz vein 0 Sphalerite Fluid-Inclusion Study of the Rock Creek Area, Nome Mining District, Seward Peninsula

solvus along the isochores. Since small amounts of salt, CH4 and(or) N2 will substantially shift a solvus to higher temperatures and pressures, the trapping pressure of 1 kbar would be the minimum trapping pressure. These pressures correspond to minimum depths of formation of 2.8 to 3.9 km, assuming lithostatic pressure. From the work of Patrick and Evans (1989), maximum temperature and pressure can be estimated to ·be 500°C and 12 kbar for peak P-T conditions of metamorphism of the. host rocks. Since the barren concordant vein quartz could not have formed at a temperature greater than 500°C and the mineralized vein quartz was clearly emplaced after peak metamorphism of the surrounding host rocks, these veins formed in the stippled region of figure 7, assuming immiscibility did not occur. Microthermometric data and estimates of the fluid density suggest a path of evolution of the fluid along the solvus from A to B in figure 7. On the other hand, bulk fluid compositions of C02and H20-dominated inclusions derived from mass spectrometry may indicate trapping of end members of an immiscible fluid that separated under lower P-T conditions. This implies a parent fluid having a composition (/) that falls between the end members' composltton, as seen in figure 6. The parent fluid would have been trapped above the C02 solvus. It is not possible to precisely estimate the bulk composition of the unmixed parent fluid. From homogenization temperatures from the gold-bearing vein quartz, minimum pressure have been estimated to be 1.5 kbar, but the presence of dissolved salts and CH4 and (or) N2 will shift the solvus to higher temperatures. DISCUSSION Fluid-inclusion data from the Rock Creek area indicate that this deposit is similar to a number of greenschist-hosted lode-gold deposits in southern Alaska, such as deposits in the Juneau Gold Belt (Goldfarb and others, 1989) and those of the Chugach-Kenai Mountains (Goldfarb and others, 1986). These data are also similar to those from California Mother Lode deposits (Bohlke, 1989) and mesothermal Canadian Cordillera gold deposits (Nesbitt and Muehlenbachs, 1989). Indeed, the similarity between the fluid-inclusion data from different --- 2.6 weight percent equivalent Na Cl - C02 (/) I 400 w u (/) w w a::

w 300 z w 0::

a: w a.. :!: w -- pure H2 0- C0 2

C02 COMPOSITION 1.5 kbar ' Figure 6. Selected solvi in the C02-H 0-NaCI system at various pressures (Crawford, 1981 ). Below a solvus, two immiscible fluids coexist; a~ove a solvus, a single homogeneous fluid is present. Addition of salts and other volatiles such as N2 greatly expands two-phase field. At 270°C and pressure greater than or equal to 1.5 kbar, a single homogeneous fluid is present. Reduction in the confining pressure to 0.5 kbar results in fluid immiscibility with end-member compositions at points A and B. Reduction in confining pressures can be achieved by uplift and erosion, structural dilation of fracture by tectonic activity, or by hydrofracturing, among other means (leach and others, 1987). C02 composition is in weight percent. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

mesothermal gold deposits is one of the most consistent features of these deposits. High C02 contents were seen in fluid inclusions associated with ore deposition at the Alaska-Juneau, Ibex , and Reagan deposits in the Juneau Gold Belt (Goldfarb and others, 1989). At these deposits, unmixing of the C02-rich parent fluid is assumed to have accompanied ore deposition. The ore fluids are defined as being lowsalinity (<5 equivalent weight percent NaCl) aqueous fluids containing 3 to 10 mole percent C02±(CH4, H2S, N2) at temperatures of zoooc to 350°C. In the Mother Lode area of California, gold-bearing quartz veins are also found in greenschist-facies metamorphic terranes. These deposits are similar to the Rock Creek deposit. Bohlke ( 1989) described them as structurally discordant synmetamorphic to postmetamorphic gold-quartz veins with low base metals contents that formed from low-salinity, C02-rich, aqueous fluids at approximately 250°C to 450°C and 0.5 to 3+ kbar. The lower C02 contents (maximum of 10 mole percent) noted in the Mother Lode deposits by Bohlke (1989) may be the result of fluids derived from chemically different host rocks than in the Rock Creek area. Calcareous rocks of the Nome Group may account for the higher C02 content in the inclusions from Rock Creek. (f) a:: 2,000 al z w a:: ::J en 1,000 en w a:: a.. TEMPERATURE, IN DEGREES CELSIUS Figure 7. Pressures and temperatures of homogenization for a fluid of 60 mole percent C02 . Heavy line is solvus, with a onephase fluid to right and a two-phase fluid to left. Light lines are isochores, in grams per cubic centimeter (Shepherd and others, 1985). If fluids were one-phase when they were trapped, they would fall to the right of solvus, whereas a fluid that was twophase attimeoftrappingwouldfall to the leftofsolvus. Stippled area represents most likely region of conditions of formation. Mesothermal Canadian Cordillera gold deposits exhibit characteristics similar to other mesothermal gold deposits, having formation temperatures of 250°C to 350°C, pressures of 1±0.3 kbar, and salinities from 1 to 5 equivalent weight percent NaCl. C02 is found in inclusions in all of the Canadian Cordillera mesothermal gold deposits, ranging from 5 to 100 mole percent C02, with average values of 10 to 20 mole percent C02. Coexisting C02-rich and H20-rich inclusions in these deposits indicate that phase separation has occurred (Nesbitt and Muehlenbachs, 1989). The trapping of two end members of an immiscible fluid may explain the bimodal nature of the data from the Rock Creek area. Alternatively, several generations of inclusions with an initial C02-rich liquid may have been trapped. The spectrum of values for C02-H20 compositions between the modes may represent leakage or necking of the inclusions. The presence of high-C02 inclusions in sphalerite and in the gold-bearing quartz veins may indicate either that only one end member of an immiscible fluid was trapped or that the ore fluid may have contained high amounts of C02. Touret (1981) suggested that high-density, carbonic fluids are ubiquitous in high-grade metamorphic rocks and granulites. However, such fluids are not recognized to be associated with mesothermal gold veins, perhaps owing to a lack of gold-complexing H2S in the fluids. Fluids with a composition (X) of C02 0.6 at Rock Creek argue for fluid production occurring during highgrade metamorphism, such as amphibolite- or granulitefacies conditions. Rocks of this metamorphic grade are widespread in the Kigluaik Mountains 40 km to the north and could underlie the Rock Creek area at depth. Alternatively, high-grade rocks that are not recognized in the Rock Creek area may be absent at depth, and in that case the C02-rich inclusions could indicate that immiscibility occurred. CONCLUSIONS The results of this study indicate that in the Rock Creek area, Nome Mining District, the gold-bearing quartz veins were deposited at minimum temperatures between 184 oc and 272°C. The mineralizing fluid was a C02 rich-H20±CH4±N2 fluid. C02, CH4, and N2 gases ·present in the inclusions may have been derived from the carbonaceous matter often seen in the wallrock. Minimum pressure has been estimated at approximately 1 kbar for 60 mole percent C02 inclusions observed in this study. This corresponds to a minimum depth of formation of approximately 2. 8 km. If the 60 mole percent C02 inclusions represent an end member of an unmixed fluid, minimum P-T estimates would shift to lower values. Fluid-Inclusion Study of the Rock Creek Area, Nome Mining District, Seward Peninsula

Microthermometric studies show minimum trapping temperature of inclusions in concordant quartz veins with Th of 259°C to 312°C, and of inclusions in gold-bearing vein quartz with Th of 184 °C to 272°C. Differences in trapping pressures or composition may explain the temperature differences between the metamorphic and gold-bearing vein quartz, as the Th values are minimum temperatures. The bimodal nature of C02-H20 compositions indicates that immiscibility may have been present during ore deposition. The high C02 content of the inclusions may not indicate immiscibility if the fluid was produced during high-grade metamorphism, although evidence of this is not present in the Rock Creek area. Acknowledgments.-! wish to thank G.P. Landis for the use of the quadrupole mass spectrometer. This work could not have been possible without the help and support of C.B. Gillette (Placer Dome U.S.), T. Eggleston and G.P. Parsley (Tenneco Minerals), and R.V. Bailey (Aspen Exploration). REFERENCES CITED Bohlke, J.K., 1989, Comparison of metasomatic reactions between a common C02-rich vein fluid and diverse wall rocks: Intensive variables, mass transfers, and Au mineralization at Alleghany, California: Economic Geology, v. 84, p. 291-327. Collins, P.L.F., 1979, Gas hydrates in C02-bearing fluid inclusions and the use of freezing data for estimation of salinity: Economic Geology, v. 74, p. 1435-1444. Crawford, M.L., 1981, Fluid inclusions in metamorphic rocks-Low and medium grade, in Hollister, L.S., and Crawford, M.L., eds., Short course in fluid inclusions applications to petrology: Mineralogical Association of Canada Short Course Handbook, v. 6, p. 157-181. Gillette, C.B., 1989, Geology and mineralization of the Rock Creek Deposit, Nome Alaska: Northwest Mining Association 95th Annual Convention and Trade Show, December · 6-8, 1989, Spokane Wash., p. 34-41. Goldfarb, R.J., Leach, D.L., Miller, M.L., and Pickthorn, W.J., 1986, Geology, metamorphic setting, and genetic constraints of epigenetic lode-gold mineralization within the Cretaceous Valdez Group, south-central Alaska, in Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Keppie, J.D., Boyle, R.W., and Haynes, S.J., eds., Turbidite-hosted gold deposits: Geological Association of Canada Special Paper 32, p. 87-105. · Goldfarb, R.J., Leach, D.L., Rose, S.C., and Landis, G.P., 1989, Fluid inclusion geochemistry of gold-bearing quartz veins of the Juneau gold belt, southeastern Alaska: Implication for ore genesis: Economic Geology, Monograph 6, p. 363-375. Landis, G.P., Hofstra, A.H., Leach, D.L., and Rye, R.O., 1987, Quantitative analysis of fluid-inclusion gases-application to the study of ore deposits [abs.]: U.S. Geological Survey Circular 995, p. 38-39. Leach, D.L., Goldfarb, R.J., and Light, T.D., 1987, Fluid inclusion constraints on the genesis of the Alaska-Juneau gold belt, in Elliot, I.L., and Smee, B.W., eds., GeoExpo 86: Exploration in the North American Cordillera: Rexdale, Ontario, Canada, Association of Exploration Geochemistry, p. 150-159. Nesbitt, B.E., and Muehlenbachs, K., 1989, Geology, geochemistry, and genesis of meso thermal lode gold deposits of the Canadian Cordillera: Evidence of ore formation from evolved meteoric water: Economic Geology, Monograph 6, p. 553-563. Patrick, B.E., and Evans, B.W., 1989, Metamorphic evolution of the Seward Peninsula blueschist terrane: Journal of Petrology, v. 30, p. 531-555. Read, J., and Meinert, L.D., 1986, Gold-bearing quartz vein mineralization at the Big Hurrah Mine, Seward Peninsula, Alaska: Economic Geology, v. 81, p. 1760-1774. Roedder, E., 1984, Fluid inclusions: Mineralogical Society of America Reviews in Mineralogy, v. 12, 644 p. Shepherd, ·T.J., Rankin, A.H., and Alderton, D.H.M., 1985, A practical guide to fluid inclusion studies: London, Blackie, 239 p. . Touret, J .L.R., 1981, Fluid inclusions in high grade metamorphic rocks, in Hollister, L.S., and Crawford, M.L., eds., Short course in fluid inclusions applications to petrology: Mineralogical Association of Canada Short Course Handbook, v. 6, p. 182-208. Till, A.B., 1984, Low grade metamorphic rocks of the Seward Peninsula, Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 16, no. 5, p. 337. Werre, R.W., Jr., Bodnar, R.J., Bethke, P.M., and Barton, P.B., Jr., 1979, A novel gas-flow fluid inclusion heating/freezing stage [abs.]: Geological Society of America Abstracts with Programs, v. 11, p. 539. Reviewers: David L Leach and Richard j. Goldfarb

Geochemistry of Lode-Gold Deposits, Nuka Bay District, Southern Kenai Peninsula By]. Carter Borden, Richard ]. Goldfarb, Carol A. Gent, Robert C. Burruss, and Bruce H. Roushey Abstract Gold-bearing quartz veins in the Nuka Bay district of the southern Kenai Peninsula fill joints and faults within graywacke of the Upper Cretaceous Valdez Group and within Tertiary felsic dikes. The veins commonly contain abundant free gold and anomalous concentrations of Ag, As, Bi, Cd, Co, Cu, Hg, Fe, Pb, Sb, Te, and Zn-rich sulfidebearing phases. 0180 values for the quartz in these veins range from 14.0 to 17.4 per mil; o34S data for the sulfide minerals range from -1.2 to + 1 .8 per mil. Fluid-inclusion studies indicate that the ore-forming fluid had an approximate composition of 90 to 92 mole percent H 20, 6 to 8 mole percent C02 , and 1 to 2 mole percent CH 4 and N2 The veins were formed at temperatures of at least 250°C to 300°C and pressures of at least 2.3 to 3.0 kbar. Data are compatible with an ore fluid produced during dehydration and decarbonation reactions of metasedimentary rocks deeper within the accretionary prism. INTRODUCTION The Nuka Bay gold district (fig. 1) is located along the eastern side of the southern Kenai Peninsula, approximately 55 km east of the town of Homer. Small, east-west trending, gold-bearing quartz veins are widespread in metasedimentary rocks near Surprise Bay, the North Arm of Nuka Bay, Quartz Bay, Yalik Bay, and Beauty Bay. These veins compose the southernmost of a series of small lode-gold districts scattered throughout the Chugach-Kenai Mountains of south-central Alaska. Other significant and similar districts include Hope-Sunrise (Tuck, 1933), Moose Pass (Tuck, 1933), Port Valdez (Johnson, 1915), Port Wells (Johnson, 1914), and Girdwood (Park, 1933). Epigenetic lodes in all districts are hosted in structurally deformed turbidite deposits of the Late Cretaceous Valdez Group, the major component of the Chugach terrane. The gold-bearing veins in the Nuka Bay district have been described in Grant and Higgins (1910), Shepard (unpublished data, 1925), Pilgrim (unpublished data, 1931), Smith (1938), and Richter (1970). The veins were discovered in the early 1900's, and peak mining activity occurred in the 1930's. Dollar estimates by Richter ( 1970) suggest combined gold production of 5,000 to 6,000 oz. from five mines. There has been no recorded production from any of the mines for the past 50 years. However, active claim work is still ongoing at the Goyne prospect and the Sonny Fox Mine near Surprise Bay and at the Little Creek prospect near the head of Beauty Bay. This paper presents new geochemical data for some of the larger gold-bearing veins in the Nuka Bay district. Atomic absorption analyses of gold-bearing vein samples provide trace-element signatures for the vein deposits. The first stable-isotope and fluid-inclusion analyses for samples from the Nuka Bay district are also presented. These data provide important constraints on the pressure, temperature, and composition conditions of mineralization. GEOLOGY The Nuka Bay district occurs within marine turbidites of the Upper Cretaceous Valdez Group. The Valdez Group consists mainly of turbiditic graywacke and slate, which has been tightly to isoclinally folded, thrust faulted, and in places disrupted to form type I melange. Metamorphic biotite in thin sections from the North Arm of Nuka Bay (T.M. Kusky and D.C. Bradley, oral commun., 1992) indicates mid-greenschist-facies metamorphism in the Nuka Bay Mining District. The Valdez Group is the major component of the Chugach terrane, the Mesozoic part of south-central Alaska's accretionary prism. It has been interpreted as having been deposited in a deep-sea trench during the latest Cretaceous (Nilsen and Zuffa, 1982); within a few million years of deposition, the Valdez Group was incorporated into the Chugach accretionary prism. Flysch sequences of the Orca Group, composing the Prince William terrane, were thrust below rocks of the Chugach terrane during Paleogene time (Winkler and Plafker, 1975; Plafker and others, 1977). Geochemistry of Lode-Gold Deposits, Nuka Bay District, Southern Kenai Peninsula

Granodiorite dikes and, less commonly, sills cut metasedimentary rocks of the Valdez Group throughout the study area. The igneous intrusions are part of a belt of Paleogene granites, granodiorites, and tonalites termed the Sanak-Baranof plutonic belt (Hudson and others, 1979), which rims much of the Gulf of Alaska. These igneous rocks have not been isotopically dated in Nuka Bay but are probably Eocene in age. Monazite from a pluton about 15 km east of Nuka Bay yielded a 56Ma U-Pb age, and amphiboles from an intermediate dike to the west near Seldovia have an Ar-Ar age of 57 Ma (W. Clendenen, quoted in Bradley and Kusky, 1992). GOLD-BEARING VEINS Gold-bearing quartz veins of the Nuka Bay district cut graywacke and the slate of the Valdez Group; locally, they also cut dikes and sills. Richter (1970) noted a spatial association of the veining with more competent metagraywacke and igneous rocks relative to the less competent fine-grained metasedimentary rocks. Veining is observed to·both parallel cleavage and to cut the foliation. The latter observation indicates mineralization postdates regional metamorphism of the immediate host rocks. Mineralized veins consistently strike east-west at 5 KILOMETERS a high angle to regional strike, and they dip steeply to vertically. East-west-trending veins may occupy tensional cross joints, as suggested by Richter (1970). In addition, veins may occupy late brittle cross faults of the set described by Bradley and · Kusky (1992) in the McHugh Complex of the Chugach terrane. Bradley and Kusky determined dextral motion on the structures to be coeval with felsic and intermediate dike emplacement; such extension may also have been coeval with the gold veining. The gold-bearing veins are irregular and discontinuous, with pinching and swelling suggesting that some deformation postdates mineralization. Larger quartz veins tend to reach about 100 m in length and 3 m in width. Arsenopyrite with lesser pyrite, chalcopyrite, galena, sphalerite, and free gold are commonly visible. Smith (1938) noted that the visible gold was often spectacular and abundant, with up to 15 oz/ton of gold in one 5-ton ore shipment from the Sonny Fox Mine. Tetrahedrite, covellite, chalcocite, sylvanite, native silver, and native copper have also been observed in the veins (Smith, 1938). Carbonization, sulfidization, and silicification characterize wall rock adjacent to the veins. Sericitization of metasedimentary wall rocks is common at the Little Creek prospect. An 40 Arf39 Ar age of 55.6±0.17 Ma was obtained on muscovite from within 150°30' 30 40 KILOMETERS 1. UttJe Creek prospect 2. Beauty Bay prospect 3. Rosness and Larson Mine 4. Robert Hatcher prospect 5. Goyne prospect 6. Sonny Fox Mine Figure 1. Locations of mines and prospects within Nuka Bay gold district. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

the vein quartz at the Little Creek prospect (Larry Snee, oral comrnun., 1992) and is believed to represent the age of gold mineralization. Richter ( 1970) presented atomic absorption analyses for gold and tellurium, colorimetric analyses for arsenic, and semiquantitative emission spectrographic data for Ag, B, Ba, Co, Cr, Cu, Ni, Pb, Sb, V, and Zn. Results of analyses show samples with as much as 304 ppm Au, 6,000 ppm As, 30 ppm Ag, 200 ppm Cu, 1,500 ppm Pb, 300 ppm Sb, and 700 ppm Zn. More sensitive atomic absorption analyses for many of the ore-related elements are listed in table 1. Sulfide-rich quartz-vein grab samples from mines and prospects around Nuka Bay were determined to contain as much as 48 ppm Au, 200 ppm Ag, 8.7 percent As, 107 ppm Bi, 220 ppm Cd, 200 ppm Co, 1,000 ppm Cu, 7.6 ppm Hg, 30 percent Fe, 2 percent Pb, 260 ppm Sb, 19 ppm Te, and 6,500 ppm Zn. Elements B, Ba, Mo, and W do not exceed background levels in any of the ore samples. ISOTOPE GEOCHEMISTRY Hydrothermal quartz separates from goldbearing veins were collected from the Sonny Fox and Rosness and Larson Mines, and from the Goyne, Robert Hatcher, and Little Creek prospects. In addition, a quartz sample was obtained from small, unmineralized veinlets that cut a granodiorite dike on the north side of Pilot Harbor. Oxygen for o180 measurements was liberated by reaction of the quartz separates with ClF 3 in nickel bombs at 550°C, as described by Borthwick and Harmon (1982). Sphalerite, pyrite, and arsenopyrite were separated from goldbearing quartz collected at the Rosness and Larson Mine and at the Goyne and Little Creek prospects. o34S values were determined following oxidation of these sulfide minerals to sulfur dioxide by cupric oxide combustion (Grinenko, 1962; Fritz and others, 1974). The standard error for each analysis is approximately ±0.2 per mil for both oxygen and sulfur. Oxygen Isotopes Oxygen-isotope values for gold-bearing quartz from mines and prospects of the Nuka Bay district range from 14.0 to 17.4 per mil (table 2). Samples from two different veins at the Little Creek prospect are statistically identical. At the lrl lrl lrl lrl v z . v v "<tONON 0\0000N -N00000 oooooo NNN zzzzz ssszz o.o.c:o.o. Ot'oo~- oN :"! z lrl v z z z

0 o. lrl lrl lrl lrlO NN lrlO M -N zz 00lrl"<t N00 O.lrl lrl lrlO

1.00\0N t

lrlC:0.-

N zzzz ZZ'QZ

1.01.0~0 N ozoo o. o. o. o. lrl lrl [' r--:o.n.n o-- --o :c--i~ lr)01rllrl OOirl z zzzz lrl lrl f'- 00 lrl z ·z zz-z lrlOON -o.f'V -lrlN- o o ·o -ooNlrlOOO

N "<t Geochemistry of Lode-Gold Deposits, Nuka Bay District, Southern Kenai Peninsula

Table 2. Oxygen-isotope data for quartz-vein samples Occurrence o18o (per mil) Goyne prospect Sonny Fox Mine Little Creek prospect 16.1, 16.2 Robert Hatcher prospect 17 .4, 17 .0, 17 .2, 17.0 Rosness and Larson Mine Vein cutting felsic dike, Pilot Harbor Robert Hatcher prospect, quartz separates were obtained from adjacent bands of white, flesh-colored, and sulfiderich blue quartz. Oxygen-isotope analysis of each of the three distinct quartz bands resulted in statistically indistinguishable values, suggesting that pulses of vein forming fluids were in isotopic equilibrium with rocks of similar compositions and that temperatures of vein formation remained constant for each pulse. Oxygen-isotope values for quartz from the Nuka Bay district are identical to those from other gold districts within the Chugach terrane. Quartz o180 values range from 13.9 to 17.0 per mil for other gold districts in the Kenai-Chugach Mountains (Goldfarb and others, 1986) and from 16.5 to 17.3 per mil for gold-bearing quartz from the Chichagof district in southeastern Alaska (Goldfarb and others, 1988). Differences in o 180 of 3.4 per mil between vein quartz from the different gold occurrences in the Nuka Bay district reflect either differences in ore fluid compositions or in temperatures of quartz precipitation. A temperature difference of about 1 oooc between fluids responsible for quartz precipitation at the Goyne prospect (0 180 uartz 14.0 per mil) and at the Robert Hatcher prospect (b 180quartz 17.4. per mil) would explain the range in the data. The most gold- and sulfide-rich samples were obtained from the Little Creek and Goyne prospects and the Sonny Fox Mine, and these samples have slightly lighter o180 values. The heavier o 180 values for samples from the Robert Hatcher prospect and Rosness and Larson Mine could reflect lower temperature conditions that are less favorable for significant gold and base-metal sulfide precipitation. Conversely, differences in actual ore-fluid isotopic values may reflect differences in fluid source areas or fluid pathways. A fluid originating within, or having greater contact along its pathway with, 180-rich argillite of the Valdez Group would be slightly heavier than a fluid that is derived from, or passes through, graywacke or conglomerate. If one assumes vein formation at 300°C and applies the quartz-water fractionation relationship of Matsuhisa and others (1979), then o180 values of 6.0 to 9.4 per mil are calculated for vein-forming fluids. This range is permissive of a fluid with a crustal origin, including both a magmatic and a metamorphic source. These data are not consistent with a meteoric fluid source, as Tertiary mete16 Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Table 3. Sulfur-isotope data for sulfide minerals Occurrence Mineral Rosness and Larson Mine sphalerite Rosness and Larson Mine pyrite Little Creek prospect pyrite Little Creek prospect arsenopyrite Goyne prospect arsenopyrite o34s (per mil) 0.1, 0.2 1.6, 1.8, 1. 7 oric water in south-central Alaska had a o 180 value of approximately -15 per mil as estimated from oD contours originally shown in Taylor (1974). If the ore fluids were meteoric in origin, they would have had to have undergone an oxygen shift of approximately 20 to 25 per mil requiring unlikely water-to-rock ratios of less than 0.1. The relatively narrow range in the o180quartz values is indicative of a fluid-dominant ore system with significantly higher water-to-rock ratios (Kerrich, 1987). Quartz veins hosted by granodiorite dikes at the Goyne prospect and near Pilot Harbor have similar isotopic compositions to those hosted by metasedimentary rocks and would have noticeably lighter isotopic ratios if igneous rock-dominant conditions prevailed. Sulfur Isotopes Eight sulfide minerals from three of the gold-bearing vein occurrences in the Nuka Bay district have a narrow o34S range between -1.2 and + 1.8 per mil (table 3). This is consistent with the only other sulfur-isotopic data from gold veins hosted by rocks of the Valdez Group; these data were obtained by Pickthorn ( 1982) from the Port Valdez district. Eight sulfide minerals from four mines in that district also had o34S values tightly clustered around 0 per mil, ranging from -2.3 to +2.2 per mil. Under mesothermal vein-forming temperatures, cogenetic pyrite and sphalerite should have a A34Spy-sl value of about + 1 to +2 per mil using the relationship of Sakai (1968). The determined A34Spy-st value of -0.5 per mil indicates a lack of isotopic equilibrium between the sulfide phases at the Rosness and Larson Mine and suggests either deposition at different temperatures or postdepositional sulfur-isotopic exchange. The narrow range for sulfide o34S values is consistent with an isotopically uniform sulfur source. The thick sequence of Valdez Group flysch is a very likely source. ·Leaching of sulfides, or desulfidization reactions associated with prograde metamorphism, could release sulfur into ore-carrying fluids. The abundance of arsenopyrite in the gold-bearing veins and the absence of significant negative o34S values favor relatively reducing conditions for ore deposition.

FLUID-INCLUSION GEOCHEMISTRY Microthermometry Microthermometric data were obtained from doubly polished thin sections of gold-bearing quartz veins from five mines and prospects (Little Creek prospect, Robert Hatcher prospect, Rosness and Larson Mine, Sonny Fox Mine, and Goyne prospect). Measurements of fluid inclusions in the quartz were made on a modified U.S. Geological Survey gas-flow heating and cooling stage. Temperatures were determined, whenever possible, for C02 .melting (TmC02), C02 homogenization (ThC02), ice melting (Tmice), clathrate melting (Tmclath), and final homogenization (Thfinal). All measurements were reproducible to ±0.2°C, except for final homogenization temperatures, which were reproducible to ±3°C. Two types of fluid inclusions were found in all samples except for those from the Robert Hatcher prospect, which had inclusions too small to be examined. Type I inclusions (fig. 2) are isolated toward the center of clear quartz crystals, range from 2.5 to 10 J.lm in diameter, and tend to be rectangular to oval in shape; these observations suggest that the inclusions formed during precipitation of the bulk of the quartz. Type II inclusions are aqueous, range in size up to 8 J.lm, have smaller bubbles, show no evidence of gas, and are found in late trails and closer to the edges of quartz crystals. These inclusions appear to have been trapped paragenetically late and long after the bulk of the vein material formed. All examined sections exhibit fluid inclusions that were trapped above the appropriate solvus in a onephase field. Variable gas-to-liquid ratios, where present, reflect necking down or autodecrepitation of inclusions; there is no evidence for fluid immiscibility. Figure 2. Typical type I inclusion from Little Creek prospect, showing three phases at -1 0°C. Type I inclusions contain 90-92 mole percent H20, 6-8 mole percent C02, 1 mole percent CH4, and mole percent N2. Inclusion is 10 1-1m in diameter. In all the samples, type I inclusions are two-phase at room temperature. However, when cooled below room temperature, inclusions from the Little Creek prospect and Rosness and Larson Mine exhibit three phases consisting of liquid water and separate C02-dominant liquid and gaseous phases. TmC02 values are between -63.6°C to -59.1 °C (table 4), suggesting that C02 is the dominant gas phase but with significant contamination from CH4 and (or) N2 (Swanenberg, 1979). One inclusion from the Rosness and Larson Mine melted an additional solid phase at -1 09°C and homogenized to a liquid at -85°C. The latter temperature suggests that the major contaminant is CH4 owing to its proximity to the methane critical temperature of -82°C. Further indication of the presence of CH4 is the range of Tmclath values between 11.3°C and 21.6°C (table 4); these values are significantly higher than the C02-H20 hydrate melting temperature of 1 0.0°C. Values for ThC0 2 range between -16.5°C and +9.8°C (table 4), and homogenization is consistently into the liquid phase. Calculated densities would be between 0.88 and 1.00 g/cm3, assuming pure C02 Higher density values in· this range reflect increased contamination of the C02 phase with other gases. This conclusion is supported by the fact that the only two inclusions with Th C02 above ooc are among the least contaminated inclusions; that is, they have relatively high TmC0 2 values. Therefore, the lower end of the range of possible densities is interpreted to best reflect true C02 densities in type I inclusions. Due to the presence of gas-hydrate formation in type I fluid inclusions, observations of ice melting were generally obscured. In three inclusions, T mice values of -1.4 oc to -1.2°C (table 4) was obtained in the presence of coexisting solid clathrate. Such values indicate a fluid salinity of 2.5 weight percent equivalent NaCI. However, the presence of coexisting clathrate, trapping water in an additional solid phase, suggests that the salinity is only a maximum value, and that the true value will be significantly lower (Collins, 1979). The Tiinal of the aqueous solution in type I inclusions varies between 155°C and 248°C (table 4). Difficulties in measuring Thfinal arose from stretching of inclusions in quartz upon repeated heating or, more commonly, from decrepitation of fluid inclusions prior to final homogenization. Most type I inclusions in a sample would decrepitate during a single heating cycle, rendering the sample useless for additional measurements. The consistent decrepitation activity prior to final homogenization and the presence of empty or deformed, autodecrepitated inclusions indicate high internal pressures. Depending on the fluid-inclusion size, decrepitation occurred between 242°C and 299°C (table 4). Type II inclusions range up to 8 J.lm in diameter and have smaller vapor bubbles than type I inclusions. Type Geochemistry of Lode-Gold Deposits, Nuka Bay District, Southern Kenai Peninsula

Table 4. Fluid-inclusion data from Nuka Bay mines and prospects [All data, except volume percent, are in degrees Celsius] Mine or prospect Tn-{:.02 T ,rice Tmelath ThC02 Vol% T!Jfinal Type I inclusions Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Rosness-Larson Rosness-Larson Rosness-Larson Rosness-Larson Rosness-Larson Rosness-Larson 1 Sonny Fox Sonny Fox Sonny Fox Sonny Fox Sonny Fox s,onny Fox Sonny Fox Goyne Goyne Goyne Goyne Type II inclusions Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Little Creek Goyne 1 An inclusion containing CH4 phase; T mCH4 is -I 09°C and ThCH4 is -85°C. 2Temperatures that indicate decrepitation prior to final homogenization. II inclusions in samples from the Rosness and Larson and Sonny Fox Mines could not be measured because the inclusions were too small, appeared to have experienced leaking, or consistently exhibited metastable be18 Geologic Studies in Alaska by the U.S. Geological Survey, 1991 havior. The Tmice in measured inclusions ranged from3.0°C to -1.5°C (table 4), suggesting salinities from 2 to 5 weight percent equivalent NaCl (Roedder, 1962). Many samples had T mice that oscillated upon repeated

beatings from -1 oc to +3°C (table 4), indicating that the inclusions are metastable. The occurrence of metastable superheated ice at high "negative pressures" is not unusual for low-salinity, low to moderate-temperature aqueous inclusions (Roedder, 1984, p. 298). Metastable inclusions were most prevalent throughout the samples from the Sonny Fox Mine. The Thfinal of the type II inclusions were 140°C to 196°C (table 4). Pressure-Temperature-Composition Conditions of Vein Formation Microthermometric data indicate that type I inclusions are in the H20+C02+CH4+NaCl±N2 system. Type I inclusions are clearly "more primary" than type II inclusions and appear to have been trapped during precipitation of the bulk of the quartz and metals. Estimates of 85 to 92 mole percent water (XH 0 =0.85-0.92) in type I inclusions for the Little CreeK prospect and for the Rosness and Larson Mine were determined using the graphical method of Burruss (1981) and temperatures of C02 homogenization assuming the vein-forming fluid is in a binary H20-C02 system. The lowest estimates are likely an artifact of calculations based on the latter assumption. Inclusions with highest apparent C02 densities also have significant XcH , which causes errors in calculation of XH 0 . Therefor'b, ore fluids most likely had a composition 2of 90 to 92 mole percent H20, 7 to 9 mole percent gas, and no more than 1 to 2 mole percent salt (3-6 weight percent equivalent N aCl). INCLUSION DIAMETER, IN MICRONS Figure 3. Internal pressure required for decrepitation offluid inclusions in quartz according to size of inclusion at 1 atm (curve from Bodnar and others, 1989). A 5-micron inclusion decrepitating prior to final homogenization will have been trapped at a pressure of at least 2.3 kbar; decrepitation of a 2.5-micron inclusion suggests a minimum trapping pressure of 3 kbar. The dominant species in the gas phase of the ore fluids is clearly C02; clathrate melting data suggest that CH4 is the dominant contaminant. Values of 0.14 to >0.30 mole percent CH4 in the gas phase were determined using both Swanenberg's (1979) technique, which plots XcH 4 versus degree of fill, and a Vbutk -XcH 4 diagram (Burruss, 1981) using the ThC0 2 values. Both methods for calculating XcH assume that the fluid is a binary C02-CH4 system an~ that the presence of H20 has a nominal effect upon the volatile phase. J_.asar Raman microprobe spectroscopy was used to confirm the presence of all volatiles within the inclusions in the quartz crystals. A few selected inclusions were scanned with the microprobe for C02, CH4, H2S, and N2 Resulting data indicate that CH4 makes up 14 to 15 mole percent of the gas phase. Nitrogen was detected but never exceeded 1 percent of the gas phase. Hydrogen sulfide was not identified in any o( the examined type I inclusions. A quadrupole mass s~ectrometer system (Landis and others, 1987) was used to more quantitatively determine the volatile composition of the inclusions from the Little Creek prospect. Analyses of gases released from individual or small groups of inclusions during thermal decrepitation indicated roughly equivalent amounts of CH4 and N2 in the ore fluids. w a:

w a: a.. B liquid+ xco2 =0·10 vapor XNaCF0.0193 OL A soo TEMPERATURE, IN DEGREES CELSIUS Figure 4. Solvus and isochores for fluids approximate in composition to type I fluid inclusions (from Bowers and Helgeson, 1983). All inclusions were trapped above experimentally determined solvus. Line A indicates that fluid inclusions with minimum trapping temperatures of 250°C are characterized by minimum trapping pressures of 2.5 kbar. Line B indicates that temperature estimates of 275°C suggest minimum trapping pressures of 1.5 kbar. An inclusion exhibiting Thfinal 275°C would have been trapped somewhere along the 0.95 g/cm3 isochore, thus extending above the solvus. Geochemistry of Lode-Gold Deposits, Nuka Bay District, Southern Kenai Peninsula

H2S was also detected in the analysis of the largest inclusion burst. Two independent approaches can be used to estimate minimum trapping pressures for the ore fluids. Type I inclusions as small as 4. 7 JJm 3 in samples from the Little Creek prospect and as small as 2.5 Jlffi in diameter in samples from the Sonny Fox Mine decrepitated at 278°C and 299°C, respectively, prior to final homogenization (table 4). Using the relationship between inclusion size and the internal pressure needed to decrepitate inclusions in quartz under one atmosphere confining pressure (Bodnar and others, 1989), one can determine that the decrepitated inclusions would have been trapped at pressures in excess of 2.3 to 3.0 kbar (fig. 3). An alternative approach uses appropriate isochores on a pressure-temperature diagram and minimum trapping temperatures determined from microthermometry to estimate minimum trapping pressures. Based on calculated isochores for the H20-C02NaCl system determined by Bowers and Helgeson (1983), a fluid similar in composition to that of the Nuka Bay district ore fluids would have minimum trapping pressures of 2.5 kbar and 1.5 kbar for homogenization (or decrepitation prior to homogenization) temperatures of 250°C and 275°C, respectively (fig. 4). The consistent decrepitation of type 1 fluid inclusions between 250°C and 300°C throughout all the samples gives a minimum estimate of vein formation temperatures. The few measured homogenizations (table 4) occur at significantly lower temperatures, but these are rare observations and likely reflect deformed and (or) leaked inclusions. Minimum trapping pressures of 2.3 to 3.0 kbar from the quartz decrepitation data also indicate minimum trapping temperatures of 250°C using the compositional relationships shown in figure 4. DISCUSSION Stable-isotope and fluid-inclusion data from Nuka Bay are essentially identical to those from other districts hosted by rocks of the Valdez Group (Pickthorn, 1982; Goldfarb and others 1986, 1988.) The ore solutions were low-salinity C-0-H-N-S fluids typical in composition of those observed in low- to medium-grade metamorphic environments (Crawford and Hollister, 1986). Our data suggest that H20 in the ore-forming fluids is a product of dehydration reactions within the metasedimentary rock pile. Minerals. such as chlorite and epidote release significant volumes of H20 during prograde greenschist-facies metamorphic reactions. The C02 in the ore fluids is similarly controlled by prograde decarbonation reactions. Significant amounts of CH4 and N2 measured in Nuka Bay samples might reflect the late stage breakdown of organic materials during the meta20 Geologic Studies in Alaska by the U.S. Geological Survey, 1991 morphism of the metasedimentary rocks. Alternatively, these species could be derived locally at the site of vein deposition through reduction of C02 in the ore fluid and oxidation of ammonia in silicates to form N2. Oxygen- and sulfur-isotope compositions are also consistent with derivation from metamorphic rocks in the Nuka Bay area. Although hydrogen-isotope data have not yet been obtained for hydrogen-bearing minerals from the Nuka Bay deposits, similar gold districts in rocks of the Valdez Group are characterized by isotopically heavy fluids of -20 to -35 per mil (Pickthom and others, 1987). Such fluids are normally incompatible with a meteoric source and are more indicative of derivation from metamorphic country rocks. A quartz sample with relatively few type II secondary inclusions from the Little Creek prospect was crushed to release fluid inclusion waters. Isotopic analysis of these waters yielded a bD value of -65 per mil. Whereas such a value clearly reflects a mixture of primary ore fluids and secondary waters trapped during later uplift (Pickthom and others, 1987), it is still about 50 per mil heavier than local meteoric water. This provides further support for a deep crustal origin for the ore fluids. In addition to the isotopic data and fluid chemistry, the minimum trapping pressures also weigh strongly against the involvement of meteoric waters in the oreforming process. Under the assumption of lithostatic pressure, the veins are inferred to have formed at depths of at least 8.0 to 10.5 km, the deepest estimates to date for any of the Valdez Group-hosted gold deposits. It is unlikely, especially when noting the lack of any major extensional structures in the Nuka Bay area, that surface water would have circulated to such depths within the metasedimentary rocks. The available data best support a model in which oreforming fluids were produced during regional mediumgrade metamorphic reactions in sedimentary rock. Such reactions are possible either at deeper levels within the rocks of the Valdez Group or in units of the underthrusted Prince William terrane. Coeval ages for veining and igneous activity suggest that both were products of the same, relatively high temperature tectonic episode. Ore fluids migrated upward into retrograding parts of the Valdez Group, either immediately after devolatilization or subsequent to periods of ponding below impermeable units. In the latter case, a reduction of the confining pressure associated with uplift of the Kenai Mountains could have led to hydraulic fracture and fluid release. CONCLUSION Mesothermal gold-bearing quartz veins of the Nuka Bay district were emplaced into metasedimentary rocks of the Valdez Group at 55.6 Ma. The C-0-H-N-S ore fluids had a b180 composition of 6.0 to 9.4 per mil and a

34S composition of about 0 per mil. The data are consistent with fluids produced via prograde metamorphic reactions within rocks of either the Valdez Group or the underthrusted Orca Group. They migrated upward to retrograding parts of the metamorphic pile at depths no shallower than 8 to 10.5 km and precipitated quartz and metals at temperatures of at least 250°C to 300°C. REFERENCES CITED Bodnar, R.J., Binns, P.R., and Hall, D.L., 1989, Synthetic fluid inclusions-VI. Quantitative evaluation of the decrepitation behavior of fluid inclusions in quartz at one atmosphere confining pressure: Journal of Metamorphic Geology, v. 7, p. 229-242. Borthwick, J., and Harmon, R.S., 1982, A note regarding ClF3 as an alternative to BrF5 for oxygen isotope analysis: Geochimica et Cosmochimica Acta, v. 46, p. 1665-1668. Bowers, T.S., and Helgeson, H.C., 1983, Calculation of the thermodynamic and geochemical consequences of nonideal mixing in the system H20-C02-NaCl on phase relations in geological systems-Metamorphic ~quilibria at high pressures and temperatures: American Mineralogist, v. 68, p. 1059-1075. Bradley, D.C., and Kusky, T.M., 1992, Deformation history of the McHugh Complex, Seldovia quadrangle, south-central Alaska, in Bradley, D.C., and Ford, A.B., eds., Geological studies in Alaska by the U.S. Geological Survey, 1990: U.S. Geological Survey Bulletin 1999, p.17-32. Burruss, R.C., 1981, Analysis of phase equilibria in C-0-H-S fluid inclusions: Mineralogical Association of Canada Short Course Handbook, v. 6, p. 39-74. 'Collins, P.L.F., 1979, Gas hydrates in C02-bearing fluid inclusions and the use of freezing data for estimation of salinity: Economic Geology, v. 74, p. 1435-1444. Crawford, M.L., and Hollister, L.S., 1986, Metamorphic fluids-The evidence from fluid inclusions, in Walther, J.V., and Wood, B .J .,eds., Fluid-rock interactions during metamorphism, v. 5 of Advances in physical geochemistry: Springer-Verlag, New York, p. 1-35. Fritz, P., Drimmie, R.J., and Nowicki, V.J., 1974, Preparation of sulfur dioxide for mass spectrometer analyses by combustion of sulfides with copper oxide: Analytical Chemistry, v. 46, p. 164-166. Goldfarb, R.J., Leach, D.L., Miller, M.L., and Pickthorn, W.J., 1986, Geology, metamorphic setting, and genetic constraints of epigenetic lode-gold mineralization within the Cretaceous Valdez Group, south-central Alaska, in Keppie, J.D., Boyle, R.W., and Haynes, S.J., eds., Turbidite-hosted gold deposits: Geological Association of Canada Special Paper 32, p. 87-105. Goldfarb, R.J., Leach, D.L., and Pickthorn, W.J., 1988, Accretionary tectonics, fluid migration, and gold genesis in the Pacific Border Ranges and Coast Mountains, southern Alaska, in Kisvarsanyi, G., and Grant, S.K., eds., North American Conference on Tectonic Control of Ore Deposits and the Vertical and Horizontal Extent of Ore Systems, 1988 Proceedings: Volume, Rolla, Mo., University of Missouri-Rolla, p. 67-79. Grant, U.S., and Higgins, D.F., Jr., 1910, Preliminary report on the mineral resources of the southern part of the Kenai Peninsula: U.S. Geological Survey Bulletin 442-D, p. 166178. Grinenko, V .A., 1962, Preparation of sulfur dioxide for isotope analysis: Zhurnal Neorganicheskoi Khimii., v. 7, p. 24782483. Hudson, T., Plafker, G., and Peterman, Z.E., 1979, Paleogene anatexis along the Gulf of Alaska margin: Geology, v. 7, p. 573-577. Johnson, B.L., 1914, The Port Wells gold-lode district: U.S. Geological Survey Bulletin 592, p. 195-235. ---1915, The gold and copper deposits of the Port Valdez district: U.S. Geological Survey Bulletin 622, p. 141-187. Kerrich, R., 1987, The stable isotope geochemistry of Au-Ag vein deposits in metamorphic rocks: Mineralogical Association of Canada Short Course Handbook, v. 13, p. Landis, G.P., Hofstra, A.H., Leach D.L., and Rye, R.O., 1987, Quantitative analysis of fluid-inclusion gases-Application to the study of ore deposits (abs.): U.S. Geological Survey Circular 995, p.38-39. Matsuhisa, Y., Goldsmith, J.R., and Clayton, R.N., 1979, Oxygen isotopic fractionation in the system quartz-albite-anorthite-water: Geochimica et Cosmochimica Acta, v. 43, p. 1131-1140. Nelson, S.W., Dumoulin, J.A., and Miller, M.L., 1985, Geologic map of the Chugach National Forest, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF1645-B, 18 p., 1 sheet, scale 1:250,000. Nilsen, T.H., and Zuffa, G.G., 1982, The Chugach terrane, a Cretaceous trench-fill deposit, southern Alaska: in Legget, J.K., ed., Trench-forearc geology-Sedimentation and tectonics on modern and ancient active plate margins: London, Blackwells, p. 213-227. Park, C.F., Jr., 1933, The Girdwood district, Alaska: U.S. Geological Survey Bulletin 849-G, p. 381-424. Pickthorn, W.J., 1982, Stable isotope and fluid inclusion study of the Port Valdez district, southern Alaska: Los Angeles, Calif., University of California, M.S. thesis, 66 p. Pickthorn, W.J., Goldfarb, R.J., and Leach, D.L., 1987, Comment on "Dual origin of lode gold deposits in the Canadian Cordillera": Geology, v. 15, p 471-472. Plafker, G., Jones, D.L., and Pessagno, E.A., Jr., 1977, A Cretaceous accretionary flysch and melange terrane along the Gulf of Alaska margin, in Blean, K.M., ed., The U.S. Geological Survey in Alaska-Accomplishments during 1976: U.S. Geological Survey Circular 751-B, p. B41B43. Richter, D.H., 1970, Geology and lode-gold deposits of the Nuka Bay area, Kenai Peninsula, Alaska: U.S. Geological Survey Professional Paper 625G, 16 p. Roedder, E., 1962, Studies of fluid inclusions 1-Low temperature application of a dual-purpose freezing and heating stage: Economic Geology, v. 57, p. 1045-1061. ---1984, Fluid inclusions: Mineralogical Society of America Reviews in Mineralogy, v. 12, 644 p. Sakai, H., 1968, Isotopic properties of sulfur compounds in hydrothermal processes: Geochemical Journal, v. 2, p. 29-49. Geochemistry of Lode-Gold Deposits, Nuka Bay District, Southern Kenai Peninsula

Smith, P.S., 1938, Mineral industry of Alaska in 1936: U.S. Geological Survey Bulletin 897-A, p. 1-107. Swanenberg, H.E.C., 1979, Phase equilibria in carbonic systems and their application to freezing studies in fluid inclusions: Contributions to Mineralogy and Petrology, v. 68, p. 303-306. Taylor, H.P., Jr., 1974, The application of oxygen and hydrogen isotope studies to problems of hydrothermal alteration and ore deposition: Economic Geology, v. 69, p. 843-883. Tuck, R., 1933, The Moose Pass-Hope district, Kenai Peninsula, Alaska: U.S. Geological Survey Bulletin 849-1, p. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Tysdal, R.G., and Case, J.E., 1979, Geologic map of the Seward and Blying Sound quadrangles, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map 1-1150, 12 p., 1 sheet, scale 1:250,000. Winkler, G.R. and Plafker, G., 1975, The Landlock fault-Part of a major early Tertiary plate boundary in southern Alaska in Yount, M.E., ed., U.S. Geological Survey Alaskan Program, 1975: U.S. Geological Survey Circular 722, p. 49. Reviewers: Lori Apodoca and john E. Gray

Placer Gold of the Kenai Lowland By Barrett A. Cieutat, Richard j. Goldfarb, Dwight C. Bradley, and Bruce H. Roushey Abstract A geochemical survey of the Kenai lowland, Cook Inlet basin, Alaska, produced stream-sediment and heavymineral-concentrate samples that were consistently anomalous in gold. Stream-sediments with as much as 2.8 ppm Au and heavy-mineral-concentrate samples with up to 700 ppm Au are believed to reflect reworked material from Quaternary glacial deposits and (or) from the upper part of the Tertiary Kenai Group. Gold grains are generally concentrated in the -35-mesh to +230-mesh size fractions. Concentrations of heavy minerals from an outcrop of Quaternary or Tertiary partly lithified conglomerate along the Fox River reveal that at least some of the gold was derived from the Chugach terrane to the east. Although no specific placer or lode occurrences that have resource potential have been identified, geomorphic processes in the Kenai lowland may have formed small, locally economic, undiscovered placer concentrations. INTRODUCTION Metallic mineral resources are largely unrecognized within the southern Kenai lowland of south-central Alaska (fig. 1). The only confirmed mineral production in the study area (fig. 1) is from a small placer gold occurrence mined intermittently from 1889 to 1911 at the mouth of the Anchor River (Cobb, 1973). Fine gold was found in thin layers of gravel m below the surface of the beach (Martin and others, 1915). Platinum was rumored to occur with the gold (Cobb, 1973). Martin and others (1915) also reported placer activities in the Kenai lowlands to the north of the present study area, near Ninilchick and between the Killey River and Kenai, but the success of these operations is unknown. Presently, placer-gold claims exist along Cook Inlet at Anchor Point and about 2 km north of the mouth of Diamond Creek (U.S. Bureau of Mines MAS deposit listings). A lo·de-gold claim is also recorded from the beach within 1 km of the mouth of Diamond Creek. Little is known about this claim, but it most likely includes rocks of the Tertiary Kenai Group and (or) overlying Quaternary gravels exposed in the beach cliffs. As part of the Alaska Mineral Resources Appraisal Program (AMRAP) study of the Seldovia 1° by 3° quadrangle, we conducted geological and geochemical studies in the southern Kenai lowland. Results from the studies indicate a widespread distribution of detrital gold grains in stream alluvium. This report details and discusses the significance and possible origin of widespread, anomalous gold values in stream-sediment and rock samples from this area. These findings are surprising because previous reports of gold in this populated region are restricted to the beach claims described above. REGIONAL GEOLOGY The Kenai lowland is part of the Cook Inlet Basin, an active forearc basin of the present day AleutianAlaska Range subduction zone (Magoon and Egbert, 1986). The basin is flanked to the west by active volcanoes that are built on older arc basement and to the east by an uplifted Mesozoic accretionary prism (the Chugach terrane) composed mainly of mafic volcanic rocks, chert, argillite, and graywacke. The Kenai lowland is characterized by gentle to flat topography_, except near shoreline escarpments, and by an abundance of fair to poorly drained marshland. The lowland is covered by Quaternary alluvial and glacial deposits that are underlain by sedimentary rocks of the Tertiary Kenai Group (Magoon and others, 1976). Only the two youngest formations of the Kenai Group-the Beluga Formation and the overlying Sterling Formation-are exposed within our study area. The Miocene Beluga Formation consists of up to 1,525 m (5,000 ft) of claystone, siltstone, thin sandstone, and subbituminous coal (Hartman and others, 1972). The Beluga Formation crops out in cliffs along the shore of Kachemak Bay to the northeast of Homer and along the shore of Cook Inlet northwest of Homer. Only the upper few hundred meters are exposed; the remainder is known from subsurface studies (Hayes and others, 1976). Paleocurrent directions are toward the west-northwest (Rawlinson, 1984). The most abundant sandstone Placer Gold of the Kenai Lowland

framework grains in the Beluga Formation are weakly metamorphosed sedimentary rocks comparable to those exposed in the Kenai Mountains to the east (Hayes and others, 1976). The heavy-mineral assemblage, which is dominated by epidote, garnet, apatite, and zircon, also is consistent with an easterly source (Hayes and others, 1976). Flores and Stricker (this volume) interpreted the Beluga Formation as having been deposited by an anastomosing fluvial system; alternatively, Hayes and others (1976) and Rawlinson (1984) interpreted it as having been deposited by braided streams. The late Miocene and Pliocene Sterling Formation consists of more than 3,050 m (l 0,000 ft) of massive sandstone and conglomerate, which are interbedded with thin claystone and lignite (Hartman and others, 1972). Rocks of the Sterling Formation are intertongued with those of the underlying Beluga Formation. Extensive exposures of the Sterling Formation occur along Kachemak Bay, along the Fox River north of Kachemak Bay, and around Epperson Knob and Lookout Mountain in the Anchor River drainage. Only the lowest 700 m of the Sterling Formation crops out, the remainder being known in the subsurface. Evidence summarized by COOK INLET Hayes and others (1976) indicates that the Sterling Formation was mainly derived from sources to the west. West-northwesterly paleocurrent directions measured at Kachemak Bay near the eastern basin margin, however, also indicate a sediment source in the Kenai Mountains (Rawlinson, 1984 ). Sandstone framework grains in the Sterling Formation consist of quartz, plagioclase, biotite, and glassy volcanic rock fragments and suggest derivation from the magmatic arc to the west (Hayes and others, 1976). The heavy-mineral assemblage, which is dominated by hornblende and pyroxene, also suggests an arc source (Kirschner and Lyon, 1973; Hayes and others, 1976). A meandering stream depositional environment has been widely accepted for the Sterling Formation (Hayes and others, 1976; Rawlinson, 1984: Flores and Stricker, this volume). Quaternary surficial deposits form a discontinuous mantle that unconformably overlies the Kenai Group. These deposits of till, outwash, and glaciolacustrine mud vary in thickness from 0 to 320 m; the thickest Quaternary section is in an exploratory well near Kenai (Calderwood and Fackler, 1972). At least five major Pleistocene and two minor post-Pleistocene glaciations lnllchlk Dome KENAI LOWLAND KENAI Anchor Point Figure 1. Site location map for sluice-concentrate and rock samples in Kenai lowland. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 PENINSULA KENAI MOUNTAINS 10 Mlt.ES 15 KILOMETERS

have been identified in the Kenai low land (Karlstrom, 1964 ). In the Seldovia quadrangle, direct evidence for four Pleistocene ice advances is provided by moraines and related ice-marginal deposits of the Caribou Hills, Eklutna, Knik, and N aptowne (listed from oldest to youngest; Karlstrom, 1964 ). Tertiary strata in the Kenai lowland generally can be discriminated from the Quaternary deposits on the basis of two characteristics-the presence of coal, and partial to complete lithification. However, a problematic succession of variably lithified, gold-bearing gravel and conglomerate occurs in 50-m-high bluffs along the upper part of the Fox River at location S1 (fig.1); these deposits deserve special mention. Weakly statified, clastsupported gravels occur in beds up to 10 m thick and are associated with thick beds (several meters thick) and thinner lenses of weakly lithified to unlithified sand. Some conglomerate beds are sufficiently lithified to occur as talus blocks at the foot of the bluff. Clasts (the largest are about 20 em long) include melange, graywacke, and greenstone; chert of the McHugh Complex; and hornfelsed metapelite of the Valdez Group. The clasts indicate an easterly sediment source in the Kenai Mountains. Bedding dips about 15 ° W. Similar deposits that fill paleovalleys among the the southeastern shore of Kachemak Bay have been assigned to the Tertiary Kenai Group on the basis of abundant plant fossils (Magoon and others, 1976). However, published maps of the regional bedrock geology (Magoon and others, 1976) and surficial geology (Karlstrom, 1964) have assigned the unfossiliferous upper Fox River outcrops to the Quaternary. In light of bedding dips, degree of lithification, and similarities to known Tertiary strata along the Kenai Mountain front, we favor the interpretation that the upper Fox River gravels are Tertiary in age and represent proximal alluvial-fan deposits that are laterally equivalent to fluvial deposits of the Beluga and (or) Sterling Formations. However, the possibility that the gold-bearing conglomerates and gravels are indeed Quaternary remains viable. REGIONAL GEOCHEMISTRY Stream-sediment and heavy-mineral-concentrate samples were collected at 75 sites within the Kenai lowland as part of an extensive regional geochemical reconnaissance survey of the Seldovia 1° by 3° quadrangle. Samples were collected between Cook Inlet and the Fox River, with the majority located south of Ninilchik Dome along first- and second-order tributaries to the Chakok and Anchor Rivers. At least five grab samples were collected at each site along a 9-m stretch of stream channel. The grab samples were composited into a single sample, air-dried, sieved in the laboratory using a stainless-steel 80-mesh screen, and pulverized prior to chemical analysis. Heavy-mineral-concentrate samples were collected at all sediment sites using a 14-in.-diameter gold pan. Typically, 3 to 4 kg of composited sediment were collected and panned, yielding the desired 30 to 60 g of concentrate. Remaining lightweight material was separated by floatation in bromoform (specific gravity 2.86), and the resulting heavy-mineral fraction was separated into magnetic, semimagnetic, and nonmagnetic fractions using a Frantz Isodynamic Separator. Both the stream-sediment sample and the nonmagnetic heavy-mineral-concentrate fraction were analyzed for minor and trace elements by emission spectrography according to the method outlined by Grimes and Marranzino (1968). Owing to the relatively high lower limits of detection for emission spectrography, more sensitive analytical methods were 1JSed on stream-sediment samples. They were analyzed for Ag, As, Au, Bi, Cd, Cu, Mo, Pb, Sb, and Zn by inductively coupled plasma spectroscopy (ICP) following the method of Motooka (1988). Lower determination limits for the method for gold and silver are 0.045 and 0.15 ppm, respectively. Gold concentrations were also determined by flame atomic absorption (Thompson and others, 1968) for about half the stream-sediment samples and by graphite furnace atomic absorption spectrophotometry (Meier, 1980) for the remaining stream-sediment samples. Samples ·that registered below the lower determination limit for flame atomic absorption were analyzed using the graphite furnace method. Lower determination limits are 0.05 and 0.002 ppm for the two methods, respectively. Anomalous concentrations of gold were identified in stream-sediment samples from 27 of the 75 sampled sites. Graphite furnace atomic absorption data indicate an anomaly threshold value of 0.007 ppm for gold in stream sediments from the Kenai .lowland. Fourteen of the sediment samples contain at least 0.012 ppm Au, and eight of these contain 0.3 to 2.8 ppm Au. Highest values were found in samples collected just above the beach in Mutnaia Gulch (1.0 ppm), along a tributary to the Anchor River in sec. 22, R. 14 W., T. 5 S. (1.8 ppm), and from another tributary to the Anchor River along the east . side of the Sterling highway about 1.5 km north of the Anchor River campground (2.8 ppm). No other anomalous metals were found consistently in stream-sediment samples containing anomalous gold. Silver concentrations for all 75 stream-sediment samples were at background levels, ranging from not detected at the 0.045-ppm lower determination limit to 0.080 ppm. A few samples with anomalous gold values contained 30 to 40 ppm As, but most arsenic values were below 20 ppm. Sixteen of 71 heavy-mineral-concentrate samples contained anomalous gold concentrations ranging from 20 ppm to 700 ppm. In addition, three other samples had detectable gold but at levels below the 20 ppm Placer Gold of the Kenai Lowland

lower determination level. Small flakes of gold were commonly visible in these samples when viewed with a 1 OX hand lens in the field. The majority of these samples were from sites that lacked anomalous gold concentrations in corresponding stream sediments. Similarly, most sites with anomalous gold concentrations in stream-sediment samples yielded heavy-mineral-concentrate samples that lacked anomalous gold. The lack of corrc:lation suggests that (1) gold grains are relatively fine and easily lost during the panning process, and (2) gold grains likely occur in many other heavy-mineralconcentrate samples but in concentrations below the 20ppm lower determination level. Anomalous silver values of 1 to 20 ppm in heavy-mineral-concentrate samples characterize most of the sites with gold anomalies in stream-sediment or heavy-mineral-concentrate samples. Molybdenum, tin, and tungsten are commonly found at anomalous levels in many of the heavy-mineral-concentrate samples from throughout the Kenai low land. Most of these samples contain microscopically visible scheelite; molybdenite was identified in one sample (Richard Tripp, unpub. data). Whereas these metals show no distinct correlation with the gold and· silver anomalies, they do suggest a strong igneous component within the Tertiary sedimentary rocks that are the source for much of the stream sediment. The regional geochemical data indicate a widespread distribution of gold grains in the Kenai lowland. The distribution, as well as the commonly delicate morphology of gold flakes viewed with a hand lens, suggest an extensive, locally derived source. Brooks (1911) speculated that the gold was derived either from rocks of the Tertiary Kenai Group or from overlying glaciofluvial gravels; our d.ata allow both possibilities. FOLLOWUP INVESTIGATION Sediments Three stream channel localities that contained samples with anomal:ous gold collected during the regional investigation were revisited for more detailed study of the placer material. Bulk samples were col-· lected at Mutnaia Gulch (site S2), on Chakok River (site S3), and on Twitter Creek (site S4). These sites (fig.l) had yielded ·samples with anomalous concentrations of gold in sediment only, both concentrate and sediment, and concentrate only, respectively. In an effort to better estimate the amount of gold at the three sites, sixteen 14-in.-diameter gold pans of alluvium were sluiced at each site. The sixteen pans approximately equal one-sixteenth of a cubic yard of alluvium. This quantity is sufficient to remove the "nugget effect" from analyses (S. Fechner, oral commun., Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Table 1. Concentrations of gold in nonmagnetic fractions of sluice concentrates from selected stream sample sites in the Kenai lowland, Alaska [Values in parts per million] Sample Mesh size: -10 to + 18 <0.002 -18 to +35 <0.002 -35 to +80 920 -80 to +230 840 -230a avalues include combination of magnetic, semimagnetic, and nonmagnetic fractions because -230-mesh fraction was too small for effective use of the magnetic separator. 1991). The alluvium was sieved using a 10-mesh (2.0mm) stainless-steel screen prior to being placed into the sluice. The sluice was arranged in each stream so that water velocity carried a 1-in.-diameter pebble through at a slow, steady rate. Material that did not wash away was saved for analysis. A thorough cleaning of the sluice between samples ensured minimal contamination. Site S2, S3, and S4 samples were air dried in the laboratory before being sieved into five size fractions (-10-mesh to +18-mesh, -18-mesh to +35-mesh, -35mesh to +80-mesh, -80-mesh to +230-mesh, and -230mesh). Lightweight material from each fraction was separated by floatation in bromoform (specific gravity 2.86), and the resulting heavy-mineral fraction was split into magnetic, semimagnetic, and nonmagnetic fractions using a Frantz Isodynamic Separator. Semimagnetic and magnetic fractions were analyzed for gold by emission spectrography using the method outlined by Grimes and Marranzino (1968). The nonmagnetic fraction was analyzed for gold using flame atomic absorption (Thompson · and others, 1968). The lower determination limits for the two are 20 and 0.050 ppm, respectively. The semimagnetic and magnetic fractions contained no detectable gold at any site in any of the five sieve catagories. The nonmagnetic fraction, however, contained gold in three of the size fractions for sites S2, S3, and S4: -35-mesh to +80-mesh, -80-mesh to +230mesh, and -230-mesh (table 1). The -35-mesh to +80mesh and the -80-mesh to +230-mesh ·fractions contained the bulk of the gold, whereas no gold was found in the -10-mesh to +18-mesh and the -18-mesh to +35-mesh fractions. Sample site S2 had the greatest concentrations of gold with 920, 840, and 8.6 ppm in the -35-mesh to +80-mesh, -80-mesh to +230-mesh, and -230-mesh fractions, respectively. All of the sluice samples contained visible gold when viewed with a lOX hand lens in the field. Site S2 is located at the mouth of Mutnaia Gulch on the shore of Cook Inlet. It is a small, steplike stream

( m across) of moderate gradient with water depth never exceeding 0.3 m in the sampling area. The sample was obtained from the drainage above possible tidal influence. The Beluga Formation crops out at the beach and on both sides of the gulch. Site S3 on Chakok River predominantly drains Quaternary sediments. However, within its drainage basin to the east is Epperson Knob, which consists of the Sterling Formation. The sampled stream is 3 to 4 m across, with depths to 1.5 m on some cutbanks. It has a low gradient, meanders, and has alluvium ranging in size from silt to cobble. Site S4 is located on Twitter Creek just north of Homer. It mainly drains the Sterling Formation of Ohlson Mountain, Crossman Ridge, and Lookout Mountain, but the Beluga Formation is present on some of the lower stretches near the Anchor River. It is of low to moderate gradient, 1.5 to 3 m across, and 0.3 to 1 m deep. A B Figure 2. Gold grains from two sluice-concentrate sites within Kenai lowland. A, Grains from sample site 52. Longest gold grain is approximately 1 mm across. B, Grains from sample site 51. Table 2. Concentrations of gold of rock samples from the Kenai lowland, Alaska [Values in parts per million. N, not detected at indicated lower determination limit; L, detected but below indicated lower determination limit; Do., ditto] Sample Concentration Location Source R 1-1 L(.002) T.3 S., R. 9 W., conglomerate R 1-2 L(.002) sec. 16, SE4 cong. matrix R 1-3 L(.002) Do. R 1-4 L(.002) Do. R 1-5 L(.002) Do. R2B-1 T. 2 S., R. 9W., conglomerate R2B-2 sec. 24, NW~·4 cong. matrix R2B-3 Do. R2B-4 Do . R2B-5 Do. R2A-1 Do . R2A-2 Do . R2A-3 Do. R3-1 T. 5 S., R. 14 W., claystone R3-2 N(.002) sec. 26, NW~4 coaly coarse ss R3-3 N(.002) coal R3-4 L(.002) 2 mm clactics R3-5 L(.002) 2 mm clastics Prior to analysis the gold grains from site S2 were hand picked (fig. 2A) for examinations of gold morphology; the gold was returned to the sample before chemical analysis. Site S2 grains are generally flat and rounded to irregular, with the largest grain being 0.95 mm at its longest dimension. With increasing distance of mechanical transport, gold-grain morphology changes from delicate, through irregular and abraded, to rounded, while grain size is progressively diminished (Grant and others, 1991 ). The studied gold grains in all likelihood have undergone at least secondary fluvial and possibly glacial transport. The range in shape from round to irregular suggests either that the gold has weathered out at different locations along the flowpath of the stream, or that morphologic differences among gold grains are inherited from prior fluvial histories, or both. Rocks The only available background gold values for the Kenai Group are from three conglomerates analyzed in the present study (table 2). Owing to relatively weak lithification, we were able to remove coarse material (cobbles to boulders) by hand. The samples then were sieved and the +35-mesh fraction was discarded. Sample R1 is from the Sterling Formation along the Fox River. The outcrop consists of conglomerate with a coarse sandstone matrix and interlayered siltstones and Placer Gold of the Kenai Lowland

claystones. The -35-mesh fraction was crushed, ground, and analyzed for low-level gold by graphite furnace atomic absorption spectrophotometry (Meier, I980). None of the five splits from site RI measured above the lower determination limit. Sample R3 is from the Beluga Formation about I 0 km northwest of Homer along the south bank of the Anchor River. The outcrop consists of conglomerate, claystone, coaly iron-stained coarse sandstone, and coal. The conglomerate matrix contained detectable gold below the determination limit. The claystone contained 0.0 I4 ppm gold, the highest value for any rock sample. No gold was detected in the sandstone or coal. Sample R2 is from the undifferentiated (Tertiary or Quaternary) gravel and coglomerate along the upper Fox River (fig. 3). Eight splits of the sample were analyzed for gold; values ranged from 0.002 to 0.006 ppm. Sample S I was taken from the same location and was crushed prior to being sieved, sluiced, prepared, and analyzed in similar fashion to the sediment samples S2, S3, and S4 mentioned previously. For sample S I, the semimagnetic and magnetic fractions from all sieve categories were barren of detectable gold at the 20-ppm determination limit using an optical emission spectrograph according to the method outlined by Grimes and Marranzino (I968). Also barren were the -IO-mesh to +I8-mesh, -I8-mesh to +35-mesh, and -35-mesh to +80mesh sieve sizes of the nonmagnetic fraction. The -80mesh to +230-mesh size fraction contained 5.6-ppm gold, while the -230-mesh size fraction contained 0.3-ppm gold. Gold from the sluice concentrate from site S I was hand picked (fig. 2B) to examine morphology. Gold grains are smaller and more delicate than those from site S2. The largest grain is approximately 0.45 mm on its longest dimension. Since the sample at site S I was taken directly from outcrop, the gold has not undergone reworking by present fluvial activity. Therefore, barring chemical transport and reprecipitation of gold, it appears likely that the lack of present-day mechanical transport is responsible for the more delicate nature of gold grains at site S I. DISCUSSION AND CONCLUSIONS Placer gold is extensive throughout the Kenai lowland. Of 75 sampled drainages, 27 were anomalous for gold in stream sediment and I6 were anomalous for gold in heavy-mineral concentrates from stream sediment. A sluice box was used to assess the placer-gold potential of three streams. Of the three stream locations (sites S2, S3, and S4) where the sluice w~s used to concentrate larger volumes of material, none contained significant gold quantities. The fine size (generally <0.2 mm) and lack of concentration of the gold deflate any hope of a minable resource at this time. Whether the gold has been reworked from the Quaternary deposits or the Tertiary Kenai Group, or both, remains unresolved. The two bedrock sites in unequivocal Sterling and Beluga Formations lack significant gold, Figure 3. Outcrop ofT ertiary or Quaternary conglomerate and gravel along upper Fox River where sluice sample 51 and rock sample R2 were taken. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

but more sampling would likely be necessary to locate significant quantities. In addition, it is possible that the placer gold was derived from the coarser fraction of conglomerates-the fraction that was discarded prior to chemical analysis. Although gold was recovered from bluffs of gravel and conglomerate along the upper Fox River (samples R2, S1), the stratigraphic position is ambiguous. Regardless of whether the strata are Tertiary or Quaternary, the occurrence of paleoplacer gold in alluvium derived from the Kenai Mountains does indicate that at least some of the placer gold was ultimately derived from the east. An analagous situation to the gold in the Kenai lowland may be the placer deposits along the Gulf of Alaska coastline from Cape Yakataga to Icy Bay. There, the Yakataga Formation is the principal source of gold, which is being reworked and concentrated by recent fluvial systems and by longshore currents (Reimnitz and Plafker, 1976; Eyles, 1990). Eyles (1990) cited glaciers as playing a major role in transporting detrital gold over such large areas and noted that postglacial shallow-marine and fluvial activity concentrated the gold. Eyles (1990) also p~inted out that gold shows a noticeable size reduction as it moves away from its source river along the beach. REFERENCES CITED Boss, R.F., Lennon, R.B., and Wilson, B.W., 1976, Middle Ground Shoal oil field, Alaska, in Braunstein, Jules, ed., North American oil and gas fields: American Association of Petroleum Geologists Memoir 24, p. 1-22. Brooks, A.H., 1911, The Mount McKinley region, Alaska, with descriptions of the igneous rocks and of the Bonnifield and Kantishna districts~ U.S. Geological Survey Professional Paper 70, 234 p. Calderwood, K.W., and Fackler, W.C., 1972, Proposed stratigraphic nomenclature for Kenai Group, Cook Inlet basin, Alaska: American Association of Petroleum Geologists Bulletin, v. 56, no. 4, p. 739-754. Cobb, E.H., 1973, Placer deposits of Alaska: U.S. Geological Survey Bulletin 1374, 213 p. Byles, N., 1990, Glacially derived, shallow-marine gold placers of the Cape Yakataga district, Gulf of Alaska: Sedimentary Geology, v. 68, p. 171-185. Byles, N., and Kocsis, S.P., 1989, Sedimentological controls on gold in a late Pleistocene glacial placer deposit, Cariboo Mining District, British Columbia, Canada: Sedimentary Geology, v. 65, p. 45-68. Grant, A.H., Lavin,O.P., and Nichol, 1.,1991, The morphology and chemistry of transported gold grains as an exploration tool: Journal of Geochemical Exploration, v. 40, p. 73-94. Grimes, D.J., and Marranzino, A.P., 1968, Direct-current arc and alternating-current spark emission spectrographic field methods for the semiquantitative analysis of geologic materials: U.S. Geological Survey Circular 591, 6 p. Hartman, D.C., Pessel, G.H., and McGee, D.L., 1972, Preliminary report on stratigraphy of Kenai Group, upper Cook Inlet, Alaska: Alaska Division of Geological and Geophysical Sun;eys, Special Report 5, 4 p., 7 maps, scale 1 :500,000, 1 pl. Hayes, J.B., Harms, J.C., and Wilson, T.W., 1976, Contrasts between braided and meandering stream deposits, Beluga and Sterling Formations (Tertiary), Cook Inlet, Alaska, in Miller, T.P., ed., Recent and ancient sedimentary environments in Alaska: Anchorage, Alaska Geological Survey, p.J1-J27. Karlstrom, T ., 1964, Quaternary geology of the Kenai lowland and glacial history of the Cook Inlet region, Alaska: U.S. Geological Survey Professional Paper 443, 69 p. Kirschner, C.E., and Lyon, C.A., 1973, Stratigraphic and tectonic development of Cook Inlet petroleum province, in Pitcher, M.G., ed., Arctic geology: American Association of Petroleum Geologist's Memoir 19, p. 396-407. Magoon, L.B., Adkinson, W.L., and Egbert, R.M., 1976, Map showing geology, wildcat wells, Tertiary plant-fossil localities, K-Ar age dates, and petroleum operations, Cook Inlet area, Alaska: U.S. Geological Survey Miscellaneous Geologic Investigations Map I-1 019, 3 sheets, scale 1:250,000. Magoon, L.B., and Egbert, R.M., 1986, Framework geology and sandstone compositions, in Magoon, L.B., ed., Geologic studies of the lower Cook Inlet COST No. 1 Well, Alaska outer continental shelf: U.S. Geological Survey Bulletin 1596, p. 65-90. Martin, G.C., Johnson, B.L., and Grant, U.S., 1915, Geology and mineral resources of the Kenai Peninsula, Alaska: U.S. Geological Survey Bulletin 587, 243 p. Meier, A.L., 1980, Flameless atomic absorption determination of gold in geologic materials: Journal of Geochemical Exploration, v. 13, p. 77-85. Motooka, J .M., 1988, An exploration geochemical technique for the determination of preconcentrated organometallic halides by ICP-AES: Applied Spectroscopy, v. 42, p. 1293-1296. Rawlinson, S.E., 1984, Environments of deposition, paleocurrents, and provenance of Tertiary deposits along Kachemak Bay, Kenai Peninsula, Alaska: Sedimentary Geology, v. 38, p. 421-442. Reimnitz, E., and Plafker, G., 1976, Marine gold placers along the Gulf of Alaska margin: U.S. Geological Survey Bulletin 1415, 16 p. Thompson, C.E., Nakagawa, H.M., and VanSickle, G.H., 1968, Rapid analysis for gold in geological materials, in Geological Survey research 1968: U.S. Geological Survey Professional Paper 600-B, p. B130-B132. Reviewers: Robert Eppinger and Greg lee Placer Gold of the Kenai Lowland

Summary of Results of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 o by 3° Quadrangles, Alaska By Thomas P. Frost, Stephen E. Box, and Elizabeth j. Moii-Stalcup Abstract A synthesis of geologic, geochemical, and geophysical data from the Bethel and southern part of the Russian Mission quadrangles in southwestern Alaska were used to estimate the mineral potential of the region. In the map area, the potential for undiscovered gold placer deposits is high, the potential for epithermal mercury and polymetallic vein deposits is moderate, and the potential for other meta II ic mineral resources is low. Five tracts, distinguished on the basis of their basement geology, are identified as favorable for the discovery of gold placers in the eastern half of the Bethel 1 :250,000 map area. The most important geologic criterion for the delineation of the gold placer tracts is the presence of granitoid plutons. Tracts favorable for epithermal mercury or simple antimony deposits have either shallow granitoid plutons or rhyolitic volcanic rocks present in them. Tracts favorable for silica-carbonate mercury deposits contain altered serpentinites along major terrane-bounding faults. Two types of gold-bearing polymetallic quartz veins are recognized in the map area: one contains significant base metals along with gold and silver; the other is lacking in base-metal enrichments. Both gold-bearing vein subtypes are associated with granitoid plutons or rhyolitic dikes and domes. Tourmaline-quartz replacement of volcanic rocks is locally present in the map area; the replacement zones are variably enriched in Au, Ag, base metals, Sn, and W. Quartz veins with anomalous Sn and W are also present near some granitoid plutons. INTRODUCTION The Bethel region has been an important placer goldproducing region in southwestern Alaska; over 250,000 oz of gold have been produced through 1989 (Bundtzen and others, 1989). The potential for undiscovered placer gold remains high in streams that drain areas underlain by granitoid plutons and their wall rocks. No other metallic minerals have been produced from the region, although Geologic Studies in Alaska by the U.S. Geological Survey, 1991 there is low to moderate potential for epithermal mercury and polymetallic veins containing gold. A geological mapping program and reconnaissance geochemical survey were conducted as part of the Alaska Mineral Resource Assessment Program (AMRAP) during 1987-89 in the Bethel and southeastem part of the Russian Mission 1 o by 3 o quadrangles (referred to hereafter as the Bethel map area or simply map area) of southwest Alaska. The western half of the Bethel map area is underlain by unconsolidated Quaternary deposits of glacial, fluvial, lacustrine, and eolian origin; this area is not considered in this report. This report is a summary of the complete mineral resource assessment, which will appear as a separate U.S. Geological Survey Bulletin. Products to date from the Bethel AMRAP include a geologic map (Box and others, in press); gravity and aeromagnetic maps (Phillips and Morin, 1992); studies of prospects and mineral occurrences (Frost, 1990; Gray and others, 1990; Goldfarb and others, 1990; Frost and Box, 1991a, b); isotopic, petrographic, and geochemical studies (Frost and others, 1988, 1992a, b; Moll-Stalcup and others, 1989; Box and others, 1990, in press; MollStalcup and Box, 1992; Roeske and Box, 1992); geochemical sampling of mines, prospects, and altered and unaltered bedrock (Frost and others, 1992a); streamsediment and heavy-mineral-concentrate data (Bradley and Frost, in press); and tectonic, sedimentologic, a~d fossil studies (Box and Murphy, 1987; Box and Elder, 1992; Elder and Box, 1992; Box, 1992). GEOLOGIC SUMMARY Pre-Late Cretaceous Tectonostratigraphic Terranes The Bethel map area (fig.l) is underlain by five accreted tectonostratigraphic terranes, separated from

one another by major faults (Jones and others, 1987), although many boundaries are covered by sedimentary rocks of the Upper Cretaceous Kuskokwim Group (Box and others, in press). The terranes, from east to west, are briefly described below. A structurally complex assemblage of Paleozoic and Mesozoic chert and Devonian to Triassic limestone, basalt, and clastic rocks form the Tikchik terrane (Mertie, 1938; Hoare and Coonrad, 1959a; 1978; Box and others, in press). The right-latern.l Togiak fault marks the boundary between the Tikchik terrane and the Togiak terrane to the west (Box and Mmphy, 1987; Box and others, in press). Rocks of the Togiak terrane (fig. 1) are complexly deformed andesitic arc volcanic and volcaniclastic rocks of Late Triassic through Early Cretaceous age (Box and others, in press). The boundary between the rocks of the Togiak terrane and the Goodnews terrane to the west is covered by sedimentary rocks of the Upper Cretaceous Kuskokwim Group (Box and others, in press). The Goodnews terrane is exposed in erosional windows through the unconformably overlying Upper Cretaceous Kuskokwim Group (fig. 1 ). The Goodnews terrane is a structurn.lly complex assemblage of variably foliated metabasalt, low-grade schist, marble, chert, graywacke, and slate (Hoare and Coonrad, 1959a, 1978; Box and Mmphy, 1987). Fossils are Ordovician to Early Cretaceous in age (Hoare and Coonrad, 1978; Box and others, in press). Proterozoic orthogneiss, amphibolite, and quartzmica schist of the Kilbuck terrane crop out in a narrow tectonic sliver east of the Golden Gate fault (fig. 1) (Box and others, 1990). The Kilbuck terrane is unconformably overlain by conglomerates of the Upper Cretaceous Kuskokwim Group that contain clasts derived in part from rocks of the Kilbuck terrane (Box and Murphy, 1987). Middle and Upper Jurassic andesitic volcanic rocks and volcaniclastic sandstones of the Nyac terrane (Box and Murphy, 1987) are the westernmost basement rocks exposed in the Bethel map area. The contact between the Nyac terrane and other pre-Late Cretaceous terranes is not exposed. The contact with the Upper Cretaceous Kuskokwim Group to the east is marked by the Sawpit fault (fig. 1) or its equivalent in the southern part of the Bethel map area (Hoare and Coonrad, 1959a, b; Box and others, in press). Late Cretaceous and (or) early Tertiary volcanic and plutonic rocks appear to overlap and intrude, respectively, the Sawpit fault (Box and others,, in press). Cretaceous Sedimentary Rocks The Upper Cretaceous Kuskokwim Group (fig. 1) consists of a thick sequence of shale, siltstone, sandstone, and basal conglomerate that was deposited unconformably on all older rocks except those of the Nyac terrane (Hoare and Coonrad, 1959a, b, 1978; Box and Murphy, 1987; Elder and Box, 1992; Box, 1992). The Kuskokwim Group is composed predominantly of marine turbidites in most of the map area, and of shallow marine and nonmarine conglomerates and sandstones in the southwestern part of the map area (Box, 1992; Box and others, in press). Cretaceous and Early Tertiary Plutonic and Volcanic Rocks Early Cretaceous (120-102 Ma) granitoid plutons are restricted to the Nyac terrane west of the Sawpit fault (fig. 1 ); granitoid plutons and volcanic fields that intrude or overlie other terranes and the Kuskokwim Group are Late Cretaceous to Tertiary (72-55 Ma) in age (Shew and Wilson, 1981; Robinson and Decker, 1986; Frost and others, 1988, 1992b; Box and others, in press). Hornblende-biotite granodiorite through granite is most common, although augite-biotite quartz monzodiorite, quartz diorite, diorite, and gabbro also are present (Robinson and Decker, 1986; Frost and others, 1988; Box and others, in press). The rocks are porphyritic or coarse grained; many contain miarolytic cavities. Most plutons are surrounded by a biotite±cordierite-bearing hornfels zone. Textures and hornfels minern.logy suggest emplacement depths of less than 2 km to subvolcanic levels. Late Cretaceous to early Tertiary volcanic rocks crop out in several geographically distinct calc-alkalic volcanic fields (fig. 1 ). These rocks consist mainly of bedded andesitic to dacitic flows and tuffs, minor basaltic flows, volcaniclastic sedimentary rocks, and scattered felsic domes (Hoare and Coonrad, 1959a; Box and others, in press). The domes are composed of biotite rhyolite and dacite that contain partially resorbed quartz phenocrysts. Intermediate to felsic dikes cut all older rock types. Mildly alkalic Eocene basalt and andesite flows and tuffs, as well as rhyolite tuffs and domes, crop out in the Nukluk volcanic field (Moll-Stalcup and others, 1992), overlapping the Sawpit fault in the western part of the map area (fig. 1) Quaternary Deposits The western third of the Bethel quadrangle is covered by thick loess, lacustrine, outwash, and moraine deposits. All known placers in the quadrangle are localized in alluvial deposits of active rivers or streams, where they are incised in bedrock, or in terrace gravels within a few tens of meters above the present channels. Glaciation in the eastern half of the quadrangle has been of both mountain and valley type, and for the most part it appears to have disrupted preexisting placers (Hoare and Coonrad, 1959a; Hoare and Cobb, 1977). Summary of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 o by 3° Quadrangles

GEOCHEMICAL STUDIES A total of 1,486 stream-sediment, 1,104 heavymineral-concentrate samples from first- and second-order streams, and 1, 773 rock samples were collected and analyzed (Bradley and Frost, 1992; Frost and others, 1992a). These sample media provide a representative

TAYLOflt.ATNS OllliNGHAt.A rl! ... Area of map 20 KILOMETERS composite of the transported material of a stream or drainage basin; the composition of the samples ideally represents the bedrock exposed in the drainage basin. Both unaltered and altered rock samples were collected to identify background chemical signatures of the various geologic units and to identify element suites associated with mineralization processes. Figure 1. Simplified geologic map of Bethel area. Geographic names as used in this report are indicated. Shaded area in inset shows area of this and subsequent figures. Adapted from Frost and others (1988), Frost (1990), and Box and others (in press). Pluton descriptions in tables 1-3. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

MINERAL RESOURCES OF THE BETHEL MAP AREA This mineral resource assessment follows in principle the protocols developed by the U.S. Geological Survey in assigning to geographic tracts varying degrees of certainty of discovery of mineral resources with geologic and geochemical similarities to known mineral deposits (for example, Cox and Singer, 1986). A mineral occurrence is defined as a "concentration of a mineral that is considered valuable by someone somewhere, or that is of scientific or technical interest" (John and others, in press). A mineral deposit is a "mineral occurrence of sufficient size and grade that it might, under the most favorable circumstances, be considered to have economic potential" (John and others, in press). Permissive tracts are areas in which the geologic, geochemical, and geophysical evidence does not eliminate the possibility of occurrence of a given mineral deposit type. Favorable tracts are areas in which there is positive indication that mineral deposits may be found. For example, presence of known occurrences, presence of plutons known to be associated with a given deposit type, or presence of an appropriate suite of anomalous elements in geochemical data are positive indicators that mineralization processes have occurred in a tract. In the interest of brevity, only favorable tracts for most deposit types are discussed in the following sections. EXPLANATION SUPERJACENT UNITS D Surficial deposits (Quaternary), undivided D ' D

PLUTONIC ROCKS Early Cretaceous plutons (some units may be Jurassic or Late Cretaceous) (west of Sawpit fault) Plutons indicated by number: 1. Nyac, 2. Bonanza Creek, 3. Sawpit, 4. Fox Creek, 5. Slate Creek, 6. Dry Creek, 7. Columbia Creek, 8. Little Kasigluk River Cretaceous to early Tertiary plutons (mostly east of Sawpit fault) Plutons indicated by number: 9. Shiping Dome, 10. Mt. Plummer, 11. Marvel Creek, 12. Fisher Dome, 13. Loco Creek, 14. Cripple Mountains, 15. North Fork pluton, 16. Aniak Lake, 17. Gemuk Mountain, 18. Crooked Mountains, 19. Canyon Creek, 20. west Canyon Creek, 21. Eek River VOLCANIC FIELDS Nukluk volcanic field (Eocene)-Alkali rhyolite and basalt Eek volcanic field (Paleocene and Late Cretaceous?)-Calc-alkaline andesite and dacite flows; subordinate rhyolite domes Kipchuk volcanic field (Late Cretaceous)-Calcalkaline andesite flows and tuff; subordinate rhyolite domes Tulip volcanic field (early Tertiary and Late Cretaceous)-Dacite and andesite flows and pyroclastic rocks; abundant rhyolite domes Swift Creek volcanic field (early Tertiary? and Late Cretaceous)-Calc-alkaline rhyolite ash-flow tuffs and andesite breccias Figure 1. Continued. SEDIMENTARY ROCKS Kuskokwim Group (Late Cretaceous)-Shallow- to deep-marine sandstone, shale, and minor conglomerate TERRANES Nyac terrane (Early Cretaceous? and Jurassic)- Calc-alkaline andesite and volcaniclastic sandstone Goodnews terrane (Mesozoic and Paleozoic)- Greenschist-facies metabasalt, argillite, and chert Togiak terrane (Early Cretaceous to Late Triassic)- Sandstone, argillite, chert, and minor basalt Tikchik terrane (Mesozoic and Paleozoic)-Chert, argillite, basalt, marble, and melange Kilbuck terrane (Early Proterozoic)-Orthogneiss and pelitic schist Contact Fault, dotted where inferred Localities indicated by letter: N Nyac RC Rainy Creek GL Gold Lake KL Kisarilik Lake AL Aniak Lake Summary of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 o by 3° Quadrangles

Gold Placers The Bethel region has produced over 250,000 oz of placer gold (Bundtzen and others, I989), mostly from streams draining areas underlain by granitoid plutons. Placer gold remains the commodity with the greatest potential for undiscovered deposits. Most placer occurrences that have had production were discovered before I930 (Hoare and Cobb, I977). Known occurrences and mines are discussed for each favorable tract below. Favorable Tracts Five geologically or geochemically distinct tracts are favorable for undiscovered gold placers (fig. 2). Known gold placers, and most stream sediment or heavy-mineralconcentrate samples that contain gold, are in streams that drain areas underlain by granitoid plutons and their hornfelsed wall rocks. Gold-bearing quartz veins associated with the plutons (discussed below) appear to be a significant source of gold in the favorable tracts. A summary of the criteria for the tracts is shown in table I. Tract ·PI (fig. 2, table I) consists of two geologically similar areas, separated by thick unconsolidated deposits. The tract is underlain by Jurassic to Early Cretaceous island arc volcanic and volcaniclastic rocks of the Nyac terrane, which are cut by I02- to I20-Ma granitoid plutons (Box and others, in press). Gold-bearing quartz veins locally are present in the tract (see later discussion). Placers in the tract, mostly along the Tuluksak River and its tributaries (fig. 2), have yielded at least 250,000 oz of gold through I989 (Bundtzen and others, I989). Twenty-three percent of heavy-mineral-concentrate samples in the tract contain gold; such samples are mostly in and near the Tuluksak River or its tributaries, or in the southern subarea near Columbia Creek (fig. 2). Tract PI is considered to have high potential for undiscovered gold placers. Tract P2 (fig. 2, table I) consists of two subareas, separated by unconsolidated Quaternary deposits. The tract has several placers with recorded production, including Marvel, Eureka, Cripple, and Rainy Creeks, and the Salmon River (Hoare and Cobb, I977). The tract is underlain by sandstones and shales of the Upper Cretaceous Kuskokwim Group; Late Cretaceous and early Tertiary felsic granitic plutonic and hypabyssal felsic rocks intrude the Kuskokwim Group at Mt. Plummer, Fisher Dome, and the Cripple Mountains (figs. I, 2). Gold-bearing quartz veins are present cutting the Mt. Plummer pluton and in quartz-stibnite veins cutting the biotite granite porphyry of Fisher Dome (figs.I, 2) (Frost, I990). Gold content in stream sediment and heavy-mineral-concentrate samples (Bradley and Frost, in press) correlates with presence or absence of plutons in well-integrated drainages. Tract P2 Geologic Studies in Alaska by the U.S. Geological Survey, 1991 is considered to have high potential for undiscovered gold placers. Tract P3 (fig. 2, table I) is similar to tract P2 except that it contains volcanic fields of Late Cretaceous to early Tertiary age in addition to sedimentary rocks of the Kuskokwim Group and felsic to intermediate plutonic rocks (Box and others, in press). Quartz-vein samples (see following section) contain as much as 0.3 ppm Au (Frost, I990). Twenty percent of heavy-mineral-concentrate samples contain at least 0.5 ppm Au (Bradley and Frost, in press). Owing to the lack of generally well-integrated drainages, and to the presence of extensive glacial outwash and moraine deposits in much of the area, we infer that tract P3 has a low potential for undiscovered gold placers. Tract P4 (fig. 2, table I) contains the placer mine at Canyon Creek and several prospects. The tract is underlain by Upper Triassic to Lower Cretaceous marine volcaniclastic rocks of the Togiak terrane (Box and others, in press). Late Cretaceous granodiorite and granite plutons are present, as well as subordinate augite-biotite quartz diorite through augite gabbro intrusions (Frost and others, I988). Quartz veins associated with pyritiferous rhyolite dike complexes contain variable amounts of Ag, Au, Pb, and Zn (Frost, I990). Gold is present in stream-sediment and heavy-mineral-concentrate samples, especially in unglaciated valleys near the Canyon Creek pluton (figs. I, 2), and in 25 percent of heavy-mineral-concentrate samples in the northern half of the tract (Bradley and Frost, in press). Pleistocene mountain and valley glaciation modified most valleys in the high country underlain by most of tract P4. Largely owing to the effects of glaciation, the potential for placers is regarded as low. Tract P5 (fig. 2, table I) is underlain by sandstones, slates, and greenschist-facies metachert and phyllite of Paleozoic to Mesozoic age of the Goodnews terrane, unconformably overlain by basal conglomerates and shales of the Upper Cretaceous Kuskokwim Group (Box and others, in press). Gold is present in several heavymineral-concentrate samples in the area, as are scattered anomalous concentrations of Ag, Hg, Sn, Pb, and Zn (Bradley and Frost, in press). The narrow canyons and low amount of alluvium in most canyons suggest a low probability of placer deposits in tract P5. Silica-Carbonate Mercury Mercury deposits fitting the silica-carbonate model of Cox and Singer (1986) are localized in altered serpentinites and sedimentary rocks in and near major faults in accreted terranes. Cinnabar and native mercury, along with minor base-metal sulfides, are common ore mineral assemblages which are present along with quartz and carbonate. No serpentinite-hosted mercury deposits fitting the silica-carbonate model of Cox and Singer

20 KILOMETERS EXPLANATION Favorable tracts for gold placer deposits Figure 2. Simplified map showing favorable tracts for gold placer deposits. Favorable tracts are as indicated in table 1 and discussed in text. Tracts P1 and P2 each have northern (N) and southern (5) subarea; for each tract, it is probable that geologic units that define subareas are continuous under intervening Quaternary cover. Geologic units as in figure 1. Summary of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 o by 3° Quadrangles

w 0"1 C') 0 Q (JQ Q. ;;-

Di' "'

Ai C"'

:;. 1-n C') 0 Q (JQ

Table 1. Summary of favorable tracts for placer gold [Tracts and localities shown on figures I and 2. See text for discussion] Tract Area Host rock types Terrane Criteria Mines and prospects Magmatism Composition Age (Mal References PI Nyac volcanic-arc Nyac I ,2,3,5,6,7,8,9 N yac district Nyac p. hbl-bio gr, grd 102-120 (K-Ar) 1 ,2,3,4,5,6,7 andesites, Sawpit p. bio gr Early Cretaceous(?) volcaniclastic Bonanza Ck. p. bio-hbl qtz mzd, grd Early Cretaceous(?) sandstones Fox Creek p. gr, gab, dio Early Cretaceous(?) Slate Creek p. bio-hbl grd, qtz mzd Early Cretaceous(?) Dry Creek p. rhy. porphyrry Jurassic to Tertiary Nukluk v.f. basalt-rhy. 55 (Ar-Ar) Columbia Ck. area Columbia Ck. p. hbl-bio grd, qtz mzd 115 (K-Ar bio) Kasigluk p. cpx gab, dio Early Cretaceous(?) Shining Dome p. cpx-bio qtz dio. Late Cret. to early Tert. P2 Kuskokwim sandstone Kuskokwim Group 1 ,2,4,5 ,6, 7,8 Marvel, Eureka, Mt. Plummer p. bio-hbl grd, qtz mzd 65-67 (K-Ar) 1 ,2,3,4,5,6,7 Mtns. and shale Cripple, Fisher, Marvel Creek p. bio gr porphyry Late Cret. to early Tert. Salmon River Rainy Cks., Fisher Dome p. bio gr Late Cret. to early Tert. Salmon R. Loco Creek p. bio gr porphyry Late Cret. to early Tert. Cripple Mtns p. bio grd, gr 62 (K-Ar) W. Canyon Ck. p. bio grd, mzd Late Cretaceous(?) P3 Tulipvolcanic flows Tulip-Kipchuk v.f. 1 ,2,4,5,6,7 ,8 Anvil Ck. North Fork p. cpx qtz dio, bio grd, 64 (Ar-Ar) 1 ,2,3,4,5,6,7 Kipchuk and domes Aniak Lake p. bio gr 61 (Ar-Ar) sandstone Kuskokwim Group Crooked Mtns p. hb-bio grd, bio gr 70 (Ar-Ar) and shale Kipchuk v.f. and, dac 70 (Ar-Ar) Tulip v.f. and, rhy Late Cret. to early Tert. P4 Togiak marine clastic Togiak 1 ,2,5,7 ,8,9 Canyon, Kapon Cks. Canyon Creek p. cpx-bio qtz dio, grd 70 (Ar-Ar) 1 ,2,3,4,5,6,7 rocks, chert, Crooked Mtns p. hb-bio grd, bio gr 70 (Ar-Ar) and basalt Swift Creek v .f. and, rhy 74 (Ar-Ar) P5 Kuskokwim Mtns. sandstone Kuskokwim Group 7,8 dikes intermediate Late Cret. to early Tert. 3,4 Kisarilik River and shale Criteria: I, known deposits; 2, known prospects; 3, Early Cretaceous granites; 4, Late Cretaceous granites; 5, hydrothermally altered rock; 6, gravity low; 7, gold in pan-concentrate samples; 8, gold in stream-sediment samples; 9, gold in quartz veins in bedrock. Abbreviations used: bio, biotite; hbl, hornblende; cpx, clinopyroxene; qtz, quartz; gr, granite; grd, granodiorite; mzd, monzodiorite; dio, diorite; gab, gabbro; and, andesite; dac, dacite; rhy, rhyolite; p., pluton; v.f., volcanic field. References: l, Frost (1990); 2, Box and others (in press); 3, Frost and others (1988); 4, Bradley and Frost (1992); 5, R. Tripp, U.S. Geological Survey, written commun. (1988-89); 6, Phillips and Morin (1992); 7, Hoare and Cobb (1977).

(1986) have been identified in southwestern Alaska to date. In the Bethel map area, a favorable tract for silicacarbonate Hg is present along the northern part of the Sawpit and Golden Gate faults (S-C on fig. 3). Present along the presumed trace of the faults are scattered unvegetated, reddish-orange- to black-weathering areas as large as 1 km2. The areas are composed of frostheaved blocks of silica-carbonate rock, gabbro, serpentinite, and peridotite, apparently in a scaly serpentinite-melange matrix. In silica-carbonate rock, protolith assemblages were replaced by serpentine-group minerals, followed by partial to complete replacement by quartz, calcite, and (or) iron-rich carbonate, followed by brecciation and quartz veining (Frost, 1990). Altered silica-carbonate rock samples from the tract (S-C, fig. 3) contain as much as 1 ppm Ag, 450 ppm As, 2,000 ppm Cr, 1,500 ppm Ni, 1,000 ppm Zn, and 41 ppm Sb (Frost, 1990; Frost and others, 1992a). Mercury content in these samples averages 4 ppm, but the Hg content of several samples exceeds the detection limit of 36 ppm (Frost and others, 1992a). The tract has moderate potential for silica-carbonate mercury deposits. Epithermal Mercury Southwestern Alaska epithermal mercury deposits are believed to form in shallow ( km) hydrothermal systems (Miller and others, 1989; Frost, 1990; Goldfarb and others, 1990) related to Late Cretaceous to early Tertiary magmatism (Gray and others, 1992). Such deposits contain cinnabar, pyrite, and native mercury, with or without arsenic and antimony sulfides, in quartz veins and disseminations in altered host rocks. The alteration and sulfide mineral assemblages in southwestern Alaska epithermal mercury deposits are similar to those of the silica-carbonate mercury deposit model of Cox and Singer (1986), although serpentinites are not associated with Alaskan examples. The Red Devil deposit in the Sleetmute quadrangle (fig. 1) is the largest of such deposits in Alaska, having shipped about 32,000 flasks of mercury (Miller and others, 1989). No other deposits in southwestern Alaska have shipped more than a few hundred flasks (Sainsbury and MacKevett, 1965). Criteria for the delineation of favorable tracts are summarized in table 2. Known Occurrences The Rainy Creek prospect (fig. 3), hosted in shales and sandstones of the Upper Cretaceous Kuskokwim Group, is the only mercury lode in the map area (Rutledge, 1948; Sainsbury and MacKevett, 1965; Berg and Cobb, 1967; Frost, 1990; Goldfarb and others, 1990). There has been no production, although 2,000 lb of cinnabar were concentrated from the gold placer downstream (Sainsbury and MacKevett, 1.965). Two distinct textural types of mercury-bearing quartz veins are present: dilational quartz veins as wide as 10 em that contain minor cinnabar and stibnite, and anastomosing realgar- and orpiment-rich veins (as much as 75 percent sulfides) and stockworks with minor cinnabar (Frost, 1990). Gold is present at levels as high as 0.05 ppm in the As-Sb-dominated veins (Frost, 1990; Gray and others, 1990). Favorable Tracts Tract ep-Hg1 (table 2, fig. 3) is defined by Late Cretaceous to early Tertiary felsic domes and flows and granitoid plutons, which unconformably overlie and intrude, respectively, marine sedimentary rocks of the Kuskokwim Group and Jurassic andesitic arc volcanic rocks of the Nyac terrane (Box and others, in press). Samples of silicic to intermediate volcanic rocks, sedimentary rocks of the Kuskokwim Group, and quartz veins contain as much as 2 ppm Hg (Frost and others, 1992a). Stream sediment samples contain as much as 7 ppm Hg (Bradley and Frost, in press). Tract ep-Hg1 is considered to have moderate potential for epithermal mercury deposits. Tract ep-Hg2 (table 2, fig. 3) has a low potential for undiscovered epithermal mercury deposits. The tract consists of greenschist- to blueschist-facies rocks of the Goodnews terrane. Lithologies include variably deformed metabasalt, metachert, marble, and phyllitic rocks of Paleozoic and Mesozoic age. Unconformably overlying the rocks of the Goodnews terrane are basal conglomerates of the Upper Cretaceous Kuskokwim Group, which grade upward in a few hundred meters to deep-water, fine-grained sandstones and shales. Altered andesite and dacite dikes are common in the Kuskokwim Group in the tract; in most dikes, plagioclase and mafic minerals are replaced by chlorite, calcite, and quartz. There is little in the lithologies or rock geochemical data (Box and others, in press; Frost and others, 1992a) in the tract to suggest the presence of epithermal mercury deposits, but cinnabar is present in some heavy-mineralconcentrate samples (R.B. Tripp, U.S. Geological Survey, written commun.), and mercury was detected at levels as high as 1.5 ppm in stream-sediment samples (Bradley and Frost, in press). Tract ep-Hg3 (table 2, fig. 3) has high potential for epithermal Hg deposits. It contains the occurrence at Rainy Creek (Sainsbury and MacKevett, 1965; Frost, 1990) and has many of the structural, stratigraphic, and volcano-plutonic associations of other Alaskan Hg deposits, including Red Devil in the Sleetmute quadrangle (fig. 1) and Cinnabar Creek in the Taylor Mountains quadrangle (Sainsbury and MacKevett, 1965; Miller and others, 1989; Frost, 1990; Goldfarb and others, 1990). Summary of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 o by 3° Quadrangles

20 KILOMETERS EXPLANATION Favorable tracts for silica-carbonate mercury deposits . Favorable tracts for epithermal mercury deposits Favorable tracts for simple antimony deposits Geologic Studies in Alaska by the U.S. Geological Survey, 1991

The southwestern part of tract ep-Hg3 consists of argillaceous melange with blocks of limestone, basalt, and chert of Paleozoic to Mesozoic age, overlain by Lower Cretaceous turbidites of the Goodnews terrane (fig. 3) (Box and others, in press). Upper Jurassic to Lower Cretaceous tuffaceous cherts and argillites and Lower Cretaceous volcaniclastic rocks of the Togiak terrane make up the eastern part of the tract (Box and others, in press). All of these rocks are unconformably overlain by the Upper Cretaceous Kuskokwim Group, which in this area is composed mostly of conglomerates, sandstones, and shales of turbidite origin (Box and others, in press). Late Cretaceous to early Tertiary felsic to intermediate plutons and batholiths intrude the older terranes and the Kuskokwim Group (Frost and others, 1988; Frost, 1990; Box and others, in press) in tract ep-Hg3 (fig. 3). Late Cretaceous to early Tertiary andesite and dacite flows and tuffs, as well as rhyolite domes, flows, and ash flows (Box and others, in press), locally overlie all older rocks. Tract ep-Hg3 does not extend as far west as do the westernmost outcrops of the Kuskokwim Group or as far east as do the easternmost parts of the Togiak terrane; the geochemical anomalies, which, in part, define the tract, appear to be associated with the magmatichydrothermal systems of the plutonic and volcanic rocks rather than the underlying rocks. Anomalous concentrations of Hg, As, Sb, and, to a lesser extent, Au, Ag, Pb, and Zn are present in samples from tract ep-Hg3 (Bradley and Frost, 1992; Frost and others, 1992a). Liesegang-banded rhyolite samples from rhyolite domes in the Tulip volcanic field in tract ep-Hg3 (fig. 3) contain as much as 15 ppm Hg, 140 ppm As, and 0.30 ppm Au (Frost, 1990). Mercury content is greater than 5 ppm in 17 stream-sediment samples (6 percent of samples) and is greater than 36 ppm in 9 samples (Bradley and Frost, in press). Cinnabar is present in 35 percent of heavy-mineral-concentrate samples (R.B. Tripp, U.S. Geological Survey, written commun.). Stream-sediment samples contain as much as 690 ppm As (Bradley and Frost, in press); corresponding heavy-mineralconcentrate samples contain arsenopyrite. Streamsediment samples contain as much as 20 ppm Sb (Bradley and Frost, in press); most of the Sb is in stibnite. Polymetallic quartz veins (see later section) are also present in the tract. Tract ep-Hg3 has high potential for epithermal Hg deposits. Tract ep-Hg4 straddles the Togiak fault (fig. 3, table 2). Mercury is present in some stream-sediment samples,

Figure 3. Simplified map showing favorable tracts for silicacarbonate mercury, epithermal mercury deposits, and simple antimony deposits. Criteria for epithermal mercury deposit tracts summarized in table 2. Fisher Dome and Rainy Creek localities shown by FD and RC, respectively. Geologic units as in figure 1. and cinnabar is present in heavy-mineral-concentrate samples (Bradley and Frost, in press). Samples of brecciated argillite along the trace of the Togiak fault, of argillite east of the fault, and of a diabase contain 0.25 to 4.0 ppm Hg and as much as 720 ppm As and 18 ppm Sb (Frost and others, 1992a). The tract has low potential for epithermal mercury deposits. Simple Antimony Veins Simple antimony veins (Cox and Singer model 27d) are present in orogenic areas and are characterized by quartz and stibnite veins or disseminations associated with pyrite, base-metal sulfides, cinnabar, silver, gold, or scheelite. A single occurrence that fits the simple Sb model is present in the Bethel map area, where quartzstibnite vein cuts the biotite quartz porphyry of Fisher Dome (figs. 1, 3). Massive and bladed stibnite as long as 7 em comprises 5 to 50 percent of the vein; the remainder is composed of vuggy quartz. Samples of the vein contain as much as 1.0 ppm Au, 7.0 ppm Ag, and 1,500 ppm As; Hg content is less than 5 ppm (Frost, 1990). Favorable Tracts Five tracts are delineated as favorable for simple Sb veins in the Bethel region (Sb on fig. 3). Indicators of mineralization in the tracts include presence of felsic plutons, anomalous contents of certain elements in heavy-mineral-concentrate samples (15,000 ppm Sb, >5,000 ppm As, 700 ppm Ag, and 10,000 ppm W in ·concentrate samples; Bradley and Frost, 1992), and gold, stibnite, arsenopyrite, and scheelite in heavy-mineralconcentrate samples (R.B. Tripp, U.S. Geological Survey, written commun., 1989). Quartz vein and host rock samples from a biotite granite porphyry at Loco Creek ( 13 on fig. 1) contain as much as 90 ppm Sb and 200 ppm As (Frost and others, 1992a). The probability of occurrence of simple Sb deposits is low. Polymetallic Veins In the Bethel map area, two types of polymetallic veins are present: a gold-silver quartz vein type without elevated base metals, and a gold-silver base-metal type. The base-metal-rich veins are referred to as the Gold Lake type after the prospect discovered in 1987 near . Gold Lake (figs. 1, 4) (Frost, 1990); known occurrences are present in rocks of pre-Late Cretaceous terranes that are overlain by sedimentary rocks of the Kuskokwim Group. Gold Lake type vein systems may crop out over areas as large as several square kilometers. The type lacking elevated base metals crops out over areas less Summary of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 ° by 3o Quadrangles

Q C')

S'" CJQ ;:;· til eQ.

:r

Dr (II

C"' rD c:

C') rD S'" OQ [ til c rD -:t loC

Table 2. Favorable tracts for epithermal mercury deposits [Tracts and localities shown on figures I and 3. See text for discussion] Tract Area Host rock types Terrane Criteria Mines and prospects Magmatism Composition Age (Ma) References ep-Hg 1 Western rhyolite to dacite Eek and Nukluk v.f. 2,3,4,5 none Eek v.f. Andesite, dacite 59 (K-Ar) 1 ,2,3,4,5,6 volcanic domes and flows Nukluk v.f. alkali rhy 55 (Ar-Ar) shale and sandstone Kuskokwim Group andesitic arc Nyac Columbia Ck. p. hbl-bio grd, qtz mzd 115 (K-Ar bio) volcanic rocks Kasigluk R. p. cpx gab, dio Early Cretaceous(?) ep-Hg2 Kisarilik River metabasalt, metachert Goodnews 3,4,5 none dikes and, dac, rhy Late Cret. to early Tert. 1,2,4,5,6 marble, phyllite conglomerate and Kuskokwim Group sandstone ep-Hg 3 Central mercury argillaceous melange Goodnews 1 ,2,3,4,5 Rainy Creek Mt. Plummer p. bio-hbl grd 65-67 (K-Ar) 1 ,2,3,4,5,6,7 belt sandstone and shale Kuskokwim Group Marvel Creek p. bio gr. porphyry Late Cret. to early Tert. chert, argillite, and Togiak Fisher Dome p. bio gr Late Cret. to early Tert. sandstone North Fork p. cpx dio, bio grd 64 (Ar-Ar) Aniak Lake p. bio gr 61 (Ar-Ar) Crooked Mtns. p. hb-bio grd, bio gr 70 (Ar-Ar) Cripple Mtns. p. bio grd 62 (K-Ar) Canyon Creek p. cpx-bio qtz dio, grd 70 (Ar-Ar) W. Canyon Ck. p. grd, mzd Late Cretaceous(?) Kipchuk v.f. and, dac 70 (Ar-Ar) Tulip v.f. and, rhy Late Cret. to early Tert. dikes and, dac, rhy Late Cret. to early Tert. ep-Hg4 Togiak fault Togiak fault zone Togiak-Tikchik 3,4,5 none dikes hbl dio Late Cretaceous(?) 2,4,5 boundary Criteria: 1, shallow plutons; 2, volcanic rocks; 3, cinnabar in concentrate samples; 4, mercury detected in stream sediments; 5, mercury detected in bedrock. Abbreviations used: bio, biotite; hbl, hornblende; cpx, clinopyroxene; qtz, quartz; gr, granite; grd, granodiorite; mzd, monzodiorite; dio, diorite; gab, gabbro; and, andesite; dac, dacite; rhy, rhyolite; p., pluton; v.f., volcanic field. References: 1, Frost (1990); 2, Box and others (in press); 3, Frost and others (1988); 4, Bradley and Frost (1992); 5, R. Tripp, U.S. Geological Survey, written commun. (1988-89); 6, Phillips and Morin (1992); 7, Hoare and Cobb (1977).

than a few tens of square meters and are referred to as the simple gold subtype. Simple gold veins may be present in any bedrock type, including the Kuskokwim Group (Frost, 1990). The absolute ages of the two types are unknown, but both are spatially associated with Late Cretaceous to early Tertiary magmatic rocks. Although the favorable tracts for the polymetallic vein types are delineated separately on figure 4, a continuum between the two types is possible. The criteria for delineation of favorable tracts for polymetallic veins in the Bethel area are summarized in table 3. Known Occurrences of Gold Lake-Type Polymetallic Veins The veins northeast of Gold Lake and north of Kisarilik Lake (fig. 4) were first described by Frost (1990). Oxidized pyrite-bearing rhyolite dikes and sulfide-bearing quartz veins cut argillite, sandstone, and volcanic and volcaniclastic rocks of the Togiak terrane at the Gold Lake and Kisarilik Lake occurrences. Most of the dike and vein networks are exposed over areas of less than a few tens of square meters, but some larger zones extend erratically over several square kilometers. The dikes are erosionally resistant and are as wide as 3 m. The dike and vein networks have anomalous concentrations of various combinations of Ag, As, Au, Cu, Hg, Mo, Pb, Sb, and W (Frost, 1990; Frost and others, 1992a). The highest concentrations of Au, Ag, and As are in quartz veins associated with rhyolite dikes (Frost, 1990). The altered rhyolite dikes contain partially resorbed quartz phenocrysts and pyrite cubes as large as 3 mm. The groundmass of the dikes is replaced by sericite or illite and quartz. Relict plagioclase phenocrysts are replaced by albite and sericite. Altered rhyolite samples contain as much as 7.0 ppm Ag, 3,000 ppm As, 0.23 ppm Au, 100 ppm Mo, and 100 ppm Sb (Frost, 1990). Several generations of quartz veins with cores containing as much as 20 per cent galena cut all rock types at the Gold Lake prospect. Argillite and sandstone wall rocks and rhyolite dikes are incipiently to extensively altered to quartz, sericite, calcite, and iron-oxide minerals near quartz veins. Quartz, quartz+calcite, quartz+(chlorite), quartz+iron oxides, quartz+pyrite, and quartz+galena are common vein assemblages. Most veins are less than 2 em wide, although some are as wide as 12 em. Quartz forms euhedral prisms as long as 1 em. Two generations of carbonate are present in the veins; both postdate crystallization of most of the quartz. The earliest carbonate is a brown, fine-grained aggregate that is intergrown with fine-grained quartz and galena. A later generation of coarse-grained, clear calcite fills vugs. Quartz veins contain as much as 7 ppm Ag, 7,000 ppm As, 2.0 ppm Au, 30 ppm Bi, 300 ppm Pb, 150 ppm Sb, and 70 ppm W (Frost, 1990; Frost and others, 1992a). Hydrothermal breccias composed of pale greenish-gray clays, quartz clasts, and oxide minerals have gold contents as high as 2.0 ppm (Frost, 1990). Favorable Tracts for Gold Lake-Type Polymeta_llic Veins Dike and vein networks similar to those at Gold and Kisarilik Lakes are present throughout the southeastern comer of the Bethel map area in rocks of the Togiak and Tikchik terranes (tract GL1, fig. 4), but most have surface expressions of less than a few tens of square meters, and most do not have anomalous contents of all the elements Ag, As, Au, Bi, Pb, Sb, and W (Frost, 1990). Contents of As and Sb are as high as 180 ppm and 28 ppm, respectively, in stream-sediment samples in drainages from the Togiak and Tikchik terranes (Bradley and Frost, in press). Heavy-mineral-concentrate samples from the same areas are as high as 5,000 ppm As, 15,000 ppm Sb, 200 ppm Pb, 5,000 ppm W, and 2,000 ppm Zn (Bradley and Frost, in press). These data indicate that additional Gold Lake vein occurrences are present in the tract, although the size of most occurrences probably are small. Anomalous As, Au, Ag, Hg, and Pb ·contents in rock and stream-sediment samples in areas underlain by the Togiak and Tikchik terranes in the Goodnews, Taylor Mountains, and Dillingham quadrangles (Coonrad and others, 1978; Hessin and others, 1978a, b; Kilburn and others, in press) suggest that these areas also are favorable for Gold Lake-type mineralization. The geochemical and mineralogical data suggest the possibility of occurrence of polymetallic veins in tract GL2 (fig. 4), although no evidence was observed in the field. The tract is underlain by arkosic sandstones and slates, and greenschist-facies metachert and phyllite of Paleozoic to Mesozoic age of the Goodnews terrane, unconformably overlain by basal conglomerates and shales of the Upper Cretaceous Kuskokwim Group. Intermediate composition dikes are common and locally constitute as much as 10 percent of exposure in outcrops of Kuskokwim Group rocks in the tract. Gold is present in several heavy-mineral concentrate samples in the area, as are scattered anomalous concentrations of Ag, Hg, Sn, Pb, and Zn (Bradley and Frost, in press). Stibnite, cinnabar, cassiterite, pyrite, barite, arsenopyrite, and scheelite are present in heavy-mineral-concentrate samples (R. Tripp, U.S. Geological Survey, written commun., 1988, 1989). Known Occurrences of Simple Gold Polymetallic Veins The simple gold polymetallic veins in the Bethel mar area are gold-bearing quartz veins that lack base-metal anoma· lies but may contain As and Sb (Frost, 1990). No example! crop out over a strike length of more than a few meters; veins are less than 35 em wide. Hand-picked specimens of pyriteand iron oxide-bearing quartz and calcite veins from prospect pits in the Nyac terrane (fig. 4) contain as much as 20 ppm Au (Frost, 1990). Gold-bearing quartz veins from the Nyac Summary of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 ° by 3° Quadrangles

20 KILOMETERS EXPLANATION Favorable tracts for simple gold polymetallic vein deposits Favorable tracts for Gold Lake-type polymetallic vein deposits Figure4. Simplified mapshowingfavorabletracts foroccurrenceof polymetallic veins. Boundaries for both Gold Lake subtype and simple subtype are indicated. Criteria for tracts are summarized in table 3. Gold Lake and Kisarilik Lake prospect areas are shown by subareas GLand KL, respectively. Other localities with gold content of at least 0.05 ppm in bedrock samples marked with a dot (from Frost, 1990). Geologic units as in figure 1. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Table 3. Favorable tracts for polymetallic veins \II [Tracts and localities shown on figures l and 4. See text for discussion]

Tract Area Host rock types Vein types Terrane Criteria Mines and prospects Magmatism Composition Age (Ma) References ;.

Gold Lake polymetallic veins s· e. GLI Gold Lake argillite, sandstone, sulfide-bearing Togiak, 1,2,3,4 Gold Lake Crooked Mtns. p. hb-bio grd, bio gr 70 (Ar-Ar) 1,2,4,5,6,7

intermediate qtz veins Tikchik Kisarilik Canyon Ck p. cpx-bio qtz dio, grd 70 (Ar-Ar)

volcanic rocks silicified Lake Gemuk Mtn p. cpx-bio qtz gab, dio Late Cretaceous(?) <ll 5; rhyolite dikes hbl dio, qtz dio Late Cretaceous ,., hydrothermal dikes rhy Late Cretaceous(?)

<ll breccias <ll

GL2 Goodnews/ sandstone, Goodnews 3,4 dikes and Late Cretaceous(?) 4,5 <ll none none

Kuskokwim metachert, a phyllite conglomerate, Kuskokwim sandstone Group

Simple gold polymetaUic veins ;.

AI SGl Nyac volcanic arc quartz veins Nyac 1,2,3,4 several Nyac p. hbl-bio gr, grd 120-101 (K-Ar) I ,2,3,4,5,6 :I Q. andesites and unnamed Sawpit p. bio gr Early Cretaceous \II clastic rocks Bonanza Creek p. bio-hbl qtz mzd, grd Early Cretaceous c ;. Fox Creek p. gr, gab,dio Early Cretaceous

Slate Creek p. bio-hbl grd, qtz mzd Early Cretaceous(?) AI Columbia Creek p. hbl-bio grd, qtz mzd 115 (K-Ar bio) :I Little Kasigluk p. cpx gab, dio Early Cretaceous(?)

Ai

SG2 Kuskokwim sandstone and shale quartz veins Kuskokwim 1,2,3,4 several Mt. Plummer p. bio-hbl grd 67-65 (K-Ar) 1 ,2,3 ,4,5 ,6, 7 Group unnamed Marvel Creek p. bio gr porphyry Late Cret. to early Tert. Fisher Dome p. bio gr Late Cret. to early Tert.

Cripple Mtns. p. hbl-bio grd, bio gr 62 (K-Ar) North Fork p. cpx qtz dio, bio grd 64 (Ar-Ar) <ll <ll iii' W. Canyon Ck. p. bio grd, mzd Late Cretaceous(?) :I Aniak Lake p. bio gr 61 (Ar-Ar)

iii' Gemuk Mtn. p. cpx-bio qtz gab, dio Late Cretaceous(?) <ll Kipchuk v.f. and, dac 60-69 (K-Ar) s· :I Tulip v.f. and, rhy Late Cret. to early Tert. C" SG3 Togiak basalt, chert, quartz veins Togiak 1,2,3,4 Crooked Mtns. p. hb-bio grd, bio gr 70 (Ar-Ar) 3,4,5

none

argillite Canyon Ck. p. cpx-bio qtz dio, grd 70 (Ar-Ar) IJ AI Criteria: l, presence of gold-bearing quartz veins; 2, felsic plutonic rocks; 3, gold in nonmagnetic heavy-mineral-concentrate samples; 4, gold detected in stream-sediment samples. eAI Abbreviations used: bio, biotite; hbl, hornblende; cpx, clinopyroxene; qtz, quartz; gr, granite; grd, granodiorite; mzd, monzodiorite; dio, diorite; gab, gabbro; and, andesite; dac, dacite; rhy, rhyolite; p., pluton; v.f., volcanic field. :I OQ References: I, Frost (1990); 2, Box and others (in press); 3, Frost and others (1988); 4, Bradley and Frost (1992); 5, R. Tripp, U.S. Geological Survey, written commun. (1988-89); 6, Phillips and Morin (1992); 7, Frost and others (l992a). ;;- <ll

w

terrane contain 0.6 to 2.3 ppm Hg; one sample contains 500 ppm As and 10 ppm Ag (Frost, 1990). Simple gold veins containing as much as 0.35 ppm Au also are present east of the Sawpit fault, cutting Late Cretaceous to early Tertiary granitoid plutons and their wall rocks, which may include sedimentary rocks of the Kuskokwim Group or low-grade metasedimentary rocks of the Togiak terrane (fig. 4) (Frost, 1990; Frost and others, 1992a). Euhedral vuggy quartz, calcite, chlorite, and, locally, pyrite are gangue minerals in the simple gold veins. Several generations of quartz veins may be present at any locality; all may contain gold. The veins are less than 30 em thick; most are anastomosing or en echelon and are exposed over strike lengths of less than 5 m. The spatial association between veins and plutons suggests that the plutons provided at least a heat source to circulate hydrothermal fluids that deposited the gold. Favorable Tract for Simple Polymetallic Gold Veins The favorable tract for simple gold veins includes all of the Nyac terrane, the Kuskokwim Group where it is cut by granitoid plutons, and part of the Togiak terrane (fig. 4, table 3). The Nyac terrane is included in the favorable tract due to the widespread presence of plutonic and volcanic rocks (Box and others, in press), gold in heavy-mineral-concentrate samples (Bradley and Frost, in press), and gold in quartz veins (Frost, 1990; Frost and others, 1992a). Kuskokwim rocks where there is field or geophysical evidence for plutons, gold in quartz veins (Frost, 1990), and gold in heavy-mineral-concentrate and stream-sediment samples, are included in the favorable tract (Bradley and Frost, in press). Areas in the Togiak terrane that host plutons are also favorable for simple gold quartz veins. There is a high probability of undiscovered simple gold vein occurrences in the favorable tract, although the small size of the known occurrences suggests that any undiscovered deposits are of similarly small size. Base- and Precious-Metal Anomalies Associated with Tourmaline Replacement Known Occurrences Partial to complete replacement of subaerial volcanic and lacustrine volcaniclastic rocks by quartz-tourmaline-iron oxide assemblages is present at two localities in the northeastern part of the map area (fig. 5). Tourmalinized rocks contain as much as 70 ppm Ag, 2,000 ppm As, 0.1 ppm Au, 300 ppm Bi, 30 ppm Cd, 700 ppm Cu, >36 ppm Hg, 11,000 ppm Pb, 1,500 ppm Sb, 100 ppm Sn, 30 ppm W, and >10,000 ppm Zn, although no individual sample is anomalous in all elements (Frost, 1990; Frost and others, 1992a). Geologic Studies in Alaska by the U.S. Geological Survey, 1991 The replacement occurrences appear to be broadly stratiform, having gradational contacts with footwall rocks and with laterally equivalent rocks. In partially replaced rocks, relict porphyritic textures are present; plagioclase and mafic phenocrysts are replaced by tourmaline and iron oxides, and the groundmass is replaced by granular aggregates of quartz+tourmaline+oxides. Modally and (or) size-graded layering 1 to 10 mm thick is common. The original attitude of the layers with respect to layering in the host rocks is not known, but the layers are tourmaline rich at their sharp "lower" contacts and grade "up" to finer grained quartz-rich rock. Layering is planar but locally may be convoluted. Orbicular patches as large as 15 em across are common and are defined by alternating quartz- and tourmaline-rich shells texturally similar to the layered rocks. Iron oxide- and quartz-cemented tourmaline-quartz breccias are present locally. Base- and precious-metal-bearing phases include, but are not limited to, iron oxides, pyrite, arsenopyrite, sphalerite, and galena. The tourmaline replacement may represent parts of hydrothermal systems related to emplacement and cooling of plutons at depth or to the volcanic host rocks (Frost, 1990). Mineralogically similar Ag-Sn-B-enriched mineralized systems with anomalous Bi, Zn, and Pb are reported from elsewhere in southwestern Alaska. Most are described as greisens, veins, shear zones, stockworks, or breccia pipes (Bundtzen and Gilbert, 1983; Bundzten and Laird, 1982, 1983, 1990; Miller and others, 1989; Nokleberg and others, 1987) and are associated with intermediate volcanic and (or) hypabyssal intermediate to felsic plutonic rocks (Bundtzen and Laird, 1990). Sulfide-tourmaline-quartz intrusive "greisen" veins in the northernmost part of the Russian Mission 1 o by 3° quadrangle, associated with the Russian Mountain pluton, yield fluid-inclusion closing temperatures of 280°C to 410°C (Bundtzen and Laird, 1990). It is not clear whether or not the tourmaline-quartz replacement zones in the Kipchuk volcanic field are similar to the veins described by Bundtzen and Laird (1990) or instead represent a distinct occurrence type. Permissive tracts The two tourmaline replacement zones are localized in intermediate volcanic rocks and volcaniclastic sedimentary rocks; similar lithologies would be appropriate exploration targets and are outlined as permissive tracts on figure 5. Lack of similar geochemical anomaly patterns in Late Cretaceous or Tertiary volcanic and plutonic rocks elsewhere in the map

Figure 5. Perr:nissiveterranes for tourmaline-quartz replacement zones, tin-tungsten veins, and Cypress massive sulfide deposits. See text for explanation. Geologic units as in figure 1.

20 Kilometers Explanation

Tourmaline-quartz replacement locality Permissive terrane for tourmaline-quartz replacement deposits Permissive terrane for tin-tungsten veins Permissive terrane for Cypress massive sulfide Summary of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 o by 3° Quadrangles

area (Frost and others, 1992a) suggests a low probability · of additional occurrences. Tin-Tungsten Veins Tungsten concentrations as high as 100 ppm are present in quartz veins cutting many plutons or the hornfels surrounding them (Frost and others, 1992a) Quartz veins that contain anomalous W may also contain as much as 100 ppm Sn. Two Sn-W -bearing quartz vein samples, one from the Columbia Creek pluton and one from the Mt. Plummer pluton (1 0 on fig. 1 ), also contain 0.1 to 0.4 ppm Au (Frost, 1990). Some of the Sn-W- . bearing quartz veins, especially those cutting rocks of the Togiak terrane, may be part of the polymetallic vein association. Pyritiferous rhyolite dikes cutting the Aniak Lake pluton (figs. 1, 5) contain as much as 100 ppm Sn, 940 ppm As, and 10 ppm Sb, without detectable ( <20 ppm) W (Bradley and Frost, in press). Cassiterite and scheelite are present in heavy-mineral-concentrate samples from streams draining all the granitoid plutons that cut shales and sandstones of the Kuskokwim Group, as well as in streams draining granitoids that cut arc volcanic and volcaniclastic rocks of the N yac terrane (R. Tripp, U.S. Geological Survey, written commun., 1988, 1989). Tin and W are present at levels exceeding detection limits (2,000 ppm and 20,000 ppm, respectively) in many heavy-mineral-concentrate samples (Bradley and Frost, in press) in a zone coincident with the locus of Late Cretaceous to early Tertiary plutonism and volcanism in the map area. The Sn-, W -, and locally Au-bearing quartz veins may represent polymetallic veins similar to the Gold Lake type. Areas permissive for Sn-W bearing quartz veins are shown on figure 5. Massive Sulfide Cypress massive Cu-Zn sulfide deposits are localized in pillow basalt sections of ophiolites (Cox and Singer, 1986, model 24a). Ophiolites represent tectonically dismembered oceanic crust and overlying sedimentary rocks incorporated in accreted terranes. Associated deposits consist of massive deposits and stockworks of pyrite, chalcopyrite, sphalerite, and (or) pyrrhotite that are commonly associated with deep-water sedimentary rocks including cherts and phyllite. Rock types permissive for the occurrence of Cypress massive sulfide deposits crop out locally in the map area. Foliated greenschist-facies metabasalt is present in the Goodnews terrane (tract C1, fig. 5). Quartz veins that cut the metabasalts contain as much as 500 ppm Cu and average 200 ppm Zn (Frost and others, 1992a). Variably foliated and deformed massive and pillow Geologic Studies in Alaska by the U.S. Geological Survey, 1991 basalts and basaltic breccias interbedded with and overlain by thin-bedded tuffaceous chert and shale of Late Triassic age (Box and others, in press) are present in the Togiak terrane (tracts C2 and C3, fig. 5). No anomalous concentrations of Cu or Zn are present in rock, streamsediment, or heavy-mineral-concentrate samples in tracts C2 and C3 (maximum for any sample media: 100 ppm Cu, 120 ppm Zn; Bradley and Frost, 1992; Frost and others, 1992a). The probability of occurrence of Cypress massive sulfide deposits is very low. DISCUSSION AND CONCLUSIONS The Bethel region has produced over 250,000 oz of placer gold, mostly from streams draining areas underlain by Early Cretaceous to early Tertiary granitoid plutons. Gold is also present in quartz veins cutting the granitoid plutons and their wall rocks. There is a strong correlation between the locus of plutonism in the eastern part of the map area and favorable tracts for placer deposits and (or) lode-gold occurrences (figs. 2, 4). Gold appears to be mobilized by and deposited from the hydrothermal systems associated with the emplacement and crystallization of the plutons. Erosion of scattered goldbearing quartz veins associated with the plutons provides a mechanism for the concentration of gold in placers. The potential for lode gold deposits that would fall on the grade and tonnage curves of Cox and Singer (1986) is low in all terranes in the map area. The potential for placer-gold occurrences that would fall on the grade and tonnage curves of Bliss and others (1987) is moderate to high. The potential for epithermal mercury deposits is high in areas underlain by Late Cretaceous to early Tertiary plutonic rocks where they intrude the Kuskokwim Group or the rocks of the Togiak terrane. Stable isotope data suggest that many mercury occurrences in southwestern Alaska, including those in the Bethel region, formed from ore fluids derived from dehydration of sedimentary rocks due either to burial metamorphism or to localized igneous activity (Goldfarb and others, 1990). The spatial association of mercury anomalies in volcanic rocks and granitoid plutons in the map area suggests that heat provided by Late Cretaceous to early Tertiary magmatism is the driving mechanism for mercury mobilization. Gray and others (1992) have obtained a 72 Ma 40 Ar-39 Ar age on hydrothermal mica in cinnabar-bearing veins in the Sleetmute quadrangle, which corroborates the magmatism-driven mineralization model. Gold is present in some Hg-As veins in the map area. The data suggest that the zonation observed at other mercury deposits-from mercury dominated at shallow levels, to antimony dominated and gold dominated at deeper levels (Gumiel and Arribas, 1987;

Seward, 1984; Miller and others, 1989)-may be telescoped in the Bethel occurrences, and (or) that higher gold contents may be present at deeper levels (Frost, 1990; Gray and others, 1990). REFERENCES CITED Berg, H.C., and Cobb, E.H., 1967, Metalliferous lode deposits of Alaska: U.S. Geological Survey Bulletin 1246, 254 p. Bliss, J.D., Orris, G.J., and Menzie, W.D., 1987, Changes in grade, volume and contained gold during the mining lifecycle of gold placer deposits: CIM (Canadian Institute of Mining) Bulletin, v. 80, no. 903, p. 75-80. Box, S.E., 1992, Evidence for basin-margin right-slip faulting during Kuskokwim Group deposition, southwestern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 24, no. 5, p.9. Box, S.E., and Elder, W.P., 1992, Depositional and biostratigraphic framework of the Upper Cretaceous Kuskokwim Group, southwestern Alaska, in Bradley, D.C., and Ford, A.B., eds., Geologic studies in Alaska by the U.S. Geological Survey, 1990: U. S. Geological Survey Bulletin 1999, p 8-16. Box, S.E., Moll-Stalcup, E.J., Frost, T.P., and Murphy, J.R., in press, Preliminary geologic map of the Bethel and southem part of the Russian Mission 1 :250,000 quadrangles, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map. Box, S.E., Moll-Stalcup, E.J., Wooden, J.L., and Bradshaw, J.Y., 1990, Kilbuck terrane: oldest known rocks in Alaska: Geology, v. 18, p. 1219-1222. Box, S.E., and Murphy, J.M., 1987, Late Mesozoic structural and stratigraphic framework, eastern Bethel quadrangle, southwestern Alaska, in Hamilton, T.D., and Galloway, J.P., eds., Geologic studies in Alaska by the U.S. Geological Survey during 1986: U.S. Geological Survey Circular 998, p. 78-82. Bradley, Leon, and Frost, T.P., in press, Analytical results and sample locality maps of stream sediment and heavy-mineral-concentrate samples from the Bethel and part of the Russian Mission quadrangles, southwestern Alaska: U.S. Geological Survey Open-File Report . Bundtzen, T.K., and Gilbert, W.G., 1983, Outline of the geology and mineral resources of the upper Kuskokwim region, Alaska: Journal of the Alaska Geological Society, v. 3, p. 101-117. Bundtzen, T.K., and Laird, G.M., 1982, Geologic map of the Iditarod D-2 and eastern D-3 quadrangles, Alaska: Alaska Division of Geological and Geophysical Surveys Geologic Report 72, scale 1:63,360. ---1983, Geologic map of the Iditarod D-1 quadrangle, Alaska: Alaska Division of Geological and Geophysical Surveys Professional Report 78, scale 1 :63,360. ---1990, Geology and mineral resources of the Russian Mission C-1 quadrangle, southwest Alaska: Alaska Division of Geological and Geophysical Surveys Public Data File 89-17, 28 p. Bundtzen, T.K., Swainbank, S.W., Deagen, J.R., and Moore, J.L., 1989, Alaska's mineral industries 1989, Alaska Division of Geological and Geophysical Surveys Special Report 44. 33 p. Coonrad, W.L., Hoare, J.M., Taufen, P.M., and Hessin, T.D., 1978, Geochem!cal analyses of rock samples in the Goodnews and Hagemeister Island quadrangles region, southwestern Alaska: U.S. Geological Survey Open-File Report 78-9-H, 1 sheet, scale 1:250,000. Cox, D.P. and Singer, D.A., eds., 1986, Mineral deposit models: U.S. Geological Survey Bulletin 1693, 379 p. Elder, W.E., and Box, S.E., in press, Late Cretaceous inoceramid bivalves of the Kuskokwim Basin, southwestern Alaska, and their implications on basin evolution: The Paleontological Society, Memoir 26. Frost, T.P., 1990, Geology and geochemistry of mineralization in the Bethel quadrangle, southwestern Alaska, in Goldfarb, R.J., Nash, J.T., and Stoeser, J.W., eds., Geochemical studies in Alaska by the U.S. Geological Survey: U.S. Geological Survey Bulletin 1950, p. C1-C10. Frost, T. P,- and Box, S.E., 1991a, Depth controls on magmatism-related gold and mercury mineralization, Bethel quadrangle, southwestern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 23, no. 2, p. 27. ---1991b, Lithologic and tectonic controls on mercury mineralization in the Bethel 1 °X3° quadrangle, southwestem Alaska [abs.], in Good, E.E., Slack, J.F., and Kotra, R.K, eds., USGS research on mineral resources-1991: Program and abstracts: U.S. Geological Survey Circular 1062, p. 29-31. Frost, T.P., Bradley, Leon, O'Leary, Rich, and Motooka, J., 1992a, Geochemical results and sample locality map for rock samples from the Bethel and southern part of the Russian Mission 1:250,000 quadrangles, Alaska: U.S. Geological Survey Open-file Report 92-316, 229 p. Frost, T.P., Calzia, J.P., Kistler, R.W., and Vivit, D.V., 1988, Petrogenesis of the Crooked Mountains pluton,. Bethel quadrangle-A preliminary report, in Galloway, J.P., and Hamilton, T.D., eds., Geologic studies in Alaska by the U.S. Geological Survey during 1987: U.S. Geological Survey Circular 1016, p. 126-131. Frost, T.P., Moll-Stalcup, E.J., and Box, S.E., 1992b, Early Cretaceous and Late Cretaceous-Paleocene plutonism in the Bethel region, southwestern Alaska: Products of two magmatic arcs [abs.]: Geological Society of America Abstracts with Programs, v. 24, no. 5, p 25. Goldfarb, R.J., Gray, J.E., Pickthorn, W.J., Gent, C.A., Cieutat, B.A, 1990, Stable isotope systematics of epithermal mercury-antimony mineralization, southwestern Alaska, in Goldfarb, R.J., and Nash, J.T., eds., Geochemical studies in Alaska by the U.S. Geological Survey, 1989: U.S. Geological Survey Bulletin 1950, p. E1-E9. Gray, J.E., Frost, T.P., Goldfarb, R.J., and Detra, D.E., 1990, Gold associated with cinnabar- and stibnite-bearing deposits and mineral occurrences in the Kuskokwim River area, southwestern Alaska, in Goldfarb, R.J., and Nash, J.T., eds., Geochemical studies in Alaska by the U.S. Geological Survey, 1989: U.S. Geological Survey Bulletin 1950, p. Dl-D6. Gray, J.E., Goldfarb, R.J., Snee, L.W., and Gent, C.A., 1992, Geochemical and temporal conditions for the formation of Summary of the Mineral Resource Assessment of the Bethel and Southeastern Part of the Russian Mission 1 o by 3° Quadrangles

mercury-antimony deposits, southwestern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 24, no. 5, p. 28. Gumiel, P., and Arribas, A, 1987, Antimony deposits in the Iberian Peninsula: Economic Geology, v. 82, p. 1453-1463. Hessin, T.D., Taufen, P.M., Seward, J.C., Quintana, S.J., Clark, A.L., Grybeck, Donald, Hoare, J.M., and Coonrad, W.L., 1978a, Geochemical and and generalized geological map showing distribution and abundance of lead in the Goodnews and Hagemeister Island quadrangles region, southwestern Alaska: U.S. Geological Survey Open-File Report 78-7-N, scale 1:250,000. ---1978b, Geochemical and and generalized geological map showing distribution and abundance of arsenic, gold, silver, and platinum in the Goodnews and Hagemeister Island quadrangles region, southwestern Alaska: U.S. Geological Survey Open-File Report 78-7-R, scale 1:250,000. Hoare, J.M., and Cobb, E.H., 1977, Mineral occurrences (other than mineral fuels and construction materials) in the Bethel, Goodnews, and Russian Mission quadrangles, Alaska: U.S. Geological Survey Open-File Report 77-156, 98 p. Hoare, J.M., and Coonrad, W.L., 1959a, Geology of the Bethel quadrangle, Alaska: U.S. Geological Survey Miscellaneous ln~estigations Map 1-285, scale 1:250,000, ---1959b, Geology of the Russian Mission quadrangle, Alaska: U.S. Geological Survey Miscellaneous Investigations Map 1-292, scale 1:250,000. ---1978, Geologic Map of the Goodnews and Hagemeister Island quadrangles, Alaska: U.S. Geological Survey OpenFile Report 78-9-B, scale 1:250,000. John, D.A., Stewart, J.H., Kilburn, J.E., Silberling, N.J., and Rowan, L.C., in press, Geology and mineral resources of the Reno 1 °X2° quadrangle, Nevada and California: U.S. Geological Survey Bulletin. Jones, D.L., Silberling, N.J., Coney, P.J., and Plafker, George, 1987, Lithotectonic terrane map of Alaska (west of the 141st meridian): U.S. Geological Survey Miscellaneous Field Studies Map MF-1874-A, scale 1:2,500,000. Kilburn, J.E., Goldfarb, R.J., Griscom, Andrew, Barnes, D.F., and Box, S.E., in press, Map showing areas of potential for mineral resources in the Goodnews Bay, Hagemeister Island, and Nushagak Bay 1 °X3° quadrangles, southwet Alaska: U.S. Geological Survey Miscellaneous Field Studies Map. Mertie, J.B., Jr., 1938, The Nushagak district, Alaska: U.S. Geological Survey Bulletin 903, 96 p. Miller, M.L, Belkin, H.E., Blodgett, R.B., Bundtzen, T.K., Cady, J.W., Goldfarb, R.J., Gray, J.E., McGimsey, R.G., Geologic Studies in Alaska by the U.S. Geological Survey, 1991 and Simpson, S.L., 1989, Pre-field study and mineral resource assessment of the Sleetmute quadrangle, southwestern Alaska: U.S. Geological Survey Open-File Report 89-363, 115 p. Moll-Stalcup, E.J., Box, S.E., and Lanphere, M.A., 1992, Eocene transition from arc to intraplate magmatism in southwestern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 24, no. 5, p. 71. Moll-Stalcup, E.J., Wooden, J.L., Box, S~E., and Frost, T.P., 1989, Preliminary Sr, Nd, and Pb isotopic evidence for magma sources in southwest Alaska: International Association of Volcanology and Chemistry of the Earth's Interior, Continental Magmatism, New Mexico Bureau of Mines and Mineral Resources Bulletin 131, p. 192. Nokleberg, W.J., Bundtzen, T.K., Berg, H.C., Brew, D.A., Grybeck, Donald, Robinson, M.S., Smith, T.E., and Y eend, Warren, 1987, Significant metalliferous lode deposits and placer deposits of Alaska: U.S. Geological Survey Bulletin 1786, 104 p. Phillips, J.D., and Morin, R.L., 1992, New aeromagnetic and gravity maps and interpretations for the Bethel 1 °X3° quadrangle and part of the Russian Mission 1 °X3° quadrangle, southwestern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 24, no. 5, p. 75. Robinson, M.S., and Decker, John, 1986, Preliminary age data and analytical data for selected igneous rocks from the Sleetmute, Russian Mission, Taylor Mountain, and Bethel quadrangles, southwestern Alaska: Alaska Division of Mining and Geological and Geophysical Surveys PublicData File 86-98, 9 p. Roeske, S.M., and Box, S.E., 1992, Metamorphism and deformation of oceanic lithologies in a Cretaceous subduction/collision setting, southwestern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 24, no. 5, p. 79. Rutledge, F.A., 1948, Investigation of the Rainy Creek mercury prospect, Bethel district, southwestern Alaska: U.S. Bureau of Mines Report of Investigations 4361, 9 p. Sainsbury, C.L., and MacKevett, E.M., Jr., 1965, Quicksilver deposits of southwestern Alaska: U.S. Geological Survey Bulletin 1187, 89 p. Seward, T.M., 1984, The transport and deposition of gold in hydrothermal systems, in Gold 82-The geology, geochemistry, and genesis of gold deposits: Geological Society of Zimbabwe Special Publication 1, p. 165-181. Shew, Nora, and Wilson, F.H., 1981, Map and table showing radiometric ages of rocks in southwestern Alaska: U.S. Geological Survey Open-File Report 81-866, 26 p. Reviewers: John E. Gray and Thomas Light

Comparison of the Effectiveness of Stream-Sediment, Heavy-Mineral-Concentrate, Aquatic-Moss, and StreamWater Geochemical Sample Media for the Mineral Assessment Study of the lditarod Quadrangle, Alaska By john E. Gray, Philip L. Hageman, and jean L. Ryder Abstract Samples of stream sediment, heavy mineral concentrate, aquatic moss, and stream water were collected during the mineral assessment study of the lditarod quadrangle. Geochemical data were evaluated to determine the effectiveness of these media in the low and swampy terrain characteristic of much of southwestern Alaska. Results also were used to investigate the effectiveness of these media in the lditarod AMRAP mineral assessment study. Anomalous concentrations of Au, Ag, Hg, As, Cu, Pb, and Zn in stream-sediment samples are the most effective pathfinder elements for delineating ground favorable for metallic-mineral deposits in the lditarod quadrangle. Anomalous concentrations of Au, Ag, Sb, As, Cu, Pb, and Zn, as well as the abundance of microscopic gold and cinnabar in heavy-mineral-concentrate samples, also are effective for delineating favorable areas for mineral deposits. These data identify 22 favorable areas for polymetallic Au vein deposits, polymetallic Cu-Ag-Sn vein deposits, or epithermal Hg-Sb vein deposits. Aquatic-moss and stream-water samples are less effective geochemical sample media for identifying areas of metallic-mineral resource potential. Anomalous concentrations for Ag, Sb, As, Cu, Pb, or Zn in moss samples correspond with only 9 of the 22 areas favorable for mineral deposits. Similarly, stream-water samples were an ineffective geochemical sample medium, where As, Cu, Zn, SO/-, or F- anomalies only indicated 9 of the 22 areas delineated by the more effective stream-sediment and heavy-mineral-concentrate media. Most importantly, aquatic-moss and stream-water samples did not delineate any new areas favorable for metallic-mineral deposits that were not already identified by the stream-sediment and heavy-mineral-concentrate geochemical data. INTRODUCTION Southwestern Alaska is generally characterized by swampy terrain and low, rolling hills with broad, sediment-filled lowlands. This extensive region of low relief presents a difficult problem not only for geologic mapping, but also for geochemical exploration. Therefore, the reconnaissance geochemical survey of the lditarod quadrangle, conducted as part of the Alaska Mineral Resource Assessment Program (AMRAP), was designed to test less conventional sampling media and evaluate their effectiveness as exploration tools in this environment. Regional geochemical reconnaissance studies typically use stream-sediment and heavy-mineral-concentrate samples to identify areas of geochemically anomalous ground favorable for mineral deposits. However, in areas of low relief, such as in the lditarod quadrangle, some streams contain little detrital sediment, making such conventional sampling techniques difficult. Thus, alternative geochemical sample media were sought for this reconnaissance study. Geochemical results from samples of aquatic moss (bryophytes) have shown promise as an exploration tool in some studies conducted near known mineral deposits (Whitehead and Brooks, 1969; Saether and Bolviken, 1983; Erdman and Modreski, 1984; Jones, 1985). Aquatic mosses are perennial plants having many small leaves that are one cell in thickness, giving them a large reactive surface in relation to the mass of the plant (Shacklette, 1984). Mosses have high ion exchange capacities and effectively scavenge metals from the streams in which they grow. Aquatic mosses appear to concentrate metals continuously over their lifetime, effectively eliminating high- and low-flow seasonal fluctuations common in other sample media (Shacklette, 1984 ). Water has also been used as a medium for geochemical exploration. Hydrogeochemical samples of stream water collected near mineral deposits have been shown to contain anomalous concentrations of ore-related cations and anions (Miller and others, 1982; Turner and Ikramuddin, 1982; Learned and others, 1985). Stream water is an advantageous sampling medium because it is widely present in humid regions, easy to collect, and needs little sample preparation prior to analysis. Comparison of Effectiveness of Geochemical Sample Media for Mineral Assessment Study of lditarod Quadrangle

Although several studies have tested specific geochemical sample media for their use in prospecting for a specific type of mineral deposit, few studies compare results for several sample media that are used over a broad region in prospecting for several mineral deposit types. Comparative reconnaissance geochemical studies on a regional scale in which several sample media are collected from the same sample sites are also rare. The purpose of this study is to compare and contrast the geochemistry of stream-sediment, heavy-mineral-concentrate, aquatic-moss, and stream-water samples collected during the reconnaissance geochemical survey of the lditarod quadrangle. GEOLOGY OF THE IDITAROD QUADRANGLE Cretaceous sedimentary rocks of the Kuskokwim Group are the dominant bedrock in the Iditarod quadrangle (fig. 1). These rocks consist of thick sequences of intercalated sandstones, shales, and conglomerates (Cady and others, 1955). The Kuskokwim Group primarily is made up of deep-water turbidite facies, with lesser amounts of shallow shoreline facies rocks (Miiier and Bundtzen, 1987). These rocks have been deformed into northeast-trending synclines and anticlines that are cut by numerous high-angle faults (McGimsey and others, 1988). Late Cretaceous to early Tertiary volcano-plutonic complexes overlie or intrude sedimentary rocks of the Kuskokwim Group at many localities (Miller and Bundtzen, 1987). These complexes consist primarily of basalt and andesite that are in contact with or overlie monzonite plutons. Volcanic- and sedimentary-rock hornfels is commonly found near the contacts with the monzonite intrusions. Other igneous rocks coeval with the volcano-plutonic complexes include (1) an extensive field of felsic to mafic volcanic rocks that covers much of the western portion of the quadrangle, (2) small mafic to intermediate dikes that intrude country rock throughout the quadrangle, and (3) peraluminous granite porphyry dikes, sills, and small stocks (Miller and Bundtzen, in press). Volumetrically minor pre-Cretaceous rocks are found in a narrow northeast-southwest-trending belt in the west-central part of the quadrangle. Metamorphic rocks of the Idono Complex with an Early Proterozoic protolith age are the oldest rocks in the quadrangle (Miller and others, 1991 ). A second unit in the belt consists of greenschist-facies metamorphic rocks of Paleozoic and Proterozoic(?) age (Angeloni and Miller, 1985). Mississippian to Triassic chert and volcanic rocks are found throughout the length of the belt (Miller and Bundtzen, in press). Finally, mafic and ultramafic rocks of the Dishna River area that are correlative with the Jurassic ophiolites of the Tozitna-Innoko belt are found Geologic Studies in Alaska by the U.S. Geological Survey, 1991 in the north-central portion of the quadrangle (Miller, 1990). MINERAL DEPOSITS Gold and Ag are the most economically significant commodities in the Iditarod quadrangle. Approximately 1, 700,000 oz of Au and 240,000 oz of Ag have been produced from the area, mostly from placer mines (Bundtzen and others, 1987). The largest placer mines in the quadrangle are those in the Flat, Moore Creek, Donlin Creek, and Ganes Creek areas (fig. 2). The source of placer gold in the lditarod quadrangle has been linked to polymetallic Au veins spatially related to monzonite stocks (Kimball, 1969; Bundtzen and others, 1985; Miller and Bundtzen, 1987). The Golden Horn mine, near Flat, is the most well-known polymetallic Au deposit in the Iditarod quadrangle. Gold and scheelite are contained in quartz-carbonate veins, with lesser amounts of arsenopyrite, pyrite, chalcopyrite, sphalerite, galena, stibnite, and cinnabar (Bull, 1988). The mineralized veins are contained within faults and shear zones in the monzonite of a volcano-plutonic complex in this area (Bundtzen and Gilbert, 1983). Approximately 2,700 oz of Au, 2,600 oz of Ag, and minor Pb, Zn, and W have been recovered from Golden Horn (Bundtzen and others, 1988b). Samples of mineralized veins are anomalous in Au, Ag, Cu, Pb, Zn, Hg, Sb, As, W, Bi, and Cd (Bundtzen and others, 1988b). The Broken Shovel lode (fig. 2) is a polymetallic Au vein deposit containing tetrahedrite, arsenopyrite, and scheelite in quartz-tourmaline veins that has been described by Bundtzen and others (1988a). Mineralized veins are hosted in a shear zone cutting an altered monzonite pluton in the Moore Creek area. Although native gold has not been identified in the lode, mineralized vein samples contain anomalous concentrations of Au and Ag, as well as Cu, Pb, Zn, As, Sb, W, B, and Bi (Bundtzen and others, 1988a). Epithermal cinnabar- and stibnite-vein deposits representing significant Hg and Sb resources are found throughout the Kuskokwim River region, which includes a large portion of the lditarod quadrangle (Sainsbury and MacKevett, 1965). These deposits are found primarily within sedimentary rocks of the Kuskokwim Group and mafic to felsic dikes, sills, or small stocks that intrude these sedimentary rocks (Bundtzen and others, 1987). Cinnabar and stibnite are the dominant ore minerals in these deposits, with lesser amounts of realgar, orpiment, native mercury, pyrite, hematite, limonite, and dickite (Sainsbury and MacKevett, 1965). The DeCourcy mine (fig. 2) is the most significant Hg mine in the lditarod quadrangle, having produced approximately 1,500 flasks of Hg (M.L. Miller, written commun., 1992).

20 MILES 25 KILOMETERS EXPLANATION -Tertiary and Late Cretaceous monzonite, syenite, quartz monzonite, granodiorite, and granite stocks and plutons ·~f1 ; Tertiary and Late Cretaceous dacite to andesite intrusions lttff~IU Tertiary and Late Cretaceous hornfels of sedimentary and volcanic rock p:::::::::::l Tertiary and Late Cretaceous andesite, dacite, rhyolite, and minor basalt flows and tuffs of the Yetna River area -Tertiary and Late Cretaceous granite porphyry sills, ~ikes, and plugs Tertiary and Late Cretaceous intermediate to mafic dikes generally less than 1 m wide

Tertiary and Late Cretaceous andesite I basalt, and dacite volcanic flows, tuffs, volcanic breccia, and vo caniclastic sandstone D Late Cretaceous sandstone, siltstone, shale, pebble conglomerate, and minor interbedded volcanic tuffs and flows, primarily of the Kuskokwim Group

Early Cretaceous colluvial chips of sandstone, tuffaceous sandstone, and siltstone containing detrital potassium feldspar E Tertiary and Cretaceous(?) amphibole-pyroxene-biotite alkali granite, "' poorly exposed []]]]]Jurassic ultramafic and gabbroic rocks of the Dishna River area, probably part of the Tozitna-lnnoko ophiolite belt (:.': .-, .: '·)Triassic 'o Mi~sissippian recr:vs.talliz~ chert, volcanic rocks; metasiltstone, and m1nor limestone correlative w1th the lnnoko terrane

Paleozoic and Proterozoic (?) greenschist-facies metaigneous and

metasedimentary rocks correlative with the Ruby terrane

Proterozoic metamorphic rocks, including orthogneiss, schist, and amphibolite

of the ldono Compfex -- Faults, dashed where inferred Figure 1. lditarod quadrangle geology (simplified from Miller and Bundtzen, in press). Comparison of Effectiveness of Geochemical Sample Media for Mineral Assessment Study of lditarod Quadrangle

Stibnite-rich veins are known near Snow Gulch in the Donlin Creek area (Cady and others, 1955), and a similar occurrence (the Wyrick lode) is found on Granite Creek (fig. 2) (Bundtzen and others, 1986). These lodes consist of small quartz-carbonate veins and vug fillings containing stibnite that are hosted in granite porphyry dikes, in the adjacent siltstone and graywacke of the Kuskokwim Group, or at contacts between the two rock types. Stibnite-rich samples from the Snow Gulch and the Wyrick lode are anomalous in Au and Ag (Cady and others, 1955; Bundtzen and others, 1986; McGimsey and others, 1988; Gray and others, 1990). Stibnite-rich veins spatially related to granite porphyry are an important deposit type in the lditarod quadrangle and throughout the Kuskokwim River region. Polymetallic Cu-Ag-Sn veins contammg chalcopyrite, sphalerite, galena, tetrahedrite, arsenopyrite, cassiterite, and wolframite were discovered in the Beaver Mountains by Bundtzen and Laird (1982). Gangue minerals include quartz, tourmaline, axinite, and fluorite. The veins are found in faults, shear zones, or breccias within the cupolas of the monzonite stocks, or in sedimentary- and volcanic-rock hornfels aureoles surrounding the igneous rocks (Bundtzen and Gilbert, 1983; Bundtzen and others, 1987). The most consistent pathfinder elements associated with these polymetallic veins are Cu, Pb, Zn, Sn, W, Au, and Ag, but anomalous Hg, Sb, As, Bi, Cd, and B may also be associated with some of the deposits (Bundtzen and Laird, 1982; McGimsey and others, 1988). 159° 156" 63°r r , ~"r rb-)

r r r Lode Mines 1 Golden Horn 2 DeCourcy x Lode Prospects 3SnowGulch 4 Wyrick / )' ) ( ~.first f ""-,_Chance c.

)

20 MILES 25 KILOMETERS EXPLANATION Stream-sediment and heavy-mineral-concentrate anomaly - - - Aquatic-moss anomaly · · · · · · Stream-water anomaly Figure 2. Geochemical anomaly map of lditarod quadrangle. As described in text and shown here, only nine areas contain moss geochemical anomalies and nine areas contain stream-water geochemical anomalies, compared with 22 anomalies defined by the stream-sediment and heavy-mineral-concentrate data. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

GEOCHEMICAL STUDIES Methods and Data Evaluation Stream-sediment, heavy-mineral-concentrate, streamwater, and aquatic-moss (bryophyte) samples were collected from active channels of perennial first- and second-order streams. The area of most drainage basins ranged from 1 mi2 (2.6 km2) to about 5 mi2 (13 km2). The stream-sediment sampling density was approximately I sample site per 9 mi2 (23 km2). The stream-sediment samples were sieved to minusSO-mesh (0.17 mm) and chemically analyzed. The panned-concentrate samples were sieved to minus-30-mesh (0.60 mm), separated in bromoform, and further separated magnetically to obtain a nonmagnetic, heavy-mineral-concentrate sample. The nonmagnetic, heavy-mineral-concentrates were split using a multiple-plate splitter. One split was hand ground for geochemical analysis, and the other split was saved for mineralogical identification. Aquatic-moss samples were preferentially collected near the current water level from boulders or dead-fall vegetation in. the active stream channel. However, when moss on the stream bottom could not be found, moss from the channel walls or overbank was collected. The moss and the sediment trapped within the moss were collected together, and the sediment was later removed by agitating the samples in water (Arbogast and others, 1991), which can be a time-consuming and labor-intensive procedure. When all of the sediment was removed, the moss samples were dried, ground in a mill, and ashed at 500°C. The ashed material was used for geochemical analysis. Stream water filtered through a 0.45-J..lm membrane was acidified with concentrated HN03 and used for cation analysis. Unfiltered water samples were also collected and used for anion analysis. No further preparation was required on the stream-water samples prior to analysis. There were 1,151 stream-sediment samples collected in this study. A total of 799 heavy-mineral concentrates were collected and microscopically examined for their mineralogical content. However, some of the heavy-mineral concentrates collected contained an insufficient amount of material necessary for geochemical analysis, and only 662 concentrates were chemically analyzed. Aquatic mosses could not be located at some sites, and thus only 863 mosses were collected. There were 1,050 stream-water samples collected during this study. Stream-sediment, heavy-mineral-concentrate, and moss samples were analyzed for multielement suites by semiquantitative emission spectrography. Stream-sediment samples were also analyzed for selected elements by inductively coupled plasma (ICP) and atomic absorption spectrophotometry (AAS) analysis. Stream-water samples were analyzed by ion chromatography, ICP, and AAS analysis. The presence of gold and cinnabar was determined microscopically in the nonmagnetic, heavy-mineralconcentrate samples. Geochemical results for the stream sediments and heavy-mineral concentrates are listed in Gray and others (1988a) and Hopkins and others (1991). Mineralogy results for this study appear in Bennett and others (1988). Geochemical results for the moss samples are found in Arbogast and others ( 1991 ), and results for the stream waters appear in Gray and others (1988b ). INTERPRETATION OF GEOCHEMICAL RESULTS Anomalous concentrations of Au, Ag, Hg, As, Cu, Pb, and Zn in stream-sediment samples and anomalous concentrations of Au, Ag, Sb, As, Cu, Pb, and Zn, as well as visual estimation of the abundance of gold and cinnabar, in heavy-mineral-concentrate samples are useful pathfinders for delineating areas favorable for mineral deposits in the Iditarod quadrangle. In aquatic-moss samples, anomalous concentrations of Ag, Sb, As, Cu, Pb, and Zn are also helpful for identifying areas favorable for mineral deposits. The moss samples were analyzed for Au, but none were found to contain Au above the lower limit of determination (2 ppm) by the spectrographic method. Thus, Au in the mosses was not useful in this study and is not discussed further. In streamwater samples, anomalous concentrations of As, Cu, Zn, SO 4 2-, and p- are the most useful. Anomalies in all data sets were defined by identifying breaks in the frequency distribution of each data set. Anomaly selections were based on the ability to explain a large population of the data with a minimum of singlesite anomalies. For example, the 98th percentile concentration of 3 ppm Ag in moss was selected as anomalous because 58 percent of the anomalous samples cluster into specific areas, whereas the 95th percentile concentration of 2 ppm Ag in moss has numerous single-site anomalies and only 39 percent of the data can be explained by clusters of samples forming geochemical anomalies. The anomalous values selected are generally near the 98th percentile. However, slightly lower percentiles were used as anomalous concentrations for some elements in the stream-sediment data and are a function of the larger number of these samples clustering into specific areas defining geochemical anomalies. A summary of the data used in this study is presented in table 1. Stream-Sediment and Heavy-MineralConcentrate Samples Stream-sediment and heavy-mineral-concentrate samples are the most useful sample media for delineating areas geochemically favorable for the mineral Comparison of Effectiveness of Geochemical Sample Media for Mineral Assessment Study of lditarod Quadrangle

U1 Table 1. Summary of stream-sediment, heavy-mineral-concentrate, aquatic-moss, and stream-water geochemical data

"' [N, not detected at the concentration shown; L, detected but less than the concentration shown; G, greater than the concentration shown;-, not applicable. All concentrations shown are parts per million (ppm), except detection ratio, N, " L, G, and the cinnabar and gold mineralogy results (in percent), and percent of anomalies in areas] Q 6'" OQ Percentile concentrations Selected Percent of !a Detection N2 G4 Minimum 50th 90th 95th 98th Maximum anomalies5 anomalies ratio1 in areas Q. iii" , :;- Stream sediments ii Au O.OOSL ,

Ag 1,084 .5N .5N .5N r:r Hg 34G

As 2.0N " Cu c:

Pb ION "' Zn 200N

6'" Concentrates OQ Au 20N 20N 20N 20N l,OOOG Ag IN lN lN lN Sb 200N 200N 200N 200N 10,000 "

As SOON SOON SOON SOON SOON 1,500 1,000 Cu lON 1,000

Pb 20N 5,000 Zn SOON SOON SOON 1,000 20,000 1,500 Gold N N N N N 1% Gold6 Cinnabar N N 1% 1-5% 1-5% 20-50% 1-5% Mosses Ag .lN Sb SON SON 50N SON As 200N 200N 200N 5,000 Cu 1,000 2,000 1,000 Pb lOL Zn lOON 1,500 2,000 2,000 7,000 2,000 Waters As .OOlN Cu .OOlN .OOlL Zn .001N so 4 2- p- .01N 1Detection ratio is the number of uncensored values divided by the total number of samples analyzed. 2N is the number of samples in which concentrations could not be detected at the lower limit of determination. 3L is the number of samples in which concentrations were reported as observable but were less than the lower limit of determination. 4G is the number of samples in which concentrations were reported as observable but were greater than the upper limit of determination. 5 Anomalous concentrations were selected by identifying breaks in the frequency distribution of each data set. 6Any amount of visible gold in heavy-mineral-concentrate samples is anomalous.

deposit types in the Iditarod quadrangle. Twenty-two areas were identified as favorable for the presence of metallic-mineral resources based solely on clusters of geochemical or mineralogical anomalies in stream-sediment or heavy-mineral-concentrate samples (fig. 2; table 2). These areas range from large, covering numerous drainage basins (for example, the Beaver Mountains), to small, covering only one or two drainage basins (for example, Ruby-Beaver Creeks). However, no areas were delineated on the basis of an anomaly at a single site. Over 75 percent of the geochemical and mineralogy anomalies in stream-sediment and heavy-mineral-concentrate samples are included in the delineated areas (table 1), except for anomalous concentrations of Au in stream-sediment samples (69 percent included in areas), Zn anomalies in concentrates (58 percent included in areas), and Cu anomalies in concentrates (56 percent included in areas). Geochemical anomalies in samples of streamsediment and heavy-mineral concentrate cluster around known placer-gold mines such as Flat, Moore Creek, and Donlin Creek. Stream-sediment and heavy-mineralconcentrate samples, with geochemical anomalies are also found downstream from polymetallic Cu-Ag-Sn veins in the Beaver Mountains and the Broken Shovel lode, in polymetallic Au vein deposits near Flat, and in the epithermal Hg-Sb deposit at DeCourcy. In addition, several new areas favorable for mineral deposits were also delineated by identifying geochemical anomalies in stream-sediment and heavy-mineral-concentrate samples in the quadrangle (table 2). Aquatic-Moss Samples Aquatic-moss samples are less useful than streamsediment or heavy-mineral-concentrate samples for identifying areas that are geochemically favorable for mineral deposits in the Iditarod quadrangle. However, some of the collected moss samples are highly metalliferous. For example, 3 moss samples contain 5,000 ppm As, 2 have 2,000 ppm Cu, and 2 contain 7,000 ppm Zn. Unfortunately, only the 3 mosses with As anomalies are found in areas delineated as geochemically anomalous by the stream-sediment and heavy-mineral-concentrate samples. The 4 mosses with the highest concentrations of Cu and Zn are found as single-site anomalies that do not coincide with any of the 22 delineated anomalous areas. Furthermore, only 2 of the 17 mosses with the highest concentrations of Zn and 5 of the 22 mosses with the highest concentrations of Cu are found in areas delineated as anomalous by the stream-sediment and heavy-mineral-concentrate data. Multiple-site anomalies for Cu, Pb, or Zn in moss samples are found in only the Beaver Mountains, Granite Mountain, and Bismarck Creek areas and are consistent with the areas defined by the stream-sediment and heavy-mineral-concentrate data that are favorable for polymetallic Cu-Ag-Sn vein deposits (table 2). These results suggest that anomalous concentrations of Cu, Pb, and Zn in moss samples are not completely reliable for identifying areas in the quadrangle that are favorable for mineral deposits. Anomalous concentrations of Ag, Sb, and As in the mosses are consistent with some areas favorable for the presence of mineral deposits in the quadrangle. Using concentrations listed in table 1, 58 percent of the Ag, 77 percent of the Sb, and 76 percent of the As anomalies in moss samples are found within the 22 delineated areas. However, Ag, Sb, or As anomalies in moss are not found in the DeCourcy, First Chance Creek, George River-Moore Creek, southeast of Twin Buttes, Moose Creek, Mosquito Mountain, north of George, Ruby-Beaver Creeks, and Yetna River areas where Au-Ag or Hg-Sb-As anomalies in stream-sediment or heavy-mineral-concentrate samples are found. The lack of moss anomalies in some of these areas may be because the mosses were not analyzed for Hg. However, most of these areas contain stream-sediment and heavy-mineral-concentrate samples anomalous in Ag, Sb, or As but do not have Ag, Sb, or As anomalies in the mosses. These results suggest that identifying areas that are geochemically favorable for epithermal Hg-Sb vein deposits at a regional reconnaissance scale is more difficult using aquatic-moss samples than it is by using stream-sediment or heavy-mineralconcentrate samples. Only 9 of the 22 areas outlined contain more than 1 moss sample with anomalous concentrations of Ag, Sb, ·As, Cu, Pb, or Zn. No areas favorable for the mineral deposit types in the lditarod quadrangle were delineated based solely on the geochemical data from the mosses. These results suggest that collecting aquatic moss samples as a geochemical exploration medium in regional reconnaissance studies in southwestern Alaska offers no particular advantage over stream-sediment ·and heavy-mineral-concentrate sampling. Stream-Water Samples Anomalous concentrations of As are found in epithermal Hg-Sb vein deposits (Gray and others, 1990) and Au and Cu-Ag-Sn polymetallic vein deposits in the quadrangle (Bundtzen and Laird, 1982; Bundtzen and others, 1988a). Polymetallic Au and polymetallic CuAg-Sn veins also contain anomalous concentrations of F- or the mineral fluorite (Bull and Bundtzen, 1987; McGimsey and others, 1988). In addition, elevated concentrations of Cu, Zn, and SO 4 2- in surface waters near ore deposits has been attributed to the oxidation of sulfide minerals (Miller and others, 1982; Learned and others, Comparison of Effectiveness of Geochemical Sample Media for Mineral Assessment Study of lditarod Quadrangle

a1::1 2._ Table 2. Summary of stream-sediment, heavy-mineral-concentrate, aquatic-moss, and stream-water anomalies of selected areas in the lditarod quadrangle

[SS, stream-sediment samples; H-M-C, nonmagnetic, heavy-mineral-concentrate samples. Data is for elements shown unless otherwise indicated. N/A, not applicable because mosses could not be located at this site and were not sampled] tNumber of anomalies (I) :;·

;- (1)

c:r

I'D c:

1::1 :g Q OQ ;:;·

I'D

location Au-Ag H-M-C Moss 1 Water2 Beaver Mountains Moore Creek Area Flat Area Mount Joaquin Tatalina Mountain Granite Mountain* DeCourcy area Donlin Creek Camelback Mountain First Chance Creek* George R.-Moore Ck. North of George* SE of Twin Buttes* Moose Creek* Dishna R.-Otter Ck. Mosquito Mtn.* Ruby-Beaver Creeks y etna River* Bismarck Creek* Takotna Mtn. Ganes Creek Fourth of July Ck. 1 Anomalous concentrations of Ag only. N/A 2 Anomalous concentrations of SO 4 2- and F-. 3 Anomalous concentrations of Hg and As. 4 Anomalous concentrations of cinnabar, Sb, and As. 5 Anomalous concentrations of Sb and As. 6 Anomalous concentrations of As only. 7 Anomalous concentrations of Cu and Zn. *New areas with no reported mines, prospects, or mineral occurrences. Hg-5b-As H-M-C4 Moss5 Water6 N/A Cu-Pb-Zn H-M-C Moss N/A Known or favorable deposit types Water7 Cu-Ag-Sn polymetallic vein Au and Cu-Ag-Sn polymetallic vein Au and Cu-Ag-Sn polymetallic vein Au polymetallic vein Cu-Ag-Sn polymetallic vein Cu-Ag-Sn polymetallic vein epithermal Hg-Sb vein Au polymetallic vein Cu-Ag-Sn polymetallic vein epithermal Hg-Sb vein epithermal Hg-Sb vein epithermal Hg-Sb vein epithermal Hg-Sb vein epithermal Hg-Sb vein epithermal Hg-Sb vein Au polymetallic vein epithermal Hg-Sb vein Au polymetallic vein Cu-Ag-Sn polymetallic vein Cu-Ag-Sn polymetallic vein Au polymetallic vein Au polymetallic vein

1985). Anomalous concentrations of As, Cu, Zn, p-, and SO 4 2- in stre~m waters should therefore be useful for identifying areas favorable for these types of mineral deposits in the Iditarod quadrangle. The concentration of Ag in stream water should also be a useful pathfinder, but the Ag content in samples of stream and lake water collected during a previously conducted National Uranium Resource Evaluation (NURE) survey was found to be ineffective for locating mineral deposits in the Iditarod quadrangle (Miller and others, in press). Therefore, the concentration of Ag in stream water was not determined during this study. The geochemistry of the stream waters is the most difficult data set to interpret. Only about 28 percent of the samples with anomalous concentrations of As, Cu, Zn, SO/-. and p- in the stream-water samples (table 1) are found in areas identified to be geochemically favorable for mineral deposits in the quadrangle. One complication is that water chemistry in this region of Alaska can be highly variable due to rapid fluctuations in discharge, related to frequent summer precipitation. Sulfate has the highest reliability, with 41 percent of the stream-water samples having concentrations greater than 10 ppm being found within areas designated as favorable for the occurrence of mineral deposits. Anomalous sol- in stream-water samples are found in the Moore Creek, Flat, Donlin Creek, Camelback Mountain, and Bismarck Creek areas, where polymetallic vein or epithermal Hg-Sb vein deposits are known (table 2). Unfortunately, no stream-water samples with anomalous SO 4 2- were found in the Beaver Mountains, where several veins containing base-metal sulfides are known. It is unclear why anomalous SO 4 2- was not found in stream-water samples collected in the Beaver Mountains, as well as in 15 other areas favorable for mineral deposits in the quadrangle. Only 33 percent of the stream-water samples with anomalous As (table 1) are found in areas geochemically favorable for mineral deposits in the quadrangle. Anomalous concentrations of As are found in stream-water samples collected in 8 areas favorable for polymetallic vein deposits, although 4 of these areas contain only 1 As anomaly. Similarly, only 26 percent of the stream-water samples with anomalous p- (table 1) are found in the areas outlined, including the Tatalina Mountain, Otter Creek-Dishna River, and Bismarck Creek areas. Mineralized veins in the Granite Mountain area contain anomalous p- (McGimsey and others, 1988), but the stream-water samples collected in this area do not. In addition, fluorite has been described in greisen veins in the Flat area, but only 1 stream-water sample contains anomalous p- in this area. Copper has the poorest reliability of the streamwater samples, with only 16 percent of the waters having concentrations greater than 0.007 ppm being found in areas designated as geochemically favorable for mineral deposits in the quadrangle. Even more discouraging for Cu is the fact that only 1 of the most anomalous 11 stream-water samples (concentrations greater than or equal to 0.009 ppm) is found in an area outlined as favorable for mineral deposits; the other samples are found as single-site anomalies that are difficult to explain. Anomalous Zn in stream water is slightly more reliable; 25 percent of the samples that contain at least 0.010 ppm Zn are located in areas favorable for mineral deposits. However, only 6 of the 16 samples of stream water with the highest concentrations of Zn (greater than 0.020 ppm) are found in areas favorable for mineral deposits. Similar to Cu, the other 10 water samples with the highest Zn concentrations are found as single-site anomalies with no apparent association to mineral deposits. In the Beaver Mountains, only 1 water sample with anomalous Cu and 2 samples with anomalous Zn are found, which is not particularly impressive for an area with several known polymetallic base-metal vein deposits. More significantly, no Cu or Zn anomalies in stream waters are found in the Camelback Mountain area, where polymetallic vein deposits containing anomalous ba.se metals are also known (Bundtzen and others, 1988a). Multiple-site stream-water Cu or Zn anomalies are consistent with only 4 of 14 areas favorable for polymetallic Cu-Ag-Sn or polymetallic Au vein deposits (table 2). Stream-water samples are not as reliable as streamsediment and heavy-mineral-concentrate samples for identifying geochemically favorable ground for mineral deposits in the Iditarod quadrangle. Similar to the aquatic-moss samples, no areas were delineated solely on the basis of geochemical data from the stream waters. CONCLUSIONS In the Iditarod quadrangle, geochemical data from stream-sediment and heavy-mineral-concentrate samples are reliable for identifying favorable areas for polymetallic Au, polymetallic Cu-Ag-Sn, and epithermal Hg-Sb vein deposits. Over 75 percent of the streamsediment and heavy-mineral-concentrate samples with anomalous concentrations of pathfinder elements are included in these areas. Data suggest that stream-sediment and heavy-mineral-concentrate samples are effective geochemical exploration media in low-energy stream environments such as the swampy and low, rolling hills terrain in southwestern Alaska. Some aquatic-moss samples collected in the Iditarod quadrangle contain high concentrations of metals. However, the moss data are difficult to interpret because mosses containing geochemical anomalies do not cluster well into areas that are favorable for the mineral deposit types in the quadrangle. Approximately 58 Comparison of Effectiveness of Geochemical Sample Media for Mineral Assessment Study of lditarod Quadrangle

percent of the Ag, 77 percent of the Sb, 76 percent of the As, and 83 percent of the Pb anomalies in the moss samples were found within areas designated to be favorable for mineral deposits. Geochemical results suggest that aquatic-moss sampling offered no particular advantage during the mineral assessment study in the Iditarod quadrangle other than to confirm the existence of anomalies already identified with stream-sediment and heavy-mineral-concentrate data. The results also suggest that had only moss sampling been conducted, over 50 percent of the areas favorable for the occurrence of mineral deposits in the quadrangle would have been overlooked. Furthermore, the aquatic-moss sampling design was adversely affected because mosses could not be found at all sites, whereas stream sediment is nearly always present. In addition, cleaning, sample preparation, and processing of mosses is time consuming and labor intensive. Stream-water samples collected during the geochemical survey of the Iditarod quadrangle contain low concentrations of metals. Stream waters with geochemical anomalies show the greatest amount of scatter of all of the geochemical sample media used in the Iditarod study. Less than 28 percent of the samples with anomalous concentrations of As, Cu, Zn, ·-sol-, and F- in the stream waters are found in areas delineated as favorable for the presence of mineral deposits. More importantly, multiple sample-site anomalies for As, Cu, Zn, Sol-, and p-in the stream waters are found in only 9 of 22 of these areas. Thus, stream waters are the least reliable medium for identifying areas favorable for mineral deposits in the Iditarod quadrangle. In no instances were aquatic-moss or streamwater samples exclusively beneficial for identifying areas favorable for mineral deposits. That is, in the Iditarod quadrangle, no geochemical anomalies were discovered solely on the basis of the aquatic-moss or stream-water data, and therefore they offered no particular advantage during the mineral assessment study of the quadrangle. This study does not imply that hydrogeochemical or aquatic-moss sampling cannot prove beneficial in some exploration programs. However, the regional reconnaissance survey used in the Iditarod quadrangle suggests stream-sediment and heavymineral-concentrate sampling are the most effective method for identifying geochemically anomalous ground in the swampy and low topographic terrain in southwestem Alaska. Acknowledgments.-The authors would like to recognize the assistance of Delmont Hopkins, John Bullock, Belinda Arbogast, Jerry Motooka, Rick Sanzolone, John McHugh, Bruce Roushey, and Walt Ficklin with chemical analyses; Cliff Taylor and Greg Bennett for mineralogy identifications; Pete Folger, James McNeal, Scott Rose, Robert Carlson, Jerry Gaccetta, Wendy Gerstel, Martha L. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Miller, and Karen D. Kelly for sample collection; and Olga Erlich, Barbara Erickson, Tom Peacock, Pete Theodorakos, and Greg Thurston for sample preparation. REFERENCES CITED Angeloni, L.M., and Miller, M.L., 1985, Greenschist facies metamorphic rocks of north-central Iditarod quadrangle, in Bartsch-Winkler, Susan, ed., The U.S. Geological Survey in Alaska-Accomplishments during 1984: U.S. Geological Survey Circular 967, p. 19-21. Arbogast, B.F, Erickson, B.M., Gray, J.E., and McNeal, J.M., 1991, Analytical results and sample locality map of moss, moss-sediment, and willow samples from the lditarod quadrangle, Alaska: U.S. Geological Survey Open-File Report 91-380-A, 101 p., 1 pl., scale 1:250,000. Bennett, G.J., Gray, J.E., and Taylor, C.D., 1988, Mineralogy and sample locality map of the nonmagnetic, heavy-mineral-concentrate samples, lditarod quadrangle, Alaska: U.S. Geological Survey Open-File Report 88-32, 37 p., 1 pl., scale 1:250,000. Bull, K.F., 1988, Genesis of the Golden Hom and related mineralization in the Flat area, Alaska: Fairbanks, University of Alaska, M.S. thesis, 149 p. Bull, K.F., and Bundtzen, T.K., 1987, Greisen and vein Au-W mineralization of the Black Creek stock, the Flat area, west-central Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 19, p. 362. Bundtzen, T.K., Cox, B.C., and Veach, N.C., 1987, Heavy mineral provenance studies in the lditarod and lnnoko districts, western Alaska, in Process Mineralogy VII, SME/ AIME joint meeting, Denver, Colorado: Metallurgical Society, p. 221-245. Bundtzen, T.K., and Gilbert, W.G., 1983, Outline of geology and mineral resources of upper Kuskokwim region, Alaska: Journal of the Alaska Geological Society, v. 3, p. Bundtzen, T.K., and Laird, G.M., 1982, Geological map of the Iditarod D-2 and eastern D-3 quadrangles, Alaska: Alaska Division of Geological and Geophysical Surveys Geologic Report 72, 1 pl., scale 1:63,360. Bundtzen, T.K., Laird, G.M., and Lockwood, M.S., 1988a, Geologic map of the lditarod C-3 quadrangle, Alaska: Alaska Division of Geological and Geophysical Surveys Professional Report 96, 13 p., 1 pl., scale 1:63,360. Bundtzen, T.K., Miller, M.L., Bull, K.F., and Laird, G.M., 1988b, Geology and mineral resources of the lditarod mining district, lditarod B-4 and B-5 quadrangles, westcentral Alaska: Alaska Division of Geological and Geophysical Surveys Public-Data File 88-19, 44 p., 1 pl., scale 1:63,360. Bundtzen, T.K., Miller, M.L., and Kline, J.T., 1985, Geology of heavy mineral placer deposits of the lditarod and lnnoko precincts, western Alaska, in Madonna, J.A., ed., Conference on Alaskan Placer Mining, 7th, Proceedings: Fairbanks, Alaska, Alaska Prospectors Publishing Company, p. 35-41. Bundtzen, T.K., Miller, M.L., and Laird, G.M., 1986, Prospect

examination of the Wyrick placer/lode system, Granite Creek, lditarod-George mining district, Iditarod B-2 quadrangle, Alaska: Alaska Division of Geological and Geophysical Surveys Public-Data File 86-29, 10 p., 1 pl., scale 1:63,360. Cady, W.M., Wallace, R.E., Hoare, J.M., and Webber, E.J., 1955, The central Kuskokwim region, Alaska: U.S. Geological Survey Professional Paper 268, 132 p. Erdman, J.A., and Modreski, P.J., 1984, Copper and cobalt in aquatic mosses and stream sediments from the Idaho cobalt belt: Journal of Geochemical Exploration, v. 20, p. Gray, J.E., Arbogast, B.F., and Hudson, A.E., 1988a, Geochemical results and sample locality map of the stream-sediment and nonmagnetic, heavy-mineral-concentrate samples for the Iditarod quadrangle, Alaska: U.S. Geological Survey Open-File Report 88-221, 69 p., 1 pl., scale 1 :250,000. Gray, J.E., Frost, T.P., Goldfarb, R.J., and Detra, D.E., 1990, Gold associated with cinnabar- and stibnite-bearing deposits and mineral occurrences in the Kuskokwim River region, southwestern Alaska, in Goldfarb, R.J., Nash, T.J., and Stoeser, J.W., eds., Geochemical studies in Alaska: U.S. Geological Survey Bulletin 1950, p. D1-D6. Gray, J.E., Ryder, J.L. Sanzolone, R.F., McHugh, J.B., and Ficklin, W .H., 1988b, Analytical data and sample locality map for stream water samples from the Iditarod quadrangle, Alaska: U.S. Geological Survey Open-File Report 88-55, 23 p., 1 pl., scale 1:250,000. Hopkins, D.M., Gray, J.E., McDougal, C.M., and Slaughter, K.E., 1991, Mercury, gold, thallium, and tellurium data and sample locality map of stream-sediment samples from the lditarod quadrangle, Alaska: U.S. Geological Survey Open-File Report 91-283-A, 37 p., 1 pl., scale 1:250,000. Jones, K.C., 1985, Gold, silver and other elements in aquatic bryophytes from a mineralised area of North Wales, U.K.: Journal of Geochemical Exploration, v. 24, p. 237-246. Kimball, A.L., 1969, Reconnaissance sampling of decomposed monzonite near Flat, Alaska: U.S. Bureau of Mines OpenFile Report 6-69, 39 p. Learned, R.E., Chao, T.T., and Sanzolone, R.F., 1985, A comparative study of stream water and stream sediment as geochemical exploration media in the Rio Tanama porphyry copper district, J?uerto Rico: Journal of Geochemical Exploration, v. 24, p. 175-195. McGimsey, R.G., Miller, M.L., and Arbogast, B.F., 1988, Paper version of analytical results, and sample locality map for rock samples from the Iditarod quadrangle, Alaska: U.S. Geological Survey Open-File Report 88-421-A, 110 p., 1 pl., scale 1 :250,000. Miller, M.L., 1990, Mafic and ultramafic rocks of the Dishna River area, north-central lditarod quadrangle, west-central Alaska, in Dover, J.H., and Galloway, J.P., eds., Geologic studies in Alaska by the U.S. Geological Survey, 1989: U.S. Geological Survey Bulletin 1946, p. 44-50. Miller, M.L., Bradshaw, J.Y., Kimbrough, D.L., Stem, T.W., and Bundtzen, T.K., 1991, Isotopic evidence for Early Proterozoic age of the ldono Complex, west-central Alaska: Journal of Geology, v. 99, p. 209-223. Miller, M.L., and Bundtzen, T.K., 1987, Geology and mineral resources of the lditarod quadrangle, west-central Alaska [abs.], in Sachs, J.S., ed., U.S.G.S. research on mineral resources, 1987-Programs and abstracts: U.S. Geological Survey Circular 995, p. 46-47. ---in press, Geologic map of the Iditarod quadrangle, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map, scale 1:250,000. Miller, M.L., Bundtzen, T.K., and Gray, J.E., in press, Mineral resource assessment of the Iditarod quadrangle, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map, scale 1 :250,000. Miller, W .R., Ficklin, W .H., and Learned, R.E., 1982, Hydrogeochemical prospecting for porphyry copper deposits in the tropical-marine climate of Puerto Rico: Journal of Geochemical Exploration, v. 16, p. 217-233. Saether, O.M., and Bolviken, B., 1983, Anomalous metal contents of aquatic bryophytes at Tverrfjellet and Snertingdal, central Norway [abs.], in Bjorklund, A., and Koljonen, T., eds., lOth International Geochemical Exploration Symposium: Espoo/Helsinki, Finland, Geological Survey of Finland, p. 69. Sainsbury, C.L., and MacKevett, E.M., Jr., 1965, Quicksilver deposits of southwestern Alaska: U.S. Geological Survey Bulletin 1187, 89 p. Shacklette, H.T., 1984, The use of aquatic bryophytes in prospecting: Journal of Geochemical Exploration, v. 21, p. Turner, L.D., and Ikramuddin, Mohammed, 1982, Electrothermal atomic absorption determination of gold, silver, and arsenic in stream waters and their relationship to gold-silver occurrences in the Republic Graben, N .E. Washington, in Levinson, A.A., ed., Precious metals in the northern Cordillera: Vancouver, British Columbia, Canada, Symposium Proceedings of the Association of Exploration Geochemists, p. 79-88. Whitehead, N.E., and Brooks, R.R., 1969, Aquatic bryophytes as indicators of uranium mineralization: Bryologist, v. 72, p. 501-507. Reviewers: Clifford J. Taylor and Richard J. Goldfarb Comparison of Effectiveness of Geochemical Sample Media for Mineral Assessment Study of lditarod Quadrangle

Geochemically Anomalous Areas in the West-Central Part of the Howard Pass Quadrangle, National Petroleum Reserve, Alaska: Evidence for Sediment-Hosted Zn-Pb-Ag-Ba Mineralization By Karen D. Kelley, J. Carter Borden, Elizabeth A. Bailey, David L. Fey, jerry M. Motooka and Bruce H. Roushey Abstract A detailed geochemical survey was conducted in 1991 in the west-central part of the Howard Pass quadrangle in the vicinity of three geochemically anomalous drainage basins that were delineated during a previous geochemical survey. These basins have similar geologic settings and geochemical signatures to those surrounding the Red Dog deposit in the westernmost Brooks Range. The Red Dog deposit is a zinc-lead-silver massive sulfide and barite deposit hosted in black shale of the Lower or Middle Pennsylvanian to Lower Mississippian Kuna Formation. The three drainage basins are characterized by highly anomalous concentrations of Ag, As, Ba, Cd, Sb, and Zn in minus-30-mesh stream-sediment samples. One of these was collected from Drenchwater Creek, which was subseguently found to contain sulfide and barite mineralization similar to that at Red Dog. Two other samples-from Twistem Creek and an unnamed tributary of the Kiligwa River-represent basin areas of approximately 20 to 26 km2 The goal of the detailed geochemical sampling conducted in 1991 was to more accurately locate the mineralized areas and to possibly locate sources of the anomalies. Several geochemically anomalous basins between Twistem Creek and the easternmost tributaries of the Kiligwa River were located that were not sampled during the regional survey; two samples, representing basin areas of less than 13 km2, contain highly anomalous concentrations of Ag (2 ppm), Ba (>5,000 ppm), Cd (2.7- 4.8 ppm), and Zn (700-1 ,000 ppm) that are comparable to many of the samples from the Red Dog and Drenchwater Creek areas. These high values in sediment samples probably reflect sulfide and barite mineralization, although sulfide minerals other than pyrite have not yet been identified. INTRODUCTION The Howard Pass quadrangle is located in the western Brooks Range, partially within the southern National Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Petroleum Reserve in Alaska (NPRA) (fig. 1). It lies within an east-west-trending fold-thrust belt of Paleozoic and Mesozoic sedimentary rocks that extends along the entire western and central Brooks Range. Mississippian rocks within this belt host several massive base-metal sulfide and barite mineral occurrences and deposits (fig. 1). Largest among these is the Red Dog deposit in the westernmost Brooks Range, a shale-hosted stratiform zinclead-silver-barite deposit now in production (Plahuta, 1978; Moore and others, 1986; Lange and Nokleberg, 1987). In 1977-78, a regional geochemical survey was conducted as part of the mineral resource assessment of the southern NPRA (Churkin and others, 1978), which included most of the Howard Pass quadrangle. Streamsediment samples were collected at a sample density of approximately 1 sample per 25 km2 (Theobald and Barton, 1978). Although not within the NPRA, nine samples were also collected from Red Dog Creek and other drainages surrounding the Red Dog deposit in order to characterize the geochemical associations and concentrations of elements characteristic of this deposit type. Several sediment samples from drainage basins within the Howard Pass quadrangle that are underlain by lithologies similar to those that host the Red Dog deposit contain anomalous concentrations of base metals and barium. One of these geochemically anomalous areas is Drenchwater Creek in the west-central part of the Howard Pass quadrangle. Followup studies led to the discovery of the .Drenchwater Creek deposit (fig. 1), a base-metal sulfide deposit in Mississippian(?) chert and shale (Nokleberg and Winkler, 1978; Churkin and others, 1978; Jansons and Baggs, 1980). The 1977-78 regional geochemical survey also identified two drainage basins east of Drenchwater Creek containing the same geochemical signature. The basins are relatively large areas (approximately 20-26 km2),

and therefore locating the source of the anomalies may be difficult, particularly if the source is poorly exposed. In the summer of 1991, detailed geochemical sampling was conducted in the vicinity of these two anomalous drainage basins in order to better define the anomalous areas, and to possibly locate sources of the anomalies. Rock samples were collected from float and outcrop, and streams were sampled at an average density of 1 sample per 7 km2, primarily from first- or second-order basins. This paper presents an interpretation of the geochemical results of the detailed sampling program. METHODS During the regional geochemical survey in 1977-78, 583 stream sediments were collected, dried, and sieved to minus-30-mesh. Heavy-mineral-concentrate samples were also collected by panning bulk stream sediment at each site. Magnetic and heavy-liquid separations were used to obtain a nonmagnetic heavy-mineral-concentrate fraction. The minus-30-mesh stream-sediment and nonmagnetic heavy-·mineral-concentrate samples were analyzed for 31 elements by semiquantitative emission spectrography (ES) (Grimes and Marranzino, 1968). Re71 50 100 KILOMETERS suits of the ES analyses are included in Theobald and Barton (1978). In 1990, the same samples'were submitted for additional analyses by inductively coupled plasma-atomic emission spectrography (ICP-AES) using an organometallic halide partial extraction method (Motooka, 1988), which was developed as an exploration method for low-level analysis of 10 ore-related elements (Ag, As, Au, Bi, Cd, Cu, Mo, Pb, Sb, and Zn). Analytical results of the ICP-AES analyses are included in Kelley and others (1990). During the detailed sampling program in 1991, 4 7 stream-sediment and nonmagnetic heavy-mineralconcentrate samples were collected from within the study area and prepared by methods similar to those used for the regional survey. In addition, 32 rock samples were collected either from float in the stream drainages or from outcrop. All samples were analyzed for 31 elements by ES, and the. minus-30-mesh stream sediments and rock samples were analyzed for 10 elements using the organometallic halide partial · extraction of Motooka (1988). In addition, the samples were analyzed for 40 elements by ICP-AES, using a low-temperature multiacid digestion (Crock and others, 1983), resulting in total dissolution of the sample as opposed to the partial extraction used in the 10-element ICP-AES method. RANGE EXPLANATION )( Massive sulfide deposit or occurrence Quadrangle boundary Figure 1. Location of study area, western Brooks Range, Alaska. Geochemically Anomalous Areas in the West-Central Part of the Howard Pass Quadrangle

GEOLOGY The Brooks Range is a fold and thrust belt trending east-west for approximately 800 km across northern Alaska that formed during an orogenic event that began in Late Jurassic time and culminated during the midCretaceous. During this orogeny, a stable continental shelf that contained Devonian to Lower and Middle Jurassic sedimentary rocks was broken up and telescoped northward by numerous thrust sheets (Mayfield and others, 1983). The presence of these stacked and folded Paleozoic and Mesozoic allochthonous thrust sequences has been recognized by several authors, including Ellersieck and others ( 1979), Mull (1982), Mayfield and others (1983), and Moore and others (1986). In the western Brooks Range, at least seven major sedimentary and igneous allochthons have been recognized (Mayfield and others, 1983; Moore and others, 1986). The five structurally lowest allochthons are composed dominantly of sedimentary rocks, whereas the two highest allochthons are composed mainly of volcanic and plutonic rocks. The allochthons dominated by sedimentary rocks are composed of Mississippian through Cretaceous chert, shale, limestone, and sandstone (Mayfield and others, 1983). The lowest is termed the Brooks Range allochthon in the western Brooks Range and the Endicott Mountains allochthon in the central Brooks Range. It is continuously exposed in erosional windows as far west as the Red Dog Creek area in the DeLong Mountains quadrangle, across the Misheguk Mountain quadrangle, and in parts of the Howard Pass quadrangle. All of the known massive . sulfide occurrences and deposits are hosted by Mississippian to Pennsylvanian chert and shale of this lowermost allochthon (Nokleberg and Winkler, 1982; Moore and others, 1986). The study area comprises mainly rocks of the lowermost allochthon, consisting primarily of Mississippian to Jurassic sedimentary rocks (fig. 2). The oldest rocks in- · elude Mississippian to Pennsylvanian dark-gray to black chert, siliceous shale, and silicified mudstone (Tailleur and others, 1966); these rocks are probably lithologically equivalent to the Kuna Formation of the Lisburne Group (Mull and others, 1982; Young and Moore, 1987), which hosts the Red Dog deposit (Moore and others, 1986). Locally in the Drenchwater Creek area, volcanic and volcaniclastic rocks are found within this Mississippian and Pennsylvanian section (Nokleberg and Winkler, 1982; Lange and others, 1985). Disconformably overlying the Mississippian and Pennsylvanian rocks are rocks of the Etivluk Group (Mull and others, 1982). Within the lowermost allochthon, the Etivluk Group consists of the Permian Siksikpuk Formation and the Lower Triassic to Middle Jurassic Otuk Formation (Mull and others, 1982), rocks formerly assigned to the Shublik Formation (Tailleur and Geologic Studies in Alaska by the U.S. Geological Survey, 1991 others, 1966). The Siksikpuk Formation is composed of a 100 to 150 m thick sequence of gray to maroon siltstone, mudstone, and chert, with an upper gray shale horizon. The overlying Otuk Formation is about 100 to 150m thick and consists of dark-gray to black shale that grades up successively into dark-gray to black, thinly interbedded, siliceous limestone ·and shale containi~g abundant pelecypod fossils (Mull and others, 1982). Locally in the southwest and northeast parts of the study area, chert and shale of Pennsylvanian to Jurassic age contain numerous mafic sills and dikes. These sedimentary and intrusive rocks are shown as a separate unit on figure 2. REGIONAL GEOCHEMICAL SURVEY During the 1977-78 regional geochemical survey of the southern NPRA, 583 minus-30-mesh stream-sediment and nonmagnetic heavy-mineral-concentrate samples were collected from the Howard Pass and Misheguk Mountain quadrangles, and 9 samples were collected from drainages surrounding the Red Dog massive sulfide deposit in the DeLong Mountains quadrangle (Theobald and Barton, 1978). The following generalizations can be made, based on data from the Red Dog area, regarding geochemical exploration specifically for stratabound sediment-hosted ZnPb-Ag deposits in the arctic environment: 1. Minus-30-mesh stream-sediment samples are better at targeting mineralized areas than nonmagnetic heavymineral-concentrate samples. None of the heavy-mineral concentrates surrounding Red Dog contain detectable Ag (1 ppm lower determination limit), two contain high concentrations of Zn (500-700 ppm), and three contain high concentrations of Pb (500-5,000 ppm) (Theobald and Barton, 1978). Stream-sediment samples, on the other hand, contain extremely high concentrations of all three elements. For example, one sediment sample collected near the discovery outcrop contains >20,000 ppm Pb and 2,000 ppm Zn (Tailleur, 1970), and immediately downstream from the deposit (less than 2.5 km), samples contain as much as 10 ppm Ag, 10,000 ppm Pb, and 6,000 ppm Zn (Theobald and Barton, 1978; Kelley and others, 1990). These data suggest that either (1) weathering of this deposit type in the arctic environment is primarily chemical, resulting in a higher degree of hydromorphic dispersion relative to mechanical dispersion, or (2) the sulfide minerals are so fine grained (clay sized) that they are removed during panning. Only minus-30-mesh stream-sediment data are considered further. 2. Zinc and Ag are the best pathfinder elements, but As, Ba, Cd, and Sb are also reasonably good. Lead is not a good pathfinder element in minus-30-mesh stream-sediment samples. Except for samples collected immediately downstream from the deposit, most sediment

samples contain background concentrations of Pb (2530 ppm) (Kelley and others, 1990). Lead is typically considered to be less mobile than Zn or Ag, and thus is not transported as readily during chemical weathering (Levinson, 1974). High concentrations of Zn and Ag, and not Pb, in minus-30-mesh stream-sediment samples is further evidence that hydromorphic processes are important in this environment. For this study, the regional-scale minus-30-mesh streamsediment data from the Misheguk Mountain and Howard Pass quadrangles were evaluated statistically and by means of histograms showing the frequency distribution of the data. Threshold concentrations for selected elements in minus-30mesh stream-sediment samples were determined by evaluation of the histograms and of the spatial distribution of the data, and by comparison with concentrations in sediment samples from the vicinity of the Red Dog deposit (table 1). Concentrations equal to or greater than these threshold concentrations are considered to be anomalous and probably related to mineralizing processes .

Permian to Jurassic sedimentary rocks---Gray, green, and red marine chert, siltstone, and shale, and minor gray to black shale and limestone; Etivluk Group Pennsylvanian to Jurassic sedimentary and igneous rocks Chert, limestone, and shale with included mafic sills and dikes Mississippian and Pennsylvanian sedimentary rocks--Dark-gray and black chert, shale, silicified limestone and mudstone, and lesser limestone and dolostone; Lisburne Group Figure 2. Generalized bedrock geologic map showing Mississippian to jurassic units in study area [com pi led by j.S. Schmidt and others, written commun. (1990) from previously published maps (Tailleur and others, 1966)]. Geochemically Anomalous Areas in the West-Central Part of the Howard Pass Quadrangle

Anomalous concentrations of Ag, As, Ba, Cd, Sb, and Zn occur in minus-30-mesh sediment samples surrounding the Drench water Creek deposit (table 1; fig. 3). The best indicators of mineralization appear to be Ag (0.72-1.4 ppm), Ba (7,000->20,000 ppm), and Zn (460800 ppm), although most of the samples also contain anomalous As (14-15 ppm), Cd (1.4-3.6 ppm), and Sb (1.9-2.7 ppm). Manganese, which is considered to be a good pathfinder element for some sediment-hosted massive sulfide deposits (Maynard, 1983, 1991 ), is also anomalous in two of the samples. Lead concentrations are mostly below threshold values, although one sample contains 90 ppm. Two additional drainages east of Drenchwater Creek are delineated by the same geochemical association (fig. 3), and are underlain by Mississippian to Jurassic shale and chert (fig. 2). These drainages are Twistem Creek and an unnamed tributary of the Kiligwa River east of Twistem Creek (fig. 3). The basin areas represented are relatively large (approximately 20-26 158°50' km2), and therefore detailed studies were conducted in order to more accurately locate mineralized areas. DETAILED GEOCHEMICAL SURVEY During the 1991 detailed geochemical survey, stream sediment samples were collected from the tributaries of Wager and Twistem Creeks and from the Kiligwa and Kuna Rivers (fig. 4). Most of the samples collected between Twistem Creek and the tributaries of the Kiligwa River are anomalous in Ag and Zn, and many also contain high concentrations of As, Ba, Cd, and Sb (table 1; fig. 4). Two samples contain 2 ppm Ag, 2.7-4.8 ppm Cd, and 700-1,000 ppm Zn; these values are comparable to many of the highly anomalous samples from the Red Dog area and are equal to or greater than those from Drenchwater Creek (table 1). These high values most likely reflect mineralization.

040 , rna EXPLANATION Basin area .& Sediment sample site SMILES I ·, ,· 1 2 3KILOMETERS

:.· / r I : . ( r r· r· ( s /

1 / r1. · ( "' ) / I

)

r1 . ) )

l ,- / / -· ( '-· . /. /., ) r 1

w~· J

A

. ).

/ .. . .

./· .I I,/ I i J .-/

.J ( ,

1.1 /.1 ./

25' ' Figure 3. Stream-sediment sampling localities from regional1977-78 geochemical survey and drainage basins anomalous in Ag, As, Ba, Cd, Sb, and Zn. (Threshold values for these elements listed in table 1.) Geologic Studies in Alaska by the U.S. Geological Survey, 1991

One of these samples is from a small first-order tributary of Twistem Creek representing a basin area of only about 1.3 km2. The other is from a tributary of the Kiligwa River that was not sampled during the regional survey. The basin area represented by this sample is about 13 km2 (fig. 4). Almost all other stream-sediment samples collected in the Twistem Creek-Kiligwa River area contain relatively lower but anomalous concentrations of Ag, As, Cd, Sb, and Zn (fig. 4; table 1 ). All of these geochemically anomalous basins also contain anomalous concentrations of Ba (:<!:5,000 ppm, the upper determination limit by ES for this study). Manganese concentrations are also anomalous (3,000-9,800 ppm) in most of these samples. High concentrations of Ni (150-300 ppm) and Co (70-150 ppm) in several samples (table 1) probably reflect the presence of abundant pyrite, which often contains anomalously high concentrations of these elements (Levinson, 1974). Highly anomalous drainage basins Anomalous drainage basins

Anomalous sediment sample sites (this study

Anomalous sediment sample sites (previous study) 5 MILES ,· 1 2 3 KILOMETERS The entire area between Twistem Creek and the headwaters of the Kuna River is underlain by shale and chert (fig. 2), rock types that are permissive for hosting massive sulfide deposits. Chert and shale samples were collected from float or outcrop in several localities (fig. 5) and analyzed geochemically. Concentrations of selected elements are listed in table 2. For comparison, average worldwide concentrations of chert and shale are listed in table 3. Most samples collected from the study area contain concentrations of Ag and Zn that are comparable to average concentrations in shale or chert as reported in the literature. However, several samples contain relatively high concentrations of selected elements. For example, one chert sample (KWR007R) collected from outcrop contains high Ag (10 ppm byES), Mn (5,000 ppm), and Zn (200 ppm). In addition, black pyritic shale collected from float in the same drainage (KWR005R) contains 1 ppm Ag, 260 ppm As and 100 ppm Cu. Silty shale collected from float farther west (

( Figure 4. Drainage basin areas delineated during this study that are geochemically anomalous in Ag, As, Cd, Sb, and Zn. All of these basins are also anomalous in Ba (0!:5,000 ppm). Values in parts per million. Geochemically Anomalous Areas in the West-Central Part of the Howard Pass Quadrangle

(KWR033R) contains 10 ppm As and 270 ppm Zn. Several samples of gray chert and greenish-gray shale contain visible barite and have correspondingly anomalous concentrations of Ba (>5,000 ppm). Sulfide minerals other than pyrite have not yet been positively identified. Petrographic studies on many of the shale and chert samples with high concentrations of base metals are in progress. CONCLUSIONS Minus-30-mesh stream-sediment samples are more effective than nonmagnetic heavy-mineral-concentrate samples in delineating stratiform massive sulfide mineralization at the Red Dog and Drenchwater Creek deposits. Silver and Zn are the most reliable pathfinder elements, although As, Ba, Cd, and Sb are also effective; lead is not an effective pathfinder element for this type of deposit in the arctic environment. ·The anomalous geochemical association (Ag-As-Ba-Cd-Sb-Zn) is also 5 MILES I ·, ,· 1 2 3 KILOMETERS present in samples collected between Twistem Creek and the easternmost tributaries of the Kiligwa River. Several geochemically anomalous basins were located that were not delineated by the regional survey owing to the low sample density. Two of these contain concentrations of Ag, As, Cd, Co, Mn, Ni, Sb, and Zn that are equal to or greater than concentrations in samples from Drenchwater Creek and similar to most samples from the Red Dog area. These high concentrations most likely reflect sulfide mineralization, although only pyrite was observed in the field. The two basin areas represented by these high concentrations are less than 13 km2 in area, compared with the previously delineated basins that were 20 to 26 km2 in area. Several chert or shale samples collected during this study contain high concentrations of Ag (up to 10 ppm), As (23260 ppm), Ba (>5,000 ppm), Mn (3,000-5,000 ppm), and Zn (200-270 ppm). These values are relatively high compared with average concentrations of these elements in unaltered and unmineralized chert and shale as reported in the literature and may reflect sulfide mineralization. Figure 5. Locations of rock samples collected from study area. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Table 1. Threshold concentrations and ranges of anomalous concentrations of selected elements in minus-30-mesh streamsediment samples from this study and comparison with results from the Red Dog and Drenchwater Creek areas [All values in parts per million (ppm); n, number of samples) Element Threshold concentration Range in concentrations in sediment samples from this study+ (n=47) Red Dog Creek area· (n=9) Drenchwater Creek area (n=4) Ag As Ba 1 5,000 Cd Co Cu M n 3,000 N i p b Sb Zn 5,000-> 5,000 3,000-9,800 300-1,000 >20,000 not anomalous not anomalous not anomalous 30-3,400 300-6,000 7,000->20,000 not anomalous 3,000 *Most anomalous concentrations are from a sample collected <2.5 km downstream from deposit where exposed at the surface. +Includes only concentration ranges equal to or greater than threshold values 1Upper limit of determination for Ba is 20,000 ppm in the regional survey (Theobald and Barton, 1978) and 5,000 ppm for this study. Acknowledgments.-We wish to thank John Gray and Gary Now Ian for their constructive and helpful reviews of this manuscript. Discussions with Gil Mull of the Alaska Division of Geological and Geophysical Surveys greatly improved our understanding of the geology of the area. We would also like to thank our helicopter pilot, Jim Tasakos, for getting us safely to and from the study area, often in spite of some typically unpredictable weather conditions. REFERENCES CITED Adachi, M., Yamamoto, K., and Sugisaki, R., 1986, Hydrothermal chert and associated siliceous rocks from the northern Pacific: Their geologic significance as indication of ocean ridge activity: Sedimentary Geology, v. 47, p. 125-148. Churkin, Michael, Jr., Mayfield, C.F., Theobald, P.K., Barton, H.N., Nokleberg, W.J., Winkler, G.R., and Huie, C., 1978, Geological and geochemical appraisal of metallic mineral · resources, southern National Petroleum Reserve in Alaska: U.S. Geological Survey Open-File Report 78-70A, 82 p. Crock, J.G., Lichte, F.E., and Briggs, P.H., 1983, Determination of elements in National Bureau of Standards Geologic Reference Material SRM 278 obsidian and SRM 688 basalt by inductively coupled argon plasma-atomic emission spectrometry: Geostandards Newsletter, v. 7, p. 335-340. Ellersieck, 1., Mayfield, C.F., Tailleur, I.L., and Curtis, S.M., 1979, Thrust sequences in the Misheguk Mountain quadrangle, Brooks Range, Alaska: U.S. Geological Survey Circular 804-B, p. B8-B9. Grimes, D.J., and Marranzino, A.P., 1968, Direct-current arc and alternating-current spark emission spectrographic field methods for the semiquantitative analysis of geologic materials: U.S. Geological Survey Circular 591, 6 p. Jansons, Uldis, and Baggs, D.W., 1980, Mineral investigations of the Misheguk Mountain and Howard Pass quadrangles, Alaska: U.S. Bureau of Mines Open-File Report 38-80, 76 p. Kelley, K.D., Slaughter, K.E., and Motooka, J.M, 1990, Results of inductively coupled plasma-atomic emission spectroscopy of minus 30-mesh stream sediment samples from within and adjacent to the National Petroleum Reserve, Alaska: U. S. Geological Survey Open-File Report 90-501, 19 p., 1 sheet, scale 1:500,000. Lange, I.M., and Nokleberg, W .J ., 1987, Geologic setting, petrology, and geochemistry of stratiform sphalerite-galenabarite deposits, Red Dog Creek and Drenchwater Creek areas, northwestern Brooks Range, Alaska-A Reply to Young and Moore, 1987: Economic Geology, v. 82, no. 4, p. 1079-1081. Lange, I.M., Nokleberg, W.J., Plahuta, J.T., Krouse, H.R., and Doe, B.R., 1985, Geologic setting, petrology, and geochemistry of stratiform sphalerite-galena-barite deposits, Red Dog Creek and Drenchwater Creek areas, northwestern Brooks Range, Alaska: Economic Geology, v. 80,p. 1896-1926. Levinson, A.A., 1974, Introduction to exploration geochemistry: Calgary, Alberta, Canada, Applied Publishing Ltd., 612 p. Geochemically Anomalous Areas in the West-Central Part of the Howard Pass Quadrangle

0"1

tD Q Q Table 2. Concentrations of selected elements in rock samples from the study area OCI ;:;· 2' [All values in parts per million (ppm); all samples from float unless otherwise noted by oc (outcrop). Abbreviations: ICP-P, inductively couplt;d plasma-atomic emission spectrography, partial digestion; lCP-T, inductively coupled plasmaa. ;· atomic emission spectrography, total digestion; ES, emission spectrography; N, not detected at the lower limit of determination shown] ii Sample Number Ag(ICP-P) Ag(ES) As(ICP-P) As(ICP-T) Ba(ES) Cd(ICP-P) Cu(ICP-T) Mn(ICP-T) Sb(ICP-P) Pb(ICP-P) Zn(ICP-T) Rock description "' :11:" AI cr 1HP311R1 N .5 <10 1,500 N .67 chert; oc tD c: 1HP315R N .067 N .5 <10 shale

1HP321R1 <10 shale tD 1HP323R2 <10 1,500 N .67 chert Q Q 1HP335R N .067 N .5 N .67 <10 N .67 chert OCI ;:;· 1HP336R N .67 <10 N .67 black chert; oc KWR001R N .067 N .5 <10 N.67 black chert 5; KWR002R <10 N .67 black chert

tD KWR003R N .067 silicified rock with pyrite KWR004R <.5 <10 1,500 N .67 boxwork silica/chert fragments loC KWR005R black shale with pyrite

KWR006R <10 1,500 N .67 chert KWR007R chert with FejMn stain; oc KWROOBR <10 black chert; oc KWR009R <10 chert with Fe/Mn stain; oc KWR010R <10 .n gray-blue chert; oc KWR011R N.5 <10 N .020 chert with Fe stain; oc KWR013R N .5 N .020 gray chert; oc KWR016R N.5 <10 N .67 cherty limestone; oc KWR017R N .67 black chert with Fe stain KWR018R N .067 N.5 N .67 <10 N .67 gray chert KWR019R N.5 <10 N.67 gray chert KWR020R N .067 N .5 N .67 <10 N .020 N .67 siltstone KWR021R N .067 N.5 N .67 <10 N .67 greenish gray silty shale; oc KWR022R1 N .067 N .5 N .67 <10 N .67 shale; oc KWR024R N.5 <10 1,500 N .020 N .67 chert with Fe stain KWR025R N .067 N.5 N .67 <10 N .67 chert; oc KWR030R N .067 N.5 N .67 <10 N .67 chert with pyrite KWR031R N .067 N.5 N .67 <10 N .67 chert KWR033R N .067 N .5 N .67 silty shale KWR037R N .067 N.5 <10 chert KWR039R N.5 <10 N .67 chert with pyrite; oc

Table 3. Average worldwide abundances of selected trace elements in unaltered and unmineralized shale and chert [All values in parts per miilion (ppm). nd, no data. References: (I) Levinson, 1974; (2) Vine and Tourtelot, 1970; (3) Maynard, 1991; (4) Adachi and others, 1986) Rock type and reference Ag Ba Shale (1) Black shale (2)

nd nd Host shale of Pband Zn-rich deposits (3)- nd nd 6,550 nd Host chert of Pband Zn-rich deposits (3)- nd nd 5,047 nd Oceanic chert (4) nd nd nd Continental chert (4)--- nd nd nd Mayfield, C.F., Tailleur, I.L., and Ellersieck, 1., 1983, Stratigraphy, structure, and palinspastic analysis of the western Brooks Range, northwestern Alaska: U.S. Geological Survey Open-File Report 83-779, 58 p. Maynard, J.B., 1983, Geochemistry of sedimentary ore deposits: Springer-Verlag, New York, 305 pp. ---1991, Shale-hosted deposits of Pb, Zn, and Ba: Syngenetic deposition from exhaled brines in deep marine basins, in Force, E.R., Eidel, J.J., and Maynard, J.B., eds., Sedimentary and diagenetic mineral deposits: A basin analysis approach to exploration: Reviews in Economic Geology, v. 5., p. 177-183. Moore, D.W., Young, L.E., Modene, J.S., and Plahuta, J.T., 1986, Geologic setting and genesis of the Red Dog zinclead-silver deposit, western Brooks Range, Alaska: Economic Geology, v. 81, p. 1696-1727. Motooka, J .M., 1988, An exploration geochemical technique for the determination of preconcentrated organometallic halides by ICP-AES: Applied Spectroscopy, v. 42, p. 1293-1296. Mull, C.G., 1982, Tectonic evolution and structural style of the Brooks Range, Alaska: An illustrated summary, in Powers, R.B., ed., Geological studies of the Cordilleran thrust belt: Denver, Colorado, Rocky Mountain Association of Geologists, v. 1, p. 1-45. Mull, C.G., Tailleur, I.L., Mayfield, C.F., Ellersieck, 1., and Curtis, S., 1982, New upper Paleozoic and lower Mesozoic stratigraphic units, central and western Brooks Range, Alaska: American Association of Petroleum Geologists Bulletin, v. 66, p. 348-362. Nokleberg, W.J., and Winkler, G.R., 1978, Geologic setting of Co Cr Cu Mn Ni Pb Sb Zn nd nd nd nd nd nd nd nd nd the lead and zinc deposits, Drenchwater Creek area, Howard Pass quadrangle, western Brooks Range, Alaska: U.S. Geological Survey Open-File Report 78-70C, 16 p. ---1982, Stratiform zinc-lead deposits in the Drenchwater Creek area, Howard Pass quadrangle, northwestern Brooks Range, Alaska: U.S. Geological Survey Professional Paper 1209, 22 p., 2 sheets, scale 1:19,800. Plahuta, J.T., 1978, Geologic map and cross sections of the Red Dog prospect, DeLong Mountains, northwestern Alaska: U.S. Bureau of Mines Open-File Report 65-78, 11 p., scale 1:20,000. Tailleur, I.L., 1970, Lead-, zinc, and barite-bearing samples from the western Brooks Range, Alaska: U.S. Geological Survey Open- File Report 70-445, p. 1-16. Tailleur, I.L., Kent, B.H., Jr., and Reiser, H.N., 1966, Outcrop geologic maps of the Nuka-Etivluk region, northern Alaska: U.S. Geological Survey Open-File Report 66-128, scale 1:63,360. Theobald, P.K., and Barton, H.N., 1978, Basic data for the geochemical evaluation of National Petroleum Reserve, Alaska: U.S. Geological Survey Open-File Report 78-70D, 15 p. Vine, J.D., and Tourtelot, E.B., 1970, Geochemistry of black shale deposits-A summary report: Economic Geology, v. 65, p. 253- 272. Young, L.E., and Moore, D.W., 1987, A discussion of "Geologic setting, petrology, and geochemistry of stratiform sphalerite- galena-barite deposits, Red Dog Creek and Drenchwater Creek areas, northwestern Brooks Range, Alaska": Economic Geology, v. 82, no. 4, p. 1077-1079. Reviewers: John E. Gray and Gary Nowlan Geochemically Anomalous Areas in the West-Central Part of the Howard Pass Quadrangle

A Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/Windfall Harbor and Gambier Bay Areas, Admiralty Island, Southeast Alaska By Cliff D. Taylor, Barrett A. Cieutat, and Lance D. Miller Abstract A followup geochemical study was carried out on Admiralty Island to more accurately locate and, if possible, to identify the sources of anomalies characteristic of Kuroko-type volcanogenic massive sulfide deposits similar to the nearby Greens Creek and Pyrola deposits. Detailed rock, stream-sediment, and panned-concentrate sampling in the Ward Creek/Windfall Harbor area delineates a narrow belt of weakly anomalous samples that are restricted to a sequence of Permian to Late Triassic mafic volcanic and flysch-type sedimentary rocks exposed along· most of the eastern shore of the island. We interpret the distribution of anomalous samples to indicate that high background concentrations of ore-related elements are a characteristic of this rock sequence. Several highly anomalous samples may indicate the presence of small mineralized occurrences. Detailed sampling in the Gambier Bay area led to the discovery of a structurally controlled CuZn-Ag-Ba mineral occurrence of unknown extent. The occurrence is located on the north shore of Gambier Bay at the transition between the basal sedimentary to mafic volcanic units of the Late Triassic Hyd Group. INTRODUCTION Admiralty Island is located 15 km south of Juneau in southeast Alaska (fig. 1). The island is 135 km by 50 km and is elongate in a north-south direction. Statistical analysis of the stream- and lake-sediment geochemical data collected during the National Uranium Resource Evaluation (NURE) program identified three anomalous areas that have geochemical characteristics interpreted to . represent volcanogenic massive sulfide (VMS) mineral occurrences (Kelley, 1990). One of these areas encompasses the drainages .in the vicinity of the Greens Creek Kuroko-type VMS deposit (fig. 1 ). The other two areas-Ward Creek/Windfall Harbor and Gambier/Pybus Bay-contain no known VMS occurrences. All three Geologic Studies in Alaska by the U.S. Geological Survey, 1991 areas lie within a sequence of Permian and Triassic mafic volcanic and flysch-type sedimentary rocks. This sequence trends northwest across the length of the island and is part of a belt of Permian and Triassic arc-related volcano-sedimentary rocks extending 300 km from Juneau to Ketchikan that hosts numerous VMS occurrences and deposits (Berg and Grybeck, 1980; Goldfarb and others, 1987). This paper presents the results of detailed followup geochemical sampling programs designed to further constrain and, if possible, to locate the source of geochemical anomalies in the Ward Creek/Windfall Harbor and Gambier/Pybus Bay areas. We also describe a structurally controlled Cu-Zn-Ag-Ba mineral occurrence of unknown extent discovered during the followup work in the Gambier/Pybus Bay area. GEOLOGY Most of Admiralty Island is part of the allochthonous Alexander terrane (Monger and Berg, 1987). The Gravina-Nutzotin belt overlap assemblage (fig. 1) is adjacent to and overlies the Alexander terrane and crops out on the Glass Peninsula and the islands in the Seymour Canal along the eastern side of the island (Berg and others, 1978; Rubin and Saleeby, 1991}. The Alexander terrane consists of Ordovician to Late Triassic marine sedimentary, volcanic, and plutonic rocks formed during intermittent volcanic arc activity along a convergent continental margin (Gehrels and Saleeby, 1987; Rubin and Saleeby, 1991). The Gravina-Nutzotin belt (Berg and others, 1972) consists of Late Jurassic to midCretaceous marine sedimentary rocks, intertonguing andesitic to basaltic volcanic rocks, and quartz diorite to dunite and peridotite plutonic rocks. They were formed adjacent to and upon the Alexander terrane following a brief period of Late Triassic rifting. Rifting is thought to have occurred as a result of back-arc spreading (Rubin

134" 30' t , J 5 KILOMETERS er - sao 56' D I" '

EXPLANATION Tertiary rocks-Andesitic basalt flows; Admiralty Island Volcanics Tertiary rocks-Conglomerate, sandstone, shale, coal; Kootznahoo Formation Cretaceous rocks-Intermediate, mafic, and ';, ultramafic plutonic rocks

mm lliJj

Cretaceous and Jurassic rocks-Argillite, conglomerate, and graywacke; Seymour Canal Formation Cretaceous and Jurassic rocks-Basaltic or andesitic flow breccia; Douglas Island Volcanics Triassic and Permian rocks-Volcanic and sedimentary rocks Triassic rocks-Chert, argillite, graywacke, carbonate rocks, and intermediate to mafic volcanic rocks. Consists of the Hyd Group, as mapped, and the Retreat Group and Gambier Bay Formation Permian rocks-Argillite, graywacke, chert, and minor conglomerate, limestone, and volcanic rocks; Cannery Formation Devonian rocks-Marble enclosed tectonically within Gambier Bay Formation Silurian and Ordovician rocks-Argillite, chert, graywacke, and minor volcanic rocks; includes Ordovician Hood Bay Formation and sedimentary rocks Mesozoic or Paleozoic rocksMetasedimentary and metaigneous rocks Ag-Au-Pb-Zn deposit Cu (±Au) occurrences Figure 1. Geologic map of Admiralty Island. Generalized from Lath ram and others (1965) and Karl (1989); modified from Kelley (1990). Inset shows location of Admiralty Island and contact between Alexander terrane and Gravina-Nutzotin belt. Modified from Goldfarb and others (1987) and Monger and Berg (1987). Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/ Windfall Harbor and Gambier Bay Areas, Admiralty Island

and Saleeby, 1991) or possibly pull-apart rifting in response to transcurrent (right lateral) movement of the volcanic arc prior to its accretion to the continental margin (Ford and Brew, 1988). For the purposes of this paper, only the Early Permian to Late Triassic rocks favorable for VMS-type mineralization are described. However, rocks on the island range in age from Ordovician to Tertiary and are volcanic, plutonic, and sedimentary. Brief descriptions and the distribution of these rocks on the island are provided in figures 1-4. Excellent descriptions of all the rocks on Admiralty Island can be found in ~oney (1964), Lathram and others (1965), and Ford and others (1989). The Early Permian Cannery Formation consists of thin-bedded argillite, graywacke, chert, and minor volcanic flows and breccias representing a deep-water depositional environment and a period of submarine deposition (Ford and others, 1989). The Cannery Formation is present at the headwaters of Pack and Greens Creeks, in the vicinity of Windfall Harbor, and as a belt along the east side of Hasselborg Lake extending south from the head of Windfall Harbor (figs. 1-3). In the area of Pybus and Gambier Bays, the Cannery Formation is overlain by dolomite of the Permian Pybus Formation (fig. 4, inset). This unit is a light-brownish-gray, fossiliferous, cherty dolostone that is inferred to indicate a shallow-marine depositional environment (Ford and others, 1989). Two probably correlative parts of a discontinuous sequence of Triassic rocks form most of the central and western parts of the island and stretch from the Mansfield Peninsula to the southern shore of Gambier Bay. A Cretaceous batholith at Thayer Lake divides the Retreat Group to the north from the Gambier Bay For- ·mation to the south (Lathram and others, 1965). Several varieties of schist and lesser phyllite and argillite compose the rocks of this belt; the most common lithic types are chlorite-albite-epidote and calc-silicate schist. These rocks have been intensely deformed and variably metamorphosed up to greenschist facies. Originally thought to be Late Triassic(?) to Early Cretaceous(?) in age (Barker, 1957), the age of these rocks was revised to Middle(?) Devonian on the basis of poorly preserved fossils in the prominent marble exposures north and south of Hood Bay and north of Gambier Bay that were thought to be in depositional contact with the Gambier Bay Formation (Loney, 1964). Recent identification of fossils in argillites of the Retreat Group suggests a Late Triassic age for these rocks (Ford and others, 1989), and recent work suggests that the Devonian marbles may be olistostromal blocks (Brew and others, 1991). These rocks are thought to represent renewed volcanism and deposition of volcanic rocks and flysch-type sedimentary rocks in a relatively deep-water arc-slope environment (Ford and others, 1989). Geologic Studies in Alaska by the U.S. Geological Survey, 1991 The Late Triassic Hyd Group consists of a relatively undeformed sedimentary and volcanic section unconformably overlying Permian rocks along the eastern side of the island. The lithic components of the sedimentary and volcanic section are unequally distributed; volcanic rocks compose most of the section on the north shore of Gambier Bay. The sedimentary sequence consists of a basal sedimentary breccia of white chert and dolostone or green and red chert, overlain by a middle interval of brown limestone and an upper interval of thin-bedded argillite, chert, limestone, and minor spilitic lavas. The volcanic rocks are characterized by massiveto thick-bedded, jasper-bearing, red and green, amygdaloidal, altered, basic lava flows that thicken rapidly northward through Windfall Harbor. The sedimentary rocks of the section are most extensive in the Gambier/ Pybus Bay area and thin rapidly northward (Lathram and others, 1965). The Hyd Group rocks are the youngest rocks of the· Alexander terrane and are interpreted to represent a period of uplift and shallow-water carbonate sedimentation followed by rifting and basaltic volcanism prior to deposition of Gravina-Nutzotin belt sediments (Brew and Ford, 1984; Gehrels and Saleeby, 1987; Ford and others, 1989). METHODS During this study, 48 stream-sediment and pannedconcentrate samples were collected from first- and second-order streams in the Ward Creek/Windfall Harbor area at a sampling density of 1 sample per 13 km2. At each site, a composite sample from a 5- to 10-m section of the active channel was sieved through a 10-mesh (2.0 · mm) screen. Approximately 1 to 2 kg was saved as a stream-sediment sample. A panned concentrate was produced by panning 6 to 8 kg of the screened material to remove most of the quartz, feldspar, and organic and clay-sized material. In addition, 64 rock samples with visible alteration, veining, and mineralization were collected from float and outcrop in an attempt to identify possible sources of the geochemical anomalies. In the Gambier/Pybus Bay area, 4 stream-sediment and pannedconcentrate samples and 43 rock samples from float and outcrop were collected. Of the rock samples, only the 41 samples collected at the newly discovered mineral occurrence are discussed. In the laboratory, stream sediments were sieved to -200 mesh (0.075 mm). Rock samples were first crushed and then pulverized using ceramic-plate grinders. The panned-concentrate samples were sieved to -35 mesh (0.5 mm) and separated magnetically and by heavy liquids to produce a nonmagnetic heavy-mineral-concentrate. The concentrate was split, and one fraction was hand ground for chemical analysis. The second split

Ki KJsc 'Rh 'Rr Pc MzF2m MzF2u

//

./ ! Explanation Intrusive rocks (Cretaceous) Seymour Canal Formation (Cretaceous and Jurassic) Hyd Group (Triassic) Retreat Group (Triassic) Cannery Formation (Permian) Metamorphic rocks (Mesozoic and Paleozoic) Sedimentary, metamorphic, and intrusive rocks, undivided (Mesozoic and Paleozoic) NURE Sample site Fault Contact ,

Swan

ISLAND Kilometers Factor 3 As-Ba-Zn-Ni Factor 5 Pb-Zn-Cu Factors 3 and 5 57° 59' Figure 2. Location of geochemical anomalies resulting from statistical analysis of NURE stream- and lake-sediment data in Ward Creek;VVindfall Harbor area. R-mode factors from Kelley (1990); geology generalized from Karl (1989). Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/ Windfall Harbor and Gambier Bay Areas, Admiralty Island

was saved for detrital-grain mineral identification using a binocular microscope. The stream-sediment and rock samples were analyzed by inductively coupled plasma-atomic emission spectrography for 40 elements (ICP-AES 40) using a low-temperature multi-acid total digestion (Crock and others, 1983), and for 10 elements (ICP-AES 10) using an organometallic halide partial-extraction method (Motooka, 1988). Gold (0.002-ppm lower limit) was determined by graphite furnace atomic absorption spectrophotometry (HGA-AAS; Thompson and others, 1968) and Hg (0.02-ppm lower limit) by cold vapor atomic absorption spectrophotometry (CV -AAS; Kennedy and Crock, 1987). The nonmagnetic heavy-mineral-concentrate samples were analyzed semiquantitatively for 37 elements by DC-arc atomic emission spectrography (DC-AES; Grimes and Marranzino, 1968). The lower limits of determination for each method and selected subsets of the elements analyzed for are shown in tables 1-3. Percentages of ore and ore-related minerals in the heavy-mineral-concentrate samples are based on visual estimates. Mineralogical identification was augmented by single-grain X-ray diffraction (XRD) analyses. Occurrences and percentage estimates of ore and orerelated minerals are listed in table 2. No galena or sphalerite was found in any of the concentrates. However, sphalerite is especially difficult to identify due to its resemblance to rutile. In the rock samples from the mineral occurrence in the Gambier/Pybus Bay area, identification of bornite and silver-rich tetrahedrite (tetrahedrite slightly shifted toward friebergite; S. Sutley, oral common., 1991) was made by single-grain XRD analyses. Identification of other phases was made by reflected-light microscopy. GEOCHEMICAL STUDIES IN THE WARD CREEK/ WINDFALL HARBOR AREA The Ward Creek/Windfall Harbor area is roughly 625 km2 in size and is characterized by steep, mountainous terrain. Most of the area is composed of Permian to Late Triassic volcanic and sedimentary strata favorable for VMS-type mineralization (figs. 2, 3). During the NURE sampling program, 42 samples were collected from this area at a density of about 1 sample every 15 km2 Anomalous samples contained up to 517 ppm Zn, 267 ppm Cu, 72 ppm Pb, and 216 ppm As. A few samples also contained elevated concentrations of Ba, Fe, Mn, Ni, Cr, Co, Sb, and Bi (Kelley, 1990). R-mode factor analysis (Davis, 1986) identified two element associations that were interpreted to be indicative of VMStype mineralization. One was As, Ba, Ni, and Zn, and the other was Pb, Zn, and Cu (factors 3 and 5, respec- . tively, in fig. 2; Kelley, 1990). Anomalous scores for Geologic Studies in Alaska by the U.S. Geological Survey, 1991 these two assoctatwns plotted at 1:63,360 occur in samples from the headwaters of Ward and Pack Creeks and in samples from creeks draining into Windfall Harbor (fig. 2). Fourteen minor and trace elements that are expected to represent elements indicative of Kuroko-type VMS mineralization (Lambert and Sato, 1974; Kelley, 1990) were selected from the stream-sediment and nonmagnetic heavy-mineral-concentrate geochemical data sets for univariate statistical analysis. These elements are Fe, Mn, Ba, Ag, As, Bi, Cd, Cu, Mo, Pb, Sb, Zn, Au, and Hg. Threshold concentrations for each element were selected by comparing the 90th-percentile value to the corresponding histogram (histograms not shown). Univariate statistical summaries of the stream-sediment and nonmagnetic heavy-mineral-concentrate sample data and the threshold concentration selected for each element are shown in table 4A and 4B. The concentrations of the selected elements in anomalous stream-sediment samples and nonmagnetic heavy-mineral-concentrate samples from the Ward Creek/Windfall Harbor area are shown in tables 1 and 2, respectively. Only sites with two or more anomalous elements in the same sample media were considered in defining anomalous drainages. The resulting anomalous areas and sample-site locations are shown in figure 3. Results and Interpretation of Geochemical Data Interpretation of the data shows a distinct northwestto southeast-trending zone of anomalous samples. From the headwaters of Pack Creek to the peninsula southeast of Windfall Harbor, stream-sediment and nonmagnetic heavy-mineral-concentrate samples are characterized by elevated values of Cu, Ni, Fe, Mn, Ag, Mo, Cd, and (or) Zn. Also, there appears to be a very subtle enrichment of As, Sb, Pb, and Ba in the northwest and Hg and Au in the southeast part of the study area. To show this graphically, sample sites along the anomalous zone were projected onto a NW-SE-oriented transect (A-A', fig. 3), and the stream-sediment concentrations of the six elements mentioned were plotted as a set of individual bar graphs (fig. 5). The slight zonation in the enrichment of these elements along the NW -SE transect also corresponds with the observed distribution of the mineralogy in the nonmagnetic heavy-mineral-concentrate samples as discussed below. The belt of anomalous samples is narrower and the zonation more apparent in the streamsediment geochemistry. The anomalous area indicated by the heavy-mineral-concentrate geochemistry is' wider and is less well zoned (fig. 3). There is a distinct correlation between the trend of anomalous sample sites and the occurrence of ore and ore-related minerals in the heavy-mineral-concentrate

Ki KJsc "Rh "Rr Pc 46 Explanation Intrusive rocks (Cretaceous) Seymour Canal Formation (Cretaceous and Jurassic) Hyd Group (Triassic) Retreat Group (Triassic) Cannery Formation (Permic~m) Metamorphic rocks (Mesozoic and Paleozoic) Sedimentary, metamorphic, and intrusive rocks, undivided (Mesozoic and Paleozoic) Sample site number and site location Fault Contact

Kilometers Anomalous stream sediments Anomalous concentrates Anomalous concentrates and stream sediments 57° 59' Figure 3. Location of geochemical anomalies resulting from analysis of stream-sediment and nonmagnetic heavy-mineralconcentrate sample single-element plots from Ward Creek;Windfall Harbor area. A-A' cross-section refers to fig. 5. Geology generalized from Karl (1989). Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/ Windfall Harbor and Gambier Bay Areas, Admiralty Island

samples (table 2). The concentrates contain mostly gray to flesh-colored silicates and rock fragments of phyllite and argillite reflecting the dominant lithology at the sample site. Commonly, they also contain variable amounts of semitranslucent to opaque white barite; tarnished, oxidized, and often goethite-replaced cubic and dodecahedral pyrite; and black and blood-red to yelloworange rutile. Well-formed cubic pyrite is particularly abundant in areas where pyrite-bearing black slate and argillite are the dominant rock types. From the headwaters of Pack Creek to the peninsula southeast of Windfall Harbor, the percentages of pyrite and barite in heavy-mineral-concentrate samples is 30 to 60 percent. In many of these samples cinnabar occurs in small quantities from a few grains to 2 percent. Sites 31 and 51 on the peninsula each contain a grain of chalcopyrite. All of the samples to the northeast of this belt of anomalous samples contain 30 to 60 percent barite and several contain >30 percent pyrite. The occurrence of cinnabar and barite is distributed throughout the anomalous trend. However, the highest percentages of barite in concentrates occur at the northwest end of the trend, and the only samples containing more than 10 to 12 grains of cinnabar occur on the peninsula southeast of Windfall Harbor. With a few notable exceptions, we regard the anomalous trend to be an indication of high background concentrations in the Permian and Triassic rocks rather than an indication of proximity to significant VMS-type mineralization. This conclusion is based on the observation that the geochemistry of nearly all the samples located in areas of Permian and Triassic rocks have a low level and fairly uniform enrichment of the elements examined in comparison to the surrounding country rock. As discussed, there may be subtle variations in the nature of elemental enrichment within this package of rock, but nearly all of the samples collected in areas characterized by country rock other than the Permian through ... riassic rock package are not geochemically anomalous. The exceptions are characterized by slightly above threshold concentrations of five to seven elements that may indicate small mineral occurrences are present. The sample from site 7 contains 2,200 ppm Mn, 350 ppm As, 93 ppm Cu, 34 ppm Pb, 5.2 ppm Sb, 190 ppm Zn, and 0.008 ppm Au in stream sediment and 1,000 ppm As and Pb in the heavy-mineral concentrate. Sample 33 contains 2,700 ppm Mn, 120 ppm As, 2.8 ppm Cd, 110 ppm Cu, 49 ppm Pb, 390 ppm Zn, and 0.014 ppm Au in stream sediment and 50 percent Fe, 7 ppm Ag, and 1,000 ppm Pb in the heavy-mineral concentrate. These two sites represent upper tributaries of the same stream and together give a strong suggestion that mineralization is present. A sample (site 6) taken at the mouth of the main stream 3 km distant contains 99 ppm As, 24 ppm Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Pb, and 210 ppm Zn in stream sediment and 3 ppm Ag in the heavy-mineral concentrate. Sample 19 contains 2,000 ppm Mn, 2,100 ppm Ba, 120 ppm Ni, 0.67 ppm Ag, 9.8 ppm Sb, 180 ppm Zn, and 0.85 ppm Hg in stream sediment and 100 ppm Pb in the heavy-mineral concentrate. Sample 50, which was collected from an iron-stained area of sparse vegetation (a kill zone), contains 160 ppm Ni, 0.61 ppm Ag, 170 ppm As, 1.6 ppm Cd, 130 ppm Cu, 6. 7 ppm Mo, 6.5 ppm Sb, and 0.55 ppm Hg in stream sediment and 100 ppm Pb in the heavy-mineral concentrate. Sample 51 contains 3,400 ppm Mn, 0.80 ppm Ag, 2 ppm Cd, 9.9 ppm Mo, and 180 ppm Zn in stream-sediment and 3 ppm Ag and 300 ppm Mn in the heavy-mineral concentrate. Although microscopic examination of the heavy-mineral-concentrate samples failed to identify sphalerite, the combination of 2,000 to 5,000 ppm Zn and 50 to 500 ppm Cd in samples 12, 17, 36, 37, 41, and 43 probably indicates its presence. Geochemical analyses (not presented) obtained for rock samples collected in the Ward Creek/Windfall Harbor Area show that for the most part, the rock samples are unmineralized. However, there are a few exceptions that may relate to the anomalous trend shown by the stream-sediment and heavy-mineral-concentrate samples. A traverse was made from south to north along the divide between sites 13 and 15. At the south end, equigranular, fine crystalline, Cretaceous diorite is in contact with a monotonous sequence of undivided Paleozoic and Mesozoic argillites and phyllites. Pyrite is commonly disseminated throughout the diorite and is locally abundant. The intruded metamorphic rocks are iron stained and locally altered near the contact. At the northern end of the traverse, intrusive rocks are absent but the metamorphic rocks are similarly iron stained. Examination of the iron-altered material shows abundant pyrite on fracture surfaces. Both the pyritic diorite and the overlying metamorphic rocks contain anomalous concentrations of Cu (55 ppm) and Zn (45 ppm), and one sample of the diorite is enriched in As (170 ppm). The pyrite in the diorite and in the altered metamorphic rocks is probably the source of the anomalies shown at sites 13 and 44. The area around site 50 is characterized by intense iron staining and lack of vegetation on flat rocky patches and along several small streams. The argillites(?) are brecciated and veined. Dissolution of the argillite clasts produces a boxwork texture of quartz and calcite. Intense iron staining is associated with pyrite on fracture surfaces, and analyses show that these rocks contain 2.2 ppm Ag and 83 ppm As. At site 43, near Swan Cove, silicified material was found in float that contains minor chalcopyrite and sphalerite. The sample contains 370 ppm Zn, 110 ppm Cu, and 0.5 ppm Ag and is probably the source of the anomalous heavy-mineral-concentrate sample. Site 43 is

Table 1. Concentrations of selected elements in anomalous stream-sediment samples from the Ward Creek/Windfall Harbor area [Mn, Ba, and Ni analyses performed by ICP-AES 40; Au analyses by HGA-AAS, and Hg analyses by CV -AAS. All other analyses are by ICP-AES 10. Brackets indicate lower limit of determination. N, not detected at lower limit;<, detected at lower limit] Sample Mn Ba Ni Ag As Bi Cd (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) [4) [1] [2] [0.045] [0.60] [0.60] [0.05) N0.67 N .67 N .67 N .67 N .67 N .67 N .67 I 0 IS N .67 N .67 N .67 N .67 N .67 N .67 N .67 N .67 N .67 N .67 ISOO IOO N .67 N .67 N .67 N .67 N .67 N .67 N .67 N .67 N .67 N .67 N .67 in the vicinity of a chalcopyrite-pyrrhotite occurrence in the Hyd Group reported by Lathram and others (1965). These workers also reported another occurrence in thr. volcanic rocks of the Hyd Group on the southeast shore of Windfall Harbor between sites 1 and 2 (fig. 3). Small amounts of chalcopyrite in iron-rich quartz-carbonate veins and stockworks were found cutting mafic volcanic rocks. A sample collected from the mineralized stockwork contains 0.46 percent Cu. GEOCHEMICAL STUDIES IN THE GAMBIER/ PYBUS BAY AREA AND THE DISCOVERY OF THE NORTH GAMBIER CUZN-AG-BA OCCURRENCE During the followup geochemical sampling in the Gambier/Pybus Bay area, we conducted an examination of a NURE stream-sediment anomaly containing 2,366 ppm Ba, 361 ppm Zn, 103 ppm Cu, 27 ppm Pb, 55 ppm As, 9 ppm Sb, and 5 ppm Bi. The presence of chalcopyrite and barite in stream cobbles led to an outcrop discovery of CuCu Mo Pb Sb Zn Au Hg (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) [0.05] [0.09) [0.60) [0.60) [0.05) [0.002) [0.020) IIO N .67 liO .OI6 Zn-Ag-Ba mineralization that, to the best of our knowledge, has not been previously reported. The occurrence, here informally named the North Gambier occurrence, is located in the northwest comer of section 27 (Sitka B-1, U.S. Geological Survey 15-minute quadrangle), approximately 2 km north from the mouth of the stream that flows into Gambier Bay from the north shore (fig. 4). The earliest reports on the region (Wright and Wright, 1904; Wright, 1906, 1907) noted the occurrence and minor working of several claims near the west end of Gambier Bay. The Brown's prospect is located at 1,000-ft elevation on the north side of Cave Mountain, which separates the north and south arms of the bay. Mineralization consists of "brecciated limestone partly replaced by quartz stringers and by small masses of pyrite and chalcopyrite" along a northwest-striking "ledge" or structure that is continuous for more than 200 ft (Wright, 1906). Preliminary assays indicated low Cu-Au values. Two other Cu-Au-bearing "ledges" were also reported on the south (Cook Claim) and northeast slopes of Gambier Mountain, north of the Brown's prospect and several kilometers to the northwest of the North Gambier occurrence. Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/ Windfall Harbor and Gambier Bay Areas, Admiralty Island

North Gambier Occurrence A Mapped by L. Miller and Fe-rich ooze along joints ITtilli] Green to purple, arnygdaloidal, spilitized mafic volcanic rocks C. Taylor SampleAA Brunton and tape Prominent joint set in creek bed Hematitic graywacke with interbedded angular breccia and conglomerate Scale: 100 FEET 3' wide altered zone with joints spaced SampleR Sedimentary breccia with greenstone clasts and graywacke matrix Red, finely layered, hematitic chert Red, polymictic breccia with clasts of chert, graywacke, and greenstone in a hematitic matrix Finely layered, well-foliated quartzmuscovite schist Unit contact E.-W. joints on shear zone.\ Breccia clasts of calcite Joint to 1 .5" subhorizontal slickensides. Sample S

Shear zone or fault + 1000ft Several iron-stained joints Map position in feet SampleK 8' wide gully with joints spaced 1 ' apart Shear with horizontal slickensides 100ft. Two 3-6" wide faults E.-W. cross cuts NNE. Sample AD 42 ~oliation in quartz-muscovite schist Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Sample location Strike and dip of joints Strike and dip of bedding

Table 2. Concentrations of selected elements in anomalous nonmagnetic heavy-mineral-concentrate samples from the Ward Creek/Windfall Harbor area [All analyses by DC-AES. Brackets indicate lower limit of determination. N, not detected at lower limit;<, detected at lower limit. Abbreviations: bt, barite; py, pyrite; en, cinnabar; cpy, chalcopyrite] Sample Fe Ag As Cd Cu (percent) (ppm) (ppm) (ppm) (ppm) [0.1) [1) [500) [SO) [10) 01 N 1 N500 N 50 02 N 1 N500 N 50 03 N 1 N500 N 50 04 N 1 N500 N 50 06 N500 N 50 07 1 N 50 08 N 1 N 50 10 N500 N 50 12

N 500 13 1 N 50 15 N 1 N 500 N 50 16 N1 N5QO N 50 17 N500 18 N 500 N 50 19 N 1 N500 N 50 20 N 1 N500 N 50 21 N 1 N 500 N 50 N 500 N 50 33 N 500 N 50 36 N 500 <50 37 39 1 N500 N 50 41 43 N 50 N500 N 50 45 N500 N 50 50 N 500 N 50 51 N500 N 50 53

N500 <50 55

N 500 N 50 These historic occurrences were noted in the reconnaissance mapping of Lathram and others ( 1965) and in the more detailed mapping of Pybus and Gambier Bays by Loney ( 1964 ). Numerous other publications (Berg and Cobb, 1967; Cobb, 1972a, b; Berg and others, 1981) list these as well. In addition they list several Cu-Ni prospects hosted by rocks of the Hyd Group on the north shore of Gambier Bay, downstream and east of the North Gambier occurrence. The Kloss and Patty prospects, de-

Figure 4. Geology and locations of rock sample sites, North Gambier Cu-Zn-Ag-Ba occurrence. Inset: Gambier Bay area showing location of the North Gambier occurrence. Map units: Dgm, Devonian marble (olistostromal blocks) enclosed within Gambier Bay Formation (Triassic); Pp, dolomiteofPybus Formation (Permian); Pc, Cannery Formation (Permian); "Rg, Gambier Bay Formation (Triassic); "Rh, Hyd Group (Triassic); "RPu, sedimentary, metamorphic, and intrusive rocks, undivided (Triassic and Permian); Kjsc, Seymour Canal Formation (Cretaceous and jurassic). Dashed lines indicate faults; black box indicates area of the main part of figure 4 (modified from Lath ram and others, 1965). Mn Ni Pb Zn Ore mineralogy (ppm) (ppm) (ppm) (ppm) values in [20] [10] [20] [500] percent <20 N 500 40bt N500 50py, en N20 2en N 500 90bt 35 py. 40 bt 60 bt, en N 500 30py N 500 N 500 N500 N500 30py. 30 bt 30 py, 30 bt, en 30 bt, en N 500 30bt N500 60 py. 30 bt, 2 en, epy N500 30 py. 30 bt 50py 75 PY 50bt 30 bt 30 py. 50 bt N 500 60py N 500 30 py. 50 bt 40 bt, epy en N 500 en scribed in the literature as possibly being different names for the same occurrence, are cited to contain disseminated Cu-Ni oxides in a 150- to 200-ft-wide shear zone in the Hyd Group and a nearby parallel shear zone containing Cu mineralization (Berg and Cobb, 1967; Cobb, 1972a, b, 1978a, b). A detailed stream-sediment geochemistry program conducted by the Alaska Division of Mines and Minerals in the Gambier/Pybus Bay area reported anomalous concentrations of Cu, Pb, Zn, and Mo in samples taken along the north shore of Gambier Bay in the vicinity of the North Gambier occurrence and suggested that Pb-ZnCu mineralization might be present. This report also noted Cu-Ni occurrences in the Hyd Group on the ridge to the east and niade a vague reference to prospectors reporting a few copper occurrences in the area (Herbert and Race, 1965). Berg and others ( 1981) made reference to an unpublished Bear Creek Mining Company report written in 1978, entitled "Significant Mineral Deposits and Anomalies, Southeast Alaska." Four occurrences are "located approximately" at different places on the south end of Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/ Windfall Harbor and Gambier Bay Areas, Admiralty Island

C"l

Table 3. Concentrations of selected elements in rock samples from the North Gambier occurrence s QQ r;· [Shaded block indicates analyses perfonned by ICP-AES 40; Au analyses by HGA-AAS, and Hg analyses by CV-AAS. All other analyses are by ICP-AES 10. Brackets indicate lower limit of detennination. N, not detected at lower limit;

<,detected at lower limit. Abbreviations: Slcfd, silicified; brx, brecciated; vole, volcanic; rx, rocks; mnlzd, mineralized; bt, barite; cpy, chalcopyrite; py, pyrite; tetr, tetrahedrite; qtz, quartz; sph, sphalerite; gal, galena; bor, bornite; cc, calcite; cov, covellite; carb, carbonate; aspy, arsenopyrite]

:r

Sample Fe Mn Ni Ag As Bi Cd Cu Mo Pb Sb Zn Au Hg Sample description and ore mineralogy ;- (percent) (ppm) ippm) · (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) (ppm) s (0.0051 (0.0451 (0.601 (0.601 (0.051 (0.051 (0.091 (0.601 (0.601 (0.051 (0.0021 (0.0201 c:r A Float, slcfd bt veined mafic volcanic, cpy, py I'D c: B S.3 3S Nl.O NIO N10 N0.30 NIO NIO Polymictic brx w/ clasts of chert and vole. rx., py

c NI.O NIO N10 N.30 N10 N10 Float, green chert w/ Fe carbonate, py, cpy C"l D SA NIO N10 Float, qtz-carbonate-bt veined wacke, cpy, py, tetr I'D E N10 NIO Slcfd brx mafic volcanic w/ qtz, bt, sph, tetr, cpy

s F NIO NIO Brx mafic volcanic w/py rimmed clasts QQ G NIO 2-in qtz-carbonate-bt vein in mafic volcanic, cpy r;· H N10 NIO Slcfd brx w/ bladed bt., cpy, cut by qtz-sph-tetr-gal til I Nl.O NIO NIO N10 Float, bladed bt w/ Fe carbonate, cpy, py matrix J Nl.O NIO N10 N.90 NIO NIO N.02 Ferruginous graywacke

6s

K Nl.O N10 NIO N.30 N.90 NIO NIO N.02 Green, altered wacke or tuff L N10 NIO N.90 NIO Brx mafic volcanic, carbonate-bladed bt matrix, cpy M NIO NIO N.90 NIO Float, bladed bt in brx Fe carbonate and green chert

N Nl.O NIO N10 N.30 N10 N10 Float, massive white vein qtz w/ graphite partings 0 N10 2-in qtz-carb vein in volcanics w/ cpy, py selvage p 5S Nl.O NlO N10 N.90 NIO .005' Float, volcaniclastic w/ green chert-Fe carb matrix Q NIO NIO Float, slcfd qtz-carbonate brx w/ cpy R NlO Qtz-carbonate vein in volcanics, cpy, bor, cov, tetr s NIO NIO N.90 NIO Slcfd brx w/ chert and volcaniclasts, qtz-carb-bt-cpy T NIO N10 NIO Float, slcfd mafic volcanic qtz w/ carb-py-sph-gal-tetr u NIO NIO N.90 N10 NIO Fe carbonate altered mafic volcanic, bladed bt veins v NIO NIO N.90 NlO N.02 Purple amygdular mafic volcanic w/ cc fillings w 2() NIO N10 NlO Slcfd brx w/ qtz-carbonate-bt-cpy matrix X lO .· ·.5200 55·. N10 N10 N.90 NIO N.02 l-in cc vein in amygdular mafic volcanic y NlO NIO Foot thick Fe-rich precipitate from seep z Nl.O NlO Float,brx chert w/ py and aspy? AA NlO >iooooo Massive flow banded cpy-bor-py-qtz-cov-tetr AB Nt.o NlO NIO N.02 Slcfd brx w/ bar, cpy AC Nl.O NlO NIO Slcfd brx w/ bar, cpy AD Nl.O NlO NlO NIO NlO N.02 Altered mafic volcanic AE lO Nl.O N10 N10 Rock chip sample, bt, cpy, sph, tetr, py minlzd AF Nl.O N10 N10 NlO N.02 Rock chip sample, unminlzd mafic volcanics AG Nl.O N10 NlO NlO Rock chip sample, bt, cpy, sph, tetr, py minlzd AH Nl.O NIO N10 NlO N10 N.02 Altered wacke or tuff AI Nl.O NIO NIO N.30 N10 Finely layered ferruginous chert AJ Nl.O NlO N.30 Ferruginous polymictic sedimentary breccia AK 4;1 IS Nl.O NIO N10 N.30 Foliated and finely layered ferruginous qtz-musc schist AL Nl.O NIO N10 N.30 N10 N10 N.02 Altered mafic volcanic or tuff AM NIO N10 llO NIO N10 Slcfd brx, volcaniclasts in bt-cpy matrix AN N.067 N.67 Ferruginous greywacke AO 2W N.67 N.67 Altered amygdular mafic volcanic w/ cc fillings

Admiralty Island, one of which is on the north shore of Gambier Bay. An excerpted quote from the unpublished report describes the four occurrences as "Major stratabound massive sulfide Cu-Zn prospects; streamsediment anomalies up to 2 percent Zn; active claims and exploration; southward continuation of major stratabound massive sulfide belt containing deposit at locality (Greens Creek Mine)." This suggests that prior ::J

ffi2oo f/) b: if 100

::J 30 a: w 20

10

a: 50 w 40 f/) 30 a: 20 10 .r::l A Distribution of arsenic Distribution of antimony Distribution of lead 11 i I; 13? WINDFALL HARBOR A' documentation of a major stratabound Cu-Zn massivesulfide occurrence in the same general area as the North Gambier occurrence may lie buried in old company files. During this study we spent a day and a half mapping 1,500 ft of the creek where the North Gambier occurrence is located. Thirty-one outcrop and 10 float rock samples were collected for petrographic examination and geochemical analyses. The geologic map of the North z g3000

a: f/) b:

a: w a: ,d Ill A Distribution of barium Distribution of gold l1 h l.h Distribution of mercury 11 i I; WINDFALL HARBOR A' SAMPLE SITE NUMBERS Figure 5. Distribution of selected elements from stream-sediment analyses in Ward Creek,/\Nindfall Harbor area projected onto northwest to southeast transect (A-A', fig. 3 ). Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/ Windfall Harbor and Gambier Bay Areas, Admiralty Island

Gambier occurrence with outcrop rock-sample locations is presented in figure 4. The position of features described below are given as distance in feet from the arbitrarily selected starting point of the map. Selected element concentrations of rock samples and brief descriptions are shown in table 3. At the south end of the mapped section, the basal unit consists of hematitic, green-streaked, brick-red to purple, finely layered, well-foliated quartz-muscovite schist (sample AK). The unit has tight, centimeter-scale W -shaped folds, and elongation lineation plunges 42° to the west. This unit is in unconformable contact with a 250-ft-thick (measured horizontally up the stream bed) section of brick-red to purple, matrix-supported, polymictic breccia with subangular clasts up to several centimeters in size. The clasts consist of chert, graywacke, and mafic volcanic rocks in a dark red, hematitic matrix (sample AJ). This unit is in contact with a 10-ft-thick section of relatively unfolded, brick-red, finely laminated, hematitic chert (sample AI). Upstream from the chert is 75 to 80 ft of reddish sedimentary breccia consisting of clasts of mafic volcanic rocks in a graywacke matrix. The breccia unit is divided by a 20-ft-thick section of soft, reddish-green, chlorite-altered mafic volcanic rocks or tuff (sample AL). Bedding in the sedimentary breccia has an irregularly laminated appearance and is folded and possibly faulted. This unit is in contact with a 120-ft-thick section of red, hematitic graywacke (samples J and AN) with interbedded layers of angular breccia (sample AM) and stretched, grain-supported chert-pebble conglomerate (sample B). Orientation of tension cracks and elongation of the chert pebbles indicate stretching in an eastwest direction. The hematitic graywacke is overlain by 80 ft of altered mafic volcanic rocks (sample AO), which are in turn overlain by another 60 to 70 ft of hematitic graywacke (samples K and AH) and interbedded conglomerate layers. From the 640-ft position to beyond the northern end of the mapped area, the country rock consists of a massive section of green to purple, aphanitic to porphyritic-aphanitic, amygdaloidal, spilitized, mafic volcanic rocks. The greenstone becomes silicic and ferruginous (as a result of the weathering of introduced sulfides) in brecciated areas and in proximity to the numerous shears and joints that occur along the creek. Amygdules are filled with calcite and an outer margin of a blue-green clay mineral. Graded bedding occurs in the sedimentary section at the south end of the mapped area and seems to indicate an up directton w the north. However, the area is structurally complicated by faulting and folding that makes the sedimentologic information equivocal. Identifiable bedding surfaces are rare and are complicated by structural deformation and probable overturning. In general, the dip angle of bedding is steep (>60°). Based on our Geologic Studies in Alaska by the U.S. Geological Survey, 1991 observations and their similarity to descriptions of the Gambier Bay Formation and the basal sedimentary unit of the Hyd Group (Loney, 1964; Lathram and others, 1965; Rubin and Saleeby, 1991, p. 14,556), we suggest that the rocks in this area represent the lower sedimentary and middle volcanic intervals of the Hyd Group, which rest unconformably upon the Gambier Bay Formation. There are three dominant structural trends. The oldest and most penetrative fabric trends northwest parallel to the regional structural trend. Previous mapping on the north shore of Gambier Bay (Loney, 1964; Lathram and others, 1965) shows a northwest-trending fault along the creek.· The previously mentioned mineral occurrences are located on or near northwest-trending structures. This structural trend is exhibited by numerous faults and shear zones less than 1 ft to 10 ft wide. Rocks within these shear zones commonly have horizontal slickensides, gouge, and silicified, thoroughly ferruginized breccia. The second structural trend is oriented predominantly east-west and is present as small shear zones and joint sets with associated horizontal slickensides and breccia as described above. Shearing and brecciation are most intense in the central part of the mapped area. The intensity diminishes northward, where iron-stained joint sets predominate. The third, less pervasive structure strikes north to northeast and is present as small faults, joints, and calcite veins. At one location (site AD near the 900-ft position), an east-trending fault displaces a north-northeast-trending fault . Mineralization occurs primarily within sheared and brecciated zones in the mafic volcanic rocks. Chalcopyrite and barite are the major constituents with lesser bornite, pyrite, sphalerite, galena, covellite, and silverrich tetrahedrite. Chalcopyrite occurs ( 1) as clots up to several centimeters in diameter with associated coarse crystalline, bladed, white barite, (2) as millimeter-sized grains disseminated in the greenstone, and (3) as massive, flow-banded material that is intimately intergrown with bornite, covellite, tetrahedrite, and fine-grained pyrite and quartz. Covellite occurs as fine, hairlike needles and laths within the bornite. Tetrahedrite occurs as small blebs and fracture fillings within the chalcopyritebornite assemblage. Quartz-carbonate veins 1 to 10 em wide are commonly present in the mineralized shear and breccia zones. The veins contain pyrite, chalcopyrite, and bladed barite in their interiors. The barite is partially to completely replaced by quartz and iron-rich dolomite. The sulfides are present as thin (1 mm), deformed lines of crystals, similar in appearance to stylolites, near the outer edges of the veins or in selvages at the vein/country-rock contact. The veins range from concordant to highly discordant to the host rock. A second, distinct mineral assemblage consists of breccia- and fracture-filling quartz, sphalerite, tetrahe-

Table 4A. Univariate statistical analysis of selected elements in -200-mesh stream-sediment samples and the threshold values selected for single-element plots [All values in parts per million. Abbreviations: DV, number of determinant (unqualified) variables; N, number of variables below the lower limit of detection; L, number of variables detected at the lower limit of detection; G, number of variables greater than the upper limit of detection.] Element DV N Concentration Percentile Threshold value Minimum Maximum 50th 90th Ba Ni Ag As Cd Cu Mn Mo Pb Sb Zn Au Hg Table 48. Univariate statistical analysis of selected elements in nonmagnetic heavy-mineral-concentrate samples and the threshold values selected for single-element plots [All values in parts per million unless otherwise noted] Element DV N G Concentration Percentile Threshold Minimum Fe(%) Ni Ag As Cd Cu Mn Pb Zn drite and very minor galena. This assemblage has also been noted in float as disseminated grains in a quartzveined and silicified mafic volcanic rock. In several well-brecciated, mineralized samples where crosscutting relationships can be seen, matrix material consisting of fine-grained chalcopyrite and large bladed crystals of barite are clearly crosscut by veinlets of quartz, sphalerite, tetrahedrite, and minor galena. This assemblage is present in the two prominent structures at the 700- and 800-ft positions near the transition from volcanic rocks to sedimentary rocks, and in a float sample (sample T) from the creek bed at the northern end of the mapped area. The geochemistry of the mineralized samples (table 3) closely reflects the mineralogy. Most samples contain 0.1 to 1 percent Cu, 0.1 percent Zn, several ppm Ag, and highly anomalous concentrations of Mn, As, Cd, Mo, Pb, Sb, and Hg. The two most highly mineralized samples (samples AA and R) are examples of the first type of mineral assemblage and contain anomalously high Cu value Maximum 50th 90th (20 and 6.8 -percent, respectively), Ag (84 and 38 ppm, respectively), and highly anomalous As, Mo, and Hg. Samples in which the second mineral assemblage is present contain high Zn and Pb concentrations with minor Cu. Sample T is the most extreme example of this, with 1.1 percent Zn, 0.65 percent Pb, and 7.4 ppm Ag. In order to test the extent of mineralization within and away from the mineralized structures, we collected a composite chip sample from each of two mineralized shear zones that cross the creek at the 700- and 800-ft positions (samples AE and AG, fig. 4). A third composite chip sample (sample AF) was collected from the massive altered mafic volcanic rocks in-between. The analyses (presented in table 3) indicate strong enrichment of Cu, Pb, Zn, and associated trace elements in the mineralized structures and a .distinct drop in concentration of the same elements in the unmineralized mafic volcanic rocks. These samples clearly show that the mineralization is confined to the structures. Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/ Windfall Harbor and Gambier Bay Areas, Admiralty Island

DISCUSSION AND CONCLUSIONS Analysis of the geochemical data from the Ward Creek/Windfall Harbor area clearly confirms and further constrains the presence of a belt of rocks that are prospective for VMS-type mineralization. This belt, which is characterized by stream-sediment and heavy-mineralconcentrate samples with anomalous concentrations of Mn, Ni, Ag, As, Cd, Sb, Mo, Ba, Cu, Pb, Zn and Hg (±Au), closely corresponds with exposures of Permian to Late Triassic mafic volcanic and flysch-type sedimentary rocks. Although there are a few stream-sediment and heavy-mineral-concentrate samples that may indicate the presence of small mineral occurrences, we interpret the belt of anomalous samples to be a reflection of the high background concentration of these elements in the Permian and Triassic rocks. This observation implies that the slight zonation in element enrichment of As, Sb, Pb, and Ba at the northwest end of the belt to enrichment of Au and Hg at the southeast end reflects a difference in the background concentrations of these elements in the Retreat Group and the Hyd Group. .Such a difference might provide a means of distinguishing geochemically between the two groups of Triassic rocks in areas of poor exposure. It also suggests that the Retreat Group may be more likely to host Cu-Pb-ZnAg-Ba rich VMS-type deposits and the Hyd Group most likely to host Cu-Zn-Au-Ag deposits. However, further work is needed to clarify and (or) substantiate this suggestion. We cannot speculate on the genesis of the North Gambier mineral occurrence. Further work will focus on determining the extent of mineralization, confirming its stratigraphic and structural setting, determining more completely the geochemical expression of the mineralization and associated alteration of country rock, and exploring the possibility that this occurrence may be continuous with previously noted mineralization on the ridge to the southeast and (or) with the prospects on Gambier Mountain. Acknowledgments.-We wish to thank Karen Kelley and AI Hofstra for thorough reviews that greatly improved the paper; John Philpotts for his cheerful help in the field; Paul Briggs, Dave Fey, Jerry Motooka, Phil Hageman, and Bruce Roushey for their careful geochemical analyses; Steve Sutley for the XRD mineralogical determinations; and Karen Kelley for her never-ending help with the NURE data. Vivian Hoffman, the U.S. Forest Service manager for the Admiralty Island National Monument, was especial! y helpful in granting access to the study area. We also offer a special thanks to Dan Maurice and Tim Perry, our helicopter pilots. Without their steady flying this study would have been impossible. REFERENCES CITED Barker, F., 1957, Geology of the Juneau (B-3) quadrangle, Alaska: U.S. Geological Survey Map GQ-100, scale 1:63,360. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Berg, H.C., and Cobb, E.H., 1967, Metalliferous lode deposits of Alaska: U.S. Geological Survey Bulletin 1246, 254 p. Berg, H.C., Decker, J.E., and Abramson, B.S., 1981, Metallic mineral deposits of southeastern Alaska: U.S. Geological Survey Open-File Report 81-122, 1 sheet, scale 1: 1 ,000,000, 136 p. Berg, H.C., and Grybeck, D., 1980, Upper Triassic volcanogenic Zn-Pb-Ag ( -Cu-Au)-barite mineral deposits near Petersburg, Alaska: U.S. Geological Survey OpenFile Report 80-527, 9 p. Berg, H.C., Jones, D.L., and Coney, P.J., 1978, Map showing pre-Cenozoic tectono-stratigraphic terranes of southeastern Alaska and adjacent areas: U.S. Geological Survey OpenFile Report 78-1085, 2 sheets, scale 1:1,000,000. Berg, H.C., Jones, D.L., and Richter, D.H., 1972, GravinaNutzotin belt: Tectonic significance of an upper Mesozoic sedimentary and volcani-c sequence in southern and southeastern Alaska: U.S. Geological Survey Professional Paper 800-D, p. D1-D24. Brew, D.A., and Ford, A.B., 1984, Tectonostratigraphic terranes in the Coast plutonic-metamorphic complex, southeastern Alaska, in Reed, K.M., and Bartsch-Winkler, S., eds., The United States Geological Survey in Alaska: Accomplishments during 1982: U.S. Geological Survey Circular 939, p. 90-93. Brew, D.A., Karl, S.M., Barnes, D.F., Jachens, R.C., Ford, A.B., and Horner, R., 1991, A northern Cordilleran oceancontinent transect: Sitka Sound, Alaska, to Atlin Lake, British Columbia. Canadian Journal of Earth Sciences, v. 28, no.6, p. 840-853. Cobb, E.H., 1972a, Metallic mineral resources map of the Sitka quadrangle, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-467, 1 sheet, scale 1:250,000. ---1972b, Metallic mineral resources map of the Sumdum quadrangle, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-425, 1 sheet, scale 1:250,000. ---1978a, Summary of references to mineral occurrences (other than mineral fuels and construction materials) in the Sitka quadrangle, Alaska: U.S. Geological Survey OpenFile Report 78-450, 124 p. ---1978b, Summary of references to mineral occurrences (other than mineral fuels and construction materials) in the Sumdum and Taku River quadrangles, Alaska: U.S. Geological Survey Open-File Report 78-698, 64 p. Crock, J.G., Lichte, F.E., and Briggs, P.H., 1983, Determination of elements in National Bureau of Standards geologic reference material SRM 278 obsidian and SRM 688 basalt by inductively coupled argon plasma-atomic emission spectrometry: Geostandards Newsletter, v. 7, p. 335-430. Davis, J .C., 1986, Statistics and data analysis in geology (2d ed.): New York, Wiley, 646 p. Ford, A.B., and Brew, D.A., 1988, The Douglas Island Volcanics: Basaltic rift not island arc volcanics of the "Gravina-Nutzotin belt," northern southeastern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 20, no. 7, p. 111. Ford, A.B., Karl, S.M., Duttweiler, K.A., Sutphin, D.M., Finn, C.A., and Brew, D.A., 1989, Sitka quadrangle, Alaska-

An AMRAP preassessment planning document: U.S. Geological Survey Administrative Report, 92 p. Gehrels, G.E., and Saleeby, J.B., 1987, Geologic framework, tectonic evolution and displacement history of the Alexander terrane: Tectonics, v. 6, p. 151-173. Goldfarb, R.J., Nelson, S.W., Berg, H.C., and Light, T.D., 1987, Distribution of mineral deposits in the Pacific Border Ranges and Coast Mountains of the Alaskan Cordillera, in Elliot, I.L., and Smee, B.W., eds., Geoexpo/ 86-Exploration in the North American Cordillera: Vancouver, British Columbia Canada, Association of Exploration Geochemists, p. 19-41. Grimes, D.J., and Marranzino, A.P., 1968, Direct-current arc and alternating-current spark emission spectrographic field methods for the semi-quantative analysis of geologic materials: U.S.Geological Survey Circular 591, 6 p. Herbert, C.F., and Race, W.H., 1965, Geochemical investigations of selected areas in southeastern Alaska, 1964 and 1965: Alaska Division of Mines and Minerals Geochemical Report 6, 65 p. Karl, S.M., 1989, Preliminary geologic map of the Sitka quadrangle, Alaska in Ford, A.B., Karl, S.M., Duttweiler, K.A., Sutphin, D.M., Finn, C.A., and Brew, D.A., Sitka quadrangle, Alaska-An AMRAP preassessment planning document: U.S. Geological Survey Administratiye Report, plate 2, scale 1 :250,000. Kelley, K.D., 1990, Interpretation of geochemical data from Admiralty Island, Alaska-Evidence for volcanogenic massive sulfide mineralization, in Goldfarb, R.J., Nash, J.T., and Stoeser, J.W., eds., Geochemical studies in Alaska by the U.S. Geological Survey, 1989: U.S. Geological Survey Bulletin 1950, p. A1-A9. Kennedy, K.R., and Crock, J.G., 1987, Determination of mercury in geological materials by continuous flow, cold vapor, atomic absorption spectrophotometry: Analytical Letters, v. 20, no. 6, p. 899-908. Lambert, I.B., and Sato, T., 1974, The Kuroko and associated ore deposits of Japan: A review of their features and metallogenesis: Economic Geology, v. 69, p. 1215-1236. Lathram, E.H., Pomeroy, J.S., Berg, H.C., and Loney, R.A., 1965, Reconnaissance geology of Admiralty Island, Alaska: U.S. Geological Survey Bulletin 1181-R, 48 p. Loney, R.A., 1964, Stratigraphy and petrography of the PybusGambier area, Admiralty Island, Alaska: U.S. Geological Survey Bulletin 1178, 103 p. Monger, J.W.H., and Berg, H.C., 1987, Lithotectonic terrane map of western Canada and southeastern Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF1874-B, scale 1:2,500,000. Motooka, J.M., 1988, An exploration geoche.nical technique for the determination of preconcentrated organometallic halides by ICP-AES: Applied Spectroscopy, v. 42, p. 1293-1296. Rubin, C.M., and Saleeby, J.B., 1991, The Gravina Sequence: Remnants of a mid-Mesozoic oceanic arc in southern southeast Alaska: Journal of Geophysical Research, v. 96, no. B9, p. 14,551-14,568. Thompson, C.E., Nakagawa, H.M., and Van Sickle, G.H., 1968, Rapid analysis for gold in geologic material: U.S. Geological Survey Professional Paper 600-B, p. B130B132. Wright, C.W., 1906, A reconnaissance of Admiralty Island: U.S. Geological Survey Bulletin 287, p. 138-161. ---1907, Lode mining in southeastern Alaska, 1907: U.S. Geological Survey Bulletin 314, p. 47-72. Wright, F.E., and Wright, C.W., 1904, Economic developments in southeastern Alaska: U.S. Geological Survey Bulletin 259, p. 47-68. Reviewers: Karen D. Kelley and AI Hofstra Followup Geochemical Survey of Base-Metal Anomalies in the Ward Creek/ Windfall Harbor and Gambier Bay Areas, Admiralty Island

Experimental Abrasion of Detrital Gold in a Tumbler By Warren Yeend Abstract Gold grains were tumbled with gravel for up to 150 h (200 km) at 1.33 and 3.00 km/h. The eroded products of both the gold and gravel were noted, and curves were plotted showing the breakdown of both gravel and gold with time (distance of travel). Comparison of these data with data gathered from an earlier study seems to indicate that the breakdown of gold increases markedly between a travel rate of 0.85 km/h and 1.33 km/h. The higher rate of gold breakdown associated with higher tumbler rates can be negated by the presence of friable gravel clasts, in that there appears to be an inverse relation between the breakdown of gravel and the included gold fragments-the faster the gravel breaks down, the slower the gold breaks down, and vice versa. A gold breakdown rate of 0.25 percent/km seems to be the average when tumbling at 1.33 to 3.0 km/h. INTRODUCTION Originally conceived in the early 1970's, this study sought to learn about the physical breakdown of nativegold fragments in a simulated high-energy, alluvial-like environment. It was thought that the malleable nature of gold would probably lead to much different rates of physical breakdown and produce different shapes as compared with the breakdown of readily fractured silicate-rich minerals and rocks. The results of this study should be useful for interpreting the geologic history of gold-containing alluvial deposits. The first phase of this study was published in 1975 (Yeend, 1975). Subsequently, it was recognized that more work was needed to better understand the physical breakdown of gold in a high-energy environment. Although this study adds to the small body of knowledge dealing with physical breakdown of native gold, there is certainly more to learn: And, as is so typical of studies of this nature, more questions are raised than answered, and related subjects that should be studied become evident. The first study used a 5-gal polyethylene bottle commonly referred to as a "carboy," with indented handholds in the sides as the tumbler. In an attempt to remove any erosive effects produced by the tumbler Geologic Studies in Alaska by the U.S. Geological Survey, 1991 itself, a rock-polishing tumbler with a rubber liner was obtained for the present study. In the original experiment, reconstructions of various combinations of sand, pebbles, and cobbles were used as the erosive material. In the current experiments, actual samples of sand and gravel collected from placer-rich river systems in Alaska were used as the erosive material. It was thought that this would more accurately simulate actual streamerosion conditions. The study was conducted in the office using gold fragments obtained from placers in the Circle Mining District in Alaska. The methods used and the experimental results from five separate sample "runs" in the tumbler are described. Graphs show both the loss of weight (unrecoverable by panning) of the original gold fragments and the weight loss of the >2-mm-size gravel fraction with duration of tumbler travel. Photos show the before-and-after appearance of the gold fragments used in the tumbling experiments. Finally, an attempt is made to interpret the experimental data. PREVIOUS WORK I am unaware of any field or laboratory studies on the physical breakdown of gold since my earlier report (Yeend, 1975). Mention is made in that paper of abrasion studies done on minerals such as quartz, garnet, tourmaline, apatite, and hornblende, among others, and of work done on the chemical and collodial properties of gold (particularly its solubility versus insolubility). Several studies of a somewhat related nature to the present one have been published. Hallbauer and Utter (1977) studied the relationship between the transport distance and changes in the surface morphology as well as changes in the chemical composition of gold particles during transport. They were able to detect characteristic changes in the morphology to allow an estimate of transport distances in the 10- to 30-km range. Parker (1974), sampling the <1-mg gold flakes along mountain creeks in Colorado found a downstream gold-fragment size decrease of from 5 to 23 percent per kilometer. How much of this decrease was a result of physical abrasion and

how much was due to selective sorting because of a lessened carrying capacity of the stream as a result of a flatter gradient is not known. METHODS Gold flakes were tumbled with sand, gravel, and water for measured time intervals of 5 to 150 h and then recovered by panning. The gold fragments and gravel were separately weighed and the weight loss noted. Changes were made to the different runs in the tumbler; these included varying the tumbler speed, the amount of gold, and the gravel components, as is subsequently described. Graphs show the weight loss of gold fragments and gravel with progressive distance of tumbler travel. The tumbler used in this study has a 40-lb, 4-gal capacity and uses a 1/3-hp (horsepower) electric motor (fig. 1 ). The tumbler rotation speed was altered by changing pulley size on the tumbler rollers, and, correspondingly, the drive-belt length running off the motor pulley. The two tumbler speeds used were 28 and 64 rpm (revolutions per minute). The tumbler interior is hexagonal in outline, 29 em in depth, and 13 em wide on each side of the hexagon, giving an interior circumference of 78 em. Using this circumference and the 28and 64-rpm tumbler speeds, travel rates for the material tumbled can be calculated at 1.33 and 3.00 km!h. The tumbler possesses a heavy rubber liner. One end of the tumbler is removable for easy loading and unloading. The removable end is covered on the interior by a heavy rubber gasket. Because of the complete rubber lining, all abrasive effects were a result of the interaction of the gravel, sand, and clay. When running at the slow (1.33 Figure 1. Tne 40-lb, 4-gal tumbler with 1 /3-hp electric motor used in this study. Metal end and rubber gasket have been removed, showing hexagonal shape and rubber liner, having 78-cm internal circumference. kmlh) rate, the sound of the gravel being tumbled was barely discernible. At the faster rate (3 kmlh), however, the noise level bordered on being objectionable in an office environment. Gold fragments used in this study were obtained from streams in the Circle Mining District in-east-central Alaska. Size of the fragments ranged from 1 to 5 mm in diameter. They were generally bright and ·shiny with no adhering minerals. For each run, 10 to 12 gold fragments were used. The total gold fragment weights used in each run ranged from 48 to 40 I mg determined to the nearest milligram. Photos were taken of the gold fragments from selected runs both before and after tumbling to show the changes in shape and number of recoverable fragments. Sand and gravel used in the tumbler runs was collected from the floodplains of several different goldbearing creeks in the Circle Mining District in east central Alaska (Y eend, 1991 ). The gravel possessed subangular to subrounded clasts up to 7 em in diameter. The clasts were predominantly quartzite and quartz-mica schist, with some white quartz and granite. In the initial run of this study, no records were kept of the weight loss of the gravel. Subsequently, it was determined that this information would be useful, so on the ensuing runs, the weight percentages of the gravel fraction greater than 2 mm was noted both before and after tumbling. The amount of abraded products produced provided some comparison of the energy levels of the different runs. The schist clasts broke down the fastest. The tumbler runs of gravel and gold were designed to test the abrasion rates of gold in several different environments. The tumbler was rotated both at 1.33 km/h and at the increased travel rate of 3.00 km/h in order to test the effects of increased velocity as might be present in a steep-gradient creek or of increased volumes of runoff in storm conditions. My earlier work (Yeend, 1975) had shown a tenfold increase in abrasion rates with a fourfold increase in velocity. The rotation rates used in the current study were determined by the pulley diameters available. The tumbler was stopped at selected intervals during a run (anywhere from 5 to 40 h), and the gold separated by panning from the gravels. The recovered gold fragments were counted and weighed. Fragments smaller than 0.25 mm were not recovered; thus, gold finer than this was considered the abraded, lost gold. In several of the runs, the gravel with all its abraded products was used throughout the run. In others, at selected intervals, the old gravel and its eroded products were replaced by a new batch of gravel. · This, in effect, introduced coarse particles and removed the fine-grained sand, silt, and clay that was produced during tumbling. The effect of this change would ·be somewhat analogous to that caused by tributary creeks contributing a coarser fraction of Experimental Abrasion of Detrital Gold in a Tumbler

Table 1. Gold breakdown rate and averages for various tumbler runs, including previous study (Yeend, 1975) Run /batch number Rotation Gold Gold (travel distance in km) rate breakdown breakdown- (km/h) rate during rate each averages run/batch (percent/ (percent/ km) km) Run 2 (198} Run 3 0 80) - -- -- - - - Run4 batch 1 ( 45) - - - - -- - 3.00 batch 2 ( 45) - - - - - - batch 3 (45)--- 3.00 batch4(45)--- --- total (18();) --- -- --- 3.00 Run 5 batch 1 ( 40) - - - - - - batch 2 ( 46) - - - - - - batch 3 (40) batch 4 ( 40) - - - - - - total (160) - - - - - - - All runs at - - - - - - - All runs at - - - - - - Yeend stud::r {1975}--- gravel periodically to the trunk creek. These details are discussed in the description of the individual tumbler runs. Graphs show the distance of tumbler travel plotted against gold and gravel weight loss percent. Because different weights of gold and gravel were used in the different runs, this method of graphical representation allows comparison between runs. In each run, 2 to 3 liters of tap water were used to completely cover the gravel and contained gold. The same muddy water was used throughout a run, and no attempt was made to add clean water. Each run lasted from 60 to 150 h, the shorter times corresponding with the faster rotation rates. The total tumbler distance traveled in all runs varied from approximately 150 to 200 km. This seemed a sufficient distance to establish breakdown rates for the gold. The results of the tumbler runs are presented in the order in which they were done. A summary of the gold breakdown rates and comparison with the earlier study (Yeend, 1975) are presented in table 1. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 TUMBLER RUNS Run #1 Ten gold flakes 3 to 5 mm in diameter weighing 324 mg were tumbled with 4.437 kg of cobble gravel obtained from Mastodon Creek in the Circle Mining District, Alaska. The tumbler was rotated at a rate of 1.33 km/h. It was periodically stopped and the gold separated and weighed at the following time (distance) intervals: 5 h (6.6 km), 15 h (20.0 km), 35 h (46.2 km), 75 h (99 km), and 115 h (151.8 km). At the end of the run the recoverable, very flattened golO consisted of 9 fragments 2.0-4.0 mm; 2 fragments 1.0-2.0 mm; 4 fragments 0.5-1.0 mm; and 2 fragments 0.25-0.5 mm. The recovered gold weighed 236 mg, representing a 27.1 percent loss of weight, or 0.18 percent/km. This total weight loss was at a more-or-less uniform rate over the duration of the run, as the slope of the weight-loss curve shows (fig. 2). Run #2 Small particles of gold weighing substantially <run #1 were used in run #2. Ten gold flakes 1 to 2 mm in greatest dimension weighed 48 mg. The gravel, obtained from Eagle Creek in the Circle Mining District, Alaska, was classified in two size fractions->2-mm diameter, 3.83 kg; and <2-mm diameter, 0.43 kg. The tumbler was rotated at the rate of 1.33 krnlh for 150 h. Gold was separated and weighed five times during the total run of 198 km. The ending gold weight of 24 mg was exactly 50 percent of the original weight and gives a weight-loss rate of 0.25 percentlkm (fig.2). At the end of the run, the gold consisted of 11 fragments 1.0-2.0 mm; 5 fragments 0.5-1.0 mm; and 3 fragments 0.5-0.25 mm. The coarse fraction of the gravel (>2 mm) had lost 55 percent of its weight (0.28 percent/km), most of TUMBLER TRAVEL, IN KILOMETERS Figure 2. Curves showing weight loss of gold fragments with distance of tumbler travel for five different tumbler runs.

which was eroded to fine silt or clay-sized particles (fig. 3). The <2-mm-size fraction had increased by 6 times, and 95 percent of this was fine silt and clay. Run #3 The rate of tumbler travel was increased to 3 km/hr in run #3. Gravel from Bottom Dollar Creek, Circle Mining District, Alaska, was tumbled with 10 fragments of gold that were 2-5 mm in greatest dimension and weighed 401 mg. The gravel clasts included 1.30 kg of <2-mm size, and 3.49 kg of greater than 2-mm size. Five of the gold fragments were obtained from the previous run and were very flat in shape; five fragments were newly introduced and were not particularly flattened but did have rounded edges. The gold-particle breakdown with progressive tumbling is as follows: 4 h (12 km), 10 fragments 2-5 mm, 4 fragments mm; 12 h (36 km), 6 fragments mm, 2 fragments 1-2 mm, 6 fragments mm; 28 h (84 km), 8 fragments mm, 3 fragments 1-2 mm, 4 fragments 0.25-1.0 mm, several hundred fragments <0.25 mm; 60 h (180 km), 10 fragments 1.5-3 mm, 11 fragments 0.25-1.5 mm, several hundred fragments <0.25 mm. The gold lost 46 percent of its original weight, or 0.25 percentlkm. The >2-mm-size gravel fraction lost 73 percent of it original weight, or 0.40 percentlkm. The <2-mm-size fraction increased 300 percent in weight, 98 percent of which was silt and clay. Run #4 Run #4 varied from the previous runs by the introduction of a new batch of gravel at selected intervals. Gravel from the Ketchem Creek drainage, Circle Mining District, Alaska, was divided into four approximately E--z La u " g:

til so g

TUMBLER TRAVEL, IN KILOMETERS Figure 3. Curves showing gravel (>2-mm size) weight loss with distance of tumbler travel. equal batches. After 15 h ( 45 km) of tumbler .travel, the gold and gravel were separated and both were weighed. The gravel was discarded, and a new batch of gravel was added for the next 15-h (45 km) run. This would simulate the introduction of gravel by tributaries along a stream system. The same gold was used throughout the run. Twelve gold fragments 1.5-3.0 mm in diameter and weighing 190 mg were used for this run. At the beginning of the run, the 12 gold fragments consisted of 8 thin and very flattened fragments, 2 thicker but definitely flattened fragments, and 2 fragments that were ragged in outline, but not flattened. Batch #1 The gravel-size fractions used consisted of 0.59 kg of <2-mm size and 1.96 kg of greater than 2-mm size. After 15 h ( 45 km) of tumbler travel, the gold weighed 165 mg, which represented a 13 percent weight loss, or 0.29 percentlkm. The coarse gravel (>2 mm) weighed 1.18 kg, having lost 40.0 percent of its weight, or 0.89 percent/km, about 3 times the rate of the gold weight loss. The gold had broken down as follows: 6 fragments 1.5-3.0 mm; 6 fragments 1.0-1.5 mm; 15 fragments 0.51.0 mm; 15 fragments 0.25-0.5 mm; and 50 to 100 fragments <0.25 mm. Batch #2 The gravel-size fractions were 1.79 kg of >2-mm size and 0.35 kg of <2-mm size. Following 15 h (45 km) of tumbling, the gold weighed 143 mg, having lost · 13 percent of its starting weight, or 0.29 percent/kin. The coarse-gravel fraction (>2 mm) had lost 40.0 percent of its original weight, or 0.89 percent/km. The gold particles included 5 fragments 1.5-3.0 mm; 8 fragments 1.0-1.5 mm; 10 fragments 0.5-1.0 mm; 13 fragments 0.25-0.5 mm; and 50 to 100 fragments <0.25 mm. Batch #3 The starting gravel weights were 1.83 kg of >2-mm size and 0.45 kg of <2-mm size. After 15 h (45 km) of tumbling, the gold weighed 126 mg, which is equivalent to a 12 percent weight loss, or 0.27 percent/km. The coarse gravel lost 40.3 percent of its original weight, or 0.89 percent/km. The gold showed the following size classification: 5 fragments 1.5-3.0 mm; 6 fragments 1.0-1.5 mm; 20 fragments 0.5-1.0 mm; 6 fragments 0.25-0.5 mm; and 50 to 100 fragments <0.25 mm. Batch #4 The recoverable gold in batch #4 was reduced to 103 mg which represented a loss of 18.3 percent from Experimental Abrasion of Detrital Gold in a Tumbler

the starting weight at the beginning of the batch run. This was the equivalent weight loss of 0.41 percentlkm. The gravel lost 46.0 percent of its starting weight, or 1.02 percentlkm. The final gold products were: 4 fragments 1.5-3.0 mm; 5 fragments 1.0-1.5 mm; 9 fragments 0.5-1.0 mm; 11 fragments 0.25-0.5 mm; and several hundred fragments <0.25 mm. Photos of the gold fragments before they were run with the first batch of gravel and after they were tumbled with the last batch of gravel are displayed in figure 4. For the entire length of run #4 (60 h, 180 km) the gold lost 45.0 percnet of its original weight, or 0.26 percent/km, and the coarse gravel averaged 41.6 percent weight loss, or 0.92 percentlkm. A B lO J

iiHte; ) S c.Xt") ertul~l'/ f (:,(.Jey Figure 4. Gold fragments used in run #4. A, Before tumbling. B, After 180 km of tumbler travel. Scale numbers are centimeters. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Run#5 Run #5 was characterized by slowing the tumbler back to the 1.33 kmlh rate, and, similar to run #4, introducing equal amounts of fresh gravel and removal of the old gravel after each 30 h (40 km) of tumbling. Gravel composed exclusively of schist clasts from Switch Creek, Circle Mining District, Alaska, was used in the first two batches. Schist and granite gravel from an alluvial bench in Deadwood Creek, Circle Mining District, Alaska, was used in the final two batches. Ten gold fragments were introduced at the beginning of the run, and the same gold was used throughout the run. Batch #1 The coarse-gravel fraction (>2 mm) weighed 1.71 kg. Ten flattened gold fragments weighed 206 mg and were all 2 to 4 mm in diameter. After tumbling for 30 h ( 40 km), the gold had lost only 1.0 percent of its weight, while the coarse gravel (>2 mm) had lost 31.0 percent of its weight. The gold had broken down as follows: 10 fragments 2.0-4.0 mm; and 1 fragment 0.5-1.0 mm. Batch #2 A fresh batch of Switch Creek gravel was introduced to the tumbler with the gold fragments from batch #1. After 30 h (40 km) of tumbling the coarse gravel fraction (>2 mm) had lost 26.5 percent of its ·weight, and the gold had lost only 0.5 percent of its weight. The gold fragments were made up of 10 fragments 2.0-4.0 mm; and 2 fragments 1.0-2.0 mm. Batch #3 The gravel used in this batch was collected from a high-level bench in Deadwood Creek valley, Circle Mining District, Alaska. It was composed predominantly of schist with a few granitic and quartz clasts. The coarse fraction of 1.25 kg lost 42.0 percent of its weight following 30 h (40 km) of tumbling. The gold fragments lost 2.0 percent of their weight and were composed of 10 fragments 2.0-4.0 mm; and 2 fragments 1.0-1.5 mm. Batch #4 The same kind of gravel as used in batch #3 was used here (Deadwood Creek bench gravel). The coarse fraction had a starting weight of 1.33 kg and lost 46.0 percent of its weight in 30 h ( 40 km) of tumbling. The gold lost 3 percent of its weight and ended the run as I 0 fragments 2.0-4.0 mm; and 2 fragments 1.0-1.5 mm. For all of run #5, the gold lost 6.5 percent of its weight,

or 0.04 percent/km. The coarse gravel from Switch Creek used in batches #1 and #2 averaged a loss of 0.72 percentlkm, while the coarse gravel obtained from the bench in Deadwood Creek used in batches #3 and #4 averaged a loss of 1.1 percentlkm. DISCUSSION This study revealed two important. variables in the physical breakdown rate of the gold fragments: the rate of tumbler rotation, and the durability of the coarse clasts making up the gravel. The data are summarized in table 1, which allows comparison of rotation rates with averages of gold breakdown including the 1975 (Yeend) study. There may be some threshold velocity above which the abrasion rate is markedly accelerated.ln table 1, an eightfold increase in gold abrasion between the 0.85 kmlh tumbler rotation and the 1.33 kmlh tumbler rotation rate is observed; such a threshold velocity would seem to be within this range. There appears to be an inverse relationship between the breakdown of the gravel and the gold-the faster the gravel breakdown, the slower the gold breakdown. This is logical in that a gravel with friable clasts composed mostly of mica schist breaks down to silt- and clay-size fragments quickly, and large clasts are not available to pound the gold fragments. In contrast, gravels rich in durable clasts such as quartz, quartzite, and granite break down slowly and retain their original coarse texture for a longer period of time, thus providing ample "pounding power" to break apart the gold fragments. This relationship is demonstrated in runs #2 and #3, which possessed the most durable gravel clasts. The gravels in these two runs broke down the slowest, as can be seen by the more gentle slope of the curve in figure 3. The gold in runs #2 and #3 broke down rather quickly, as can be seen in the steepness of their curves in figure 2 as compared with the other runs. Likewise, run #5 possesses very friable gravel and broke down extremely fast as the steepness of the curve in figure 3 shows. The gold in run #5 broke down extremely slowly (fig. 2). There is no explanation for this extremely slow abrasion rate of gold in run #5 other than the friability of the gravel. It should be pointed out that both the gravel and the gold in batches #3 and #4 broke down at a faster rate than the gravel or gold in batches #1 and #2. This would seem to have been caused by the presence of the durable granite clasts introduced in the gravel in batches #3 and #4. Batches #1 and #2 used gravel composed of only schist. It seems clear that the granite pulverized the schist and pounded the gold more thoroughly than did the schist clasts in batches #1 and #2. Although the original aim of the study was to learn about the physical breakdown of gold, it became apparent that the breakdown of the gravel was equally as interesting, and the factors affecting its breakdown were looked for. Certainly rate of tumbling was a factor. How the increase in rate of travel affects the breakdown of gravels with identical clast compositions and textures would be a worthwhile investigation. No attempt was made to do this in this study; however, another study focusing on such factors is contemplated. It seems quite clear that the velocity of travel does produce an increase in the gravel breakdown, but this correlation can be overwhelmed by the soft nature of the gravel clasts. This was demonstrated in runs #3 and #5. Here the faster tumbling rate of run #3, at 3.00 kmlh, produced a lower rate of physical breakdown of the gravel composed predominately of quartzite and quartz and 10 percent or less schist-0.4 percentlkm as compared to run #5 at 1.33 kmlh and a breakdown rate of 0.8 percentlkm. This was due to the highly friable nature of the gravel used in run #5, which was composed of both pure schist and a schist/granite mixture. The use of friable gravel here was not purposefully done but, rather, was ascertained after the fact as the factor responsible for this seemingly anomalous result. Comparisons between the results of this study and those of the 1975 (Yeend) study are suspect for the following reasons. A plastic carboy was used as the tumbler in 1975, and the gravel used in that study was not a "real" gravel collected from a creek bed, but rather it was constructed from two size classes of clasts: cobbles (3-6 em) and granules (2-5 mm). The extremely low abrasion rates of gold obtained in the 1975 experiment is, therefore, suspect when compared with the results of the current study, where a true tumbler and actual gravel collected from the floodplains of creeks were used. Perhaps the internal shape and hardness of the tumbler was a primary factor. The carboy used in 1975 was hard and round, allowing the contents to slide easily along the bottom during the rotation with little, if any, turbulence of the clastic particles. The tumbler used in the current study was hexagonal in outline with a soft rubber liner (fig. 1 ). During a run, the gravel and its contained gold would be carried by friction part way up the inside of the rotating barrel, causing the contents to fall back to the bottom. This action could have produced a higher energy environment than within the plastic carboy and may have been responsible for the higher abrasion rates obtained. CONCLUSIONS 1. A threshold tumble rate above which the breakdown of gold fragments increases substantially (8 times) seems to be in the range of 0.85 kmlh to 1.33 kmlh. 2. Although a higher rate of tumbling generally produces Experimental Abrasion of Detrital Gold in a Tumbler

a higher rate of breakdown of gold, this is not always true. The higher rate of gold breakdown associated with higher tumbler rate can be negated by the presence of friable gravel clasts, thereby producing a low gold breakdown rate. 3. An average gold breakdown rate of approximately 0.25 percentlkm is produced when tumbling at the 1.33 and 3.0 km/h rate with the quartzite schist-rich gravel, the predominant gravel type in creeks within the Circle Mining District; the gravel seems to break down at about 2 to 4 times the gold breakdown rate (0.59 to 1.00 percentlkm). 4. There is an inverse relationship between the breakdown of gravel and the breakdown of associated gold fragments-the faster the gravel breakdown, the slower the gold breakdown, and vice versa. 5. The internal shape of the tumbler and the composition of the liner seem to be important factors in the breakdown rate of tumbled rocks and minerals. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 REFERENCES CITED Hallbauer, K., and Utter, T. 1977, Geochemical and morphological characteristics of gold particles from recent river deposits and the fossil placers of Witwatersrand: Mineralium Deposita, no. 12, p. 293-306. Parker, B.H.Jr., 1974, Gold placers of Colorado: Quarterly of the Colorado School of Mines, no. 3, p. 37. Yeend, W.E., 1975, Experimental abrasion of detrital gold: U.S. Geological Survey Journal of Research, v. 3, no. 2, p. ---1991, Gold placers of the Circle district, Alaska-Past, present, and future: U.S. Geological Survey Bulletin 1943, 42 p., scale 1:63,360. 1 pl. Reviewers: Robert A. Loney and Bela Csejtey, Jr.

MINERAL RESOURCE STUDIES-GEOLOGIC NOTES Gold in the Usibelli Group Coals, Nenana Coal Field, Alaska By Gary D. Stricker, Richard B. Tripp, john B. McHugh, Ronald H. Affolter, and John B. Cathrall INTRODUCTION NENANA COAL FIELD Few data are available on gold content and distribution in U.S. coal, and even fewer data are available on the distribution or the occurrence of gold (Au) in Alaskan coal. Stone (1912) reported Au values as high as 17 parts per million (ppm) in coal ash from northeastern Wyoming. Finkleman ( 1981) detected 11 ppm Au . in one sample of the Waynesburg coal ash from West Virginia. He also observed several flecks of Au in polished sections of the· Upper Freeport coal from Pennsylvania and of the Ferron coal from Utah. Chyi (1982) reported that Au contents from 65 bituminous coal samples collected from western Kentucky, on a whole-coal basis, ranged from below the detection level to 1.8 parts per billion (ppb). Rao (1968, p. 2) noted a concentration of 4 ppm Au in the coal ash from one sample of the No. 2 bed at the Nenana coal field. Gold contents in coal are commonly below the analytical detection limit that was used over 10 years ago. The present lower detection limits for the U.S. Geological Survey's coal-analysis package is 3 ppm on coal ash (Golightly and Simon, 1989). Prior to the early 1980's, the detection limit was 50 ppm on coal ash. The majority of Alaskan coal samples were analyzed at the 50 ppm detection limit. Therefore, all Au analytical values for Alaskan coal stored in the National Coal Resource Data System have been reported as "less than" or "not detected." Recent improvements in analytical methods have lowered the detection limits to 0.001 ppm (1 ppb) on coal ash (O'Leary and Meier, 1986). Since 1984, we have been collecting face channel samples from the Nos. 3 and 4 beds of the Suntrana Formation that are exposed in the surface mine in the Lignite Creek coal basin, a small basin within the Nenana coal field, near Healy, Alaska (fig. 1). Twelve selected coal samples from these beds along with fly ash from the Golden Valley Electric Association, Inc. (GVEA) power plant at Healy were analyzed for Au content. Fly ash can be defined as all solids that are carried in the gas stream that are the result of the combustion of coal. These particles are collected in the stack recovery system at GVEA. Since the first written record of utilization of Alaskan coal in 1786, more than half of the coal mined in Alaska has been produced from the Nenana coal field (Sanders, 1981 ). This field consists of 10 synclinal coal basins partially or completely separated by erosion. These basins extend about 200 km along the northern foothills of the Alaska Range and include, from west to east, the western Nenana, Healy Creek, Lignite Creek, Rex Creek, Tatlanika Creek, Mystic Creek, Wood River, West Delta, East Delta, and Jarvis Creek coal basins. The coals in these basins are in the U sibelli Group (Wahrhaftig, 1987), a sequence of five formations of poorly consolidated continental sedimentary rocks of late Eocene and early to late Miocene age (fig. 2). The Usibelli Coal Mine is presently surface mining the Nos. 3, 4, and (occasionally) 6 beds from the Suntrana Formation and the Caribou and Moose beds from the Healy Creek Formation (fig. 2). The Caribou and Moose beds are not exposed in the type section near Suntrana Creek and are not shown on figure 2. These beds are found in the upper part of the Healy Creek Formation near the F and E beds (fig. 2). The bulk of the coal mined at the Usibelli Coal Mine is from the Nos. 3 and 4 beds. The specific coal that was burned ·to produce heat energy and fly ash is and will remain unknown. The power plant utilizes whichever coal is being mined by the Usibelli Coal Mine at the time the power plant's coal stockpile requires additional fuel. ANALYTICAL PROCEDURE Gold analyses for this investigation were determined by the procedure reported by O'Leary and Meier (1986). Depending upon the content, sufficient coal to produce 5 to 10 g of organic-free ash was ashed at 680°C for a 24-h period. The ash was then digested with concentrated hydrobromic acid (a 0.5-percent bromine solution), forming a · gold-bromide complex that was extracted with methyl isobutyl ketone. This extract was analyzed with electrothermal Gold in the Usibelli Group Coals, Nenana Coal Field

atomic-absorption spectrophotometry to determine gold to a lower detection limit of 1 ppb. Approximately 3,000 g of fly ash were collected at GVEA and three random 10 g splits were heated to 680°C to remove any remaining organic material. This organic-free fly ash was then analyzed for Au using the same method as described for coal ash. ANALYTICAL RESULTS Gold contents in coal ash for the 12 selected samples from the Nos. 3 and 4 beds of the Suntrana Formation range from 0.001 to 0.018 ppm (1 to 18 ppb). The arithmetic mean is 0.007 ppm with a standard deviation of 0.005 ppm. The analytical results for ash, gold values in FAIRBANKS ,, ' ' '

' ' ' ANCHORAGE ' Figure 1. Location map of Lignite Creek and Healy Creek coal basins, Nenana coal field (modified from Wahrhaftig and others, 1969). Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Nenana Gravel (Neogene) CD CD (,) "0 as CD CD (,) 0 e Grubstake Formation Lignite Creek o. Formation :I

a; .Q "' ;:) Suntrana Formation Q) Q) (,) 0 ·- Suntrana ::::!: Formation "0 r::: as Q) Q) (,) 0 w , Q. :I (; Q) :e "' A 50 METERS Sanctuary Formation B Coalbed No.6 Coalbed No. 5 Coalbed No. 4 Coalbed No. 3 Coalbed No. 2 Coalbed No. 1 G coalbed Suntrana Formation Sanctuary Formation Q) Q) (,) "0 r::: as Q) Q) (,) 0 w Healy Q. :I Creek ~Formation

-Q) .Q "' ;:) Pelitic and F coalQuartzose bed Schist (Lower Paleozoic and Precambrian?) EXPLANATION Coal (showing bone or clay parting)

Bony coal IJ Bone

Claystone and

shale D Siltstone Coalbed No. 2 Coalbed No. 1 G coalbed F coalbed E coalbed D coalbed C coalbed B coalbed A coalbed ""·L)·~~ Pebbles and conglomerate Schist (unconformity at top) Figure 2. Generalized stratigraphic section of Usibelli Group, Nenana coal field, at Suntrana, Alaska. Sections A and B were measured along Suntrana Creek, and section C was measured on north bank of Healy Creek (modified from Wahrhaftig, 1987, and Frost and Stanley, 1991 ). Gold in the Usibelli Group Coals, Nenana Coal Field

Table 1. Ash and gold contents, Nos. 3 and 4 coal beds, Suntrana Formation, Lignite Creek coal basin, Alaska Coal Depth from Ash Gold in Gold in bed* top of bed percent ash whole coal (in meters) (ppm) (ppb) 3 3 3 3 3 3 4 4 4 4 4 4 *The No.3 bed is 5.4 m thick and the No.4 bed is 7.8 m thick where sampled in 1987. coal ash, and gold values in samples of whole coal from the Nos. 3 and 4 beds are shown in table 1. A correlation coefficient was computed between ash content and gold values in whole coal for the 12 samples in order to determine relationships of the Au to coal. The coefficient between ash content and Au content in whole coal is 0.92 (which is significant at the 95 percent level). Gold values for the three fly-ash splits are 0.076, 0.12, and 0.87 ppm. Individual grains of gold were identified on the scanning electron microscope (SEM). These grains were recovered from the GVEA fly ash. The gold appears to be a placer particle in that it is rough and irregular in outline (fig. 3). Parts of the particle are coated with bubbles of fused calcium- silicate glass, a result of the particle's trip through the GVEA' s boiler. Other areas of the gold particle show new-growth gold in the form of sharply defined filament crystals and fine wires up to 7 long. The energy-dispersive system of the SEM determined that the gold wires are approximately 95 percent Au with 5 percent silver impurity. The SEM photograph of a placer-shaped gold particle and a correlation coefficient of 0.92 between ash and Au content in whole coal suggest that there is some detrital Au in the Nos. 3 and 4 coal beds. However, it seems unreasonable that most of the Au in these coals would be detrital in origin. A vigorous flood of water capable of carrying detrital Au from the nearby alluvial system into the mire would also carry sufficient detritus to dilute the organic material and produce a carbonaceous shale. Gold is present in low-ash coal samples (table 1), and thus we must conclude that some of the Au presently found in the Nos. 3 and 4 beds must have entered the peat swamp in solution from ground or surface water during the peat-accumulating stage and precipitated within the peat/ coal. Also, gold may have entered the coalbed both Geologic Studies in Alaska by the U.S. Geological Survey, 1991 during the coalification phase, when large volumes of water moved within the peat, and after coalification when ground water moved along cleats and fractures within the coal. Gold mobilization and crystallization is suggested by the existence of new-growth Au on the detrital grain (fig. 3). When the Au in solution entered the Nos. 3 and 4 beds is presently unknown. A B Figure 3. Scanning electron micrographs of detrital gold found in fly ash, GVEA, Healy, Alaska. A, Detrital gold grain showing (A) new-growth gold filament and (B) bubbles of fused calciumsilicate glass. 8, Enlargement of new-growth gold filament (A) shown in A.

CONCLUSIONS This study of gold content of two coal beds of the Suntrana Formation utilizing a lower detection limit for Au provides a preliminary evaluation of gold in Alaska coals. With this improved detection limit, Au was detected in all samples studied from the Nos. 3 and 4 beds from the Lignite coal basin and in fly ash of the GVEA. The Nos. 3 and 4 coal beds of the Lignite Creek basin may have economic potential for the byproduct recovery of Au. Bagby and others (1986) report that the range for Au content in carbonate-hosted gold and silver deposits is from 0. 7 to 8 ppm. Gold content of one fly-ash sample (0.87 ppm) is within the range for the abovementioned carbonate-hosted Au and Ag deposits. Our study has not addressed the vertical variability within the coal bed nor the modes of occurrence of the nondetrital Au. Work by Affolter and Stricker (1987) indicates that the coal beds at Healy have a significant · vertical variability in the content of trace elements. This limited study indicates that further investigation into the occurrence and origin of Au in the Usibelli Group coals is warranted. REFERENCES CITED Affolter, R.H., and Stricker, G.D. 1987, Variation in element distribution of coal from the Usibelli Mine, Healy, Alaska, in Rao, P.O., ed., Focus on Alaska's coal, '86: Fairbanks, Alaska, Mineral Industry Research Laboratory Report 72, p. 91-99. Bagby, W.C., Menzie, W.O., Mosier,D.L., and Singer, D.A., 1986, Grade and tonnage model of carbonate-hosted AuAg, in Cox, D.P., and Singer, D.A., eds., Mineral deposit models: U.S. Geological Survey Bulletin 1693, p. 176177. Chyi, L.L., 1982, The distribution of gold and platinum in bituminous coal: Economic Geology, v. 77, p. 1592-1597. Finkleman, R.B ., 1981, Modes of occurrence of trace elements in coal: U.S. Geological Survey Open-File Report 81-99, 301 p. Frost, G.M., and Stanley, R.G., 1991, Compiled geologic and Bouguer gravity map of the Nenana Basin area, central Alaska: U.S. Geological Survey Open-File Report 91-562, 30 p., 2 sheets, scale 1 :250,000. Golightly, D.W., and Simon, F.O., 1989, Methods for sampling and inorganic analysis of coal: U.S. Geological Survey Bulletin 1823, 72 p. O'Leary, R.M., and Meier, A.L., 1986, Analytical methods used in geochemical exploration, 1984: U.S. Geological Survey Circular 948, 48 p. Rao, P. Dharma, 1968, Distribution of certain minor elements in Alaskan coal: Fairbanks, Alaska, Mineral Industry Research Laboratory, University of Alaska, Report 15, 47 p. Sanders, R.B., 1981, Coal resources of Alaska, in Rao, P.D., and Wolff, E.N., eds., Focus on Alaska's coal '80: Fairbanks, Alaska, Mineral Industry Research Laboratory Report 50, p. 11-31. Stone, R.W., 1912, Coal near Black Hills Wyoming-South Dakota: U.S. Geological Survey Bulletin 499, 66 p. Wahrhaftig, Clyde, 1987, The Cenozoic section at Suntrana, Alaska, in Hill M.L., ed. The Cordilleran section of the Geological Society of America: Boulder, Colo., Geological Society of America, Centennial Field Guide 1, p. Wahrhaftig, Clyde, Wolfe, J.A., Leopold, E.B., and Lanphere, M.A., 1969, The coal-bearing group in the Nenana coal field, Alaska: U.S. Geological Survey Bulletin 1274-D, p. Reviewers: Michael E. Brownfield and Romeo M. Flores Gold in the Usibelli Group Coals, Nenana Coal Field

Rare Earth Minerals in "Thunder Eggs" from Zaremba Island, Southeast Alaska By John Philpotts and John R. Evans INTRODUCTION Fluorite veining occurs at a locality (lat 56°16' N., long 132°56' W.) on the west side of Zaremba Island adjacent to Snow Passage in southeast Alaska (fig.1). Elevated Y and Nb levels were found in stream-sediment nonmagnetic heavy-mineral concentrate from this site in the field geochemistry assessment of Cathrall and others ( 1983). Yttrium minerals invariably have signi.ficant contents of rare earth elements (REE). In broader definitions, Y is often included as a REE; in this note, however, REE refers to the 15-element sequence from La to Lu. Fluorite veins are often associated with REE mineralization (for example, Salvi and Williams-Jones, 1990). Rare earth element mineralization is known to occur at several localities on the eastern side of Prince of Wales Island (fig.l), including Salmon Bay, which lies just 10 km to the west of the fluorite locality, on the west side of Clarence Strait (Grybeck and others, 1984). It was in this context that the fluorite locality was visited in the summer of 1991, and a suite of samples was collected for the specific purpose of looking for evidence of REE mineralization. The general geology has been described by Buddington ( 1923, p. 54-55) as consisting of Tertiary volcanic flows, breccias, tuffs, and agglomerates, with some interbedded sedimentary rocks (sandstones and conglomerates) at the base (see also Brew and others, 1984 ). The lavas include basalt, andesite, and rhyolite. Amygdules, with fillings of chalcedony, chlorite, calcite, and epidote, are abundant in many of the lavas. The volcanic rocks, including amygdaloidal rhyolites, are host to numerous narrow breccia zones and seams filled with mammillary surfaces and drusy coatings of fine-grained quartz crystals or, less commonly, fluorite (Buddington, 1923, p. 75). Basalt fragments as large as a foot in diameter have been coated with chalcedony and then with pale-green fluorite. One fluorite vein ranges up to several feet in thickness. Grybeck and others (1984, p. 67) describe the fluorite as sparse, filling veins and coating chalcedony-encrusted breccia fragments over an area of at least 100 m in diameter in Tertiary volcanic rocks. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 During the visit of July 31, 1991, both fluorite-silica vein samples and amygdaloidal samples were collected. The latter were found to be common among beach pebbles. A collection of such pebbles is shown in figure Individual spheroids can be several centimeters across; many structures are compound. The spheroids resemble the small, siliceous, geodelike bodies that occur in welded tuff in Oregon and are known by the popular term "thunder eggs" (Staples, 1965). Based on examination of hand-specimens, two of these samples and three fluorite-silica samples were selected for further work. They were slabbed using a water-cooled diamond saw and prepared for study on a scanning electron microscope (SEM) equipped with an energy-dispersive Xray fluorescence system (EDS). This note reports the results of the SEM study. ANALYTICAL METHODS Samples were hand polished on unused abrasive paper and then carbon coated. The SEM used is an Etec Autoscan with a Kevex Delta Microanalysis EnergyDispersive System and Quantex software. Operating conditions of the SEM were accelerating voltage of 30 kV, filament current of 125 J!A, and working distance of 10 to 15 mm. Output was in the form of backscattered-electron images, in which grains of higher average atomic number tend to appear brighter, and X-ray spectra plots of relative intensity versus energy. Spectral assignments were made using the system's integral software. Because thick slabs were used in this SEM study in the interest of rapid sample preparation, there is very little control or knowledge of what is being analyzed at depth below the surface. Many of our EDS spectra are no doubt reflecting compositional data from more than one mineral. For this reason we have not attempted to make the corrections to raw peak intensities necessary to convert these into quantitative element abundances. Because of the geometric constraint, the lack of corrections to the intensity data, and other problems inherent to SEM-EDS analysis (especially for elements of lower atomic number),

c!l BOKAN Mol Figure 1. Locations of Snow Passage fluorite locality on Zarembo Island and of known rare earth element mineralization on neighboring Prince of Wales Island, southeast Alaska. Area covered by map shown on inset. Rare Earth Minerals in ''Thunder Eggs" from Zarembo Island

all mineral identifications and any intimations of exact compositions must be considered tentative (although hopefully reasonable) assignments. SEM RESULTS The three fluorite-silica samples examined revealed complex textures. Some textures suggest replacement or coprecipitation involving the two major phases. Other textures provide unambiguous evidence of a fluoriteprecipitating phase invading brecciated silica. In hand specimen, some fluorite has a faint purple tint while most is greenish gray. EDS analysis revealed only Ca and F in the spectra of both types. A variety of accessory minerals were found in these samples, including fairly abundant Fe-sulfide in the silica phase. However, no REE minerals were identified. The amygdaloidal pebble samples proved more mineralogically diverse. The first sample examined contains amygdules in a light-green matrix. A cut section through this sample is shown in the upper left corner of figure 2. A low-magnification backscattered-electron image is shown in figure 3. The image shows the nature of the amygdules and the texture of the matrix. EDS analysis indicates that both the amygdules and the surrounding matrix are high in silica, usually with some showing of alkali feldspar. In addition to silica and Figure 2. "Thunder egg" beach pebbles from Zarembo Island. Note centimeter scale. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

alkali feldspar, EDS analysis indicates the presence of a number of accessory phases, including a matrix grain composed of Fe-sulfide and containing small blebs of galena. A titaniferous phase located at an amygdule margin contained minor Fe and Nb. Of more interest for the present study are grains containing REE. The bright patch displayed at the center of figure 3, lying in the matrix near an amygdule margin, is rich in Y, Zr, and heavy REE (HREE-that is, REE with higher atomic mass or number). The patch is about 30 J..lm in diameter; it is composed of numerous small grains, several micrometers across, that appear to show variation in Y, Zr, and HREE content. The EDS spectrum for one of these grains is shown in figure 4.

Figure 3. Scanning electron microscope backscattered-electron image of amygdaloidal volcanic rock. Small brighter patch in center of image is composed of Y-HREE-Zr minerals. Field of view is approximately 4 mm. Vertical= 1051 counts Si (Ka) O AI (Ka) (Ka) y (La 1 ) HREE (L- series) keV y (Ka) Zr (Ka) Figure 4. X-ray spectrum of relative intensity versus energy for a grain in figure 3 bright patch. Significant peaks are labeled in terms of major contributing X-ray lines. Rare Earth Minerals in "Thunder Eggs" from Zarembo Island

Many of the peaks shown in figure 4 are composite and are identified only by the element believed to be making the major contribution. The HREE spectrum for this sample appears to be dominated by Gd, Dy, Er, and Yb. REE of even atomic number tend to be more abundant than neighboring odd atomic number REE because of their greater cosmic abundance. This particular grain may be a carbonate such as Y -bastnaesite [(Y,REE)(C03)F] or possibly a silicate such as thalenite [Y 3Si30 10(0H,F)] or tombarthite [Y iSi,H4)012_x(OH)4+2x]. We did not find any light REE (LREE) in the amygdaloidal sample. The other sample examined is a "thunder egg" containing a V -shaped central cavity in cut section (the lower of the two samples in the center of fig. 2). Based on EDS analysis the bulk of this sample is composed of silica and alkali feldspar. Accessory phases tentatively identified include apatite, zircon, and a Ti02-polymorph. These accessory minerals occur within the body of the "egg"; apatite also occurs in the central vug. The zircons show minor U content. Neither the apatite grains nor the zircon grains showed any detectable Y or REE. However, a Ti02-polymorph, about 75 J.lm in diameter, contains at its contact with siliceous matrix numerous· small blebs, each a few micrometers across, that show Y and HREE abundances very similar to those found in the amygdaloidal sample. Elsewhere in the sample, an apparent assemblage of Ti02 and potassium feldspar also shows Y and REE. Here the REE are more abundant relative toY, and they are dominated by the intermediate REE Nd, Sm, and Dy. This appears to be a different mineral than the Y -rich one that yielded the spectrum shown in figure 4. The second assemblage also contains Nb, presumably in the titaniferous phase. The REE phosphate, monazite, is also tentatively identified within this "thunder egg." The REE spectrum is dominated by Ce with lesser La and Y. Yet another REE abundance pattern occurs in grains and in areas both near the vug and in the interior of this second "thunder egg" sample. An example of the occurrence of these grains is shown in figure 5. The triangular shape occupied by the grains suggests that this area might be a filled pore space., The EDS spectrum is shown in figure 6. The REE are dominated by Ce. The only other REE detected is La, which shows up as a shoulder on the left flank of the major Ce peak. The Cerich mineral may be a carbonate. The presence of Kfeldspar is also indicated by the spectrum. Ca is also present and may be in either or both of these phases. A strikingly different LREE mineral occurs in and around the vug. Examples are the bright minerals shown in figure 7 along with dark quartz crystals lining the cavity. The EDS spectrum of the largest of the bright crystals is given in figure 8. This mineral also appears to be a LREE carbonate. However, in this case, the REE are Geologic Studies in Alaska by the U.S. Geological Survey, 1991 dominated by La and Nd. The small peak to the left of the main La peak is also contributed by La. As indicated by the trough between the two main La peaks, Ce, significantly, is absent. DISCUSSION At the Zarembo Island fluorite locality, we found REE minerals not in the fluorite-silica vein material, but rather in beach pebbles of an amygdaloidal rock and of a concretionlike or geodelike "thunder egg" similar to the amygdules. The amygdaloidal rock is presumed to be volcanic. EDS analyses show the matrix surrounding the amygdules to be high in silica, and if the composition is primary, the rock would perhaps be a rhyolite. However, the amygdules likely have formed from a fluid that may also have altered the original composition of the matrix. The REE minerals within these samples show a variety of REB-abundance patterns. Rare earth patterns reflect both the REE abundances of the environment in which a mineral forms and any fractionation due to crystal-chemical constraints. Yttrium is similar in ionic radius to the HREE. Enrichment of HREE is expected, therefore, in Y -minerals such as the amygdule-matrix mineral represented by the figure 4 spectrum. The LREE-enriched patterns in the samples are not as readily explained. An EDS spectrum of a monazite grain that displayed La and Ce peaks has been mentioned. Rare earth elements heavier than Ce are not present in this monazite at levels above EDS detection limits. This spectrum is what might be expected, in general, for a phase having enrichment of the LREE. In contrast, the REE spectrum shown in figure 6 shows only Ce and barely detectable La. This might represent a standard LREE-enrichment abundance pattern displaying a sharp maximum at Ce. Another explanation for the figure 6 spectrum, however, is suggested by the REE data shown in figure 8. This spectrum shows no Ce at all, but only La and Nd. Crystal-chemical constraints are not sufficiently discriminating to provide a REE pattern with bimodal abundance maxima at La and Nd and a minimum sufficient to exclude detection for Ce which is intermediate in ionic radius between La and Nd. This spectrum strongly suggests aCe anomaly. In igneous rocks, the REE represent a coherent geochemical group of trivalent ions that display small, regular differences in properties, principal of which is ionic radius. However, in sufficiently oxidizing sedimentary environments, the coherence breaks down as Ce, alone among the REE, becomes tetravalent.· Tetravalent Ce is then the only REE entering Mn nodules and crusts in any great quantity, and this accounts for the negative Ce anomaly of seawater, fishbones, phos- .

phorites, and so on. We believe that the REE data shown in figure 8 indicate that this mineral also has a negative Ce anomaly. This would require that Ce was tetravalent at some stage in the development of this mineral. A positive Ce anomaly would explain the high relative abundance of Ce indicated for the mineral yielding the spectrum in figure 6. The abundance data for this phase do riot of themselves make as conclusive an argument for anomalous Ce as do the figure 8 data. However, the Ce-rich and Ce-deficient minerals do occur in close proximity within the "thunder egg," and it is likely that the unusual complementary abundances of Ce in both minerals have the same explanation (that is, stabil~ ity of tetravalent Ce) rather than resulting from independent causes. In any case, we consider the evidence for a

Figure 5. Scanning electron microscopebackscattered-electron image. Bright area is composed of LREE-enriched grains that may be filling pore space. Field of view isapproximately40 ~m. Vertical 3350 counts Ce (La 1) (Ka) Si (Ka) Ca (Ka) La · (La1) keV Ce Figure 6. X-ray spectrum for a grain within figure 5 bright area. Ce is by far most abundant REE. Rare Earth Minerals in "Thunder Eggs" from Zarembo Island

tetravalent-Ce effect to be very strong for the Cedeficient mineral. As mentioned above, Ce anomalies are features normally encountered in oxidized sedimentary environments, not in igneous rocks, so the questions arise as to what they represent within this "thunder egg" and at what stage they formed. The large crystal shown in figure ?appears to demonstrate clearly that the Ce-deficient phase is not merely a surface coating within the vug but rather has a crystal form that is well defined within the enclosing siliceous matrix. It is possible that, during a period of oxidative weathering, Ce was preferentially leached as a tetravalent ion from the mineral. If this were the case, then these rocks might constitute a ready source of leachable Ce and one that might form eco-

Figure 7. Scanning electron microscope backscattered-electron image of REE minerals (bright) and quartz crystals (dark) around a "thunder egg" vug. Field of view is approximately 0.7 mm. Vertical 6272 counts La (La 1) keV Nd (La 1) Figure 8. X-ray spectrum for large bright crystal in northeast quadrant of figure 7. Note absence of Ce. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

nomic beach deposits. However, the occurrence of such an event is uncertain. For one thing, it would require that the nearby Ce-rich minerals have escaped such Ce loss. Further, this "thunder egg" is a typically well indurated sample, and it is not at all clear that its interior was open to weathering. For these reasons we currently favor the interpretation that Ce was tetravalent at the time that this phase formed, and that the Ce-deficient mineral crystallized at the same time as its siliceous surroundings during Tertiary or Quaternary hydrothermal activity. The identification of anomalous Ce in this process has genetic significance because it indicates hydrothermal conditions sufficiently oxidizing to stabilize tetravalent Ce. Although we did not find REE minerals within samples of the fluorite-silica material, the presence of this veining offers at least the possibility of (re)mobilization of REE. Within the "thunder eggs," EDS analysis of minerals indicates that Ti, Nb, and Zr were mobile in addition to REE and Y. Very similar assemblages of mobile elements have been identified by Belkin (1991) for late-stage magmatic brine activity within a diabase in Pennsylvania and by Willis ·and others (1990) and Flohr (1991) for quartz-Ti02 veins near Magnet Cove, Arkansas. This type of metasomatism also may have had a role in the enrichment of Y, Zr, and REE at major deposits such as Strange Lake, QuebecLabrador (Salvi and Williams-Jones, 1990), and Thor Lake, N.W.T. (Pinckston and Smith, 1988). REFERENCES CITED Belkin, H.E., 1991, Hydrothermal transport of zirconium in mafic igneous rocks: Evidence from fluid inclusions in ilmenite and implications for petrogenetic interpretation [abs.]: Plinius, v. 5, p. 19. Brew, D.A., Ovenshine, A.T., Karl, S.M., and Hunt, S.J., 1984, Preliminary reconnaissance geologic map of the Petersburg and parts of the Port Alexander and Sumdum 1:250,000 quadrangles, southeastern Alaska: U. S. Geological Survey Open-File Report 84-405, 43 p. Buddington, A.F., 1923, Mineral deposits of the Wrangell District, in Brooks, A.H., and others, Mineral resources of Alaska: U.S. Geological Survey Bulletin 739, p. 51-75. Cathrall, J.B., Day, G.W., Hoffman, J.D., and McDanal, S.K., 1983, A listing and statistical summary of analytical results for pebbles, stream sediments, and heavy-mineral concentrates from stream sediments, Petersburg area, southeast Alaska: U.S. Geological Survey Open-File Report 83-420-A, 279 p. Flohr, M.J.K., 1991, Titanium-, vanadium-, and niobium mineralization at the Christy deposit, Magnet Cove alkaline igneous rock complex, Arkansas [abs.]: Geological Society of America Abstracts with Programs, v. 23, no. 5, p. A292. Grybeck, D.J., Berg, H.C., and Karl, S.M., 1984, Map and description of the mineral deposits in the Petersburg and eastern Port Alexander quadrangles, southeastern Alaska:U.S. Geological Survey Open-File Report 84-837, 87 p. Pinckston, D.R., and Smith, D.G.W., 1988, Mineralogy of the Lake Zone, Thor Lake, N.W.T. [abs.]: Geologic Association of Canada Program with Abstracts, v. 13, p. 99. Salvi, S., and Williams-Jones, A.E., 1990, The role of hydrothermal processes in the granite-hosted Zr, Y, REE deposit at Strange Lake, Quebec/Labrador: Evidence from fluid inclusions: Geochimica et Cosmochimica Acta, v. 54, p. 2403-2418. Staples, L.W., 1965, Origin and history of the thunder egg: The Ore Bin, v. 27, no. 10, p. 195-204. Willis, M.A., Pasteris, J.D., and Shock, E.L., 1990, Hydrothermal transport of titanium as exemplified by quartz-titanium dioxide veins near Magnet Cove, Arkansas [abs.]: Geological Society of America Abstracts with Programs, v. 22, no. 7, p. A363. Reviewers: Marta j.K. Flohr and Philip A. Baedecker Rare Earth Minerals in "Thunder Eggs" from Zarembo Island

GEOLOGIC FRAMEWORK STUDIES-ARTICLES Upper Devonian Shallow-Marine Siliciclastic Strata and Associated Fauna and Flora, Lime Hills D-4 Quadrangle, Southwest Alaska By Robert B. Blodgett and Wyatt G. Gilbert Abstract A 1 03.6-m-thick section of Frasnian (lower Upper Devonian) siliciclastic strata in the Lime Hills D-4 quadrangle, southwestern Alaska, contains fauna and flora consistent with a shallow-water, nearshore depositional environment. This section occurs within the lower part of the Mystic sequence and represents a period of shallowwater deposition that followed deposition of deeper water shale and chert. The section is part of the Farewell terrane, which encompasses the Nixon Fork, Dillinger, and Mystic terranes of former usage. The fossil fauna of the measured section is dominated by brachiopods, followed secondarily in abundance by gastropods. Several species of the megafauna are also found in Frasnian fauna from the Shellabarger Pass area, northwestern Talkeetna quadrangle. Biogeographically, this fauna is allied with other Frasnian faunas from western North America, the Russian Platform, Novaya Zemlya, Taimyr, and Kolyma. Locally, the uppermost strata of the underlying Dillinger sequence include deep-water turbiditic sandstone and shale, with minor carbonate debris flows yielding fauna of Lochkovian to Pragian (early Early to middle Early Devonian) age. The base of the overlying Mystic sequence is represented by Emsian (upper Lower Devonian) shallow-water platform carbonates. These are succeeded by deep-water shale and chert, followed by black shale, and succeeded in turn by Frasnian greenish-gray shale and interbedded orange siltstone and black shale of the measured section. Regional correlation of Frasnian strata of the Lime Hills and McGrath quadrangles indicates the presence of a carbonate platform in the area of the Lyman Hills, which is correlative to the siliciclastic strata to the south and southeast in the Lime Hills D-4 quadrangle. The Frasnian siliciclastic strata may represent an orogenic clastic wedge derived from nearby tectonic highlands, probably situated to the southeast. INTRODUCTION In this paper we describe a measured stratigraphic section in Frasnian (lower Upper Devonian) siliciclastic strata in the Lime Hills D-4 quadrangle and discuss both Geologic Studies in Alaska by the U.S. Geological Survey, 1991 its contained fossil biota and its implications for the regional tectonic setting and correlation. Faunal data are also presented on some subjacent lithologic units. The research reported here results from a cooperative study between geologists of the U.S. Geological Survey (USGS) and the Alaska Division of Geological and Geophysical Surveys. Field work was conducted in the study area (Lime Hills D-4 and D-5 quadrangles) by W.G. Gilbert during the summers of 1989 and 1990; that of R.B. Blodgett was undertaken during the summer of 1990. TERRANES AND SEQUENCES The Paleozoic strata of the Lime Hills D-4 quadrangle are part of the Farewell terrane (Decker and others, in press), which encompasses the earlier described but obviously genetically related Dillinger, Nixon Fork, and Mystic terranes (fig. 1). Previously, strata of this quadrangle were included within the Dillinger terrane (Jones and Silberling, 1979; Jones and others, 1981, 1987). The Dillinger terrane was described by Jones and others (1981, p. 6-7) as a "coherent but complexly folded assemblage of lower and middle Paleozoic graptolitic shales, sandstone turbidites, and basinal limestones of known Ordovician to Devonian age." The Mystic terrane was defined by Jones and others (1981, p. 6) to consist of "highly folded and partly disrupted shallow-water marine clastics and carbonate ro~ks of Late Devonian age overlain by radiolarian cherts and flyschlike graywacke, argillite and conglomerate of late Paleozoic age." Also included were Ordovician graptolitic shale associated with pillow basalt, blocks of Silurian platform rocks, and pillow basalts of presumed Triassic age. Gilbert and Bundtzen ( 1984) considered the Dillinger and Mystic terranes to represent a single stratigraphic succession of Paleozoic to Triassic age, preferring to apply the term "sequence" to each. They considered the underlying Dillinger sequence to be a Cambrian to Lower Devonian deep-water succession that is followed depositionally

by the Mystic sequence, which consists of laterally variable Devonian to Triassic(?) shallow-water to nonmarine sedimentary rocks and intrusive and extrusive mafic and ultramafic rocks. Both sequences are well exposed in the Lime Hills D-4 and D-5 quadrangles. The Nixon Fork terrane is now recognized to represent a predominantly 100 MILES 1 00 Kl LOM ETERS Figure 1. Generalized geologic map of southwestern Alaska showing location of cited place names and component "subterranes" (Dillinger, Nixon Fork, and Mystic) of the Farewell terrane. Map modified from jones and others (1987). Dillinger "subterrane" here includes East Fork terrane of Dutro and Patton (1982). Map symbols: 1, study area; 2, Lyman Hills; 3, headwater region of the Gagaryah River; 4, VABM Steep; 5, North Lime Lake; 6, Door Mountains; 7, St. johns Hill; 8, Farewell Mountain; 9, Shellabarger Pass; 10, East Fork Hills; NX, Nixon Fork terrane; DL, Dillinger terrane; MY, Mystic terrane. Upper Devonian Shallow-Marine Siliciclastic Strata and Associated Fauna and Flora, Lime Hills D-4 Quadrangle

shallow-water carbonate platform that is laterally equivalent to deeper water Cambrian to Devonian rocks of the Dillinger terrane exposed to the south (Blodgett, 1983; Blodgett and Gilbert, 1983; Bundtzen and Gilbert, 1983; Gilbert and Bundtzen, 1983; Blodgett and Clough, 1985). DILLINGER SEQUENCE The deep-water lower and middle Paleozoic strata of the Dillinger sequence occur throughout west-central and southwestern Alaska; this region includes the southeastern McGrath quadrangle (B undtzen and others, 1982; Kline and others, 1986; Bundtzen and others, 1987; Gilbert and others, 1988, 1989), the northern Lime Hills quadrangle (Gilbert, 1981; Gilbert and others, 1990), and as far south as the western end of the Door Mountains (loc. 6 in fig. 1), east of the Hoholitna River in the Sleetmute A-2 quadrangle (R.B. Blodgett, personal observations, 1984 and 1985). Similar deep-water strata occur in the East Fork Hills (loc. 10 in fig. 1) in the southern part of the Medfra quadrangle (East Fork Hills Formation of Dutro and Patton, 1982) and have recently been recognized as far north as the ·Kantishna River C-5 quadrangle (R.B. Blodgett and W.G. Gilbert, personal observations, 1991). In the study area (loc. 1 in fig. 1), the upper part of the Dillinger sequence includes deep-water turbiditic sandstone, siltstone, and shale. Limestone is uncommon and occurs either as carbonate turbidites (debris flows) or as platy, thin-bedded lime mudstone. The latter carbonate type probably represents times when siliciclastic sediment input was strongly reduced. One limestone debris flow (USGS locality 11940-SD) was sampled for conodonts. It consists of a thin limestone lens reaching up to 0.15 m (0.5 ft) in thickness that is laterally persistent for 4.5 m (15 ft). This lens is in an outcrop of steeply inclined, thinly interbedded shale and sandstone situated along the left bank of an unnamed northwestflowing creek in SE~NW~NW~ sec. 7, T. 20 N., R. 28 W., Lime Hills D-4 quadrangle (lat 61°50'41" N., long 154°25'50" W.). A predominantly pelagic fauna is found in the limestone bed (USGS locality 11940-SD), consisting of abundant dacryoconarid tentaculitids, orthoconic nautiloids, and an undetermined bivalve. Conodonts recovered from this limestone all belong to a single species, Pandorinellina optima, indicative of a Lochkovian (but not earliest Lochkovian) to Pragian age (Anita G. Harris and Robert Stamm, USGS, oral commun., 1991). The contact with the strata of the immediately overlying Mystic sequence has not been observed yet in the study area. MYSTIC SEQUENCE The type area of the Mystic sequence near Shellabarger Pass (loc. 9 in fig. 1) in the Talkeetna Geologic Studies in Alaska by the U.S. Geological Survey, 1991 quadrangle (Jones and others, 1981) includes Upper Devonian sandstone and limestone, uppermost Devonian and Mississippian radiolarian chert, and Pennsylvanian chert and argillite, succeeded structurally (and probably depositionally) by flyschlike rocks, including conglomerate that contains clasts of limestone and black cherty argillite (Jones and others, 1982). Middle Pennsylvanian plant fossils (identified by S.H. Mamay) were collected from a conglomerate lens at one locality; hence, these flyschlike rocks were presumed to be of late Paleozoic age. Older (pre-Devonian) lithologies reported by Jones and others ( 1981) are part of the Dillinger sequence (T.K. Bundtzen and W.G. Gilbert, personal observation, 1983). Late Paleozoic shallow-marine and nonmarine clastic rocks and Triassic(?) pillow basalt form isolated exposures that overlie the Dillinger sequence in the McGrath quadrangle (B undtzen and others, 1982; Bundtzen and others, 1987; Gilbert and others, 1989). The lowermost unit recognized in the Mystic sequence of the study area is a shallow-water, platform carbonate (Dl unit of Gilbert and others, 1990) that is at least 100 m thick. It consists of dark-gray lime mud-, wacke-, and grainstone bearing stromatoporoids, Amphipora, tabulate corals, and crinoid ossicles. This unit has been only cursorily examined for fossils. One locality that yielded both megafossils and conodonts is USGS locality 11941-SD, which is at the southeast end of a large limestone exposure atop a small hill (elevation 2001 ft) in NE~SE~SW~ sec. 15, T. 20 N., R. 29 W ., Lime Hills D-5 quadrangle (lat 61 °49'08" N., long 154°30'48" W.). The presence here of distinctive twohole crinoid ossicles indicates an Emsian to early Eifelian age. Conodonts from this locality include polygnathids and Pelekysgnathus similar to Emsian species (Robert Stamm, USGS, oral commun., 1991). The combined ranges suggest an Emsian (late Early Devonian) age for strata of this locality. This limestone is succeeded by the Dsc unit of Gilbert and others (1990), which consists of very· thinly bedded chert, cherty argillite, and shale, and which hosts the Gagaryah barite deposit (Bundtzen and Gilbert, in press). This unit appears to have been deposited under deep-water conditions, in contrast to both the subjacent and superjacent units. The next succeeding unit is the Ds unit of Gilbert and others (1990), which is gradational with the underlying Dsc unit. Unit Ds includes the strata of the described measured section in its upper part. Stratigraphic Section The measured stratigraphic section is located (fig. 2) in SE~SW~SE~ sec. 12, T. 20 N., R. 29 W., Lime Hills D-4 quadrangle. The section extends uphill along a northwest-southeast-trending transect. Strata of this section be-

long to the upper part of the Ds unit of Gilbert and others (1990). The base of the measured section (fig. 3) was placed at the transition between underlying unfossiliferous, black shale and superjacent thin-bedded, grayish-green shale. It is situated at an elevation of approximately 2,850 ft (lat 61 °49'56" N., long 154°26'51" W.). The top of the measured section is at the base of a thick, lithic sandstone interval that begins around an elevation of 3,200 ft (lat 61 °49'52" N., long 154°26'38" W.) and forms the base of the uPzs unit of Gilbert and others (1990). The lower part of the measured section consists of 32.0 m ( 105 ft) of greenish-gray shale (locally contorted). This shale lacks visible fossils for the most part, except near the top where crinoid ossicles and small brachiopods were recovered (USGS localities 11928-SD, 11929-SD). At 32.0 m above the base of the section, the greenish-gray shale grades into an overlying interval consisting of predominantly orange-brown-weathering siltstone interbedded with minor black shale. The latter lithology tends to predominate in the upper part of this interval. This interval extends from 32.0 m (1 05 ft) to 103.6 m (340 ft) above the base of the section [with a covered interval between 34.1 m (112 ft) and 39.0 m (128ft)]. A silty limestone lens (USGS locality 11938SD) was found at 86.6 m (284 ft) above the base. Openmarine fossils are found thoughout this interval, both in R. 29 W. R. 28 W. 2 MILES 2 KILOMETERS Figure 2. Location of measured section (see fig. 3) of unnamed lower Upper Devonian siliciclastic marine strata in Lime Hills D-4 quadrangle. Stippled pattern shows distribution of Ds unit of Gilbert and others (1990). Base at northwest end of section. Base map from U.S. Geological Survey, 1958 (minor revisions 1975), Lime Hills D-4 1 :63,360 topographic map. outcrop and in talus rubble. Brachiopods are the most common faunal element, followed in abundance by gastropods. Brachiopods include Eleutherokomma n. sp. (fig. 4.1-4.5), Schizophoria sp. (fig. 4.1, 4.6, 4.7), and Spinatrypa sp. (fig. 4.1 ). Gastropods include the following species: Aglaoglypta n. sp. (fig. 4.8), n. gen. aff Acanthonema, n. sp. (fig. 4.1 0, 4.11 ), and Orecopia cf. 0. mccoyi (Walcott, 1884) (fig. 4.9). Gyrogonites of the charophyte genus Sycidium (a green alga) (fig. 4.12, 4.13) are abundant at USGS locality 11930-SD. A 15.3-m (50-ft) covered interval separates the fossiliferous siltstone and shale interval from an overlying interval of medium- to thick-bedded, fine- to medium-grained, brown, locally laminated lithic sandstone. The sandstone is predominantly fine grained, with subangular grains, and contains numerous small, angular shale chips. No visible fossil remains were noted from these beds. This upper lithic sandstone interval has been mapped by Gilbert and others ( 1990) as the lowermost part of their uPzs unit and was shown here to rest upon their Ds unit. Age An early Late Devonian (Frasnian) age is indicated by both the megafauna and conodonts collected from the strata of the measured section. The brachiopod genus Eleutherokomma Crickmay, 1950, occurs in strata of late Givetian (latest Middle Devonian) to middle Frasnian (middle early Late Devonian) age. This genus is represented here by a new species (fig. 4.1-4.5) that shows the rather distinctive microomament of the genus, consisting of strong, closely spaced concentric lamellae and subordinate radial capillae. This species differs from most of the described species in being somewhat larger than typical for the genus and in being considerably less mucronate. The new species appears to most closely resemble Eleutherokomma reidfordi Crickmay, 1950, from the basal part of the Hay River Formation of the District of Mackenzie, Northwest Territories. The latter species is of middle Frasnian age. The close relationship between these two species is supported by direct comparison with material of E. reidfordi that has been deposited in the National Museum of Natural History. They both share common features such as relatively large size, more numerous costae, and development of specimens that are weakly mucronate. The new species can be distinguished from E. reidfordi in having a more strongly rounded dorsal fold and in being less transverse. The brachiopod genera Schizophoria and Spinatrypa are both relatively long ranging, and until species-level affinities are elucidated, they provide no critical age resolution. The gastropod genus Orecopia is recognized only from the Frasnian. Among North American species, the Lime Hills form illustrated here (fig. 4.9) is most similar Upper Devonian Shallow-Marine Siliciclastic Strata and Associated Fauna and Flora, Lime Hills D-4 Quadrangle

to the type species, Orecopia mccoyi (Walcott, 1884) from the Devils Gate Limestone of central Nevada and its stratigraphic equivalents elsewhere in the Great Basin. The species resemblance is especially strengthened by the mutual occurrence of a strong angulation at the juncture between the upper and outer whorl surfaces. According to Pedder (1966, p. 144), the Nevada species ranges in age from early Frasnian (argentarius Zone) to possibly middle Frasnian. The gastropod genus Aglaoglypta is nearly wholly restricted to strata of Frasnian age, though a single species occurs in the late Givetian (late Middle Devonian) of Germany. Biogeographically, the megafauna is closely allied with other Frasnian faunas from western North America, the Russian Platform, Novaya Zemlya, Taimyr, and Kolyma, as shown by the presence of the genera Eleutherokomma and Orecopia. Several of the Frasnian megafauna species occurring in the upper part of the measured section are identical to species found in USGS fossil collections from Frasnian strata of the Shellabarger Pass region, northwestern Talkeetna quadrangle (R.B. Blodgett, personal observation, 1991 ). Conodonts recovered from 1. 7 kg of limestone at USGS locality . 11938-SD yielded specimens identified by Robert Stamm (USGS, written common., 1991) as Polygnathus aff. P. aspelundi Savage and Funai [four platform (Pa) elements with broken blades, six platform (Pa) elements with broken platforms, and one platform Fine- to medium grained, brown lithic sandstone, locally laminated EXPLANATION 11937-SD 11935-SD Orange-brown-weathering siltstone (predominant) interbedded with minor black shale, which predominates near top of interval Greenish-gray shale, beds contorted, with small crinoid ossicles common near top of interval Unfossiliferous, black, thin-bedded shale [§] Sandstone Limestone lens

Not exposed ..,_11938-SD Fossil locality Figure 3. Columnar measured section of unnamed lower Upper Devonian siliciclastic marine strata. Location of section shown on figure 2. Arrows to right of column indicate USGS fossil localities (see descriptions in appendix). Geologic Studies in Alaska by the U.S. Geological Survey, 1991

(Pa) blade fragment]. All specimens were slightly abraded, with argillaceous and silt material attached. According to Stamm, "the complete stratigraphic and consequent age of P. aspelundi is poorly constrained. The faunal assemblage of Savage and Funai ( 1980) was assigned a probable early Frasnian age (Lower asymmetricus Zone) based on occurrences of P. aff. P. dengleri and Pandorinellina insita. Klapper and Lane (1985) feel that Savage and Funai' s collections are of late Frasnian age (Lower to Upper gigas Zones) based on the occurrence of P. unicomis which was misidentified asP. aff. P. d(mgleri by Savage and Funai." Environment of Deposition The abundance, but low diversity, of the brachiopods, along with the relative abundance and diversity of gastropods, suggest an extremely shallow-water, openmarine, nearshore environment for that part of the section above 32.0 m. The presence of brachiopods and crinoid ossicles throughout this interval indicates conditions of normal marine salinity. The abundance of the charophyte Sycidium (a green alga) at USGS locality 11930-SD indicates very shallow (upper end of the photic zone) depths. Although charophytes are strictly found in nonmarine or brackish-water environments today. some of their earliest Devonian antecedents appear to have inhabited shallow-water, marine environments (Racki, 1982). Large, fragmentary vascular plant remains were noted in abundance near and at the levels of USGS localities 11931-SD and 11935-SD. It seems plausible that their abundance would require the nearby presence of a land area from which such vegetation could have been introduced. The lower part of the measured section, from 0 to 32~0 m above the base, is interpreted to represent relatively deeper water, as it is much finer grained (being a shale) and thinner bedded. This interpretation is supported by the absence of fossils from the lower, greater part of this interval, as well as by the transition further below into unfossiliferous black shale, and then even further below into shale and chert (Dsc unit of Gilbert and others, 1990). REGIONAL CORRELATION Laterally equivalent to the Frasnian siliciclastic strata in the study area are platform carbonates that are . exposed 30 to 65 km to the west and northwest in the Lyman Hills (loc. 2 in fig. 1). Most, if not all, of the thick, carbonate platform succession (uDl unit of Gilbert, 1981) appears to be of Frasnian age. These strata consist almost wholly of carbonate rocks, but a thin basal sandstone bed has been recognized at several localities in the McGrath A-5 quadrangle (R.B. Blodgett, personal observation, 1979) (loc. 2 in fig. 1 ). This basal sandstone may represent a distal tongue of the thicker, clastic succession recognized to the southwest in the Lime Hills D-4 quadrangle. It and the overlying thick Frasnian carbonate succession form the base of the Mystic sequence in the Lyman Hills and appear to rest .unconformably upon older deep-water strata of the Dillinger sequence. No Emsian and (or) Middle Devonian strata typical of the basal Mystic sequence in the study area and else-· where have been recognized in the Lyman Hills. The lower Upper Devonian carbonate platform succession crops out nearly continuously along the south side of the Cheeneetnuk River from the McGrath A-4 quadrangle in the northeast (15 to 20 km northeast of loc. 2 in fig. 1), southward to the vicinity of V ABM Steep (loc. 4 in· fig. 1) in the Lime Hills D-8 quadrangle. Other areas of exposure within the vicinity of the Lyman Hills include prominent exposures just east of North Lime Lake (loc .. 5 in fig. 1) (Lime Hills B-6 and B-7 quadrangles) and also in a southwest-northeast-trending synclinorium in the headwater r!gion of the Gagaryah River (loc. 3 in fig. 1) (Lime Hills C-6 quadrangle). North of the Farewell fault this same succession can be recognized in the area of Farewell Mountain (loc. 8 in fig. 1) and St. Johns Hill (loc. 7 in fig. 1 )(McGrath B-2 and B-3 quadrangles). Fossil algae have been described from this succession (Mamet and Plafker, 1982) based on five sample localities, three situated in the northern Lyman Hills, and the remaining two localities being situated approximately 6 km south-southwest of the town of Farewell (loc. 7 in fig. 1 ). Also, some limited discussion of a silicified ostracode fauna from this sequence in the McGrath A-5 quadrangle was given by W.K. Braun in Blodgett (1983, p. 128). CONCLUSIONS The measured section in the Lime Hills D-4 quadrangle documents the existence of Frasnian shallow-water siliciclastic strata equivalent to platform carbonate rocks to the west and northwest in the Lyman Hills. This contrast in facies is further evidence for the heterogeneity of the ¥ystic sequence in southwestern Alaska· (Bundtzen and Gilbert, 1983; Gilbert and Bundtzen, 1984). The presence of older (pre-Frasnian) Mystic sequence strata in the study area, but absent in the Lyman Hills, implies more continuous marine sedimentation during the Devonian in the former area. The ·Frasnian siliciclastic strata may represent an orogenic clastic wedge derived from nearby tectonic highlands. The presence of a coeval Frasnian carbonate platform to the north in the Lyman Hills, as well as the regional trend of Upper Devonian Shallow-Marine Siliciclastic Strata and Associated Fauna and Flora, Lime Hills D-4 Quadrangle.

this siliciclastic stratal succession (with strikingly similar rocks and fauna occurring in Shellabarger Pass, northwestern Talkeetna quadrangle), suggest that these highlands were probably located to the southeast. REFERENCES CITED Blodgett, R.B., 1983, Paleobiogeographic affinities of Devonian fossils from the Nixon Fork terrane, southwestern Alaska, in Stevens, C.H., ed., Pre-Jurassic rocks in western North American suspect terranes: Los Angeles, Calif., Society of Economic Paleontologists and Mineralogists, Pacific Section, p. 125-130. Blodgett, R.B., and Clough, J.G., 1985, The Nixon Fork terrane-Part of an in-situ peninsular extension of the Paleozoic North American continent [abs.]: Geological Society of America Abstracts with Programs, v. 17, no. 6, p. 342. Blodgett, R.B., and Gilbert, W.G., 1983, The Cheeneetnuk Limestone, a new Early(?) to Middle Devonian formation in the McGrath A-4 and A-5 quadrangles, west-central Alaska: Alaska Division of Geological and Geophysical Surveys Professional Report 85, 6 p., 1 sheet, scale 1:63,360. Bundtzen, T.K., and Gilbert, W.G., 1983, Outline of geology and mineral resources of the upper Kuskokwim region, Alaska: Journal of the Alaska Geological Society, v. 3, p. ---in press, The geology and geochemistry of the Gagaryah barite deposit, western Alaska Range, Alaska, in Reger, R.D., ed., Short notes on Alaskan Geology-1991: Alaska Division of Geological and Geophysical Surveys Professional Report 111. Bundtzen, T.K., Kline, J.T., and Clough, J.G., 1982, Preliminary geology of the McGrath B-2 quadrangle, Alaska: Alaska Division of Geological and Geophysical Surveys Open-File Report 149, 22 p., 1 sheet, scale 1:63,360. Bundtzen, T.K., Kline, J.T., Smith, T.E., and Albanese, M.D., 1987, Geologic map of the McGrath A-2 quadrangle, Alaska: Alaska Division of Geological and Geophysical Surveys Professional Report 91, 20 p., 1 sheet, scale 1:63,360. Crickmay, C.H., 1950, Some Devonian Spiriferidae from Alberta: Journal of Paleontology, v. 24, p. 219-225. Decker, John, Bergman, S.C., Blodgett, R.B., Box, S.E., Bundtzen, T.K., Clough, J.G., Coonrad, W.L., Gilbert, W.G., Miller, M.L., Murphy, J.M., Robinson, M.S., and Wallace, W.K., in press, Geology of southwestern Alaska: Boulder, Colo., Geological Society of America, Geology of North America, v. F1, chapter II-F. Dutro, J.T., Jr., and Patton, W.W., Jr., 1982, New Paleozoic formations in the northern Kuskokwim Mountains, westcentral Alaska: U.S. Geological Survey Bulletin 1529-H, p. H13-H22. Gilbert, W.G., 1981, Preliminary geologic map and geochemical data, Cheeneetnuk River area, Alaska: Alaska Division of Geological and Geophysical Surveys OpenFile Report 153, 10 p., 2 pis. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Gilbert, W.G., and Bundtzen, T.K., 1983, Paleozoic stratigraphy of Farewell area, southwest Alaska Range, Alaska [abs.]: Alaska Geological Society Symposium, New Developments in the Paleozoic Geology of Alaska and the Yukon, Anchorage, Alaska, 1983, Program and Abstracts, p. 10-11. ---1984, Stratigraphic relationships between Dillinger and Mystic terranes, western Alaska Range, Alaska [abs]: Geological Society of America Abstracts with Programs, v. 16, no. 5, p. 286. Gilbert, W.G., Bundtzen, T.K., Kline, J.T., and G.M. Laird, 1990, Preliminary geology and geochemistry of the southwest part of the Lime Hills D-4 quadrangle, Alaska: Alaska Division of Geological and Geophysical Surveys Report of Investigations 90-6, 1 sheet, scale 1:63,360. Gilbert, W.G., Solie, D;N., and Kline, J.T., 1988, Geologic map of the McGrath A-3 quadrangle, Alaska: Alaska Division of Geological and Geophysical Surveys Professional Report 92, 2 sheets, scale 1:63,360. Gilbert, W.G., Solie, D.N., Kline, J.T., and Dickey, D.B., 1989, Geologic map of the McGrath B-3 quadrangle, Alaska: . Alaska Division of Geological and Geophysical Surveys Professional Report 102, 2 sheets, scale 1:63,360. Jones, D.L., and Silberling, N.J., 1979, Mesozoic stratigraphy-The key to tectonic analysis of southern and central Alaska: U.S. Geological Survey Open-File Report 791200,41 p. Jones, D.L., Silberling, N.J., Berg, H.C., and Plafker, George, 1981, Map showing tectonostratigraphic terranes of Alaska, columnar sections, and summary description of terranes: U.S. Geological Survey Open-File Report 81792, 20 p., 2 sheets, scale 1:2,500,000. Jones, D.L., Silberling, N.J., Coney, P.J., and Plafker, George, 1987, Lithotectonic terrane map of Alaska (west of the 141st Meridan): U.S. Geological Survey Map MF-1874-A, 1 sheet, scale 1 :2,500,000. Jones, D.L., Silberling, N.J., Gilbert, Wyatt, and Coney, Peter, 1982, Character, distribution, and tectonic significance of accretionary terranes in the central Alaska Range: Journal of Geophysical Research, v. 87, p. 3709-3717. Klapper, Gilbert, and Lane, H.R., 1985, Upper Devonian (Frasnian) conodonts of the Polygnathus biofacies, N.W.T., Canada: Journal of Paleontology, v. 59, p. 904951. Kline, J.T., Gilbert, W.G., and Bundtzen, T.K., 1986, Preliminary geologic map of McGrath C-1 quadrangle, Alaska: Alaska Division of Geological and Geophysical Surveys Report of Investigations 86-25, 1 sheet, scale 1:63,360. Mamet, B.L., and Plafker, G., 1982, A Late Devonian (Frasnian) Microbiota from the Farewell-Lyman Hills area, west-central Alaska: U.S. Geological Survey Professional Paper 1216-A, p. A1-A10. Pedder, A.E.H., 1966, The Upper Devonian gastropod Orecopia in western Canada: Palaeontology, v. 9, p. 142147. Racki, Grzegorz, 1982, Ecology of the primitive charophyte algae; a critical review: Neues Jahrbuch ftir Geologie und Palaontologie Abhandlungen, v. 162, p. 388-399. Savage, N.M., and Funai, C.A., 1980, Devonian conodonts of probable early Frasnian age from the Coronados Islands of

southeastern Alaska: J oumal of Paleontology, v. 54, p. Walcott, C.D., 1884, Paleontology of the Eureka District: U.S. Geological Survey Monograph 8, 298 p. Reviewers: Bruce M. Gamble and J.G. Johnson APPENDIX-FOSSIL LOCALITIES AT MEASURED SECTION Collections of R.B. Blodgett USGS locality 11928-SD (field station 90ABd4), 27.4 m (90 ft) above base of section: Eleutherokomma n. sp., crinoid ossicles. USGS locality 11929-SD (field station 90ABd5), 30.2 m (99 ft.) above base of section, collection from small outcrop: undetermined rhynchonellid brachiopod. USGS locality 11930-SD (field station 90ABd6), 33.2 m (1 09 ft) above base of section, collection mostly from in situ rubble: Spinatrypa sp., Eleutherokomma n. sp., n. gen., n. sp. of gastropod (a similar, probably conspecific form occurs in the Frasnian of Shellabarger Pass, northwestern Talkeetna quadrangle), crinoid ossicles, Sycidium sp. USGS locality 11931-SD (field station 90ABd7), 34.1 m (112 ft) above base of section, collection from locally derived rubble: Eleutherokomma n. sp., undetermined high-spired gastropod, undetermined solitary rugose coral, crinoid ossicles, vascular plant debris. USGS locality 11932-SD (field station 90ABd8), 39.0 m (128 ft) above base of section, collection from locally derived rubble: Schizophoria sp., Spinatrypa sp., Eleutherokomma n. sp., undetermined gastropods. USGS locality 11933-SD (field station 90ABd9), 46.3 m (152 ft) above base of section, collection from locally derived rubble near base of prominent resistant knoblike exposure: Schizophoria sp., Eleutherokomma n. sp., Aglaoglypta n. sp., bellerophontid gastropod with spiral ornament, Straparollus (Straparollus) sp., S. (Euomphalus) sp., undetermined bivalve, crinoid ossicles. USGS locality 11934-SD (field station 90ABd10), 49.4 m (162 ft) above base of section, collection from thin, laterally continuous, fossiliferous lens in lower part of resistant exposure: Schizophoria sp., Eleutherokomma n. sp., crinoid ossicles. USGS locality 11935-SD (field station 90ABdll), 59.4 m (195 ft) above base of section, collection from fossiliferous lens from upper part of resistant exposure: Schizophoria sp., Eleutherokomma n. sp., crinoid ossicles, vascular plant debris. USGS locality 11936-SD (field station 90ABd12), 71.6 m (235 ft) above base of section, collection made from locally derived rubble: Schizophoria sp., Spinatrypa sp., Eleutherokomma n. sp., crinoid ossicles. USGS locality 11937-SD (field station 90ABd13), 79.2 m (260 ft) above base of section, collection made from more or less in place lens: Schizophoria sp., Spinatrypa sp., Eleutherokomma n. sp., undetermined bellerophontid gastropod, Straparollus (Straparollus) sp., Orecopia cf. 0. mccoyi (Walcott, 1884), undetermined gastropods, undetermined bivalve, crinoid ossicles. USGS locality 11938-SD (field station 90ABd14), 86.6 m (284 ft) above base of section, collection made in locally derived rubble: Schizophoria sp., Eleutherokomma n. sp., indeterminate bellerophontid gastropod, crinoid ossicles. Collection of W.G. Gilbert USGS locality 11939-SD (field station 89WG 112): collection. made from rubble between 32.0 to 62.5 m (1 05 to 205 ft) above base of section: Schizophoria sp., Spinatrypa sp., Eleutherokomma n. sp., crinoid ossicles. Upper Devonian Shallow-Marine Siliciclastic Strata and Associated Fauna and Flora, Lime Hills 0~4 Quadrangle

Figure 4. Early Late Devonian (Frasnian) fossils from measured section in SE,4SW\SE~4 sec. 12, T. 20 N., R. 29 W., Lime Hills D-4 quadrangle. Latex replica of siltstone block, USNM 460756, x1.5, containing impressions of Eleutherokomma n. sp., Spinatrypa sp., and Schizophoria sp., USGS loc. 11939-SD. 2-5. Eleutherokomma n. sp. 2. Latex replica of a brachial valve, x4.0 (same specimen as in upper-left-hand corner of fig. 4.1), USGS loc. 11939-SD. 3. Latex replica of pedicle valve, USNM 460757, x2, USGS loc. 11937-SD. 4. Latex replica of two partially exposed pedicle valves, USNM 460758, x2, USGS loc. 11934-SD. 5. Latex replica of pedicle valve, USNM 460759, x2, USGS loc. 11934-SD. 6, 7. Schizophoria sp. 6. Partially decorticated pedicle valve showing internal structure, USNM 460760, x2, USGS loc. 11937-SD. 7. Latex replica offree valve, showing external ornamentoffine radial costellae, USNM 460761, x2, USGS loc. 11934-SD. 8. Latex replica of Aglaoglypta n. sp., USNM 460762, x8, USGS loc. 11933-SD. 9. Latex replica of Orecopia cf. 0. mccoyi (Walcott, 1884), USNM 460763, lateral view, x5, USGS loc. 11937-SD. 10, 11. N. gen. aff. Acanthonema, n. sp., USGS loc. 11937-SD. 10. Apertural view of latex replica, USNM 460764, x6. 11. Same view, x1 0. 12, 13. Latex replicasofthegyrogonitesofthecharophyte Sycidium, all specimens from a single hand specimen, USNM 460765, USGS loc. 11930-SD. 12. Apical view of a single gyrogonite, x9. 13. Scattered gyrogonites in differing orientations, x9. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Upper Devonian Shallow-Marine Siliciclastic Strata and Associated Fauna and Flora, Lime Hills D-4 Quadrangle

Petrography and Provenance of Sandstones from the Nation River Formation (Devonian) and the Step Conglomerate (Permian), Kandik Region, East-Central Alaska By Thomas Brocculeri, Michael B. Underwood, and David G. Howell Abstract The Kandik region of east-central Alaska can be subdivided into two major fault-bounded geologic units (herein referred to as the Tatonduk "belt" and the Kandik River "belt"), plus three smaller tectonostratigraphic terranes. Within this overall framework are sandstones of the Devonian Nation River Formation and the Permian Step Conglomerate. These sandstones are composed mostly of monocrystalline quartz grains and chert clasts, together with modest percentages of sedimentary and metasedimentary rock fragments. All of the petrographic data and detrital modes are consistent with a recycled-orogenic tectonic provenance, and the detritus probably was eroded from a northern source area. The Step Conglomerate displays a slight increase in compositional maturity, which may be related to intrabasinal recycling. However, because of the petrographic similarities between the two formations, detrital modes do not provide definitive criteria for tectonostratigraphic assignment. INTRODUCTION Most of east-central Alaska is a composite of tectonostratigraphic terranes (Churkin and others, 1982; C~mey and Jones, 1985). The Kandik region (fig. 1) has been described and interpreted by some workers within the conceptual framework of terrane analysis (for example, Howell and Wiley, 1987; Laughland and others, 1990; Howell and others, 1992). Conversely, other geologists regard most or all of ·these rocks as thrustfaulted continental-margin deposits of North American affinity (Dover, 1990; D. Bradley, written commun., 1992). Regardless of which concept is favored, there are two major belts of strata to consider, herein referred to as the Tatonduk "belt" and the Kandik River "belt" (following the suggestion of D. Bradley, written commun., 1992). A southeast-verging, younger-over-older thrust Geologic Studies in Alaska by the U.S. Geological Survey, 1991 fault (Glenn Creek fault zone) serves as the structural boundary between the two belts (fig. 1; Brabb and Churkin, 1969; Dover and Miyaoka, 1988). Both belts are overlain unconformably by unit TKs of Brabb and Churkin (1969), and both belts appear to be parautochthonous with respect to North America. In this paper, we make no inferences regarding the distance of tectonic transport for either belt with respect to the Proterozoic edge of North America. The primary purpose of this brief paper is to present compositional data for sandstones assigned to the Nation River Formation (Devonian) and the Step Conglomerate (Permian). Detrital modes are used to describe the generic tectonic provenance. Comparisons between the Devonian and Permian formations also serve as a means of assessing temporal variations among the inferred detrital sources, as well as possible differences between the Paleozoic depositional histories of the Tatonduk and Kandik River belts. Within this report, the formational assignments of Brabb and Churkin (1969) and Foster (1976) have been retained. We note, however, that ambiguities exist in the tectonostratigraphic identity of certain samples (Churkin and others, 1982; Dover and Miyaoka, 1988; Howell and others, 1992). GEOLOGIC SETTING The Tatonduk belt includes basement rocks of the Tindir Group (Middle Proterozoic to Lower Cambrian) and extends upsection through Triassic to Lower Cretaceous strata of the Glenn Shale (Churkin and others, 1982; Howell and Wiley, 1987; Howell and others, 1992; see Underwood and others, this volume, for stratigraphic column). Stratigraphic and facies relations are consistent with a continental rifting event, followed by progressive deepening of depositional environments from

Late Proterozoic through Late Devonian time; lower Paleozoic carbonate strata, which include intervals of carbonate turbidites, grade upsection into ribbon chert and siliciclastic turbidite deposits (Payne and Allison, 1981; Howell and Wiley, 1987). The Nation River Formation is a Devonian deepmarine turbidite sequence consisting of rhythmically interbedded chert-rich conglomerate, sandstone, and mudstone (Brabb and Churkin, 1967; Howell and Wiley, 1987). This diverse succession has been interpreted as a submarine-fan complex sandwiched between a deep-water chert (McCann Hill Chert) and a siliceous shale unit known as the Ford Lake Shale (Brabb, 1969; Brabb and Churkin, 1969; Payne and Allison, 1981). Nilsen and others (1976) described occurrences of west-directed paleocurrent indicators and linked the sediment source to uplift in the Northwestern Cordillera, coupled with westward progradation of a deep-sea fan system. In contrast, Howell and Wiley (1987) documented a radial pattern of current indicators and concluded that paleoflow was directed predominantly toward the south and southeast (that is, toward the North American craton). Accordingly, a northern landmass may have been the sediment source for the Nation River Formation. The Kandik River belt is dominated by Jurassic to Lower Cretaceous deep-marine deposits assigned to the Glenn Shale, Keenan Quartzite, Biederman Argillit~, and Kathul Graywacke (see Underwood and others, this v-olume, for stratigraphic column; see Howell and others, 1992, for preliminary petrographic data for Cretaceous sandstones). Most of these rocks display a pronounced slaty cleavage, and Laughland and others (1990) docuKANDIK RIVER BELT Indian G ra ve i\lounlai·n EXPLANATION

Limit of paleothermal anomaly --Contact -- Fault, dashed where approximate Strike-slip fault, arrows indicate direction of relative movement 15 MILES

15 Kl LOMETERS Thrust fault, sawteeth on upper plate +Syncline -t- Anticline Figure 1. Geologic index map of east-central Alaska showing Kandik River belt (no pattern) and Tatonduk belt (horizontal lined pattern). Hachured line southeast of Glenn Creek fault delineates boundary of thermal-maturity anomaly (see Underwood and others, this volume). Also shown are unitTKs of Brabb and Churkin (1969), and Woodchopper Canyon (WC), Slaven Dome (50), and Tacoma Bluff (TB) terranes of Churkin and others (1982). Rocks southwest of Tintina fault zone are assigned to Yukon-Tanana composite terrane (Coney and jones, 1985). Modified from Brabb and Churkin (1969) and Foster (1976). For relevant fossil control and alternative interpretations of structural geology, see Dover and Miyaoka (1988) and Miyaoka (1990). Petrography and Provenance of Sandstones from the Nation River Formation and Step Conglomerate, Kandik Region

mented much higher levels of thermal maturity in Mesozoic rocks of the Kandik River belt as compared with older strata of the Tatonduk belt. In the western portion of the study area (fig. 1 ), the Kandik River belt is in fault contact with the Woodchopper Canyon terrane, the Slaven Dome terrane, and ·the Takoma Bluff terrane of Churkin and others (1982). Pre-Mesozoic sandstones and conglomerates occur in two general regions to the northwest of the Glenn Creek fault zone: (1) in the core of the Step Mountains anticline, and (2) west of the Mardow Creek fault (fig. 1). Based upon a single occurrence of poorly preserved spores (Miyaoka, 1990), Devonian conglomerates have been identified in the core of the Step Mountains anticline (Brabb and Churkin, 1969; Dover and Miyaoka, 1988). These strata, however, may correlate with poorly dated rocks of the Takoma Bluff terrane of Churkin and others (1982), rather than the widespread Nation River turbidite successions to the southeast of the Glenn Creek fault zone (Howell and others, 1992). The Step Conglomerate (Lower Permian) consists primarily of massive, chert-rich sandstone and chertpebble conglo.merate, plus lenses of bioclastic limestone. The type locality of this formation is located in the Step Mountains (Brabb, 1969), but most of the mapped localities of the Step Conglomerate (Brabb and Churkin, 1969) are located within the Takoma Bluff terrane of Churkin and others (1982). The correlation between the Takoma Bluff terrane and the type section in the Step Mountains remains uncertain. Moreover, portions of the Takoma Bluff terrane have }?een remapped as Devonian (Nation River Formation?) by Dover and Miyaoka (1988). Locally, the Step Conglomerate lies above an angular unconformity that resulted from regional uplift and erosion prior to the Early Permian Epoch (Payne and Allison, 1981). A northern source has been inferred for the Permian gravels, but no specific provenance has been identified (Nilsen and others, 1976). METHODS Outcrop samples of sandstone (fine to coarse grained) were collected from a field area bounded to the southwest by the Tintina fault zone, to the north and west by the loess deposits of the Yukon Flats, and to the east by the international border with Canada (fig. 2). Two samples from Canada were provided by ARCO Alaska, Inc. Detrital modes were determined for 33 sandstone samples from the Nation River Formation and 7 sandstone samples from the Step Conglomerate (table 1 ). As stated previously, all of the formational assignments follow the maps of Brabb and Churkin ( 1969) and Foster (1976). Thin sections were impregnated with blue epoxy to identify pore space and stained for both Geologic Studies in Alaska by the U.S. Geological Survey, 1.991 plagioclase and potassium feldspar, using a modified version of Houghton's (1980) technique. The GazziDickinson point-counting method was followed, in which only aphanitic grains are classified as lithic fragments (Ingersoll and others, 1984 ). This technique is advantageous because it minimizes the effect of grain size changes on modal data. At least 500 grid points were counted per specimen. PETROGRAPHIC GRAIN DESCRIPTIONS The following brief descriptions pertain to sandstone constituents in both the Nation River Formation and the Step Conglomerate. Monocrystalline quartz grains are common in all of these sandstones; extinction varies from straight to strongly undulose. Minute inclusions of sericite(?) occur within many of the quartz grains. Chert (cryptocrystalline quartz) is consistently the most abundant type of framework grain. Finely crystalline chert typically contains clay and mica impurities plus stains of metallic oxides. Outlines and spherical ghosts of radiolaria are rare. The feldspar content of some samples may be underestimated as a result of diagenetic alteration to albite, clay minerals, and calcium carbonate. Pure albite did not stain and was identified based on its characteristic cleavage. Potassium feldspar occurs as orthoclase, microcline, and perthite; the presence of K-feldspar suggests shallow burial depths and (or) minimal thermal alteration. Most of the metamorphic rock fragments are slates and phyllites with a subtle to obvious planar fabric. Foliated mica-schist fragments are relatively rare. Finegrained metaquartzites display both crude planar fabrics and strongly sutured grain boundaries. Fragments of quartz-mica tectonite exhibit undulose extinction and a foliation defined by parallel to subparallel alignment of micas. A gradation exists from strongly compacted shales to their low-grade metamorphic equivalents. Micaceous, quartz-rich siltstone fragments and clay-rich mudstones are both common. Volcanic rock fragments occur in only trace amounts; these fragments contain small plagioclase laths surrounded by strongly altered matrix, which is probably the weathering product of devitrified glass. Sandstone matrix is composed of undifferentiated clay minerals, fine-grained mica grains, dark organic material, and dispersed silt-sized quartz. The matrix content tends to increase in samples containing a relatively high percentage of sedimentary and metamorphic rock fragments, suggesting that much of the material is pseudomatrix (that is, not original detrital material). Chemical cement also is more abundant in samples enriched in metasedimentary rock fragments; for example, calcium carbonate occurs as a replacement of framework

grains, as a filling of intergranular pores, and as porelining cement. Authigenic clay minerals also fill pore cavities and partially replace chemically unstable framework grains. Chalcedony and polycrystalline quartz cements exist in minor amounts, both within intergranular pores and as microveins. Silica cementation evidently occurred subsequent to pore-lining carbonate cementation. The abundance of clay-rich lithic fragments and the presence of clay matrix and clay coatings collectively served to absorb stresses during compaction, thereby inhibiting pressure solution along grain contacts. Visible porosity (open spaces filled with blue epoxy) is extremely low for both formations (0-3 percent). Because of the abundance of pore-filling matrix and the absence of visible porosity, it appears that the Nation River and Step sandstones have rather limited potential as reservoirs for oil or gas. MJ90 K44A Snowy Peak MILES KILOMETERS PROVENANCE ANALYSIS Provenance interpretations for the Nation River Formation and the Step Conglomerate are based on standard ternary plots of framework constituents (fig. 3) and comparisons with empirical analogs (for example, Dickinson and others, 1983). QFL ternary diagrams (where Q is total quartz, F is total feldspar, and L is total unstable aphanitic lithic fragments) indicate the survivability of framework grains and the compositional maturity of sandstones. Shifting polycrystalline quartz (QP) to the Lt mode of the QmFLt diagram sometimes helps emphasize detrital provenance better than QFL modes alone (Graham and others, 1976). The amounts and types of feldspar can be valuable provenance discriminators, but they are not useful in this particular study. The limiting factor for QmPK analysis (where Pis plagioclase and K Step Mountains MU90 K7 MJ90 K1A / ) MJ90 K3 ''- KANDIK RIVER BELT / oCb MU90 K1

TB90 K59D MJ90 K74C TB90 K59E MJ90 K31C -.o 0 TB90 K59H "'. MJ90 K34B MJ90 K72B !

0\, Mj90 K71B ·,. Mj90 K69A Mj90 K69E Mj90 K70B /

UM90 K54B ·, UM90 K54D j UM90 K54E / '· 0 m :I m a. m 90VK135·1 Figure 2. Sample localities for analyses of sandstone composition. See table 1 forformational assignments of each sample. Sample numbers with an asterisk(*) are associated with the Nation River Formation (as mapped by Brabb and Churkin, 1969) northwest of Glenn Creek fault zone. See figure 1 for explanation of symbols. Petrography and Provenance of Sandstones from the Nation River Formation and Step Conglomerate, Kandik Region

t.J "' Table 1. Sandstone point-count data, Nation River Formation and Step Conglomerate, Kandik region, east-central Alaska

Q [See figure 2 for sample localities. Abbreviations: GZ, mean visual grain-size estimate (VC, very coarse; C, coarse; M, medium; F, fine; VF, very fine); Ch, chert; Ls, sedimentary rock fragments; Lm, metasedimentary rock fragments; Pore, visible pore space (blue epoxy); Cmt, chemical cement; Mx, fine-grained matrix; Alt, altered/unidentified grains. See figure 3 for definition of other abbreviations] la Q. ;· TOTAL POINTS COUNTED DETRITAl MODES (PERCENT)

SAMPLE GZ Qm Qp Ch Di p K Ls Lm Pore Cmt Mx A It Q F Qm F Lt Qm p K Qp Lsm "'

NATION RIVER FORMATION (italic indicates sample from Step Mountains anticline) C"' 90VK 12-l VC

c: 90VK 19-1 F !on 90VK43-2 F "' 90VK45-l VF

Q 90Vk 135-1 Vf

Mj90 Kia Vc Oq ;:;· Mj90 K3 M

MJ90 KBA c

MJ90 K 16B c

Mj90 K3 I C M -:: Mj90 K34B M Mj90 K35A F

MJ90 K59C F MJ90 K59D M MJ90 K59E M MJ90 K68B c MJ90 K68C F MJ90 K69A M MJ90 K69E M MJ90 K70B M MJ90 K71 B M MJ90 K72B F MJ90 K74C - M TB90 K23 M TB90 K5 1 C M TB90 K52C M TB90 K53D c TB90 K59D F TB90 K59E c TB90 K59H M UM90 K54B c UM90 K54D M UM90 K54E F MEAN DETRITAL M 0 DES MEAN PERCENT 17.5 LOW PERCENT HIGH PERCENT 32.8

.Q

E u..

z

z u.J u 01=: u.J

u.J

J

1u.J u.J 1-z 0 u 1-z J

0' Q.. E 0' £ 0'

E "' Q..

u 0' E 0' N \.J LIJ Q..

V) e .£ c. e

"'

"0 NNV'lOr<'lNO N 0000000 0 t-OO\Ot-0'10000 Nt-NV'IO'I\ON -oor-O'IOt-\0 ~uuuuuu

O'IO'<t" ! ; ;f'i IV')Nr<'l

0'100 r-:r"ir"i l--N loo- ir<'lNV'l 10\l/') 4 v.l..o.n.n (l:lN !::I,I:lzZ C:::uLI:lLI:l E-<c:::uu LI:li,I:lc:::C::: ClP..(l:lLI:l zzP..P.. LI:lLI:lo9 :l::C is potassium feldspar) is the relative depletion of monocrystalline constituents. The relative amounts and types of lithic fragments are probably the most valuable discriminators for interpreting sandstone provenance (Graham and others, 1976). The QpLvLsm diagram illustrates the predominance of QP (including chert), and sedimentary and metamorphic rock fragments (Lsm) with respect to volcanic rock fragments (Lv). The mean QFL value for the Nation River Formation is Q=62, F=5, and L=33. These data are consistent with a recycled orogenic provenance (fig. 3). Interpretation of the QmFLt diagram suggests a transitional-re- . cycled to lithic-recycled source, with a mean value of Qm=21, F=5, and Lt=74. Average polycrystalline modes are QP=55, Lv=l, and Lsm=44 for the Nation River Formation, and these data are consistent with a sedimentary and low-grade metamorphic terrain as the principal source. However, rock fragments of this type could have been eroded from a variety of tectonic settings, including a collisional suture zone, a sedimentary foldthrust belt, or an uplifted continental-margin succession with exposed metamorphic basement. The QmPK data likewise support the interpretation of a continental-block provenance· (fig. 3); there is no evidence, based on this plot, for significant input from either a dissected volcanic/plutonic terrain or comparable crystalline lithologies within continental basement. Overall, we note only minor differences among the petrographic modes of the Step Conglomerate and the Nation River Formation. Some of the differences could be due to grain size, in that Step sandstones are generally coarser grained (table 1). However, the data set is too small to be statistically reliable. The Step Conglomerate yields a modal average of Q=72, F=2, and L=26. Percentages of chert are modestly. higher in the Step Conglomerate, resulting in an averag~ modes of Qm=18, F=2, and Lt=80 and Qp=67, Lv=l, and Lsm=32. Finally, both P and K values are slightly lower, on average, in the Step Conglomerate. Data for individual samples plot in the same tectonic-provenance fields as most of those described above for the Nation River Formation (fig. 3). DISCUSSION The identification of a specific geologic or geographic provenance for sandstones of the Nation River Formation and the Step Conglomerate requires some caution given the variety of factors influencing detrital modes. For example, the effects of distance of nonmarine transport, sediment recycling, transport energy and transport processes within the depositional setting, and diagenetic alteration of framework grains all must be considered. With this caveat in mind, we emphasize the obvious enrichment in both formations of toPetrography and Provenance of Sandstones from the Nation River Formation and Step Conglomerate, Kandik Region

tal quartz and unstable lithic fragments relative to feldspar. Conversely, volcanic rock fragments and feldspar grains are rare. Collectively, these petrologic signatures define the quartzo-lithic suite (Dickinson, 1988). A logical conclusion is that sedimentary plus low-grade metasedimentary rocks dominated the provenance region. Based on the turbidite paleocurrent analysis of Howell and Wiley (1987), the source was probably located to the north. Paleoflow data of this type must be viewed with some caution, however, because turbidity F p Q Recycled Stable . continental block o Nation River Formation Step Conglomerate currents generally flow in directions parallel to a basin's axis during deposition, and that direction may not coincide with the initial sediment transport direction out of the subaerial source area. The origin of the detrital chert grains is key to any provenance interpretation. This highly resistant debris may have been recycled from older sedimentary rocks within any of several types of geologic settings, such as an uplifted subduction complex, a fold-thrust belt, or a rifted continental-margin succession (Jones and Arc orogen Collisional suture; fold-thrust belt Figure 3. Ternary diagrams showing detrital modes for sandstones of Nation River Formation (open circles) and Step Conglomerate (solid diamonds), east-central Alaska. Abbreviations are as follows: Q, total quartz; QP, polycrystalline quartz (including chert); Q~, monocrystalline quartz; P, plagioclase; K, potassium feldspar; F, total feldspar; L, unstable aphanitic rock fragments; Lt, total polycrystalline grains; Lv, volcanic rock fragments; L m' sedimentary and metasedimentary rock fragments. Boundaries for tectonic-provenance fields are fromD icki nson and others (1983) and Dickinson (1988). See table 1 for complete petrographic data. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Murchey, 1986). The overall paucity of both volcanic debris and feldspar casts serious doubt on the subduction-zone hypothesis, and the scarcity of radiolaria tests and sponge spicules (or their diagenetic ghosts) leads us to suggest that most of the chert did not form through direct biogenic accumulation of silica ooze (that is, as a normal ribbon chert). Instead, we favor significant amounts of erosion from chert nodules that formed via diagenetic segregation of silica within shallow-water carbonate sequences (Knauth, 1979). A significant difference in provenance between the Devonian Nation River Formation and the Permian Step Conglomerate is not supported by our petrographic analyses. Payne and Allison (1981) speculated that widespread Carboniferous uplift and intrabasinal erosion, including areas to the west and northwest of our study area (that is, in the vicinity of the Kaltag-Porcupine fault zone), may have provided the Step Conglomerate with some of the chert-rich gravel. The Step Conglomerate, on average, does contain greater percentages of resistant chert and slightly lesser amounts of mechanically unstable lithic fragments, so some intrabasinal recycling appears possible. Evidence for a major reduction in fragile metamorphic and sedimentary rock fragments is lacking, however, so we conclude that both units shared the same basic type of detrital source. CONCLUSIONS Detrital modes for 33 sandstones selected from the Nation River Formation and 7 from the Step Conglomerate define overlapping fields on all ternary plots of tectonic provenance. The high percentages of polycrystalline quartz (chert), sedimentary rock fragments, and metamorphic debris collectively point to a recycled-orogenic provenance for both formations. We speculate that the sediments were derived from an uplifted continental margin to the north of the Kandik region. Our data allow for a limited amount of intrabasinal recycling of older Paleozoic strata including, perhaps, the Devonian Nation River Formation as a partial source for the subtle enrichment of chert in the Permian Step Conglomerate. Because of the similarities in detrital modes, however, petrographic data cannot be used to provide definitive criteria for the assignment of individual Paleozoic sandstones and conglomerates to appropriate tectonostratigraphic units. Acknowledgments.-Mark Johnsson and Lu Haufu assisted in the field. Samples and financial support to the University of Missouri were generously supplied by ARCO Alaska, Inc. We thank Gerry Van Kooten and his ARCO colleagues for their scientific cooperation and logistical aid. Superintendent Don Chase granted sampling permits and access to the Yukon-Charley Rivers National Preserve. Acknowledgment is also made to the Donors of the Petroleum Research Fund, administered by the American Chemical Society, for partial support of this research (Grant #22773-AC2 to Underwood). Editorial suggestions by D. Bradley helped improve the manuscript. REFERENCES CITED Brabb, E.E., 1969, Six new Paleozoic and Mesozoic formations in east-central Alaska: U.S. Geological Survey Bulletin 1274-I, p. 1-26. Brabb, E.E., and Churkin, M., Jr., 1967, Stratigraphic evidence for the Late Devonian age of the Nation River Formation, east-central Alaska: U.S. Geological Survey Professional Paper 575-D, p. D4-D15. ---1969, Geologic map of the Charlie River quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Geologic Investigations Map I-973, scale 1: 250,000. Churkin, M., Jr., Foster, H.L., Chapman, R.H., and Weber, F.R., 1982, Terranes and suture zones in east-central Alaska: Journal of Geophysical Research, v. 87, p. 37183730. Coney, P.J., and Jones, D.L., 1985, Accretion tectonics and crustal structure in Alaska: Tectonophysics, v. 119, p. Dickinson, W.R., 1988, Provenance and sediment dispersal patterns in relation to paleotectonic and paleogeography of sedimentary basins, in Kleinspehn, K.L., and Paola, C., eds., New perspectives in basin analysis: New York, Springer-Verlag, p. 331-351. Dickinson, W.R., Beard, L.S., Brakenridge, G.R., Erjavec, J.L., Ferguson, R.C., Inman, K.F., Knepp, R.A., Lindberg, F.A., and Ryberg, P.T., 1983, Provenance of North American Phanerozoic sandstones in relation to tectonic setting: Geological Society of America Bulletin, v. 94, p. 222-235. Dover, J.H., 1990, Geology of east-central Alaska: U.S. Geological Survey Open-File Report 90-289, 66 pp. Dover, J.H., and Miyaoka, R.T., 1988, Reinterpreted geologic map and fossil data, Charley River quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2004, 2 sheets, scale 1:250,000. Foster, H.L., 1976, Geologic map of the Eagle Quadrangle, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map 1-922, 1 sheet, scale 1:250,000. Graham, S.A., Ingersoll, R.V., and Dickinson, W.R., 1976, Common provenance for lithic grains in Carboniferous sandstones from the Ouachita Mountains and Black Warrior basin: Journal of Sedimentary Petrology, v. 46, p. Houghton, H.F ., 1980, Refined techniques for staining plagioclase and alkali feldspars in thin section: Journal of Sedimentary Petrology, v. 50, p. 629-630. Howell, D.G., Johnsson, M.J., Underwood, M.B., Lu Haufu, and Hillhouse, J.W., 1992, Tectonic evolution of the Kandik region, east-central Alaska: Preliminary interpretations, in Bradley. D.C., and Ford, A.B., eds., Geologic studies in Alaska by the U.S. Geological Survey, 1990: U.S. Geological Survey Bulletin 1999, p. 127-140. Howell, D.G., and Wiley, T.J., 1987, Crustal evolution of Petrography and Provenance of Sandstones from the Nation River Formation and Step Conglomerate, Kandik Region

northern Alaska inferred from sedimentological and structural relations in the Kandik area: Tectonics, v. 6, p. Ingersoll, R.V., Bullard, T.F., Ford, R.L., Grimm, J.P., Pickle, J.D., and Sares, S.W., 1984, The effect of grain size on detrital modes: A test of the Gazzi-Dickinson point-counting method: Journal of Sedimentary Petrology, v. 54, p. Jones, D.L., and Murchey, B., 1986, Geologic significance of Paleozoic and Mesozoic radiolarian chert: Annual Review of Earth and Planetary Sciences, v. 14, p. 455-492. Knauth, L.P., 1979, A model for the origin of chert in limestone: Geology, v. 7, p. 274-277. Laughland, M.M., Underwood, M.B., and Wiley, T.J., 1990, Thermal maturity, tectostratigraphic terranes, and regional tectonic history: an example from the Kandik area, eastcentral Alaska, in Nuccio, V.F., and Barker, C.E., eds., Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Applications of thermal maturity studies to energy exploration: Society of Economic Paleontologists and Mineralogists, Rocky Mountain Section, Special Publication, p. 97-111. Miyaoka, R.T., 1990, Fossil locality map and fossil data for the southeastern Charley River quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2007, 45 p., scale 1:100,000. Nilsen, T.H., Brabb, E.E., and Simoni, T.R., 1976, Deep sea fan deposition of the Devonian Nation River Formation, Yukon-Kandik area, Alaska: Proceedings of the Alaska Geological Society Symposium, p. E1-E20. Payne, M.W., and Allison, C.W., 1981, Paleozoic continentalmargin sedimentation in east-central Alaska: Geology, v. 9, p. 274-279. Reviewers: H. Mclean and M. Wilson

Magnetic Susceptibilities and Iron Content of Plutonic Rocks Across the Coast Plutonic-Metamorphic Complex near Juneau, Alaska By james L. Drinkwater, Arthur B. Ford, and David A. Brew Abstract Magnetic susceptibility measurements and chemical data for granitoid samples collected from a transect across the Coast plutonic-metamorphic complex near Juneau help define and characterize 18 plutons and plutonic units that form three major belts and seven subbelts. Magnetic susceptibility values along the transect correlate moderately well with oxidation-state values but not well with total iron (FeO*) or normative magnetite. Very low magnetic susceptibility characterizes a group of 95-Ma plutons that lie along the western margin of the transect (Admiralty-Revillagigedo belt); these are ilmenite-bearing type granitoids that contain very little or no magnetite. Three small plutonic sills, which are part of the great tonalite sill belt east of the megalineament, are also ilmenite bearing and have very low ,magnetic susceptibility, but other major plutons of the same belt are magnetite-bearing types with very high magnetic susceptibility. East of the great tonalite sill belt, magnetic susceptibility generally decreases, oxidation states vary considerably, and most of the plutons are magnetite bearing. FeO* steadily decreases in the rocks as the Si02 and K-feldspar contents increase. Differences in magnetic susceptibility are due to changes in composition, oxidation states, and Fe3+ mineralogy. Between the major plutonic belts these parameters are related to the different ages of the belts and the host terranes in which the plutons were emplaced, but the causes of these differences between subbelts is not well established. They may be related to the differentiation trends of the different groups of plutons. Geochemical and petrographic characteristics of the ilmenite-bearing plutons indicate that they are 1-type granitoids that were generated either as reduced magmas or as oxidized magmas that became reduced during their magmatic history. An average oxidation value of 0.23 roughly separates ilmenite-bearing from magnetite-bearing type granitoids. Factors such as depth of magma generation, contamination, H20 and sulfide content, and tectonic stresses during pluton emplacement may influence the extent of oxidation and reduction and the crystallization of magnetite and trivalent Fe-rich ferromagnesian silicate minerals. Trivalent Fe-rich silicate minerals such as biotite, epidote, allanite, and brown hornblende are much more common in granitoids that have high FeO* and relatively high oxidation-state values but low magnetic susceptibility. Primary (magmatic) epidote is found mainly in the ilmenite-bearing (reduced) type plutons of the AdmiraltyRevillagigedo plutonic belt. INTRODUCTION Magnetic susceptibility (MS) is an easily measured rock property that provides a semiquantative measure of the magnetite content of a rock (Puranen, 1989; Ross, 1989; Tulloch, 1989). It is particularly useful in characterizing and delineating individual plutons for mapping purposes (Ross, 1989). Used in conjunction with iron content and iron oxidation states, the MS values may aid in the assessment of the magmatic and tectonic environment of emplacement (Ishihara, 1977; Takahashi and others, 1980; Tulloch, 1989; Bateman and others, 1991). Takahashi and others (1980) used MS data to separate granitic rocks into the magnetite and ilmenite series; these series, in turn, were interpreted to indicate different tectonic settings. Such interpretative use of MS data, however, may be weakened by significant differences in susceptibility that can occur between different plutons, parts of plutons, and lithologies. An example is the Idaho batholith, where the boundary between a western magnetite series and an eastern ilmenite series in the western part may partly reflect ages of intrusion (Piccoli and Hyndman, 1985). Some differences in MS are probably due to variable hydrothermal alteration (Criss and Champion, 1984). This report provides data on the MS and iron content of granitic rocks of Late Cretaceous and Tertiary age in a northeast-trending transect (Taku Inlet transect) across the informally named Coast plutonic-metamorphic complex of Brew and Ford (1984a, b) (fig. 1). These data were derived from our ongoing investigations of the chemical and petrographic variations along the transect. The transect lies southeast of Juneau between Taku Inlet Magnetic Susceptibilities and Iron Content of Plutonic Rocks Across the Coast Plutonic-Metamorphic Complex Near Juneau

TKmg Arm-Fords Terror map area (Brew and Grybeck, 1984)] KILOMETERS EXPLANATION Average magnetic susceptibility values of units

High (>3 ,000 SI units) Medium (1 ,000-3,000 S I units) CJ Low (<1,000 SI units) Map Unit Abbreviations WGs - Wright Glacier stocks TLbg - Turner Lake batholith eastern unit TLbn - Turner Lake batholith northern unit TLbc - Turner Lake batholith central unit TLbs - Turner Lake batholith southern unit TLb - Turner Lake batholith undivided TL - Tease Lakes stocks SR Speel River pluton TC - Taku Cabin pluton AL Annex Lakes pluton - Carlson Creek pluton LCG - Lemon Creek Glacier pluton MJ - Mount Juneau pluton EP - Everett Peak pluton AP - Arthur Peak pluton BP - Butler Peak pluton IP - Irving Peak pluton G I - Grand Island pluton GP - Glass Peninsula stocks TKmg - Migmatitic gneiss (Tertiary and Cretaceous) TKs - Schist (Tertiary and Cretaceous) KJ s - Metasedimentary and metavolcanic rocks (Cretaceous and Jurassic) Di - Diorite Figure 1. Generalized geologic map ofTaku lnlettransect area, southeastern Alaska, showing individual plutons and plutonic units discussed in text. Geology modified from Brew and Ford (1986). Geologic Studies in Alaska by the U.S. Geological Survey, 1991

and the Whiting River and extends from the Glass Peninsula of Admiralty Island northeastward to the Canadian border (fig. 1). The study area, or "transect area," includes MS data on the Annex Lakes pluton and for a group of plutonic sills (Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons) that lie east of Juneau and are just north of the Taku Inlet transect. Available age data for rocks in the transect are also summarized here. Descriptions of the plutons and country rocks in the transect area are found in Brew and Grybeck ( 1984 ), Brew and Ford (1986), Brew (1988), and Drinkwater and others (1989). Drinkwater and others (1990) used the informal term "Juneau sill group" for the Mount Juneau, Carlson Creek, Lemon Creek Glacier, Mendenhall Glacier, Annex Lakes and Taku Cabin plutons. We discontinue the usage of that informal term in this paper and refer to these plutons by their individual names. (The Mendenhall Glacier pluton is not shown on figure 1 and is not discussed in this paper.) Methods Magnetic susceptibility measurements were made from the same sawn slab surfaces from which modal compositions were determined. The measurements were made with a model JH-8 (Geoinstruments, Finland) hand-held susceptibility meter. The units of measurement for volume susceptibility readings, as used in this report, are dimensionless SI units (International standard units), x1o-5 One percent magnetite produces an MS of about 4,000x 1 o-5 SI units, which approximates 3,000x1o-6 cgs units (Ross, 1989). MS can be affected by grain size and chemistry of the magnetite (Tulloch, 1989) and commonly decreases with rock weathering (Puranen, 1989). Magnetite grains in samples from the Juneau transect show little size variation. For consistent results we used slabs of fresh rock with a minimum size of 10 em by 8 em by 2 em. We scanned each sample by moving the meter around the surface of the slab and then used the highest reading, which reflects the closest approach to the actual value, according to the method of Tulloch (1989). We also multiplied the results by 2 to compensate for the use of hand samples rather than outcrops, in accordance with the JH-8 manual. The MS values varied considerably in some samples, and those samples giving very heterogeneous readings were not used. Samples with anomalously high or low readings were also discarded unless they represented a distinct part of a pluton, such as the border zone or core phase. We measured MS on 148 samples, _of which 137 were used in determining ranges and averages. Three to 18 samples were measured for each granitoid unit. Total iron (FeO*) was determined by X-ray fluorescence analysis in laboratories of the U.S. Geological Survey, and FeO was determined by wet chemical methods. The presence of sulfides, refractory oxides (magnetite and ilmenite), and organic material can obstruct the accuracy of FeO determinations because of problems of Fe oxidation or reduction during acid dissolution of samples in preparation for either coulometric or potentiometric titration methods (Jackson and others, 1987). Because of this potential problem, we did not use anomalously high or low oxidation-state values from sulfide- or oxide-rich samples. Chemical analyses that included FeO determinations were available for 120 samples, and both MS measurements and iron data were available for 54 samples. GENERAL GEOLOGY The granitic rocks investigated in this report are part of the Coast plutonic-metamorphic complex, which extends the length of southeastern Alaska and beyond into Canada. The granitoids of the Taku Inlet transect are part of the three plutonic belts (table 1) that, together with the intervening metamorphic rocks and the adjacent Gravina overlap assemblage, define the complex. The plutons along the western edge of the complex are part of the Admiralty-Revillagigedo belt of Brew and Morrell (1983) and are separated from other parts of the Coast plutonic-metamorphic complex by the northwest-trending Coast Range megalineament (Brew and Ford, 1978), a large fault zone with inferred east-side-up displacement and right-lateral offset (Stowell and Hooper, 1990; Hooper and others, 1990). These plutons intrude ( 1) metavolcanic rocks of the Douglas Island Volcanics and metasedimentary rocks of the Seymour Canal Formation of Late Jurassic and Early Cretaceous age (Brew and Ford, 1986) on the western side of Stephens Passage, and (2) high-grade schist (Brew and Ford, 1984a; Brew and others, 1989) on the eastern side. Collectively these units are referred to as the western metamorphic zone of the Coast plutonic-metamorphic complex (Brew and others, 1989). They were called parts of the Taku and Gravina tectonostratigraphic terranes by Berg and others (1978) and the Taku terrane and Gravina-Nutzotin overlap assemblage by Monger and Berg (1987); we, however, do not consider them to be a separate terrane. Gehrels and others (1990) argued that the eastern part of the western metamorphic zone (that part between the megalineament and the great tonalite sill belt) may belong to the Yukon crystalline terrane of eastern Alaska. The great tonalite sill (Brew, 1988) is a northwesttrending belt of eastward-dipping plutonic sills of Late Cretaceous to early Tertiary age that forms the western boundary of the central granitic zone of the Coast pluMagnetic Susceptibilities and Iron Content of Plutonic Rocks Acr.oss the Coast Plutonic-Metamorphic Complex Near Juneau

Table 1. Major plutonic belts and subbelts of the Coast plutonic-metamorphic complex within the Taku Inlet transect, southeastern Alaska Belt Subbelt Plutons Main rock t~Qes Age AdmiraltyGrand Island Glass Peninsula stocks; Quartz diorite; Late Cretaceous Revillagigedo Grand Island pluton; Tonalite; Irving Peak pluton; Diorite; Butler Peak pluton Granodiorite Taku Harbor Everett Peak and Quartz monzodiorite Late Cretaceous Arthur Peak plutons Great tonalite sill Western subbelt Speel River pluton; Tonalite; Early Tertiary, and Mount Juneau, Carlson Quartz diorite, tonalite, Late Cretaceous Creek, and Lemon and granodiorite to early Tertiary Creek Glacier plutons Eastern subbelt Annex Lakes pluton; Granodiorite, granite; Early Tertiary Taku Cabin pluton Tonalite Central graniti·c Turner Lake batholith (western subbelt) Northern, central, and Granodiorite, and Tertiary (Eocene) southern granodiorite granite units (eastern subbelt) Undivided granite unit Granite Tertiary (Eocene) Foliated stocks Tease Lake stocks Quartz monzodiorite, Early Tertiary Wright Glacier stocks --- Two plutons tonic-metamorphic complex (Brew and Ford, 1984a, b). Within the transect area the sill belt is represented by the Speel River, Taku Cabin, and Annex Lakes plutons (Drinkwater and others, 1989, 1990) and by the largely more deformed and older Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons (Drinkwater and others, 1990). East of the great tonalite sill, granodiorites and granites of Tertiary age form most of the central granitic belt of the Coast plutonic-metamorphic complex (Brew and Ford, 1984a, b). In the transect area these granitoids are grouped as the Turner Lake batholith and other units (table 1 ). They intrude high-grade metamorphic rocks and migmatites of Late Cretaceous to early Tertiary metamorphic age (Brew and Ford, 1986). These metamorphic host rocks are considered by Berg and others (1978) and Monger and Berg (1987) as part of the Tracy Arm terrane, although the tectonic affinity of these metamorphic rocks is inconclusive because of possible links to the Stikine terrane and to the older Alexander terrane (Brew and Ford, 1983, 1984a), as well as to the Yukon crystalline terrane (Gehrels and others, 1990). Brew and Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Quartz monzonite Granite, granodiorite Late(?) Tertiary Ford ( 1984b ), Gehrels and others ( 1984 ), and Brew and others (1989) provide additional information on the timing of regional deformation, metamorphism, and pluton emplacement. Plutonic Rocks of Transect Area As noted above, the plutonic rocks of the transect area belong to three belts; from west to east they are: ( 1) the Admiralty-Revillagigedo belt of Late Cretaceous age (Brew and Morrell, 1983), (2) the great tonalite sill belt of Late Cretaceous to Paleocene age (Brew, 1988), and (3) the central granitic belt of Eocene age. The three belts are divided into seven separate subbelts (table 1). All of the plutonic rocks are calc-alkaline and I-types (Brew and Ford, 1986; Brew, 1988), and they vary in modal composition from quartz diorite to tonalite and granodiorite to granite northeastward along the transect (fig. 2). The major belts are charcterized by an increase in Si02 from west to east along the transect (fig. 3). Exceptions to this trend are the quartz monzodiorite bodies at Everett and

Arthur Peaks, which are silica-poor ( <60 percent Si02), Kfeldspar-rich rocks. The Admiralty-Revillagigedo belt is divided into two subbelts (table 1 ). The Grand Island subbelt consists of the Grand Island, Irving Peak, and Butler Peak plutons and six stocks of biotite-hornblende quartz diorite and tonalite on the Glass Peninsula. The rocks are typically well foliated to gneissic, contain primary-appearing garnet and epidote, and have traces of ilmenite and sulfides but contain no appreciable magnetite. A few thin sections of the Grand Island pluton contain traces of magnetite. Mafic and hornfelsic inclusions are common, particularly near the margins of the plutons. The epidote is magmatic according to the criteria of Zen and Hammarstrom ( 1984) and is typical of the 95-Ma plutons described by Brew and others ( 1984, 1989) and of the "Taku terrane" plutons in the Ketchikan area described by Arth and others (1988) . The Taku Harbor subbelt consists of two stocklike bodies of foliated, porphyritic hornblende quartz monzodiorite. The stocks contain abundant brown hornblende, much sphene and magnetite, traces of sulfide minerals, phenocrystic K-feldspar, and primary epidote, but they lack biotite. The great tonalite sill belt is divided into two subbelts (table 1). The Speel River, Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons form the western subbelt, and the Taku Cabin and Annex Quartz K-Feldspar Lakes plutons compose the eastern subbelt. The Speel River and Taku Cabin plutons are typical granitoids of the great tonalite sill (Brew, 1988); they are foliated, homogeneous, medium- to coarse-grained tonalites that carry much magnetite. The Mount Juneau, Lemon Creek Glacier, and Carlson Creek plutons are much more deformed and metamorphosed, show more heterogeneous textures and compositions, and lack magnetite or opaque minerals other than sulfides (Drinkwater and others, 1990). The Mount Juneau pluton, like the Grand Island subbelt, contains primary epidote and garnet. The Annex Lakes pluton consists mostly of foliated and inequigranular to porphyritic granodiorite, quartz monzodiorite, and tonalite. The pluton is very heterogeneous in texture and lithology, and it may well be a composite body of two to four individual sills (Drinkwater and others, 1990). Magnetite is commonly seen in most thin sections of the Annex Lakes pluton. The central granitic belt lies east of the great tonalite sill belt; it contains three major units, including the Turner Lake batholith, which has two subbelts. Mapping has yet to determine the exact number of plutons that occur in this part of the Coast plutonic-metamorphic complex, but three granodioritic units form a western subbelt, and undivided granite forms an eastern subbelt. These units are distinguished by field and petrographic features, and by MS values (discussed later). Plagioclase EXPLANATION CENTRAL GRANITIC BELT Turner Lake batholith Southern granodiorite unit Central granodiorite unit Northern granodiorite unit

Eastern granite unit Wright Glacier stock GREAT TONALITE SILL BELT 'Y Speel River pluton Taku Cabin pluton + Annex Lakes pluton ADMIRAL TV -REVILLAGIGEDO BELT

Everett Peak pluton & Arthur Peak pluton Grand Island pluton e Glass Peninsula stocks Figure 2. Quartz-potassium-feldspar-plagioclase diagram showing average modal composition of plutons or granitic units along Taku Inlet transect area, plotted on Streckeisen's (1973) plutonic rock classification diagram. Mount juneau, Carlson Creek, and Lemon Creek Glacier plutons are tonalites according to average modal compositions from Drinkwater and others (1990). See table 1 for list of plutons and their belts. Magnetic Susceptibilities and Iron Content of Plutonic Rocks Across the Coast Plutonic-Metamorphic Complex Near Juneau

The northern unit is intermediate in physical features and composition between the porphyritic biotite granite of the eastern subbelt and the typical massive equigranular biotite-hornblende granodiorite of the other two units of the western subbelt. K-feldspar occurs in different forms in the different subbelts and units. In the eastern subbelt, it is coarse granular to phenocrystic; in the central and southern units, it is intergranular and poikilitic; and in rocks of the northern unit, it occurs in all phases. In the western subbelt, granular clots of magnetite and apatite are typical of the central unit but are absent in the southern and northern units. Allanite is the most conspicuous accessory mineral of the eastern subbelt 10 w a.. CJj 6 ffi 4 m Admiralty-Revillagigedo belt

M n n Si02, IN WEIGHT PERCENT 8 ffi, 6 a.. CJj 4 a: w m z Great tonalite sill belt so SI02, IN WEIGHT PERCENT 14 (/) ~10 a.. CJj 8 6 w m 4 z Central granitic belt so Si02, IN WEIGHT PERCENT Figure3. Frequency distribution ofSi02 for major plutonic belts ofT aku In let transect area. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 granites, whereas sphene is more typical of the western subbelt granodiorites. Both allanite and sphene are common in the northern granodiorite unit. Several granitic stocks of late(?) Tertiary age occur within the Turner Lake batholith; they are typically massive, slightly porphyritic but otherwise fine- to medium-grained, magnetite-bearing hornblende-biotite granites. The two foliated stocks (Tease Lake stocks) are also considered part of the central granitic belt. Chemical and MS data are lacking on these foliated stocks, but generally they are heterogeneous in texture and range from granodiorite to monzodiorite in composition. Ages Granitoids of the Taku Inlet transect area are part of three major age groups that were emplaced between Late Cretaceous and Eocene time (Brew and Ford, 1986) . The oldest group includes the plutons of the Grand Island and Taku Harbor subbelts, which belong to the Admiralty-Revillagigedo linear belt of 95-Ma dioritic and quartz dioritic plutons (Brew and Morrell, 1983; Brew and others, 1989) that extends most of the length of southeastern Alaska. The next oldest group includes the plutons of the great tonalite sill belt. The sill-belt plutons range in age from at least as old as 70 Ma to 55 Ma (Brew, 1988). Within the transect area, they are probably Paleocene in age based on a K-Ar hornblende age of 60.5 Ma from the Speel River pluton (recalculated age using newer standard constants; Drinkwater and others, 1989) and from U-Pb ages of 67.0 and 60.0 Ma (Gehrels and others, 1984) from the Carlson Creek and Annex Lakes plutons, respectively. Wood and others ( 1991) report a range of 40 Arf39 Ar ages on biotite and hornblende of 63 to 53 Ma that systematically decrease from west to east across the Speel River pluton in the Holkham Bay area, which is located just south of the study area. A 66.5-Ma K-Ar hornblende age from the Mount Juneau pluton (Drinkwater and others, 1990) is considered a metamorphic age; although its emplacement age is unclear to us, we consider this body as the oldest pluton of the great tonalite sill group in this area. The youngest group consists of plutons of the Turner Lake batholith of the central granitic belt; they were emplaced from 54 to 49 Ma (K-Ar ages; Brew, 1988). A U-Pb zircon age of 50 Ma from granodiorite of the Turner Lake batholith near Taku Inlet was reported by Gehrels and others (1984 ). The westernmost of the Tease Lake stocks, which has an intrusive contact with the Speel River pluton, yielded a K-Ar hornblende age of 50.9 Ma (J.G. Smith and D.A. Brew, unpub. data); its field relationship to the Turner Lake batholith is uncertain because the Tease Lake stocks are mostly within metamorphic country rock.

MAGNETIC SUSCEPTIBILITIES The results of 137 MS measurements (figs. 4, 5) define three different distribution patterns for rocks of the plutonic belts. The MS of the granitoids initially increases from southwest to northeast along the transect to the great tonalite sill belt, beyond which it decreases (with one exception) (fig. 5). The highest MS values of 2,500 to 5,000 x1o-5 SI units (hereafter abbreviated as 2,500-5,000) are found in rocks of the great tonalite sill belt (Speel River, Taku Cabin, and Annex Lakes plutons) and central granodiorite unit of the western subbelt of the Turner Lake batholith. The lowest MS values en 25 Q. 20 15 w In 10 z Western margin plutons MS RANGE, IN Sl UNITS x 10-5 20 , ffi 15

Q.

10 a: w In z Great tonalite sill belt MS RANGE, IN Sl UNITS x 1 o-5 30 en w

20

15 a: w 10 Central granitic belt MS RANGE, IN Sl UNITS x 1 Q-5 Figure 4. Frequency distribution of magnetic susceptibility for major belts ofT aku In let transect area. Magnetic susceptibility measured in 51 units, x 1 o-5. ( <250) are found in two groups of plutons that lie along the western edge of the complex, including the Grand Island pluton and Glass Peninsula stocks, and the Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons. Rocks from these plutons rarely show opaques in thin section; where present, they are generally sulfides or ilmenite, and more rarely magnetite. The small positive MS values of magnetite-free granitoids are attributed to biotite· (Tulloch, 1989). These plutons with very low MS are similar to the ilmenite-series granitoids from the Japan area described by Ishihara (1977). The MS range and average (fig. 5) are generally similar among the plutonic subbelts, but in many cases individual plutons or units have distinct MS signatures. The high MS values of the central granodiorite unit distinguish it from other units of the Turner Lake batholith. Granitic rocks of the eastern subbelt of the Turner Lake batholith generally show a relatively small range of low to intermediate MS values, but the two Wright Glacier stocks exhibit widely different MS values. The Speel River pluton has the widest range of MS values, but all main phases have values generally greater than 1,500. The highest average MS values are found in rocks of the Annex Lakes pluton and central granodiorite unit. Interestingly, three samples of border-zone rocks of the Speel River pluton have very low MS ( <300) values similar to the ilmenite-bearing plutons. The Arthur Peak and Everett Peak plutons (Taku Harbor subbelt) show significant differences in their range and average MS values, with the Everett Peak body having the higher values. CORRELATION WITH CHEMICAL PARAMETERS The patterns of MS range values (fig. 5), surprisingly, show no systematic correlation with those of total iron (FeO*) (fig. 6) or normative magnetite content (fig. 7). The MS of the ilmenite-bearing plutons of the westem margin exhibit a positive correlation with FeO* (fig. 8, lower band of points), and as a group, the plutons of the Turner Lake batholith also show a rough positive correlation between MS and FeO* (fig. 8); however, no correlation is evident on figure 8 when all data points are considered. The granitoids with the highest FeO* and normative magnetite values (Grand Island subbelt and Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons) actually contain no or only traces of magnetite. Conversely, the central granodiorite unit of the Turner Lake batholith, which contains abundant magnetite, has relatively low to moderate amounts of FeO* and normative magnetite but yields very high MS values. This discrepancy is mostly explained by the iron oxidation state (OXS); here, OXS is calculated as the molecular ratio Fe3+J(Fe3+Fe2+) (fig. 9), which shows much better agreement with the MS trend (fig. 5) and Magnetic Susceptibilities and Iron Content of Plutonic Rocks Across the Coast Plutonic-Metamorphic Complex Near Juneau

thus is a better indicator of magnetite content than normative magnetite. A general increase in MS with increase in OXS is indicated by figure I 0 but with much scatter. Bateman and others (I99I) also determined that the regional patterns and variations of magnetic susceptibility in the central Sierra Nevada batholith correlate with iron oxidation ratios but not with total iron content. Most of the magnetite-bearing granitoids have oxidation-state values greater than 0.2I, and most ilmenitebearing (very low MS) granitoids have OXS values less than 0.23 (fig. 9); the overlap of OXS between these two pluton types is a small part of the total range of values and in part may be attributed to FeO analytical problems, as described by Jackson and others (1987). Average OXS values from figure 9 for ilmenite-bearing granitoids are 0.23 or less, and for most magnetite-bearing granitoids they are 0.23 or higher (fig. 9). Data from figure IO (based on OXS vs MS) indicate a division at about 0.2I OXS between these two plutonic types of the Stock 1 Stock2 transect. MS values between I,OOO and 200 occur less frequently in these rocks, as evident from the gaps in data distribution in figures 5 and I 0. Admiralty-Revillagigedo Belt and Mount Juneau, Carlson Creek, and Lemon Creek Glacier Plutons The granitoids along the western margin of the transect area include plutons of the Grand Island and Taku Harbor subbelts and the Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons. They are iron-rich rocks 4 percent FeO*) that have a range of very low to intermediate MS values. Although the Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons are spatially part of the great tonalite sill belt, their petrographic features (Drinkwater and others, 1990), high iron content, and very low MS values tie them (particularly the Mount Juneau pluton) to the Wright Glacier stocks A Turner (Eastern granite unit) C/) 1-z ::J

(.) 0 a: (.) z ::J

a.. (Northern granodiorite unit) Lake (Central granodiorite unit) (Southern granodiorite unit) batholith Armex Lakes pluton B Taku Cabin pluton main phase border ph:se Speel River pluton !-+ -+- Lemon Creek Glacier pluton +- Carlson Creek pluton Mount Juneau pluton + Everett Peak pluton Arthur Peak pluton -+- Butler Peak pluton + Irving Peak pluton -t- Grand Island pluton +- Glass Peninsula stocks MAGNETIC SUSCEPTIBILITY, IN Sl UNITS X 10-5 Figure 5. Ranges of magnetic susceptibility for plutons and granitic units of Taku Inlet transect area., Average magnetic susceptibility shown by vertical tick mark. See Appendex for individual sam~ data. A, Central granitic belt. 8, Great tonalite sill belt. C, Admiralty-Revillagigedo belt. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Grand Island subbelt. More discriminating evidence, such as rare earth chemistry, isotopic attributes, and additional ages, is needed to fully compare these plutons. The Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons have the lowest OXS values, as is expected from ilmenite-series plutons (Takahashi and others, 1980). However, the ilmenite-bearing (very low MS) plutons of the Grand Island subbelt have a higher range of OXS (0.21-0.26) that is comparable to the range of OXS values of the Speel River, Taku Cabin, and Annex Lakes plutons, which have the most magnetite and the highest MS values. Evidently, other factors besides iron content and oxidation state can control the crystallization of Fe3+ -rich minerals (magnetite, biotite, epidote, and possibly garnet). Granitoids of the Everett Peak and Arthur Peak plutons have the highest OXS values in the transect area as well as high total iron (4-6 percent), but they show only intermediate MS values. Much of the iron must have been incorporated into the epidote and hornblende that are common phases in these granitoids. Surprisingly, although the plutons are generally similar, the rocks of the Arthur Peak. pluton have higher OXS values but generally lower MS values than rocks of the Everett Peak pluWright Glacier stocks- ---

ton. The abundant K-feldspar and lack of biotite of these two plutons make them different from the typical 95-Ma dioritic plutons found in the western part of the Coast plutonic-metamorphic complex. Great Tonalite Sill Belt (Speel River, Taku Cabin, and Annex Lakes Plutons) The Speel River and Taku Cabin plutons are typical granitoids of the great tonalite sill belt; both show very similar MS ranges and averages as well as similar chemical, petrographic, and textural characteristics (Drinkwater and others, 1989, 1990). The iron data from the Taku Cabin pluton are within the narrow range of iron data shown for the Speel River pluton (figs. 6, 7). In contrast, the Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons differ by their much lower MS and OXS values but higher average FeO*. In comparison, the Annex Lakes pluton has a large range of FeO* with a lower average FeO* value than the other great tonalite sill belt rocks but a higher average OXS value. Two samples from the Annex Lakes pluton not included in figures 4 and 5 because of their abundant A Turner (Eastern granite unit) (Northern granodiorite unit) Lake (Central granodiorite unit) batholith (Southern granodiorite unit) C/) t: z Annex Lakes pluton

B

(.) 0 a: (.) z

Speel River and Taku Cabin plutons Lemon Creek Glacier pluton Carlson Creek pluton Mount Juneau pluton Arthur Peak and Everett Peak plutons Butler Peak pluton --+- Grand Island pluton Glass Peninsula stocks J I FeO*, IN WEIGHT PERCENT

Figure 6. Ranges of FeO* for plutons and granitic units of Taku Inlet transect. Averages shown by vertical tick marks. A, Central granitic belt. B, Greattonalite sill belt. C, Admiralty-Revillagigedo belt. Magnetic Susceptibilities and Iron Content of Plutonic Rocks Across the Coast Plutonic-Metamorphic Complex Near Juneau

coarse-grained (>2 mm) magnetite yielded the highest MS values (8,000 and 7 ,000) of the complex. Granitoids of the Central Granitic Belt East of the great tonalite sill belt, the Coast plutonic-metamorphic complex is composed of magnetitebearing calc-alkaline plutons of the Turner Lake batholith, late(?) Tertiary granitic stocks, and screens of metamorphic rock. Rocks of the Turner Lake batholith contain low FeO* (mostly percent), but have relatively high OXS values and intermediate MS values. The much higher MS shown by rocks of the central granodiorite unit of the western subbel.t reflects the abundant magnetite-apatite clots that are characteristic of this unit. Although the southern granodiorite unit has a higher average range of FeO* than other units of the Turner Lake batholith and similar OXS values as the central granodiorite unit, it contains less magnetite and lower MS values than the central granodiorite unit. The reason for this discrepancy is unknown but may be partially due to the more mafic-silicate-rich composition of the southern unit. Tulloch (1989) argued that Fe3+ may be preferentially incorporated into biotite under certain conditions. The granite unit of the eastern subbelt contains only 1 to 3 percent FeO* but has a wide range of low to high OXS values and a relatively narrow and low range of MS values. The cause of this inconsistency is unclear but may be due in part to the abundant allanite in these rocks; allanite can accommodate varying amounts of trivalent iron (Deer and others, 1975). Analytical data are sufficient for consideration here of only one late(?) Tertiary granitic stock (southern Wright Glacier stock); these rocks have low to intermediate OXS and FeO* values and very low MS values. Whether this is typical of other late(?) Tertiary stocks in the region is unknown. but preliminary petrographic examination of available thin sections from these rocks reveal very few opaques such as magnetite. DISCUSSION The three major plutonic belts of the transect are characterized by plutons and plutonic units that are of A Wright Glacier stocks en tz

0 0 a: 0 z 0 t- J a.. Turner (Eastern granite unit) (Northern granodiorite unit) Lake (Central granodiorite unit) batholith (Southern granodiorite unit) Annex Lakes pluton B Speel River and Taku Cabin plutons Lemon Creek Glacier pluton Carlson Creek pluton Mount Juneau pluton Arthur Peak and Everett Peak plutons Butler Peak pluton Grand Island pluton NORMATIVE MAGNETITE, IN WEIGHT PERCENT Figure 7. Ranges of normative magnetite for pi utons and granitic units ofT aku In let transect. Averages shown by vertical tick marks. A, Central granitic belt. B, Great tonalite sill belt. C, AdmiraltyRevillagigedo belt. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

similar age and have distinct MS values. The 95-Ma plutons in the western part of the transect are, in general, ilmenite-bearing types, and those of Tertiary age to the east are magnetite-bearing types. The larger plutons of the great tonalite sill belt, which separates the aforementioned western and eastern belts, are characterized by very high MS values and magnetite content and moderately high OXS values, whereas some (smaller ones) are more like the 95-Ma plutons. MS values, along the transect, correlate better with OXS values than with total FeO* or normative magnetite. An average OXS value of 0.23 roughly separates ilmenite-bearing plutons from magnetite-bearing plutons, but discrepancies in the MSOXS correlation indicate that other factors may influence the crystallization of magnetite and Fe3+ -rich silicate minerals. The Grand Island subbelt and Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons are not S-type granitoids according to the criteria of Chappell and White (1974), as would be expected for ilmenite-series rocks (Tulloch, 1989). They are instead Si02-poor (54-64 percent), Na20-rich (>3.2 percent) rocks with Al20i(N~O+Ca0+K20) ratios <l. These are 1-type granitoids that were generated either as reduced magmas or as oxidized magmas that became reduced during their magmatic history. The differences between the ilmenite- and magnetite-bearing type plub

A .J. A ee

A+ A e A e e ++ + A

A t

.A A en

A

d

en

::J

en u

) FeO*, IN WEIGHT PERCENT EXPLANATION e Western margin plutons (95 Ma) A Plutons of great tonalite sill belt (Spec! River, Taku Cabin, Armcx Lake plutons) + Turner Lake batholith Figure 8. Magnetic susceptibility versus FeO* for granitic samples from major plutonic belts. Lower points from western margin plutons are sample~ of ilmenite-bearing plutons. tons may bear on the nature of the bedrock terranes into which they were emplaced. The ilmenite series of granitoids are believed to represent reduced magmas (Ishihara, 1977; Takahashi and others, 1980) and are generally associated with S-type granitoids (Chappell and White, 1974; Tulloch, 1989) or strongly contaminated reduced 1-type granitoids (Ishihara and Sasaki, 1989). Magmas may be reduced at any stage-from site of generation, during ascent, or at shallow levels (Ishihara, 1977; Ishihara and Sasaki, 1989). At low oxygen fugacity, iron is not partitioned into FeTi oxides but rather becomes incorporated into the ferromagnesian silicate phases (biotite, amphiboles, epidote, and garnet) and sometimes into iron sulfides (Puranen, 1989). Some workers (Ishihara, 1977; Ishihara and Sasaki, 1989) argue that the reduction of magma at the site of generation and during ascent is attributed to incorporation of carbon and sulfur by assimilation of carbonaceous and sulfur-rich sedimentary or metasedimentary rocks, resulting in lowered oxygen fugacity and reduction of iron. In their study of the central Sierra Neveda batholith, Bateman and others (1991) found no discernible relationship between free (noncarbonate) carbon and magnetic susceptibility and explained the regional variations in magnetic susceptibility and iron oxidation state as a function of the relative abundance of sialic crustal component in the magma. Local reduction of intrusive magma at margins of magnetite-bearing plutons can also occur by assimilation of pelitic wall rock (Ishihara and others, 1985). Magmas at shallower levels may be reduced by hydrothermal solutions. According to Ishihara (1977), Takahshi and others (1980), and Ishihara and Sasaki ( 1989), the ilmenite series of granitoids were generated at shallower levels than the magnetite series; the latter formed at deep levels in tensional tectonic settings where carbonaceous material is not stable, although Puranen ( 1989) argues that magnetite formation is possible at different depths, including shallow levels from reduced magmas . The differences in MS signature between the ilmenite-bearing plutons of the Admiralty-Revillagigedo belt and the magnetite-bearing plutons of the central granitic belt may. be largely explained by the differences in their ages and in the host terranes they intruded. Differences in source regime and emplacement history for the two plutonic groups could account for different oxidation histories. The rocks of the great tonalite sill belt were emplaced in a zone of compressional deformation represented by highly deformed, sheared, and moderately to highly metamorphosed rocks that form a boundary zone between two major composite terranes (Monger and others, 1982) considered by Rubin and others (1990) to be a former continental margin ·convergent zone. Brew and Ford ( 1983 ), in contrast, consider this zone essentially as a continent-continent collision zone. The Magnetic Susceptibilities and Iron Content of Plutonic Rocks Across the Coast Plutonic-Metamorphic Complex Near Juneau

emplacement of rocks of the great tonalite sill belt occurred later and formed a suture between the two terranes; their age and MS signature are similar to those plutons of the central granitic belt. The MS variations in plutons of the central granitic belt follow more closely a pattern of decreasing MS with increasing differentiation (increasing Si02 and K20) that is probably related to extensional tectonics and the distance to the collision or convergent zone to the west. The Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons remain anomalous in being ilmenite-bearing plutons located east of the Coast Range megalineament. If these plutons represent an earlier phase of tonalitic sill plutonism and are unrelated to the 95-Ma Grand Island subbelt plutonism, then the relative spatial position of these two groups on opposite sides of the megalineament is coincidental. Low MS values are reported for deformed granitoids in metamorphic terranes (for example, Ishihara, 1977; Tulloch, 1989). In Japan, low MS (ilmenite-type) foliated granitoids occur in high TIP grade zones of metamorphic belts (Ishihara, 1977). Tulloch (1989) describes mylonitic and cataclastically deformed plutonic rocks in New Zealand that have relatively high Fe3+fFe2+ values but very low MS values. Gastil (1990), in his study of en Wright Glacier stocks Turner (Eastem granite unit) (Northern granodiorite unit) Lake (Central granodiorite unit) batholith (Southern granodiorite unit) the batholiths of southern California, also noted a correlation between grade of metamorphism of host rock and magnetic character of the intruded plutons, with nonmagnetic plutons occurring in higher grade metamorphic rocks. The Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons, which are ilmenite bearing, occur with the highest grade metamorphic rocks in the transect (Brew and others, 1989). However, in the Taku Inlet transect, an increase in MS eastward from the Grand Island subbelt to the great tonalite sill belt (Speel River pluton) corresponds with northeast progressive metamorphic grade (Brew and others, 1989). Contamination and tectonic stresses may have collectively influenced the production of the ilmenite-bearing plutons. The deep bedrock units beneath the moderately metamorphosed Gravina overlap assemblage, where the Grand Island subbelt is exposed, are uncertain but possibly include Alexander and Wrangellia terrane rocks that contain some graphite or other carbonaceous material (Brew and others, 1984; Brew and Ford, 1986; Brew and Karl, 1988). The plutons of the great tonalite sill belt were emplaced in higher grade metamorphic rocks, which we interpret to have been derived from the Alexander and Wrangellia terranes and Gravina overlap A z ::J

0 0 a: 0 z f2 ::J

a.. Annex Lakes pluton B Speel River and Taku Cabin plutons Lemon Creek Glacier pluton Carlson Creek pluton Mount Juneau pluton

Arthur Peak and Everett Peak plulons Butler Peak pluton Grand Island pluton Glass Peninsula stocks IRON OXIDATION STATE, X 10 Figure9. Ranges in oxidation state [molecularratioofFe3/(Fe3+Fe2)(x1 O)J forplutonsand granitic units ofTaku In let transect. Average shown by vertical tick marks. A, Central granitic belt. 8, Great tonalite sill belt. C, Admiralty-Revillagigedo belt. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

assemblage. The common occurrence of hornfelsed sedimentary and volcanic rock inclusions, and of sulfide minerals, which according to Ishihara and others ( 1985) are common in contaminated reduced magma types, implies that contamination was a factor in the formation of the ilmenite-bearing type plutons of the western margin. Because the ilmenite-bearing lviount Juneau, Carlson Creek, and Lemon Creek Glacier plutons in the belt are the most deformed and metamorphosed, we conclude that high compressional stresses may have also contributed to the development of reduced magmas and ilmenite-bearing plutons in the western part of the Juneau transect. An alternate explanation is that the Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons are analogous to the deformed and metamorphosed (and very low MS) border-phase rocks of the Speel River pluton, as defined by Drinkwater and others (1989). This suggests that the chemical and physical features of these plutons are a function of size more than age. The smaller size of the Mount Juneau, Carlson Creek, and Lemon Creek Glacier plutons may have allowed for more complete contamination. Many of the ilmenite-bearing granitoids of the transect contain primary (magmatic) epidote, which according to the critera of Zen and Hammarstrom (1984) indicates formation under oxidizing conditions at pressures greater than 8 kbars or at depths greater than 25 km. This conclusion conflicts with the contention of Takahashi and others ( 1980) and Ishihara and Sasaki ( 1989) that ilmenite-series granitoids, if formed at sites of magma generation, form at shallower depths and pressures under reducing conditions. If these particular 10000 .., ,.--, --""7'1 1000-: 100: / /

10-: /

. :. : /

IRON OXIDATION STATE, X 10 Figure 10. Magnetic susceptibility versus oxidation state values (x1 0) for granitic samples from Taku Jnlettransect. Long dashed line is best-fit line based on simple exponential regression equation. Short dashed line shows arbitrary division between magnetite- and ilmenite-bearing granitic samples according to oxidation state. granitoids indeed formed at great depths, then they must have been reduced during their ascent to upper crustal levels. The occurrence of magmatic epidote in the Mount Juneau pluton but not in the other two similar ilmenite-bearing sills (Lemon Creek Glacier and Carlson Creek plutons) is perplexing. All three sills consist of iron-rich granitoids, but the OXS of the Mount Juneau pluton is slightly higher, overlapping the OXS of the Grand Island subbelt plutons; this suggests a possible genetic link between the Mount Juneau pluton and plutons of the Grand Island subbelt. As far as we know, epidote-bearing ilmenite-type plutons are unique and are seldom described in the published liter:ature; it is not known whether other 95-Ma epidote-bearing plutons of this belt are ilmenite- or magnetite-bearing types, but petrographic descriptions from Burrell (1984) and Douglass and others (1989) reveal a conspicuous absence of magnetite and opaques from most rocks of ihe 95-Ma plutons in the Petersburg region (Brew and others, 1984 ). We examined many of the opaque-bearing sections from this group and observed that the opaques are chiefly sulfides or secondary magnetite. Zen and Hammarstrom (1984) report that the Ecstall and Moth Bay plutons farther away at the very southern end of this belt contain magnetite and primary epidote. This observation suggests that a large part of the 95-Ma plutonic belt in central and northern southeastern Alaska may be ilmenite-bearing type granitoids. The Everett Peak pluton is unusual in that it contains both magnetite and epidote and no primary biotite. The crystallization history of this pluton is apparently different from that of the magnetite-absent/poor, biotiterich, epidote-bearing Mount Juneau and Grand Island subbelt plutons. The fact that well-developed (primary) epidote and sulfides do not occur together in the same thin section may bear on the origin of this pluton. The lack of a hydrous silicate phase (biotite) in this pluton implies that its magma crystallized under either low water pressure or high load pressure, with the second possibility being compatible with the presence of primary epidote . The magnetite-bearing granitoids of the Coast plutonic-metamorphic complex were emplaced in a terrane of high-grade metamorphic rocks and are interpreted to have formed under conditions of higher oxidation state than the plutons along the western margin. The decrease in MS east of the great tonalite sill belt correlates with an increase in K-feldspar and Si02 content across the complex . Oxidation-state values remain relatively high, but total iron is rapidly depleted eastward, resulting in lower MS values. These trends may reflect variations in the depth of magma generation and thus the distance from the subduction or collision zone to the west. In conclusion, we believe that the magnetic susceptibility data demonstrate significant differences between Magnetic Susceptibilities and Iron Content of Plutonic Rocks Across the Coast Plutonic-Metamorphic Complex Near Juneau

the granitoids in the three major plutonic belts of the Taku Inlet transect area. Differences in total iron content may explain variations in MS within subbelts or larger suites, but they do not explain the differences in MS on a larger regional scale; rather, the presence and amount of magnetite in these rocks depend largely on the degrees of oxidation and magmatic differentiation, grade of metamorphism, and amount of deformation during synkinematic emplacement. These factors are also affected by many variables, including the depth of magma generation, composition of the host country rock, amount of crustal assimilation, and hydrothermal alteration. We are uncertain if the reduced oxidation state of the magmas that formed the ilmenite-bearing plutons is a primary magmatic feature or if it is related to synkinematic emplacement under conditions of strong tectonic stress. This question may be resolved when more physical and chemical data become available for these plutons from the entire Admiralty-Revillagigedo belt. REFERENCES CITED Arth, J.G., Barker, F., and Stem, T.W., 1988, Coast batholith and Taku plutons near Ketchikan, Alaska: Petrography, geochronology, geochemistry, and isotopic character: American Journal of Science, v. 288-A, p. 461-489. Bateman, P.C., Dodge, F.C., and Kistler, R.W., 1991, Magnetic susceptibility and relation to initial 87Sr/86Sr for granitoids of the central Sierra Nevada, California: Journal of Geophysical Research, v. 96, no. B 12, p. 19,555-19,568. Berg, H.C., Jones, D.L., and Coney, P.J., 1978, Map showing pre-Cenozoic tectonstratigraphic terranes of southeastern Alaska and adjacent areas: U.S. Geological Survey OpenFile Report 78-105, scale 1:1,000,000. Brew, D.A., 1988, Latest Mesozoic and Cenozoic magmatist.o in southeastern Alaska-A synopsis: U.S. Geological Survey Open-File Report 88-405, 41 p. Brew, D.A., and Ford, A.B., 1978, Megalineament in southeastern Alaska marks southwest edge of Coast Range batholithic complex: Canadian Journal of Earth Science, v. 15, no. 11, p. 1763-1772. ---1983, Comment on "Monger, J.W.H., Price, R.A., and Tempelman-Kluit, D.J., 1982, Tectonic accretion and the origin of two major metamorphic and plutonic welts in the Canadian Cordillera": Geology, v. 11, p. 427-429. ---1984a, Tectonstratigraphic terranes in the Coast plutonic-metamorphic complex, southeastern Alaska, in Reed, K.M., and Bartsch-Winkler, S., eds., The United States Geological Survey in Alaska: Accomplishments during 1982: U.S. Geological Survey Circular 939, p. 90-93. ---1984b, The northern Coast plutonic complex, southeastern Alaska and northwestern British Columbia, in Coonrad, W.L., and Elliott, R.L., eds., The United States Geological Survey in Alaska: Accomplishments during 1981: U.S. Geological Survey Circular 868, p. 120-124. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 ---1986, Preliminary reconnaissance geologic map of the Juneau, Taku River, Atlin and parts of the Skagway 1:250,000 quadrangles, southeastern Alaska: U.S. Geological Survey Open-File Report 85-395, 23 p., 2 sheets. Brew, D.A., Ford, A.B., and Himmelberg, G.R., 1989, Evolution of the western part of the Coast plutonic-metamorphic complex, southeastern Alaska, USA-A summary, in Daly, J.S., and others, eds., Evolution of metamorphic belts: Geological Society of America Special Publication 42, p. 447-452. Brew, D.A., and Grybeck, D., 1984, Geology of the Tracy Arm-Fords Terror Wilderness study area and vicinity, in U.S. Geological Survey and U.S. Bureau of Mines, Mineral resources of the Tracy Arm-Fords Terror Wilderness Study Area and vicinity, Alaska: U.S. Geological Survey Bulletin 1525, p. 19-52. Brew, D.A., and Karl, S.M., 1988, A reexamination of the contacts and other features of the Gravina belt, southeastern Alaska, in Galloway, J.P., and Hamilton, T.D., eds., Geologic studies in Alaska by the U.S. Geological Survey during 1987: U.S. Geological Survey Circular 1016, p. Brew, D.A., and Morrell, R.P., 1983, Intrusive rocks and plutonic belts of southeastern Alaska, U.S .A., in Roddick, J.A., ed., Circum-Pacific plutonic terranes: Geological Society of America Memoir 159, p. 171-193. Brew, D.A., Ovenshine, A.T., Karl, S.M., and Hunt, S.J., 1984, Preliminary reconnaissance geologic map of the Petersburg and parts of the Port Alexander and Sumdum 1 :250,000 quadrangles, southeastern Alaska: U.S. Geological Survey Open-File Report 84-405, 43 p., 2 pis. Burrell, P.D., 1984, Late Cretaceous plutonic rocks, Petersburg quadrangle, southeast Alaska, in Reed, K.M., and BartschWinkler, S., eds., The United States Geological Survey in Alaska: Accomplishments during 1982: U.S. Geological Survey Circular 939, p. 93-96.

Chappell, B.W., and White, A.J.R., 1974, Two contrasting granite types: Pacific Geology, v. 8, p. 173-174. Criss, R.E., and Champion, D.E., 1984, Magnetic properties of granitic rocks from the southern half of the Idaho batholith: Influences of hydrothermal alteration and implications of aeromagnetic interpretation: Journal of Geophysical Research, v. 89, no. B8, p. 7061-7076. Deer, W.A., Howie, R.A., and Zussman, J., 1975, An Introduction to the rock-forming minerals: London, Longman, 528 p. Douglass, S.L., Webster, J.H., Burrell, P.D., Lanphere, M.l., and Brew, D.A., 1989, Major-element chemistry, radiometric ages, and locations of samples from the Petersburg and parts of the Port Alexander and Sumdum quadrangles, southeastern Alaska: U.S. Geological Survey Open-File Report 89-527, 66 p. Drinkwater, J.L., Brew, D.A., and Ford, D.A., 1989. Petrographic and chemical description of the variably deformed Speel River pluton, south of Juneau, southeastern Alaska, in Dover, J.H., and Galloway, J.P., eds., Geological studies in Alaska by the U.S. Geological Survey, 1988: U.S. Geological Survey Bulletin 1903, p. 104-112. ---1990, Petrographic and chemical data for the large Mesozoic and Cenozoic plutonic sills east of Juneau,

southeastern Alaska: U.S. Geological Survey Bulletin 1918, 47 p. Gastil, G., 1990, The boundary between the magnetite-series and ilmenite-series granitic rocks in Peninsular California: University Museum, University of Tokyo, Nature and Culture, no. 2., p. 91-99. Gehrels, G.E., Brew, D.A., and Saleeby, J.B., 1984, Progress report on U-Pb (zircon) geochronologic studies in the Coast plutonic-metamorphic complex east of Juneau, southeastern Alaska, in Reed, K.M., and Bartsch-Winkler, S., eds., The United States Geological Survey in Alaska: Accomplishments during 1982: U.S. Geological Survey Circular 939, p. 100-102. Gehrels, G.E., McClelland, W.C., Samson, S.D., Patchett, P.J., and Jackson, J.L., 1990, Ancient continental margin assemblage in the northeiJl Coast Mountains, southeast Alaska and northwest Canada: Geology, v. 18, no. 3, p. Hooper, R.J., Brew, D.A., Himmelberg, G.R., Stowell, H.H., Bauer, R.L., and Ford, A.B., 1990, The nature and significance of post-thermal-peak shear zones west of the great tonalite sill near Juneau, southeastern Alaska, in Dover, J.H., and Galloway, J.P, eds., Geologic studies in Alaska by the U.S. Geological Survi?Y· 1989: U.S. Geological Survey Bulletin 1946, p. 88-94. Ishihara, S., 1977, The magnetite-series and Ilmenite-series granitic rocks: Mining Geology, v. 27, p. 293-305. Ishihara, S., Matsuhisa, Y ., Sasaki, A., and Terashima, S., 1985, Wall rock assimilation by magnetite-series granitoid at the Miyako pluton, Kitakami, northeastern Japan: Journal of the Geological Society of Japan, v. 91, no. 10, p. Ishihara, S., and Sasaki, A., 1989, Sulfur isotopic ratios of the magnetite-series and ilmenite-series granitoids of the Sierra Neveda batholith-A reconnaissance study: Geology, V. 17, p. 788-791. Jackson, L.L., Brown, F.W., and Neil, S.T., 1987, Major and minor elements requiring individual determination, classical whole rock analysis, and rapid rock analysis, in Baedecker, P.A., ed., Methods for geochemical analysis: U.S. Geological Survey Bulletin 1770, p. G1-G23. Monger, J.W.H., and Berg, H.C., 1987, Lithotectonic terrane map of western Canada and southeastern Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF1874B, scale 1:2,500,000. Monger, J.W.H., Price, R.A., and Tempelman-Kluit, D.J., 1982, Tectonic accretion and the origin of two major metamorphic and plutonic welts in the Canadian Cordillera: Geology, v. 10, p. 70-75. Piccoli, P.M., and Hyndman, D.W., 1985, Magnetite/ilmenite boundary in the western Atlanta lobe of the ·Idaho batholith: Northwest Geology, v. 14, p. 1-5. Puranen, Risto, 1989, Susceptibilities, iron and magnetite content of Precambrian rocks in Finland: Geological Survey of Finland, Report of Investigations no. 90, 45 p. Ross, D.C., 1989, Magnetic susceptibilities of modally analyzed granitic rocks from the southern Sierra Nevada, California: U.S. Geological Survey Open-File Report 89204, 53 p. Rubin, C.M., Saleeby, J.B., Cowan, D.S., Brandon, M.T., and McGroder, M.P., 1990, Regionally extensive mid-Cretaceous west-vergent thrust system in the northwestern Cordillera: Implications for continent-margin tectonism: Geology, v. 18, p. 276-280. Stowell, H.H., and Hooper, R.J., 1990, Structural development of the western metamorphic belt adjacent to the Coast plutonic complex, southeastern Alaska: Evidence from Holkham Bay; Tectonics, v. 9, p. 391-407. Streckeisen, A.L, 1973, Plutonic rocks-classification and nomenclature recommended by the I.U.G.S. subcommission on the systematics of igneous rocks: Geotimes, v. 18, no. 10, p. 26-30. Takahashi, M., Aramaki, S., and Ishihara, S., 1980, Magnetiteseries/ilmenite series vs. 1-type/S-type granitoids: Mining Geology, Special Issue no. 8, p. 13-28. Tulloch, A.J ., 1989, Magnetic susceptibilities of WestlandNelson plutonic rocks: Discrimination of Paleozoic and Mesozoic granitoid suites: New Zealand Journal of Geology and Geophysics, v. 32, p. 197-203. Wood, D.J., Stowell, H.H., Onstott, T.C., and Hollister, 1991, 40Ar/39Ar constraints on the emplacement, uplift and cooling of the Coast plutonic complex sill, Southeastern Alaska: Geological Society of America Bulletin, v. 103, p. Zen, E-an, and Hammarstrom, J .M., 1984, Magmatic epidote and its petrologic significance: Geology, v. 12, p. 515518. Reviewers: james G. Moore and Robert C. Jachens Magnetic Susceptibilities and Iron Content of Plutonic Rocks Across the Coast Plutonic-Metamorphic Complex Near Juneau

High-Pressure Amphibolite-Facies Metamorphism and Deformatio_n Within the Yukon-Tanana and Taylor Mountain Terranes, Eastern Alaska By Cynthia Dusei-Bacon and Vicki L. Hansen Abstract Ductilely deformed amphibolite-facies tectonites comprise two adjacent terranes in the eastern part of the Yukon-Tanana upland in eastern Alaska. These terranes differ in protoliths, structural level, and cooling ages. A structurally complex zone of ductilely deformed, gently north-dipping tectonites separates the two terranes. The northern, structurally higher Taylor Mountain terrane includes garnet amphibolite, biotite±hornblende gneiss, marble, quartzite, metachert, pelitic schist, and crosscutting granitoids of intermediate composition (including the Taylor Mountain batholith). Lithologic associations and isotopic data from the granitoids suggest an oceanic or marginal basin origin for the Taylor Mountain terrane. 40Arf39Ar metamorphic cooling ages from the Taylor Mountain terrane are latest Triassic to earliest Middle jurassic. The southern, structurally lower Lake George subterrane of the Yukon-Tanana terrane is made up of quartz-biotite schist and gneiss, augen gneiss, pelitic schist, garnet amphibolite, and quartzite; we interpret it to comprise a continental margin and granitoid belt built on North American crust. Metamorphic cooling ages from the Lake George subterrane are almost entirely Early Cretaceous. Geothermobarometric analysis of garnet rims and adjacent phases in garnet amphibolite and pelitic schist from the Taylor Mountain terrane and Lake George subterrane yields the following results: Taylor Mountain terrane: 7 to 12.5 kbar at 560oC to 685°C within a northern structural zone in which displacement is consistently top-to-thenortheast, and 6 to 7 kbar at 560oC to 605oC within a southern structural zone in which lineations and shearsense directions are more variable; Lake George· subterrane: 8.2 to 12.5 kbar at 625oC to 750oC. Preservation of growth zoning in garnet and the absence of partial melting in both terranes suggest metamorphic temperatures did not exceed 600oC to 650°C. Tectonites of the Lake George subterrane record dominantly top-to-the-northwest displacement in the northern part of the study area and top-to-the-southeast displacement within the southern part of the area. Where the two shear-sense directions occur together, top-to-the-northwest displacement preceded topto-the-southeast displacement and was more penetrative. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 We interpret the pressure, temperature, kinematic, and age data to indicate that metamorphism of the Taylor Mountain terrane and Lake George subterrane took place during different phases of a latest Paleozoic through early Mesozoic shortening episode, resulting from closure of an ocean basin now represented by klippen of the Seventymile-Siide Mountain terrane: High-pressure metamorphism of the Taylor Mountain terrane took place within a southwest-dipping (present-day coordinates) subduction system. High-pressure metamorphism of the Lake George subterrane occurred during continentward overthrusting of the Taylor Mountain and SeventymileSlide Mountain terranes and imbrication of the continental margin in jurassic time. The difference in metamorphic cooling ages between the Taylor Mountain terrane and adjacent parts of the Lake George subterrane is best explained by Early Cretaceous unroofing of the Lake George subterrane caused by crustal extension, recorded in its younger top-to-the-southeast fabric. Extension was probably related to the development of a northeast-dipping (present-day coordinates) subduction system outboard of the continental margin. Although our geothermobarometric and kinematic data are from the eastern Yukon-Tanana upland, several lines of evidence suggest that Early to Middle Jurassic obduction of the outboard Seventymile-Siide Mountain and Taylor Mountain terranes onto the lower-plate Lake George subterrane was also a major tectonic and metamorphic event in the western Yukon-Tanana upland. INTRODUCTION The Yukon-Tanana terrane has commonly been defined as the broad elongate band of heterogeneous metamorphic and igneous rocks that lies between the right-lateral Tintina and Denali faults in east-central Alaska and Yukon Territory (Jones and others, 1987; Monger and Berg, 1987). Numerous workers have recognized the composite nature of this complex and extensive terrane and have proposed various subdivisions of it (Churkin and others, 1982; Aleinikoff and others, 1987;

Wheeler and McFeely, 1987; Wheeler and others, 1988; Hansen and others, 1991). The origin and history of the composite Yukon-Tanana terrane are clouded not only by its heterogeneity, structural complexity, poor exposure, and paucity of fossil-age control, but also because it occupies a suspect position in the northern Cordillera, being fault bounded along most of its length and lying between autochthonous or slightly displaced North America to the northeast and the innermost far-travelled terranes to the southwest (fig. 1 ). A flap of the terrane has been offset by approximately 450 km of dextral offset along the Tintina fault and lies northeast of the Tintina fault (Roddick, 1967; Tempelman-Kluit, 1979). Recent detailed kinematic, metamorphic, and isoto~ pic studies in the Teslin suture zone of southern Yukon (fig. 1) and kinematic and isotopic studies in eastern Alaska (Hansen, 1990; Hansen and others, 1991) have resulted in a subdivision of the Yukon-Tanana terrane into two separate structural levels, with differing cooling histories, that originated on opposite sides of an ancient ocean basin, the Anvil Ocean of Tempelman-Kluit ( 1979). The Teslin suture zone forms the fundamental boundary between the parautochthonous rocks to the east and allochthonous terranes to the west; it is marked by elongate, northwest-trending belts of the Teslin-Taylor Mountain and Nisutlin terranes (fig. 1 ). Deformational fabrics of the Teslin-Taylor Mountain and Nisutlin terranes are steeply dipping within the Teslin suture zone, whereas elsewhere in Yukon and east-central Alaska, the metamorphic fabrics and terrane boundaries are generally gently dipping. The interpretation of structural levels presented in this paper follows that presented by Hansen and her coworkers; however, rather than considering the upper structural levels as subdivisions of the Yukon-Tanana terrane, we have retained the YukonTanana terrane name only for the structurally lower packages (subterranes) and consider the upper-level packages as terranes in their own right (fig. 2). Although metamorphic rocks compose the major part of the formerly, broadly defined Yukon-Tanana terrane, and although knowledge of the conditions and tectonic setting of metamorphism is crucial to unraveling the tectonic evolution of this terrane, few geothermobarometric studies have been conducted on these rocks in Alaska. Geothermobarometry of lowerlevel rocks in the western part of the Yukon-Tanana terrane in Alaska has indicated medium-pressure ( -5 kbar) metamorphism in two areas (Keskinen, 1989; Sisson and others, 1990). In southern Yukon, medium- to highpressure (5-17 kbar) metamorphism is recorded in structurally higher rocks (correlative with the Taylor Mountain terrane), and high-pressure (7-13 kbar) metamorphism is recorded in structurally lower rocks (correlative with the Lake George subterrane) (Hansen, 1992b ). In this paper, we present the preliminary results of our combined geothermobarometric and kinematic study of tectonites of the structurally higher Taylor Mountain terrane and the structurally lower Lake George subterrane of the Yukon-Tanana terrane (subterrane terminology proposed in Nokleberg and others, 1989). Our primary focus is on the pressure and temperature (P-D estimates from garnet amphibolite and pelitic schist, but the kinematic and cooling histories of these rocks also are briefly outlined because these histories are critical to our metamorphic and tectonic interpretation of the quantitative P-T calculations. We also speculate about the importance and applicability of the protracted late Paleozoic to early Mesozoic contractional episode and the Early Cretaceous extensional episode we recognize in eastern Alaska to the geologic evolution of the western part of the Yukon-Tanana upland. Age designations used in this report are based on the Decade of North American Geology geologic time scale (Palmer, 1983). GEOLOGIC FRAMEWORK The Yukon-Tanana upland, a physiographic province of moderately dissected hills and mountains bounded by the Tintina fault on the north and the Tanana River on the south, is composed primarily of greenschist- and amphibolite-facies rocks of Paleozoic and probable Proterozoic protolith age that are intruded by postkinematic granitoids of latest Triassic to Early Jurassic, mid-Cretaceous, and early Tertiary age (Foster and others, 1987). The general geology of the major part of the Yukon-Tanana upland in Alaska is shown in figure 2, which elucidates the relationships that we discuss in this paper. Mylonitic and blastomylonitic textures are common in most metamorphic rocks, and many medium-grade rocks are sufficiently foliated and lineated to be classified as L-S tectonites (Turner and Weiss, 1963). Many metamorphic-unit boundaries, as defined on a metamorphic facies map of the region (Dusel-Bacon and others, in press), also are terrane or subterrane boundaries marked by low-angle faults. Metamorphic grade changes abruptly across many of the faults, indicating that major metamorphism predated final emplacement of the fault-bounded units (Foster and others, 1987; Dusel-Bacon and others, in press). Structurally higher terranes all yield pre-Cretaceous metamorphic cooling ages both in Alaska and in presumably equivalent rocks in Yukon Territory (Hansen and others, 1991). The highest structural level, shown in figure 2, comprises a dismembered ophiolite composed of fault-bounded slices of serpentinized peridotite and weakly metamorphosed mafic volcanic and sedimentary rocks (Keith and others, 1981); sedimentary rocks are Mississippian to Late Triassic in age (Foster and others, High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

t.,J

S'" Oq ;:;·

:r

Dr

c:r

c:

S'" Oq ;:;·

loCI

E 147 ° Cretaceous granites Seventymile-Siide Mountain terrane Nisutlin terrane* Teslin-Taylor Mountain terrane* Devonian to Mississippian augen gneiss (grey) and host metasedimentary rocks of the Yukon-Tanana terrane* Inboard, (par)autochthonous terranes Dorsey marginal basin terrane

Cassiar passive continental margin Ancestral North America Outboard, allochthonous terranes

Quesnellia terrane

Cache Creek terrane

Stikinia terrane

Nisling terrane Major roads

Contact

Thrust fault--sawteeth on upper plate

MACKENZIE FOLD KILOMETERS Figure 1. Simplified terrane map of northern Canadian and Alaskan Cordillera (modified from Hansen and others, 1991; data sources listed therein). Town abbreviations: Fb, Fairbanks; Ds, Dawson; Rr, Ross River; Wh, Whitehorse; Sk, Skagway. Location of eclogite (Erdmer and Helmstaedt, 1983; Laird and others, 1984; Brown and Forbes, 1986; Erdmer, 1987) shown by stars. *denotes terranes formerly included as subterranes of Yukon-Tanana terrane. TSZ, Teslin suture zone (see text for explanation). Unlabeled area surrounding Fairbanks is alluvium and unlabeled area north of Denali fault in Alaska comprises various terranes not discussed in this paper.

EXPLANATION Surficial deposits (Quaternary) Postmetamorphic granitoids (Tertiary) Postmetamorphic granitoids (Cretaceous) Unmetamorphosed sedimentary and (or) volcanic rocks STRUCTURALLY HIGHER ROCKS Peridotite, greenstone, metalimestone, metachert, and metasedimentary rocks-- Seventymile terrane Greenschist-facies schist, quartzite, mylonitic schist, phyllite, and marble--Nisutlin terrane Intermediate-composition granitoids (Early Jurassic to Late Triassic)--Taylor Mountain terrane Amphibolite-facies gneiss, amphibolite, marble, and metachert--Taylor Mountain terrane STRUCTURALLY LOWER ROCKS (Yukon-Tanana terrane) Amphibolite-facies pelitic schist, quartzite, marble, and amphibolite--Chena River subterrane Amphibolite-facies augen gneiss, pelitic schist, amphibolite, and quartzite. Augen gneiss shown by pattern--Lake George subterrane

A A

Contact Strike-slip fault--Arrows show direction of relative movement. Dotted where concealed Thrust fault--Sawteeth on upper plate Low-angle normal fault--Sawteeth on upper plate; dotted where concealed Major rivers Major towns Figure 2. Simplified geologic map of part of Yukon-Tanana upland within Big Delta, Eagle, Tanacross, and northeastern Nabesna 1 o by 3° quadrangles (modified from Dusei-Bacon and others, in press; data sources given therein). Box outlines area of figure 3. Closed line with double ticks shows sillimanite isograd that outlines sillimanite gneiss dome referred to in text. High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

1987). These rocks are included in the Seventymile terrane in Alaska and are considered equivalent to the Slide Mountain terrane in Canada, which is composed of weakly metamorphosed Devonian to Triassic oceanic rocks and Permian arc-related rocks (Harms and others, 1984; Harms, 1985; Nelson and Bradford, 1987; Nelson and others, 1988). Most Seventymile terrane rocks are not penetratively deformed, and they structurally overlie rocks of the Nisutlin or Taylor Mountain terranes. These oceanic rocks generally have been metamorphosed under prehnite-pumpellyite- to lower greenschist-facies conditions, but at one locality, near the northern edge of the area shown in figure 2, they have been metamorphosed under blueschist-facies conditions (Foster and Keith, 1974; Dusel-Bacon and others, in press). Small, well-foliated bodies of ultramafic rocks that reach lower amphibolite-facies metamorphic grade crop out in a discontinuous belt across the southwestern part of the Lake George subterrane west of the study area (Weber and others, 1978). Only one of these bodies is extensive enough to show in figure 2 (near the southeastern edge of the large body of augen gneiss). Commonly associated with the Seventymile terrane, and structurally underlying it, is a sequence of greenschist-facies, quartz-rich clastic metasedimentary rocks and mafic and intermediate metavolcanic rocks. Common lithologies are quartz-chlorite-white mica schist, carbonaceous quartzite, calc-phyllite, mylonitic quartzofeldspathic schist, and marble. Protolith ages are unknown, but a Devonian and Mississippian age is likely on the basis of regional correlations (Foster and others, 1987) and a U-Pb age of 375 Ma on zircon from a metaandesite in the Big Delta quadrangle (J .N. Aleinikoff and W.J. Nokleberg, unpub. data, 1989, reported in DuselBacon and others, in press). Metamorphic cooling ages have not been determined for these rocks in Alaska, but at least part of this greenschist-facies sequence is thought to be correlative with the Klondike Schist in Canada, which has yielded Late Triassic to Early Jurassic metamorphic ages [175±14-Ma K-Ar age on muscovite (Tempelman-Kluit and Wanless, 1975) and 202±11-Ma Rb-Sr whole-rock age (Metcalfe and Clark, 1983)]. The greenschist-facies rocks of this generalized unit are assigned to the Nisutlin terrane as proposed by Hansen (1990) and Hansen and others ( 1991). In Canada, rare occurrences of blueschist and eclogite are associated with Nisutlin terrane rocks (Erdmer, 1987). Amphibolite-facies garnet amphibolite, biotite±hornblende±garnet gneiss and schist, marble, quartzite, metachert, and pelitic schist constitute the metamorphic rocks of the Taylor Mountain terrane. Geothermobarometric data presented in this paper and previously in abstract form (Dusel-Bacon and Douglass, 1990; DuselBacon and Hansen, 1991) indicate high-P amphibolitefacies conditions during metamorphism of this terrane. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Intrusive into these rocks are latest Triassic to Early Jurassic plutons of mostly granodioritic composition, including the Taylor Mountain batholith from which this terrane derives its name. Abundant marble, sparse manganiferous metachert, and initial common Pb ratios from plutonic rocks (Aleinikoff and others, 1987) suggest an oceanic or marginal-basin origin. Metamorphic cooling ages from 40 ArP9 Ar incremental heating experiments are latest Triassic and Early Jurassic (Hansen and others, 1991; fig. 3). Similarities in metamorphicprotolith assemblages, high-P metamorphic conditions, and metamorphic cooling ages of the metamorphic rocks of the Taylor Mountain terrane to those in southern Yukon Territory led Hansen to correlate these two areas and to refer to them collectively as the Teslin-Taylor Mountain terrane (Hansen, 1990) (fig. 1). Structurally lower rocks (fig. 2) consist of two similar amphibolite-facies subterranes of the Yukon-Tanana terrane. The Lake George subterrane is characterized by augen gneiss derived from dynamothermal metamorphism of a belt of Mississippian granitoids (Dusel-Bacon and Aleinikoff, 1985), as well as quartz-biotite schist and gneiss, pelitic schist, garnet amphibolite, and quartzite. A similar lithologic assemblage, albeit differing in the relative abundances of rock types (specifically, having very little augen gneiss and more quartzite) makes up the Chena River subterrane of the Yukon-Tanana terrane (terminology proposed in Nokleberg and others, 1989). Both of these subterranes, as well as other quartz-rich amphibolite- and greenschist-facies rocks north and northwest of the area shown in figure 2 (shown in fig. 1 as the Devonian to Mississippian augen gneiss and host sedimentary rocks of the Yukon-Tanana terrane), yield Early Cretaceous metamorphic cooling ages (Wilson and others, 1985). In Alaska, as well as in Yukon Territory, these rocks give Proterozoic U-Pb inheritance ages (Aleinikoff and others, 1984, 1986; Mortensen and Jilson, 1985; Mortensen, 1990) and SmNd model ages (Aleinikoff and others, 1981b; Bennett and Hansen, 1988) consistent with North American cratonal basement values. We interpret these lower structural-level rocks (Yukon-Tanana terrane in our restricted application of the name) to comprise a continental margin and granitoid belt built on (par)autochthonous North American basement as previously proposed by Hansen (1990). STRUCTURAL AND METAMORPHIC-COOLINGAGE SUBDIVISIONS IN THE AREA OF STUDY Our study area (fig. 3) was chosen so we could examine both the Taylor Mountain terrane and Lake George subterrane and the nature of the structural contact between the two. Our data, combined with those

from previous structural and 40 Arf39 Ar studies (Foster, 1970, 1976; Cushing and others, 1984; Cushing, 1984; Hansen and others, 1991 ), indicate two different structural and metamorphic-cooling-age zones within the Taylor Mountain terrane and two different structural zones within the Lake George subterrane. Within the northern zone of the Taylor Mountain terrane, shear-sense indicators record top-to-the-northeast tectonic displacement, and metamorphic cooling ages are Early Jurassic. The 40 Arf39 Ar hornblende plateau age (204±4 Ma) in this zone is approximately 10 m.y. older than the oldest hornblende plateau age in the southern zone of the Taylor Mountain terrane. At the same locality Figure 3. Taylor Mountain area showing location of samples and pressure-temperature estimates discussed in this report, direction of tectonic transport of upper plate rocks (arrows) determined by shear-sense indicators, and metamorphic or, in the case of the Taylor Mountain batholith, igneous cooling ages (Ma). Isotopic age abbreviations: m, b, and h indicate muscovite, biotite, and hornblende 40Arf39Ar incremental heating ages, respectively;* indicates data from Cushing and others (1984) and Cushing(1984); **indicates data from T.M. Harrison; unmarked 40Ar/39Ar data from Hansen and others from which the 204-Ma age was determined, biotite from an actinolite-biotite schist yields a 40 Arf39 Ar plateau age of 187±2 Ma. The dated actinolite-biotite schist occurs within a shear zone that is parallel to, and probably of similar origin to, the south-dipping thrust contact that placed the amphibolite-facies rocks of the Taylor Mountain terrane onto the greenschist-facies rocks of the Nisutlin terrane; hence, the metamorphic cooling age probably dates the juxtaposition of these two metamorphic packages (Cushing and others, 1984). A 186±2-Ma plateau age on muscovite from an undeformed, unmetamorphosed dike that crosscuts metamorphic foliation within this northern zone (Cushing and others, 1984; Cushing, 1984) establishes a SADL: 8.2-12.5 kbar, 625-750°C (1991 );s, U-PbspheneagefromAieinikoffandothers(1981 a); b, K-Ar, biotite K-Ar age from Nora Shew (written commun., 1985). Unit patterns are same as those shown for figure 2. Vertical-ruled overprint of Taylor Mountain terrane shows area of heterogeneous kinematics and unknown tectonictransportdirections in which azimuths of elongation lineations trend east-west, northeast-southwest, and northwest-southeast. East-west-trending line within Lake George subterrane (of Yukon-Tanana terrane) shows axis along which displacement changes from northwest- to southeast-vergent. High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

minimum metamorphic cooling age (to -300°C, muscovite blocking temperature) for this zone. The southern zone of the Taylor Mountain terrane (shown by a vertical-ruled overprint in fig. 3) is characterized by heterogeneous kinematics and metamorphic cooling ages. Only one shear-sense indicator has been determined thus far, but mineral stretching lineations are quite variable in azimuth. Metamorphic cooling ages in this zone straddle the boundary between the Early and Middle Jurassic epochs, with the exception of one older Early Jurassic age (194±2-Ma 40Arf39Ar plateau age on hornblende). Two 40 ArP9 Ar hornblende-biotite pairs indicate rapid cooling (100°C/m.y.) from 500°C (hornblende blocking temperature) at -187 Ma to 300°C (biotite blocking temperature) at -185 Ma (Hansen and others, 1991 ). We interpret the metamorphic cooling data to indicate rapid exhumation. We also define two apparent structural zones in the lower-plate Lake George subterrane to the south. North of an east-west-trending line shown on figure 3, shearsense indicators show top-to-the-northwest tectonic displacement, whereas south of the line, tectonic displacement is top-to-the-southeast. Along the line, tectonites locally record both top-to-the-northwest and top-to-the-southeast displacement. Preliminary data indicate that top-to-the-northwest displacement preceded top-to-the-southeast displacement and is associated with a more penetrative fabric. Metamorphic cooling ages from this area are Early Cretaceous (fig. 3), similar to ages elsewhere in the Lake George subterrane to the west. Comparison of 40 Arf39 Ar plateau ages on hornblende (119 Ma) and biotite (109 Ma) from nearby outcrops indicates more gradual cooling (20°C/m.y.) from -500°C to -300°C that we interpret to indicate moderately rapid exhumation. SAMPLE SELECTION AND ANALYTICAL PROCEDURES Garnet amphibolite and pelitic schist from both the Taylor Mountain terrane and the Lake George subterrane of the Yukon-Tanana terrane were evaluated for apparent textural equilibrium, presence of phases necessary to apply multiple geothermometers and geobarometers, and representative coverage of the various structural and metamorphic age domains. Five samples (three garnet amphibolites and two pelitic schists) were chosen for microprobe analysis. Samples are discussed from north to south; their locations are shown on figure 3. Major element analyses were obtained for garnet, biotite, hornblende, plagioclase, ilmenite, and muscovite using an ARL-SEMQ electron microprobe equipped with nine wavelength-dispersive spectrometers. Synthetic and natural mineral standards were analyzed before, during, Geologic Studies in Alaska by the U.S. Geological Survey, 1991 and after mineral analyses in order to monitor drift. Data were reduced using the corrections of Bence and Albee (1968) and Albee and Ray (1970). In order to increase the probability that the analyzed compositions are representative of equilibrium compositions at peak metamorphic conditions, the following analytical technique, suggested by Hodges and Spear ( 1982) was used. Two to four domains were identified in each polished section such that all minerals to be analyzed in a given domain were either in contact or less than 1 mm apart. Mineral rims were then analyzed at points of mutual contact. For each mineral grain, three to six closely spaced analyses were averaged to arrive at mean compositions at each spot. An accelerating voltage of 15 kV was used. With the exception of plagioclase, analyses of all phases were obtained from 1 0-s counting times with a sample current of 25 nA; counting durations on plagioclase were 20 s with an accelerating voltage of 20 nA. The mineral compositions reported in tables 1-4 are typical of the mineral composition in the rocks, although minor compositional variations occur. SAMPLE DESCRIPTION TH78 Garnet-biotite-amphibole gneiss was collected along the Taylor Highway within the northern zone of the Taylor Mountain terrane, in which tectonic displacement is top-to-the-northeast; asymmetric fabrics at this locality indicate N. 50° E. tectonic transport. Subhedral garnet porphyroblasts, 1.5 to 2 mm in diameter, occur in a wellfoliated and lineated matrix defined by tabular, 1- to 2mm-long porphyroblasts of reddish-brown biotite and medium-green (bluish- to yellowish-green to tan pleochroism) amphibole of ferroan pargasite composition [modified classification of Leake (1978), presented in Hawthorne (1981)]. Quartz and plagioclase occur as fine-grained (0.25-mm-long) mosaics; accessory phases are titanite, ilmenite, and rutile. TH608 Garnet-bioiite-amphibole gneiss also was collected along the Taylor Highway in the northern zone of the Taylor Mountain terrane, approximately I km south of sample 7B. Tectonites at this locality record N. 40° E. tectonic transport. This sample is similar in composition to sample 7B but contains slightly more biotite and amphibole and garnet porphyroblasts are larger (3-4 mm) in diameter. Biotite and amphibole porphyroblasts are 1.5 mm long; the former are reddish brown, and the latter show a light- to medium-bluish-green to tan pleochroism. Amphibole

Table 1. Representative analyses of garnet Taylor Mountain terrane Lake George subterrane Sample 7B-domain 2 60B-domain 2 liE-domain 3 42A-domain 2 SADL-domain 3 (amphibolite) (amphibolite) (metapelite) (amphibolilte) (metapelite) Si02 Ti02 Al20 3 FeO MnO MgO CaO Total 102.27 Formula normalized to 12 oxygens: s i 2.98 AllY AJYI 1.92 Ti Fe 1. 73 M n Mg Ca Pyrope Almandine S pessartine Grossular Table 2. Representative analyses of biotite Taylor Mountain terrane Sample 7B-domain 2 (amphibolite) 60B-domain 2 (amphibolite) Si02 3 5 .4 8 Ti02 Al20r 16.9 7 FeO 21 . 6 2 MnO MgO CaO N~O Lake George sub terrane 11 E-domain 3 SADL-domain 3 (metapelite) (metapelite) Total 9 5. 8 8 Formula normalized to 11 oxygens: s i 2.73 AI IV 1.27 AI VI Ti Fe 1.39 M n M g 1.04 Ca Na K Mg/(Mg+Fe) High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

Table 3. Representative analyses of hornblende Taylor Mountain terrane Sample 7B-domain 2 (amphibolite) Si02 39.63 Al203 15.23 Ti02 MgO FeO 21.74 MnO CaO 11.13 N~O K2 0 Total 97.96 60B-domain 2 (amphibolite) Formula normalized to 23 oxygens: s i Al1v Alv' Ti *Fe3+ Mg Fe2+ M n Ca N a(M 4) Na(A) K Fe2+/(Mg + Fe2+) -- Lake George subterrane 42A-domain 2 (amphibolite) *Fe3+ calculated according to the method of Spear and Kimbal1 (1984). Table 4. Representative analyses of plagioclase Ta~lor Mountain terrane Sample 7B-domain 2 60B-domain 2 Si02 A120 3 CaO N~O K20 Total 101.13 Formula normalized to 8 oxygens: s i AI Ca Na K An 21.1 A b 80.2 Or Geologic Studies in Alaska by the U.S. Geological Survey, 1991 composition is ferroan tschermakite. Quartz and plagioclase are fine grained (0.25 mm long) and less abundant than in sample 7B. Ilmenite and trace amounts of rutile are accessory phases. TH11E Staurolite-kyanite-garnet-quartz mica schist also was collected along the Taylor Highway but from the southern, mixed kinematic zone of the Taylor Mountain terrane; lineation has an azimuth of N. 70° E. and plunges moderately to the northeast or southwest. Garnet porphyroblasts (1-2 mm in diameter) occur in a compositionally layered matrix composed of folia of intergrown and equally abundant 1-mm-long red-brown biotite and white mica that alternate with folia of 0.5- to 1-mm-long subhedral quartz and much less abundant plagioclase. Porphyroblasts of euhedral to subhedral kyanite (1-2 mm long) and subhedral staurolite (0.5 mm long) are present in minor amounts, and ilmenite, tourmaline, zircon, and rutile occur in trace amounts. Very minor alteration of biotite or garnet to chlorite occurs in a few areas of the thin section. 83ADb42A Garnet amphibolite was collected within an area of the Lake George subterrane in which augen gneiss is the predominant rock type. This sample was collected from the zone in which kinematic indicators record top-to-theLake George subterrane liE-domain 3 42A-domain 2 SADL-domain 3

northwest tectonic transport. Euhedral to subhedral garnet porphyroblasts (1 mm in diameter) lie in a well-foliated matrix of 0.5-mm-long grains of medium-bluish-green (to yellowish-tan) amphibole (ferroan pargasite ), quartz, and plagioclase; there is no apparent compositional layering. Ilmenite, rutile, and titanite are accessory phases. SADL Staurolite-garnet-quartz mica schist was also collected within the augen gneiss-bearing part of the Lake George subterrane, but about 7 km to the southwest of sample 42A and near the boundary between northwestand southeast-vergent fabrics; lineation at this locality is essentially flat lying, and nearby outcrops record S. 25° E. tectonic transport. Subhedral to euhedral porphyroblasts of garnet (1.5-4 mm in diameter) and staurolite (1-2 mm long) are enclosed in a matrix composed of folia of redbrown biotite and white mica (equal amounts of 1-1.5mm-long grains) alternating with similar-sized grains of quartz and minor plagioclase. Ilmenite, rutile, tourmaline, and zircon are present in trace amounts. GEOTHERMOMETRY AND GEOBAROMETRY Peak metamorphic temperatures and pressures corresponding to peak temperatures were determined by geothermometry and geobarometry using compositions determined by microprobe analysis of garnet rims and adjacent matrix phases as described above. Garnet zoning profiles (Cynthia Dusel-Bacon and Neena Bashir, unpub. data, 1991) show a core-to-rim decrease in Mn and an increase in Mg indicative of growth zoning. The profiles support our interpretation that garnet-rim compositions record maximum temperatures. It has not been possible to determine P-T conditions during early garnet growth because the only inclusions observed within garnet have been quartz and ilmenite. Temperatures of final equilibration of the samples were calculated using the garnet-biotite (GT -BT) geothermometer (Hodges and Spear, 1982) and the garnet-hornblende (GT-HB) geothermometer (Graham and Powell, 1984). Pressures of final equilibration were derived using the garnet-plagioclase-hornblende-quartz (GPHQ) geobarometer for both tschermakite (TSCH) and its Fe end-member equivalent (FE TSCH) (Kohn and Spear, 1990); the garnet-rutile-ilmenite-plagioclasequartz (GRIPS) geobarometer (Bohlen and Liotta, 1986); the garnet-rutile-Al2Si05 (kyanite)-ilmenite (GRAIL) geobarometer (Bohlen and others, 1983); the garnetAl2Si05 (kyanite)-plagioclase-quartz (GASP) geobarometer (Koziol and Newton, 1988); and the garnet-plagioclase-muscovite-biotite (GPMB) geobarometer (Hodges and Crowley, 1985). Our calculations used the garnet activity model of Hodges and Spear (1982). Fe3+ in amphibole (table 3) was calculated to a minimum value consistent with amphibole stoichiometry using the method of Spear and Kimball (1984 ). Results of the geothermobarometry for all the domains of the five samples are plotted in figure 4. The Al2Si05 phase boundaries of Holdaway ( 1971) are shown for reference; the overlap in P-T results for all samples occurs in the kyanite field. Taken at face value, the following range of pressures and temperatures are indicated for each sample: Taylor Mountain terrane: garnet amphibolite, 10 to 12.5 kbar, 635oC to 685oC and 7 to 10 kbar, 560oC to 645oC; staurolite-kyanite-garnetquartz-mica schist, 6 to 7 kbar, 560oC to 605oC; Lake George subterrane: garnet amphibolite, 8.7 to 12.5 kbar, 600oC to 730oC; garnet-staurolite-quartz-mica schist, 8.2 to 12.5 kbar, 625°C to 750°C. In general, there is close agreement between the results of the various geobarometers applied to the same sample and of the same geobarometer applied to mineral suites from different domains. Agreement is poorer between the results of the GT -BT and GT -HB geothermometers where applied to the same sample and between the results of the same geothermometer applied to mineral pairs from different domains within the same sample. For the two samples in which both GT-HB and GT-BT temperatures were determined, GT-HB temperatures are higher. Generally, the highest GT-HB temperatures correspond to the hornblende with the highest Fe# (Fe2+fFe2++Mg) value. Graham and Powell (1984) warn that significant overestimates of temperature may arise in using their GT-HB geothermometer if the hornblende compositions are unusually rich in Fe3+. Potential problems related to the ferric iron content of our hornblendes could be minimized by the fact that estimates of the Fe3+ contents based only on stoichiometry tend to produce minimum values. The empirical experience of M. J. Kohn (written commun., 1990) suggests that when the Fe# is above 0.6 or 0.65, GT-HB temperatures become progressively higher than those indicated by other geothermometers and are inconsistent with phase equilibria. This is the case with the two GT-HB temperatures for sample 7B (Fe#' s=0.67 and 0.64) and the three highest GT -HB temperatures for sample 42A (Fe#' s=0.63 and 0.61 ). Additional problems with accurate temperature estimates for sample 7B come from the fact that the mole fraction of Mn in garnets (0.3) used in the application of the GT-HB geothermometer exceed the compositional range for garnet for which the calibration is recommended (XMn<O.l for GT-HB; Graham and Powell, 1984 ). If the three highest temperatures for sample 42A are ignored because of their high Fe#' s and the lower three temperatures ranging from 600-625 °C High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

en a:

w a:

en en w a: a. en a:

al

w a:

en en w a: a. en a:

w a:

en en w a: a. 3 U~--L-L TEMPERATURE, IN DEGREES CELSIUS TEMPERATURE, IN DEGREES CELSIUS TEMPERATURE, IN DEGREES CELSIUS Figure 4. Pressure-temperature (P-n diagrams showing results of geothermobarometric calculations for samples analyzed in this study. P-Testimates for each sample discussed in text are based on area of overlap of Keq (equilibrium constant) lines. Each line represents a different Keq based either on same geothermometer or geobarometer applied to multiple domains in each sample or on different geothermometers and Geologic Studies in Alaska by the U.S. Geological Survey, 1991 TEMPERATURE, IN DEGREES CELSIUS TEMPERATURE, IN DEGREES CELSIUS EXPLANATION GRIPS GASP GRAIL GPMB GPHQ (FE TSCH) GPHQ (TSCH) GT-BT GT-HB geobarometers applied to various domains; abbreviations for geothermometers and geobarometers explained in text. Heavy lines show AI 2Si05 phase diagram of Holdaway (1971 ). Abbreviations: k, kyanite; s, sillimanite, a, andalusite. A, Sample 78. 8, Sample 608. C, Sample 11 E. D, Sample 42A. E, Sample SADL. See figure 3 for location of samples.

(for which Fe#'s=0.51-0.53) are considered only, there is better correspondence between the pressures determined by the GRIPS and GPHQ geobarometers. Although there is a considerable range in calculated temperatures for most samples, the relatively flat slopes of the equilibrium constant (Keq) lines for the various equilibria on which the geobarometers are based cause the uncertainty in temperature to have little effect on overall pressure estimates. A realistic estimate of -600°C to 650°C for maximum metamorphic temperatures is suggested by the preservation of garnet zoning profiles, indicating that temperatures were not high enough for modification of chemical zonation by diffusion during cooling from peak temperatures, and by the absence of evidence for partial melting in Taylor Mountain terrane and Lake George subterrane tectonites. Another argument that suggests that maximum metamorphic temperatures were in the range of -600±50°C is that this is the approximate range of temperatures over which different geobarometers for the same sample intersect (fig. 4A-E). The simultaneous solution of the various equilibria used as geothermometers and geobarometers · in this study (represented by the areas of overlap of the Keq lines for each sample) is shown in figure 5. The temperature and pressure ranges shown in figure 4 have been expanded in figure 5 by ±30°C and 0.5 kbar, respectively. Although absolute uncertainties in the calculated P-T conditions have been suggested to be as high as ±50°C and ± 1 to 2 kbar for each of the calibrations, relative uncertainties are probably significantly smaller (Hodges and McKenna, 1987). Therefore, P-T conditions that were obtained using the same geothermobarometers can be compared. As mentioned above, the upper temperature ranges allowed by the uncertainty in temperature estimates are shown to extend well into the range of partial melting (fig. 5), a relation not seen in the field. The consistency in estimated P-T conditions for pelitic schist (SADL) and garnet amphibolite ( 42A) of the Lake George subterrane indicates that geothermobarometers using different bulk compositions can give the same result. This observation supports the COMPOSITE PRESSURE-TEMPERATURE FIELDS Cf) a:

a:l

S2

w a: :J Cf) Cf) Cf) w a: a.. 3 L TEMPERATURE, IN DEGREES CELSIUS Figure 5. Estimated pressure-temperature (P-n fields for samples 78, 608, 11 E, 42A, and SADL, expanded to allow for temperature uncertainty of 30°C and pressure uncertainty of 0.5 kbar. Dashed lines show initial melting of wet and dry pelite (lower and higher temperature lines, respectively; Thompson and Tracy, 1979). Abbreviations: LGS, LakeGeorgesubterraneofthe Yukon-Tanana terrane (P- Tfields shown in gray); N. TMT, northern zone of the Taylor Mountain terrane (P- Tfields shown in diagonal ruled pattern); S. TMT, southern zone of Taylor Mountain terrane (P- Tfield shown in stippled pattern). High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

real difference in P-T conditions indicated by our geothermobarometry of pelitic schist (liE) and garnet amphibolite (7B and 60B) from the Taylor Mountain terrane. We are presently investigating possible explanations for the differences in pressures determined for the compositionally similar samples 7B and 60B. Our current hypothesis is that the calculated pressure differences are valid, for if these tectonites were indeed metamorphosed within a subduction-zone environment as we propose, one can envision displacement paths (within the subduction channel; Cloos and Shreve, 1988) that could juxtapose tectonites of different peak P-T conditions (for example, Cloos, 1985; Hansen, 1992a). DISCUSSION High-Pressure Metamorphism in the Eastern Yukon-Tanana Upland The P-T conditions indicated by geothermobarometry reported above are consistent with high-P, moderate-T metamorphism and penetrative ductile deformation within a subduction-zone setting. Kinematic and thermochronometric (primarily 40 Ar/39 Ar incremental-heating ages) differences between Taylor Mountain terrane and Lake George subterrane rocks in the area of our study suggest that although similar P-T conditions (8-12 kbar) were determined for the northern zone of the Taylor Mountain terrane and the Lake George subterrane to the south, the timing and tectonic setting of metamorphism in tl_lese two areas differed. Ongoing P-T studies should help elucidate the tectonic model. In the northern zone of the Taylor Mountain terrane, postmetamorphic and postkinematic cooling to about 500°C (hornblende-blocking temperature) took place as early as 204±4 Ma and deformation was top-to-thenortheast, perpendicular to regional strike. Similar highP, moderate-T metamorphic conditions, orogen-normal tectonic transport, and Early Jurassic metamorphic cooling ages have been documented in correlative rocks in the Teslin suture zone of central Yukon (fig. 1) (Hansen, 1989, 1992b; Hansen and others, 1991 ). Considering our data in conjunction with those from the structurally and metamorphically similar rocks in the Teslin suture zone, we conclude that metamorphism and deformation within the northern zone of the Taylor Mountain terrane took place within the deep-seated part of a southwestdipping (present-day coordinates) subduction zone outboard of western North America. A Permian Rb-Sr muscovite cooling age on blueschist-facies rocks associated with eclogite north of the town of Ross River in central Yukon (fig. 1) places a minimum age on initiation of subduction (Erdmer and Armstrong, 1988). Geologic Studies in Alaska by the U.S. Geological Survey, 1991 The single, and significantly lower (6-7 kbar), pressure determined for the staurolite-kyanite-garnet-quartzmica schist from the southern zone of the Taylor Mountain terrane is more difficult to interpret. Rapid cooling and exhumation of this structural zone at about 188 to 186 Ma is indicated by the hornblende and mica 40 Arf39 Ar cooling ages. We interpret the cooling and exhumation to have resulted from collision and obduction of these rocks onto the margin of western North America, including the Lake George subterrane and other associated lower-plate rocks of the YukonTanana terrane. This interpretation of ages, presented previously by Hansen and others (1991 ), is supported by the 187±2-Ma 40 Arf39 Ar muscovite cooling age from the shear zone that is parallel to the thrust fault along which the Taylor Mountain terrane was emplaced above the greenschist-facies rocks of the Nisutlin terrane during collision. Kinematic analysis from the southern zone of the Taylor Mountain terrane has not been completed, but the presence of northeast-, northwest-, and east-westtrending elongation lineations suggests a complex structural history. Our tentative interpretation of the. structural data is that the rocks in this package, being closer to the plate boundary between the hanging wall (Taylor Mountain terrane) and the footwall (Lake George sub terrane), were affected both by precollision deformation (northeast-directed tectonic transport) that we assume was dominant for the rocks of the northern structural zone and by syn-collision deformation (perhaps northwest-directed tectonic transport, as discussed below). The east-west-trending lineation also observed within this southern zone, and at the locality from which this P-T determination was made, may be analogous to the lineation developed within a zone of dextral strikeslip deformation within the Teslin suture zone. Metamorphic pressures from within the dextral strike-slip zone were significantly lower (5-8 kbar) than elsewhere within the Taylor Mountain terrane-equivalent rocks of the Teslin suture zone (12±4 kbar) (Hansen, 1992b). Our lower P determination from the southern Taylor Mountain terrane zone, relative to the two determinations from the northern package, may record either a later and shallower phase of accretion or dextral transpression that was shown to be an important process during the early Mesozoic accretion of Taylor Mountain terrane-equivalent rocks in southern Yukon (Hansen, 1989), or pressure that accompanied underthrusting of the Lake George subterrane, as discussed below. High-P metamorphism of the Lake George subterrane is most reasonably explained by subduction of continental crust beneath the overriding, accreted rocks of the Taylor Mountain, Seventymile-Slide Mountain, and (probably) Nisutlin terranes. Rapid cooling and exhumation of the upper-plate Taylor Mountain terrane from about 500°C to 300°C around 186 Ma establishes a

likely age for initiation of tectonic burial of the lower plate. Crustal thickening of the Lake George subterrane undoubtedly would have continued after collision and initial obduction of upper-plate rocks, since continued contraction would have resulted in imbrication and folding of the lower plate. We interpret the Early Cretaceous metamorphic cooling ages in the eastern Lake George subterrane, and probably in much of the lowerplate Yukon-Tanana terrane as well, to reflect cooling and exhumation of these rocks as a result of crustal extension following obduction of the overlying terranes, as previously proposed by Hansen and others (1991). As mentioned above, comparison of hornblende and biotite 40 Arf39 Ar ages (119 and 109 Ma, respectively) from nearby localities in the eastern Lake George subterrane (fig. 3) indicates a cooling rate of 20°C/m.y., which, in turn suggests moderately rapid exhumation. The Lake George subterrane was apparently at the surface by -93 Ma because it forms the country rock into which calderas of that age (shown on fig. 2 as the area of unmetamorphosed rocks north of Tok) were emplaced (Bacon and others, 1990). Kinemati_c data from the eastern Lake George subterrane can be interpreted in two ways. In our first, and preferred, interpretation, the older, top-to-the-northwest, displacement records deformation during contraction accompanying accretion of obducted terranes and imbrication within lower-plate rocks, and the younger, top-to-the-southeast, deformation records shallower, subsequent extension. This interpretation is supported by a northwest-vergent shear sense recorded in quartzose rocks at the leading edge of the Yukon-Tanana terrane north of Fairbanks (fig. I) during contraction, and by an east-southeast-directed transport direction measured in greenschist-facies mylonites at the base of the hanging wall above a sillimanite gneiss dome (fig. 2) that was exposed during major crustal extension (Pavlis and others, 1988, in press). In our second interpretation, both the top-to-the-northwest and top-to-the-southeast displacement occurred synchronously during extension. However, if upper-plate rocks were shed both to the northwest and to the southeast at the same time, one might expect a broad 10 km) transitional zone dominated by coaxial shear deformation between the domains that record opposing ductile shear direction (for example, 10-20 km wide in the Albion, Grouse Creek, and Raft River Mountains, Idaho; Malavieille, 1987). However, the east-west-trending line (fig. 3) that separates opposing ductile shear fabrics in the study area is extremely sharp [in fact, opposing shear sense is recorded within a single outcrop at one locality (fig. 3, east of sample 42A)], and we also see no indication of coaxial deformation. In addition, kinematic analysis reveals that microstructures that record top-to-the-southeast shear consistently crosscut top-to-the-northwest microstructures. Evidence for High-Pressure Metamorphism, Related to Jurassic Accretion, in the Western Yukon-Tanana Upland Although our geothermobarometric and kinematic data are from the eastern part of the Yukon-Tanana upland in Alaska, several lines of evidence suggest that Early to Middle Jurassic obduction of the outboard Seventymile-Slide Mountain and Taylor Mountain terranes onto the lower-plate Lake George subterrane was also a major tectonic and metamorphic event in the western Yukon-Tanana upland. (1) Klippen of ultramafic and mafic Seventymile-Slide Mountain terrane rocks extend -230 km westward into the center of the Big Delta quadrangle, attesting to the widespread original distribution of these oceanic rocks. (2) Rocks from an eclogitebearing klippe of the Chatinika terrane northeast of Fairbanks (fig. 1) yield Middle Jurassic muscovite K-Ar cooling ages (Wilson and Shew, 1981 ). (3) The high-P assemblage garnet-hornblende-kyani te-s tauroli te occurs in schist (C. Dusel-Bacon and V.L. Hansen, unpub. data, 1991) within an area of amphibolite-facies mafic rocks that yield Early to Middle Jurassic hornblende K-Ar and 40A f39A r r coo mg ages [188±6 Ma (Wtlson and others, 1985) and 181±3 Ma (M.A. Lanphere and C. Dusel-Bacon, unpub. data, 1991), respectively], in contrast to the Early Cretaceous metamorphic cooling ages that are typical of the Lake George subterrane. We tentatively interpret this area of mafic rocks (shown on fig. 2 to occur within the large area of augen gneiss west of our area of study) as a klippe of the Taylor Mountain terrane above the Lake George subterrane. Our contention that these rocks record high-P conditions is based on P-T studies of kyanite+staurolite+hornblende amphibolites in other high-P metamorphic belts of the world, which indicate metamorphic pressures of 9 to 10 kbar (Selverstone and others, 1984, p. 514). High-P -metamorphism related to this same episode of plate convergence has been well documented in the Yukon by the high-P rocks in the Teslin suture zone (Hansen, 1989, 1992b; Hansen and others, 1991) and by isolated occurrences of eclogite (Erdmer and Helmstaedt, 1983) (fig. 1). TECTONIC MODEL The P-T estimates and kinematic and age data presented in this paper are consistent with a tectonic model proposed by Hansen (1990) and Hansen and others (1991) that is based to a large degree on the initial hypothesis of arc-continent collision made by Tempelman-Kluit (1979). A simplified two-dimensional version of this model is shown in figure 6. For clarity, no translational component High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

is shown, but we do not propose that movement was exactly normal to the continental margin. Beginning in Devonian to Mississippian time, rifting commenced along the western margin of North America and a basin was formed. At about this same time, or perhaps before and possibly related to rifting, peraluminous Devonian and Mississippian granitoids intruded the continental crust. Rifting continued until Permian time, resulting in the development of the Anvil Ocean (a basin of unknown width, floored by oceanic crust) (fig. 6A). In this model, the peraluminous plutons shown northeast of the Anvil Ocean are the protoliths of the augen gneiss of the Lake George subternine. We speculate that the rifted fragment of North American crust is the Nisling terrane, which lies outboard of the accreted terranes (fig. 1 ), and which is also intruded by Devonian and Mississippian peraluminous plutons (Gareau, 1989; McClelland and others, 1992). By Late Permian time, southwest-dipping, rightoblique subduction (as indicated by precollision dextral shear zones within the Teslin suture zone; Hansen, 1989) of the Anvil Ocean lithosphere had begun (fig. 6B). This minimum age for initiation of subduction is provided by Rb-Sr and K-Ar age determinations of approximately 250 Ma for blueschist associated with eclogite near the town of Ross River, Yukon Territory (fig. 1). An arc, which includes the Triassic to Jurassic granitoids (including the Taylor Mountain batholith of the Taylor Mountain terrane) that characterize the Stikinia terrane, developed above the subduction zone. Multiple lines of evidence (summarized in Hansen and others, 1991, p. 72) suggest a composite oceanic/continental [Nisling(?) terrane] basement for the Stikinia arc, as we show in figure 6. According to this model, protoliths of the Taylor Mountain and Nisutlin terranes, the terranes that compose the intermediate and lower levels of the subduction complex associated with the Stikinia arc, include sedimentary clastic rocks derived from erosion of crustal and arc-related rocks, intermediate and siliceous volcanic rocks associated with arc activity, and sedimentary rocks deposited within the marginal basin. We propose that meta~orphism of the Taylor Mountain terrane took place within the deep-seated part of the subduction zone where temperatures were highest. Nisutlin terrane rocks were metamorphosed and deformed at lower temperatures and, locally, at high pressures (only documented in Yukon; Hansen and others, 1991 ). The Seventymile terrane, the tectonic mixture of oceanic sedimentary rocks and crust, was only weakly metamorphosed in most areas. In Early to Middle Jurassic time ( -187 Ma), the outboard terranes were obducted onto North America (fig. 6C). Most of the obducted oceanic lithosphere (ultramafic rocks) of the Seventymile-Slide Mountain terrane were not penetratively deformed (represented by righthand area of oceanic crust; fig. 6C). We propose that a Geologic Studies in Alaska by the U.S. Geological Survey, 1991 minor amount of the oceanic lithosphere from deeper within the subduction zone (middle sliver of oceanic lithosphere above the deformed augen gneiss bodies in fig. 6C) was ductilely deformed, and that the belt of outcrops of foliated ultramafic rocks in the southwestern part of the Lake George subterrane (west of our study area and described earlier in the paper) formed in this tectonic setting. Continued contraction resulted in further underthrusting of the continental margin. High-P metamorphism and ductile deformation occurred in the lower plate (Lake George subterrane) as a result of crustal thickening caused by a combination of obduction of outboard terranes and imbrication within the lower plate. Subsequent development of a north- to northeastdipping subduction system farther outboard was probably a major factor in widespread crustal extension that unroofed the Lake George subterrane and related basement rocks (Pavlis, 1989; Pavlis and others, in press)(fig. 6D). Extension resulted from gravitational collapse of overthickened crust, slab rollback and trench retreat, or a combination of these and other factors (Dewey, 1988; Pavlis and others, in press). Kinematic data in the western part of the Lake George subterrane indicate that the mylonites that form a partial sheath enveloping a domal structure in the footwall (sillimanite gneiss dome of the Lake George subterrane; fig. 2) record a uniform east-southeast transport direction of the hanging wall . [greenschist-facies rocks of the Nisutlin terrane as defined by Hansen, 1990] (Pavlis and others, 1988, in press). Juxtaposition of these high- and lowgrade rocks suggests removal of as much as 10 km of crustal section (Pavlis and others, in press). Decompression within the lower-plate gneiss dome also is indicated by the occurrence of andalusite-bearing quartz veins (Sisson and others, 1990), andalusite-bearing partial melts developed within the core of the gneiss dome (Dusel-Bacon and Foster, 1983), and the growth of texturally late andalusite that postdates kyanite growth in pelitic schist that flanks the gneiss dome (Sisson and others, 1990). A clastic wedge (Hauterivian to Cenomanian in age) that could represent the eroded material, stripped away during the postulated period of crustal extension, was shed to the east and deposited within the MacKenzie fold and thrust belt (Pavlis and others, in press). In the eastern Lake George subterrane, the primary evidence for crustal extension is the juxtaposition of Taylor Mountain terrane tectonites with Early to Middle Jurassic cooling ages above Lake George subterrane tectonites with Early Cretaceous cooling ages, a relation best explained by the stripping away of upper-plate Taylor Mountain terrane along low-angle normal faults, resulting in exhumation and cooling of the lower-plate Lake George subterrane. This relation, as well as the

SOUTHWEST A. Devonian to Permian B. Permian and Triassic C. Early to Middle Jurassic D. Early to mid-Cretaceous EXPLANATION Seventymile-Siide Mountain terrane sedimentary and volcanic rocks Seventymile-Siide Mountain terrane oceanic crust Nisutlin terrane Taylor Mountain terrane Stikinia terrane ANVIL OCEAN Lake George subterrane (of Yukon-Tanana terrane) Devonian and Mississipian peraluminous plutons North American crust Cretaceous plutons NORTHEAST Figure 6. Simplified tectonic model for Devonian to mid-Cretaceous evolution of accreted terranes and North American crust in east-central Alaska. Modified from Hansen and others (1991 ). See text for explanation, and Hansen and others (1991) for a more complete explanation of some of the data used in developing this model. High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

downdropping of the northern belt of unfoliated oceanic crust and the southern belt of foliated oceanic crust (shown by the right-hand and left-hand klippen of oceanic crust, respectively), is schematically displayed in figure 6D. As shown in figure 6D, we speculate that mid-Cretaceous plutonism was probably associated with crustal extension, perhaps softening and further weakening the crust and thus facilitating the development of ductile shear zones such as those that we have observed in extremely attenuated rocks near the Lake George subterraneffaylor Mountain terrane contact zone in the area of our study. The main period of plutonism (10590 Ma; Wilson and others, 1985) postdates the ductile deformation but may have been related to melting facilitated by decreasing pressure during the later stages of extension-related uplift. CONCLUSIONS 1. Geothermobarometric analysis of garnet rims and adjacent phases in garnet amphibolite and pelitic schist from the structurally higher Taylor Mountain terrane and structurally lower Lake George subterrane indicate highP, moderate-T metamorphism in both tectonic units. P-T conditions were 7 to 12.5 kbar at -560°C to 650°C within the northern structural zone of the Taylor Mountain terrane in which displacement is consistently top-to the-northeast, and 6 to 7 kbar at 560°C to 605°C within the southern structural zone in which lineations and shear-sense directions are variable. Tectonites within the Lake George subterrane crystallized under conditions of 8.2 to 12.5 kbar at -600°C to 650°C. Preservation of growth zoning in garnet and absence of partial melting in both terranes suggest that temperatures did not exceed 600°C to 650°C. Within the Lake George subterrane, tectonites record dominantly top-to-the-northwest displacement in a northern zone and top-to-the-southeast displacement within a southern zone. Where the two shear-sense directions occur together, top-to-the-northwest displacement preceded top-to-the-southeast displacement and was more penetrative. 2. We interpret metamorphic and kinematic data to indicate that metamorphism of the Taylor Mountain terrane and Lake George subterrane took place during different phases of late Paleozoic to Jurassic contraction, resulting from closure of an ocean basin now represented by klippen of the Seventymile-Slide Mountain terrane. High-P metamorphism of the Taylor Mountain terrane took place within a southwest-dipping subduction system. 40 Arf39 Ar data indicate rapid cooling of the Taylor Mountain terrane at -186 Ma, our interpreted age of accretion of the outboard terranes onto the leading edge of North America (Lake George subterrane). High-P meta156 Geologic Studies in Alaska by the U.S. Geological Survey, 1991 morphism of the Lake George subterrane resulted from crustal thickening caused by overthrusting of the Seventymile-Slide Mountain, Taylor Mountain, and (probably) Nisutlin terranes and by imbrication of the continental margin during continued convergence. Subsequent development of a northeast-dipping subduction system was probably a major factor in widespread crustal extension that unroofed the Lake George subterrane basement in Early Cretaceous time. 3. Our working hypothesis to explain the kinematic data that we have obtained thus far is that the northeastvergent fabric in the northern zone of the Taylor Mountain terrane formed prior to final accretion of the terrane, that the northwest-vergent fabrics in the Lake George subterrane record subduction of the lower-plate continental crust, and that the southeast-vergent fabrics in the Lake George subterrane are extension related. Several lines of evidence suggest that high-P, moderate-T metamorphism resulting from obduction of the outboard Seventymile-Slide Mountain, Taylor Mountain, and (probably) Nisutlin terranes onto the lower-plate Lake George subterrane was also a major event in the western and northwestern parts of the Yukon-Tanana upland. Acknowledgments.-We are especially grateful to Susan Douglass and Charles Bacon for invaluable assistance in the field and microprobe analyses. We also wish to express our respect and admiration for the detailed and insightful reconnaissance mapping of the region by Helen Foster and the unselfish sharing of her knowledge of the area with us. Acknowledgment is also made to Kari Cooper for her artful and rapid preparation of the illustrations. REFERENCES CITED Albee, A.L., and Ray, L., 1970, Correction factors for electron microprobe microanalysis of silicates, oxides, carbonates, phosphates and sulfates: Analytical Chemistry, v. 42, p. 1408-1414. Aleinikoff, J.N., Dusel-Bacon, Cynthia, and Foster, H.L., 1981a, Geochronologic studies in the Yukon-Tanana Upland, east-central Alaska, in Albert, N.R.D., and Hudson, Travis, eds., The United States Geological Survey in Alaska: Accomplishments during 1979: U.S. Geological Survey Circular 823-B, p. 34-37. ---1986, Geochronology of augen gneiss and related rocks, Yukon-Tanana terrane, east-central Alaska: Geological Society of America Bulletin, v. 97, p. 626-637. Aleinikoff, J.N., Dusel-Bacon, Cynthia, Foster, H.L., and Futa, K., 1981b, Proterozoic zircon from augen gneiss, YukonTanana upland, east-central Alaska: Geology, v. 9, p. Aleinikoff, J .N., Dusel-Bacon, Cynthia, Foster, H.L., and Nokleberg, W .J ., 1987, Lead isotopic fingerprinting of tectono-stratigraphic terranes, east-central Alaska: Cana-

dian Journal of Earth Sciences, v. 24, p. 2089-2098. Aleinikoff, J.N., Foster, H.L., Nokleberg, W.J., and Dusel-Bacon, Cynthia, 1984, Isotopic evidence from detrital zircons for early Proterozoic crustal material, east-central Alaska, in Coonrad, W.L., and Elliott, R.L., eds., The United States Geological Survey in Alaska: Accomplishments during 1981: U.S. Geological Survey Circular 868, p. 4345. Bacon, C.R., Foster, H.L., and Smith, J.G., 1990, Rhyolitic calderas of the Yukon-Tanana Terrane, east central Alaska: Volcanic remnants of a mid-Cretaceous magmatic arc: Journal of Geophysical Research, v. 95, p. 21,45121,461. Bence, A.E., and Albee, A.L., 1968, Empirical correction factors for the electron microanalysis of silicates and oxides: Journal of Geology, v. 76, p. 382-403. Bennett, V .C., and Hansen, V .L., 1988, Neodymium isotopic similarities between the Yukon-Tanana terrane, Yukon Territory and continental North America [abs.]: Geological Society of America Abstracts with Programs, v. 20, no. 7, p. A111. Bohlen, S.R., and Liotta, J.J., 1986, A barometer of garnet amphibolites and garnet granulites: Journal of Petrology, v. 27, p. 1025-1034. Bohlen, S.R., Wall, V.J., and Boettcher, A.L., 1983, Experimental investigations and geological applications of equilibria in the system FeO-Ti02-AI20 3-Si02-H20: American Mineralogist, v. 68, p. 1049-1058. Brown, E.H., and Forbes, R.B ., 1986, Phase petrology of eclogitic rocks in the Fairbanks district, Alaska, in Evans, B.W., and Brown, E.H., eds., Blueschists and eclogites: Geological Society of America Memoir 164, p. 155-167. Churkin, M., Jr., Foster, H.L., Chapman, R.M., and Weber, F.R., 1982, Terranes and suture zones in east-central Alaska: Journal of Geophysical Research, v. 87, no. B5, p. 3718-3730. Cloos, Mark, 1985, Thermal evolution of convergent plate margins: Thermal modeling and reevaluation of isotopic Ar-ages for blueschists in the Franciscan Complex of California: Tectonics, v. 4, p. 421-433. Cloos, Mark, and Shreve, R.L., 1988, Subduction-channel model of prism accretion, melange formation, sediment subduction, and subduction erosion at convergent plate margins: 1. Background and description: Pure and Applied Geophysics, v. 128, p. 454-489. Cushing, G. W ., 1984, The tectonic evolution of the eastern Yukon-Tanana upland, Alaska: Albany, N.Y., State University of New York, M.S. thesis, 255 p. Cushing, G.W., Foster, H.L., Harrison, T.M., and Laird, Jo, 1984, Possible Mesozoic accretion in the eastern YukonTanana upland, Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 16, no. 6, p. 481. Dewey, J.F., 1988, Extensional collapse of orogens: Tectonics, v. 7, p. 1123-1139. Dusel-Bacon, Cynthia, and Aleinikoff, J.N., 1985, Petrology and tectonic significance of augen gneiss from a belt of Mississippian granitoids in the Yukon-Tanana terrane, east-central Alaska: Geological Society of America Bulletin,v. 96,no.4,p.411-425. Dusel-Bacon, Cynthia, Csejtey, Bela, Foster, H.L., Doyle, E.O., Nokleberg, W.J., and Plafker, George, in press, Distribution, facies, ages, and proposed tectonic associations of regionally metamorphosed rocks in east- and south-central Alaska: U.S. Geological Survey Professional Paper 1497-C, 100 ms p., 2 pis. Dusel-Bacon, Cynthia, and Douglass, S.L., 1990, New thermobarometric evidence for high-pressure, mediumtemperature metamorphism of two subterranes of the Yukon-Tanana composite terrane (YTT) in easternmost Alaska [abs.]: Geological Association of Canada Programs with Abstracts, v. 15, p. A35. Duse1-Bacon, Cynthia, and Foster, H.L., 1983, A sillimanite gneiss dome in the Yukon crystalline terrane, east-central Alaska: Petrography and garnet-biotite geothermometry: U.S. Geological Survey Professional Paper 1170-E, 25 p. Dusel-Bacon, Cynthia, and Hansen, V.L., 1991, High-pressure, medium-temperature early Mesozoic metamorphism and deformation within the Yukon-Tanana composite terrane, eastern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 23, p. 20. Erdmer, Phillipe, 1987, Blueschist and eclogite in mylonitic allochthons, Ross River and Watson Lake areas, southeastern Yukon: Canadian Journal of Earth Sciences, v. 24, p. 1439-1449. Erdmer, Phillipe, and Armstrong, R.L.,1988, Permo-Triassic isotopic dates for blueschist, Ross River area, Yukon, in Yukon Geology, v. 2: Exploration and Geological Services Division, Yukon, Indian and Northern Affairs, Canada , p. 33-36. Erdmer, Phillipe, and Helmstaedt, Herwart, 1983, Eclogite from central Yukon: a record of subduction at the western margin of ancient North America: Canadian Journal of Earth Sciences, v. 20, p. 1389-1408. Foster, H.L., 1970, Reconnaissance geologic map of the Tanacross quadrangle, Alaska: U.S. Geological Survey Miscellaneous Geologic Investigations Map I-593, scale 1:250,000. ---1976, Geologic map of the Eagle quadrangle, Alaska: U.S. Geological Survey Miscellaneous Investigations Map I-922, scale 1:250,000. Foster, H.L., and Keith, T.E.C., 1974, Ultramafic rocks of the Eagle quadrangle, east-central Alaska: U.S. Geological Survey Journal of Research, v. 2, no. 6, p. 657-669. Foster, H.L., Keith, T.E.C., and Menzie, W.D., 1987, Geology of east-central Alaska: U.S. Geological Survey Open-File Report 87-188, 59 p. Gareau, S.A., 1989, Metamorphism, deformation and geochronology of the Ecstali-Quaal rivers area, Coast plutonic complex, British Columbia, in Current Research, Part E, Geological Survey of Canada Paper 89-1E, p. 155-162. Graham, C.M., and Powell, R., 1984, A garnet-hornblende geothermometer: Calibration, testing, and application to the Pelona Schist, southern California: Journal of Metamorphic Geology, v. 2, p. 13-31. Hansen, V .L., 1989, Structural and kinematic evolution of the Teslin suture zone, Yukon: Record of an ancient transpressional margin: Journal of Structural Geology, v. 11, p. 717-733. ---1990, Yukon-Tanana terrane: A partial acquittal: Geology, v. 18, p. 365-369. High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

---1992a, Backflow and margin-parallel shear within an ancient subduction complex: Geology, v. 20, no. 1 p. 7174. ---1992b, P-T evolution of the Teslin suture zone and Cassiar tectonites, Yukon, Canada: Evidence for A- and B-type subduction: Journal of Metamorphic Geology, v.10, p. 239-263. Hansen, V.L., Heizler, M.T., and Harrison, T.M., 1991, Mesozoic thermal evolution of the Yukon-Tanana composite terrane: New evidence from 40 Art39 Ar data: Tectonics, v. 10, p. 51-76. Harms, T.A., 1985, Pre-emplacement thrust faulting in the Sylvester allochthon, northeast Cry Lake map area, British Columbia: Geological Survey of Canada Paper 85-1A, p. Harms, T.A., Coney, P.J., and Jones, D.L., 1984, The Sylvester allochthon, Slide Mountain terrane, British Columbia: A correlative of oceanic terranes of northern Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 16, p. 288. Hawthorne, F.C., 1981, Crystal chemistry of the amphiboles in Amphiboles and other hydrous pyriboles-Mineralogy: Mineralogical Society of America Reviews in Mineralogy, v. 9A, p. 1-102. Hodges, K.V .. , and Crowley, P.D., 1985, Error estimation and empirical geothermobarometry for pelitic systems: American Mineralogist, v. 70, p. 702-709. Hodges, K.V., and McKenna, L.W., 1987, Realistic propagation of uncertainties in geologic thermobarometry: American Mineralogist, v. 72, p. 671-680. Hodges, K.V., and Spear, F.S., 1982, Geothermometry, geobarometry and the Al2Si05 triple point at Mt. Moosilauke, New Hampshire: American Mineralogist, v. 67' p. 1118-1134. Holdaway, M.J., 1971, Stability of andalusite and the aluminum silicate phase diagram: American Journal of Science, v. 271, p. 97-131. Jones, D.L., Silberling, N.J., Coney, P.J., and Plafker, G., 1987, Lithotectonic terrane map of Alaska (west of the 141st Meridian): U.S. Geological Survey Miscellaneous Field Studies Map MF-1874-A, scale 1:2,500,000. Keith, T.E.C., Foster, H.L., Foster, R.L., Post, E.V., and Lehmbeck, W.L., 1981, Geology of an alpine-type peridotite in the Mount Sorenson area, east-central Alaska in Shorter contributions to general geology: U.S. Geological Survey Professional Paper 1170-A, p. A1-A9 Keskinen, M., 1989, Lithological and fluid compositional control of amphibole compositions and their potential as P-T indicators of regional metamorphism: An illustration from the Fairbanks Schist, Yukon crystalline terrane, Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 21, no. 6, p. A275. Kohn, M.J., and Spear, F.S., 1990, Empirical calibration of geobarometers for the assemblage garnet-hornblende-plagioclase-quartz: American Mineralogist, v. 75, p. 89-96. Koziol, A.M., and Newton, R.C., 1988, Redetermination of the anorthite breakdown reaction and improvement of the plagioclase-garnet-Al2S i 0 5-quartz geobarometer: American Mineralogist, v. 73, p. 216-223. Laird, Jo, Foster, H.L., and Weber, F.R., 1984, Amphibole Geologic Studies in Alaska by the U.S. Geological Survey, 1991 eclogite in the Circle quadrangle, Yukon-Tanana Upland, Alaska, in Coonrad, W .L., and Elliott, R.L., eds., The United States Geological Survey in Alaska: Accomplishments during 1981: U.S. Geological Survey Circular 868, p. 57-60. Leake, B.E., 1978, Nomenclature of amphiboles: Canadian Mineralogist, v. 16, p. 501-520. Malavieille, J., 1987, Kinematics of compressional and extensional ductile shearing deformation in a metamorphic core complex of the northeastern Basin and Range: Journal of Structural Geology, v. 9, p. 541-554. McClelland, W.C., Gehrels, G.E., Samson, S.D., and Patchett, P .J ., 1992, Proto lith relations of the Gravina belt and Yukon-Tanana terrane in central southeastern Alaska: Journal of Geology, v. 100, no.1, p.107-123. Metcalfe, Paul, and Clark, G.S., 1983, Rb-Sr whole-rock age of the Klondike Schist, Yukon Territory: Canadian Journal of Earth Sciences, v. 20, no. 5, p. 886-891. Monger, J.W.H., and Berg, H.C., 1987, Lithotectonic terrane map of western Canada and southeastern Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF1874-B, scale 1:2,500,000. Mortensen, J.K., 1990, Significance of U-Pb ages for inherited and detrital zircons from Yukon-Tanana terrane, Yukon and Alaska [abs.]: Geological Association of Canada Programs with Abstracts, v. 15, p. A91. Mortensen, J.K., and Jilson, 1985, Evolution of the YukonTanana terrane: Evidence from southeastern Yukon Territory: Geology, v. 13, no. 11, p. 806-810. Nelson, J., and Bradford, J., 1987, Geology of the area around the Midway deposit, northern British Columbia (1040/16): Canada, British Columbia Ministry of Energy, Mines and Petroleum Resources, Geological Fieldwork, 1986, Paper 1987-1, p. 181-192. Nelson, J., Bradford, J.A., Green, K.C., and Marsden, H., 1988, Geology and patterns of mineralization, Blue Dome map area, Cassiar district (104P/12): British Columbia Ministry of Energy, Mines and Petroleum Resources, Geological Fieldwork, 1987, Paper 1988-1, p. 233-243. Nokleberg, W.J., Foster, H.L., and Aleinikoff, J.N., 1989, Geology of the northern Copper River Basin, eastern Alaska Range, and southern Yukon-Tanana Basin, southern and east-central Alaska, in Nokleberg, W.J., and Fisher, M.A., eds., Alaskan Geological and Geophysical Transect, Field trip guidebook T104: American Geophysical Union, Washington, D.C., p. 34-63. Palmer, A.R., 1983, The decade of North American geology 1983 geologic time scale: Geology, v. 11, no. 9, p. 503504. Pavlis, T.L.,1989, Middle Cretaceous orogenesis in the northern Cordillera: A Mediterranean analog of collision-related extensional tectonics: Geology, v. 17, no. 10,p. 947-950. Pavlis, T.L., Sisson, V.B., Foster, H.L., Nokleberg, W.J., and Plafker, George, in press, Mid-Cretaceous extensional tectonics of the Yukon-Tanana terrane, Trans-Alaskan Crustal Transect (TACT), east-central Alaska: Tectonics. Pavlis, T.L., Sisson, V.B., Nokleberg, W.J., Plafker, George, and Foster, Helen, 1988, Evidence for Cretaceous crustal extension in the Yukon crystalline terrane, east-central

Alaska [abs.]: Eos (American Geophysical Union, Transactions), v. 69, p. 1453. Roddick, J.A., 1967, Tintina trench: Journal of Geology, v. 75, p. 23-33. Selverstone, Jane, Spear, F.S., Franz, Gerhard, and Morteani, Giulio, 1984, High-pressure metamorphism in the Tauern window, Austria: P-T paths from hornblende-kyanitestaurolite schists: Journal of Petrology, v. 25, part 2, p. Sisson, V.B, Pavlis, T.L., and Dusel-Bacon, Cynthia, 1990, Metamorphic constraints on Cretaceous crustal extension in the Yukon crystalline terrane, east-central Alaska [abs.]: Geological Association of Canada, Mineralogical Association of Canada Annual Meeting, Programs with Abstracts, v. 15, p. A-122. Spear, F.S., and Kimball, K.L., 1984, RECAMP-A FORTRAN IV program for estimating Fe3+ contents in amphiboles: Computers and Geosciences, v. 10, no. 2-3, p. 317-325. Tempelman-Kluit, D.J., 1979, Transported cataclasite, ophiolite and granodiorite in Yukon: evidence of arc-continent collision: Geological Survey of Canada Paper 79-14, 27 p. Tempelman-Kluit, D.J., and Wanless, R.K., 1975, Potassiumargon age determinations of metamorphic and plutonic rocks in the Yukon crystalline terrane: Canadian Journal of Earth Sciences, v. 12, no. 11, p. 1895-1909. Thompson, A.B., and Tracy, R.J., 1979, Model systems for anatexis of pelitic rocks. Part I. Theory of melting reactions in the systems KA102-NaA102-Al20 3-Si02-H20: Contributions to Mineralogy and Petrology, v. 70, p. 429438. Turner, F.J., and Weiss, L.E., 1963, Structural analysis of metamorphic tectonites: New York, McGraw-Hill, 545 p. Weber, F.R., Foster, H.L., Keith, T.E.C., and Dusel-Bacon, Cynthia, 1978, Preliminary geologic map of the Big Delta quadrangle, Alaska: U.S. Geological Survey Open-File Report 78-529A, scale, 1:250,000. Wheeler, J.O., Brookfield, A.J., Gabrielse, H., Monger, J.W.H., Tipper, H.W., and Woodsworth, G.J., 1988, Tectonic assemblage map of the Canadian Cordillera and adjacent parts of the United States of America (rev. ed.): Geological Survey of Canada Open File 1894, scale 1:2,000,000. Wheeler, J.O., and McFeely, P., 1987, Tectonic assemblage map of the Canadian Cordillera and adjacent parts of the United States of America: Geological Survey of Canada Open File 1565, scale 1:2,000,000. Wilson, F.H., and Shew, Nora, 1981, Map and tables showing preliminary results for potassium-argon age studies in the Circle quadrangle, Alaska, with a compilation of previous dating work: U.S. Geological Survey Open-File Report 81-889, scale 1:250,000. Wilson, F.H., Smith, J.G., and Shew, Nora, 1985, Review of radiometric data from the Yukon crystalline terrane, Alaska and Yukon Territory: Canadian Journal of Earth Sciences, v. 22, p. 525-537. Reviewers: Virginia B. Sisson and James 0. Eckert, Jr. High-Pressure Amphibolite-Facies Metamorphism and Deformation Within the Yukon-Tanana and Taylor Mountain Terranes

Some Facies Aspects of the Upper Part of the Kenai Group, Southern Kenai Peninsula, Alaska By Romeo M. Flores and Gary D. Stricker Abstract Detailed facies characteristics of the upper part of the Beluga Formation and the lower and middle parts of the Sterling Formation, Kenai Peninsula, indicate deposition in anastomosing and meandering streams, respectively. The presence of coeval channel sandstones and of abundant crevasse-overbank and flood plain mudstone-siltstone associations suggests deposition by suspended-loaq anastomosed streams. The fine-grained flood plain sediments, as well as coal-carbonaceous shale associations, laterally separate small sandstone bodies of merging and diverging fluvial channels. The abundance of narrow point-bar sandstone bodies and crevasse-flood plain sandstones, siltstones, and mudstones indicates deposition in mixed-load meandering streams. In both depositional systems, raised platforms served as sites of peat accumulation. The peat bodies formed as narrow, lenticular deposits parallel to the length of the me- , ander belt and as broad, elongate deposits in the crevasse splay of an anastomosing fluvial complex. Peat bodies, and the resulting coalbeds, in the meandering river systems of the Sterling Formation tended to be thin because of lateral migration of the sequence. Those peat bodies that developed in the anastomosing river systems of the Beluga Formation tended to be thick because of vertical accretion of the sequence. INTRODUCTION The upper part of the Kenai Group in the Cook Inlet, Alaska, which includes the Miocene Beluga Formation and the Miocene and Pliocene Sterling Formation, contains subbituminous coals and lignites (Merritt and others, 1987) and coalbed methane (Magoon, 1990). These formations are exposed in steep beach cliffs along the eastern coast of Cook Inlet near Clam Gulch, Ninilchik, Anchor Point, and Homer, and along the northern shore of Kachemak Bay (fig. 1 ). The Beluga Formation in the southern Kenai Peninsula is more than 1,500 m thick and consists of interbedded conglomeratic sandstones, sandstones, siltstones, mudstones, carbonaceous shales, and coalbeds Geologic Studies in Alaska by the U.S. Geological Survey, 1991 (fig. 2; Calderwood and Fackler, 1972). The Sterling Formation is composed of sandstones, siltstones, mudstones, carbonaceous shales, and coalbeds and is as much as 2,100 m thick (Hayes and others, 1976; Merritt and others, 1987). The upper part of the Beluga Formation and lower part of the Sterling Formation are exposed in the vicinity of McNeil Canyon along the northern shore of Kachemak Bay (fig. 1). Here, the Beluga Formation is gray and the Sterling Formation is buff or light brown. The color change occurs at about coalbed B (figs. 3, 4) of Barnes and Cobb (1959), which defines the Homerian (Upper Miocene) and Clamgulchian (Upper Miocene to Pliocene) floral boundary of Wolfe and others ( 1966). During the summer of 1991, facies characteristics that include lithofacies associations (groups of rocks with common physical properties) and facies sequences (a succession of genetically related lithofacies associations) of the upper part of the Beluga and the lower and middle parts of the Sterling were studied at 35 measured sections to assess relationships of coalbed occurrence and distribution to depositional environments. Differentiation of facies characteristics between the Beluga and Sterling Formations was utilized to compare their depositional setting. GEOLOGIC HISTORY The Kenai Peninsula is located in the southeastern part of the Cook Inlet basin-the active forearc basin of the Aleutian subduction zone. Cook Inlet basin lies between a volcanic-plutonic terrane of the Aleutian and Alaskan Ranges to the northwest and an uplifted accretionary wedge of graywacke, argillite, chert. and mafic and ultramafic igneous rocks of the Chugach and Kenai Mountains to the southeast. During deposition of the Beluga Formation, episodic detrital influx from the Aleutian and Alaskan Ranges occurred; however, the Kenai-Chugach Mountains were the primary provenance (Hartman and others, 1972;

Kirschner and Lyon, 1973; Hite, 1975; Hayes and others, 1976; Magoon and others, 1976; Rawlinson, 1984; Magoon, 1986). During deposition of the Sterling For- ?Tks NINILCHIK mation, the Aleutian and Alaskan Ranges were the primary source of sediments (Kirschner and Lyon, 1973). Reinink-Smith (1990a) suggested that these contrasting KASILOF KENAI EXPLANATION l Tksl Sterling Formation Tk b l Beluga Formation Study area QSTks --- ~PENINSULA Tkb 5g045' 1 0 KILOMETERS Figure 1. Location map of Kenai Peninsula showing outcrops of Beluga and Sterling Formations and study areas at eastern shore of Cook Inlet and along northern shore of Kachemak Bay (modified from Merritt and others, 1987). Some Facies Aspects of the Upper Part of the Kenai Group, Southern Kenai Peninsula

sources for the Beluga and Sterling Formations were overprinted by volcanic activity related to the rise of the Alaska Range. Based on volcanic-ash partings in coalbeds, Reinink-Smith (1990a) indicated that an ash fall was recorded every 8,000 to 10,000 yr during deposition of the upper part of the Beluga and once every 11 ,000 yr during deposition of the lower part of the Sterling. Previous interpretations of depositional environments of the Beluga and Sterling Formations were based on subsurface data (Hite, 1975, 1976; Hayes and others, 1976). Hayes and others (1976) suggested that deposition of the Beluga Formation was accomplished by small, high-gradient braided streams flowing on broad alluvial fans and plains, and that the Sterling Formation was deposited by large, meandering streams characterized by 9- to 14-m-thick point-bar sandstones. LITHOFACIES ASSOCIATIONS Lithofacies associations of the Beluga and Sterling Formations may be recognized as a combination of rock types with common lithologic properties (for example, color, grain size, mineral composition, internal structures, and body and ichno-fossil contents). The lithofaCies associations include sandstone-dominated, mudstone-siltstone, and coal-carbonaceous shale associations. :I:

t) a: a: w w a.. a.. w a;

CD a;~ g CD ·- (.) ~.Q a;o::

::l u ·a N

c:: Q) tO CD t: t) Q) .e; g

CICD =c. Oc. ::l a.. 0 a: (!)

c'B ·a; c:: Q)

FORMATION THICKNESS (IN METERS) Sterling Formation o-2,100 Tyonek Formation 1 ,20D-2,350 Hemlock Conglomerate DESCRIPTION Alluvium and glacial deposits Sandstone, siltstone, mudstone, carbonaceous shale, and lignites Sandstone, conglomeratic sandstone, sihstone, mudstone, carbonaceous shale, and sutlbituminous coalbeds Sandstone, mudstone, siltstone interbeds, and subbituminous coalbeds Sandstone and conglomerate OLDER TERTIARY ROCKS Figure 2. Generalized stratigraphic column of Tertiary Kenai Group in Kenai Peninsula. Beluga and Sterling Formations comprise upper part of Kenai Group (modified from Calderwood and Fackler, 1972). Geologic Studies in Alaska by the U.S. Geological Survey, 1991 The mudstone-siltstone association consists of darkgray mudstone and light- to dark-gray siltstone (figs. 35). The ·mudstone is either massive or crudely laminated, contains vertical and horizontal burrows, abundant finely macerated plant fragments, and plant root marks. The mudstone exhibits popcornlike weathered surfaces, indicating smectitic clay components (Reinink-Smith, 1990b). The siltstone exhibits current ripples, plant root marks, plant fragments, and vertical burrows. The mudstone-siltstone association may be randomly interbedded with current-rippled, rooted, burrowed, sandy siltstone; calcareous mudstone; and plantrich ferruginous or hematite-rich sandy siltstone. The mudstone-siltstone association is commonly capped by fine- to medium-grained sandstone of the sandstone-dominated association with which it constitutes a coarsening-upward sequence (figs. 3-5). The sandstone cap of the mudstone-siltstone association contains current-ripple laminations and small-scale (<7.5 em thick) trough crossbeds and is frequently locally incised by trough-crossbedded sandstones with erosional bases. Several coarsening-upward sequences are vertically stacked to as thick as 30 m. In addition, the sandstonedominated association consists of lag sandstones that fine upward (from pebbly to fine grained) and have erosional bases (figs. 3-5). Color varies from gray to light . gray for the Beluga sandstones and buff for the Sterling sandstones. The sandstone-dominated association ranges from 3 to 15 m thick and varies from a single body to vertically stacked multiscoured (internally divided by erosional surfaces) and multilateral (laterally offset) bodies. The lower part of the sandstone bodies is trough crossbedded ( 1 m high and 13 m long) as well as convolute bedded (1.5 m high and 75 m long). The upper parts of the sandstone bodies exhibit current-ripple laminations and small- to medium-scale trough crossbeds ranging from 15 to 45 em high. In places, some of the sandstone bodies are divided into smaller inclined subbodies by siltstone and mudstone drapes that extend from the margin to the interior of the body. A rare type of sandstone-dominated association is a basally erosional body consisting of inclined, rippled silty sandstone and siltstone-mudstone couplets that assumed the dominantly concave shape of the erosional surface (fig. 5), which lies about 15 m above the base of the section. The coal-carbonaceous shale association comprises subbituminous coals and lignites intercalated with, as well as underlain and overlain by, black, plant-rich carbonaceous shale (figs. 3-5). The coals are mostly dull with vitrain bands distributed throughout the beds. Coal petrology work by Merritt and others (1987) indicated that vitrinite, representing the woody parts of trees, is relatively constant in percent in both the Beluga and Sterling coals. However, the higher quality coals of the Beluga Formation are richer in liptinites than the lower

UPPER BELUGA COMPOSITE SECTION (McNEIL CANYON AREA) METERS 30 T FACIES PROFILE '/ A U A A u u u A A A A A A U A A A A U A u A LITHOLOGIC ASSOCIATIONS Coal-carbonaceous shale Sandstone-dominated Mudstone-siltstone Sandstone-dominated Coal-carbonaceous shale Sandstone-dominated Mudstone-siltstone EXPLANATION (SEE FIG. 4) FACIES INTERPRETATION Bog Levee Fluvial channel Crevasse splay Bog/flood plain Flood plain Flood plain/crevasse splay Crevasse channel Bogs/flood plain Flood plain Bogs/flood plain Levee Fluvial channel (multistacked;multiscoured) Flood plain/crevasse splay Figure 3. Facies profile, lithologic associations, and facies interpretations of upper part of Beluga Formation in McNeil Canyon area. Some Facies Aspects of the Upper Part of the Kenai Group, Southern Kenai Peninsula

quality coals of the Sterling Formation. Beluga coals are low in inertinites, and the Sterling coals are high in inertinites (Merritt and others, 1987). The upper part of some coalbeds in the Sterling Formation have petrified tree trunks in growth position. The coalbeds are as thick as 3.8 m and are laterally continuous (>3 km). Carbonaceous shales consist of fissile, mixed clay and very finely macerated plant fragments, and commonly they contain root marks. Carbonaceous shale is a common parting in coalbeds along with volcanic-ash partings (Reinink-Smith, LOWER STERLING COMPOSITE SECTION (McNEIL CANYON AREA) METERS FACIES PROFILE A A A A A u A UA B coalbed Lag conglomerate LJ Sandstone

Siltstone

Mudstone

Coal/carbonaceous shale

Fining-upward sequence LITHOLOGIC ASSOCIATIONS Coal-carbonaceous shale Sandstone-dominated Sandstone-dominated FACIES INTERPRETATIONS Flood plain Bog Crevasse splay/channel Levee Fluvial channel (multistory; multilateral) Coal-carbonaceous shale Bog Flood plain/crevasse splay Mudstone-siltstone Sandstone-dominated Crevasse splay/channel Bog EXPLANATION B Ferruginous sandstone U Burrows rv Convolute laminations A Root marks

Current ripple laminations T Tonstein Trough crossbeds Tree trunks Mudstone/siltstone drapes / Coarsering-upward sequence Figure 4. Facies profile, lithologic associations, and facies interpretations of lower part of Sterling formation in McNeil Canyon area. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

1990b ). The volcanic-ash, or tonstein, partings are white to brownish in color, consist of pumice fragments, and are as thick as 0.3 m. The coalbeds in the lower part of the Sterling Formation contain more tonsteins than coalbeds of the upper part of the Beluga Formation. Mudstone and siltstone also are common partings in the coalbeds. They range from 0.005 to 3m thick, are root marked, and contain dispersed smectite clays. VERTICAL FACIES SEQUENCES Upper Part of the Beluga Formation The uppermost 100 m of the Beluga Formation at McNeil Canyon is largely composed of mudstone-siltstone associations with subordinate sandstone-dominated and coal-carbonaceous shale associations (fig. 3). The MIDDLE STERLING COMPOSITE SECTION (NINILCHIK AREA) FACIES LITHOLOGIC FACIES METERS PROFILE ASSOCIATIONS INTERPRETATIONS A A Bogs/flood plain Coal-carbonaceous shale Levee Fluvial channel Sandstone-dominated (multilateral) A Flood plain/bogs A Mudstone-siltstone A Levee Sandstone-dominated Fluvial channel (multilateral) Coal-carbonaceous shale Bog/flood plain A A Abandoned channel Sandstone-dominated A Crevasse splay A Bog A Coal-carbonaceous shale Flood plain Bog EXPLANA Tl 0 N (SEE FIG. 4) Figure 5. Facies profile, lithologic associations, and facies interpretations of middle part of Sterling Forl}1ation in Ninilchik area. Some Facies Aspects of the Upper Part of the Kenai Group, Southern Kenai Peninsula

mudstone-siltstone associations, in the middle part of the study interval, contain caps of silty sandstone of the sandstone-dominated association, thus forming stacked, coarsening-upward facies sequences. Each silty sandstone cap includes a thin (1.5-3 m thick), fine-grained sandstone with a scoured base. This sandstone may also occur as a single body in the lower and middle parts of the coarsening-upward facies sequence. The study interval contains sandstone-dominated and coal-carbonaceous shale associations at several stratigraphic levels (fig. 3). The sandstone-dominated association occurs either as thick, stacked, multiscoured, fining-upward conglomeratic sandstones or as individual fining-upward sandstone bodies. The stacked, multiscoured sandstone facies sequence is 13 m thick. This sandstone body is overlain by a thick (0.82 m) coalcarbonaceous shale association with 0.3-m-thick tonstein parting (fig. 3). The single, fining-upward sandstone is overlain by the 1.3-m-thick coal-carbonaceous shale association of the B coalbed (fig. 3). The vertical and lateral variations of the lithofacies associations of the Beluga Formation below the B coalbed of Barnes and Cobb (1959) are shown in figure 6. The finingupward, single and multiscoured sandstone in the upper WEST B COALBED . ·tP· EXPLANATION part of the cross section is coeval. The~c sandstone facies sequences are as thick as 5.5 m and 0.4 km in width and are overlain by a thick, coal-carbonaceous shale association. Also, they merge laterally with and are underlain by thick, stacked mudstone-siltstone associations with local sandstone caps of coarsening-upward facies sequences. Lower and Middle Parts of the Sterling Formation The lower and middle parts of the Sterling Formation in the McNeil Canyon and Ninilchik areas consist largely of sandstone-dominated and coal-carbonaceous shale associations and subordinately of a mudstonesiltstone association (figs. 4, 5). The sandstone-dominated association is mainly composed of multistory (stacked sandstone bodies separated by fine-grained deposits), multilateral (laterally offset sandstone bodies), finingupward sandstone facies sequences as thick as 20 m and laterally traceable for as much as 0.4 km (figs. 4, 5, 7). The lower part of the Sterling Formation studied in the McNeil Canyon area is 60 m thick and may be grouped into several facies sequences (fig. 4). The lower part of the studied interval is characterized by a EAST I~LOCATI~N OF STRATIGRAIPHIC SECTION I Conglomerate Ripple laminations METERS~ r:-:71 Ld D Sandstone Mudstone-siltstone Coal-carbonaceous shale ::; Trough crossbeds 0.2 KILOMETERS A Root marks Figure 6. Stratigraphic cross section of uppermost part of Beluga Formation west of McNeil Canyon. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

mudstone-siltstone association with sandstone caps forming a series of well-developed coarsening-upward facies sequences. The upper part of the studied interval consists mainly of a sandstone-dominated association containing multistory, multilateral, and fining-upward sandstone facies sequences that consist of small inclined sandstone bodies separated by very thin mudstonesiltstone drapes. In addition, these facies sequences are overlain by coarsening-upward sandstone and singlebodied fining-upward sandstone facies sequences. These sequences are interbedded with a 0.3- to 1.3-m-thick coal-carbonaceous shale association and a 6-m-thick mudstone-siltstone association. Vertical and lateral lithofacies variations of the upper part of the Sterling Formation at McNeil Canyon are shown in figure 7. The sandstone bodies are multistory and multilateral in the upper part of the cross section. The sandstone bodies are stacked and are as thick as 10 m and as wide as 0.4 km. Each sandstone facies sequence contains inclined point-bar sandstones that are 3.5 m high. The sandstones grade laterally into the mudstone-siltstone association. The lower part of the cross section consists mainly of a coarsening-upward sandstone facies sequence that grades laterally into a mudstone-siltstone association. The middle part of the Sterling Formation studied in the Ninilchik area is a 75-m-thick interval and may also be grouped into several facies sequences (fig. 5). The lower part of the interval includes a coarsening-upward sandstone facies sequence. This sequence is capped by inclined beds of silty sandstone and siltstone-mudstone couplets of the sandstone-dominated association, which is marked by an erosional base. The upper part of the interval consists of fining-upward, multilateral, and multiscoured sandstone facies sequences. These facies sequences are interbedded with 0.2- to 3.8-m-thick sequence of the coal-carbonaceous shale association and 0.3- to 6m-thick sequences of the mudstone-siltstone association. FACIES INTERPRETATIONS AND CONCLUSIONS Variations in the facies characteristics of the upper part of the Beluga and lower and middle parts of the Sterling Formations reflect differences in the depositional environments. The mudstone-siltstone, coal-carbonaceous shale, and sandstone-dominated associations of these formations were all deposited in a fluvial system. In general, these lithologic associations represent overbank-flood plain, bog, and channel deposits, respectively (fig. SA, B). The abundance of mudstone-siltstone association in the upper part of the Beluga Formation suggests that the WEST EAST r:-=:1

D II I '' ' McNeil Canyon / -"!~LOCATIO~ OF STRATI~RAPHii: ~ECTION~

/

A EXPLANATION METERS~ / Mudstone0 siltstone drapes 0 · 0.2 KILOMETERS Sandstone Mudstone-siltstone Trough crossbeds Coal-carbonaceous shale X .-v Ripple laminations B Ferruginous sandstone A Root marks Figure 7. Stratigraphic cross section of upper part of study interval of Sterling Formation east of McNeil Canyon. Some facies Aspects of the Upper Part of the Kenai Group, Southern Kenai Peninsula

fluvial environment was dominated by suspended-load sediments (fig. 8A), which were mainly accreted vertically as thick overbank-flood plain deposits. The coarseningupward facies sequences of the mudstone-siltstone association, which represent deposits of crevasse-splay channel complexes, suggest that the sediment load was transported into the flood plain via a breach in the levee during flood events. The vertically accreted mudstone-siltstone association reflects rapid subsidence by sediment compaction and basin subsidence. The occurrence of thick overbank-flood plain sediments that merge laterally with crevasse-splay sediments and thick, narrow, coeval sandstone bodies with erosional bases represents deposition A B by an anastomosed fluvial system. In addition, confinement of fluvial channels by thick overbank-flood plain sediments resulted in deposition of narrow, stacked, finingupward multiscoured sandstone facies sequences. The coal-carbonaceous shale association of the upper part of the Beluga Formation represents peat accumulation in bogs that formed on topographically high platforms built by overbank-flood plain crevasse fluvial channel sediments. These bog platforms were removed from detrital influx that allowed sufficient time for the accumulation of thick peat deposits. However, bogs formed on low-lying areas were commonly inundated by floodwaters that transported flocculated clays into the bog, Crevasse-splay channel / Meandering stream Abandoned meander belt Figure 8. Diagrammatic depositional models in the McNeil Canyon and Ninilchik study areas. A, Beluga Formation. B, Sterling Formation. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

which mixed with plant material and formed carbonaceous shale. Our interpretation of deposition by anastomosing streams for the Beluga Formation at McNeil Canyon differs from that of Hayes and others (1976). They interpreted the Beluga as having been deposited by braided streams flowing on broad alluvial fans and plains. Two decades ago, Schumm (1968) suggested that braided streams refer to rivers in which flow diverges and rejoins around constantly shifting sand and gravel bars, whereas anastomosed streams reflect a river that divides into more permanent subchannels that are separated by fine-grained flood plain sediments. The largely finegrained sediments juxtaposed with thin, narrow, coeval channel sandstones in the Beluga Formation at McNeil Canyon suggest deposition in anastomosed streams as defined by Schumm (1968). In addition, the finegrained sediments formed in subaqueous flood plains, as indicated by associated animal burrows. The intervening flood plains of the anastomosed streams also formed low-lying bogs, as reflected by the coal-carbonaceous shale association interbedded with the mudstone-siltstone association. These depositional settings were described by Smith and Smith (1980) and Smith (1983) in their discussion of the anastomosing streams of the upper Columbia and Saskatchewan Rivers of Canada and the Magdalena River in Colombia. In the lower and middle parts of the Sterling Formation, the development of thick, coarsening-upward sequences of the mudstone-siltstone association and moderately thick, fining-upward, multistory and multilateral sandstone facies sequences suggests deposition by a mixed-load fluvial system (fig. 8B). The mud, silt, and sand were deposited in small meandering fluvial systems, as indicated by moderately thick channel sandstones of limited width. Lateral migration developed by erosion on the cutbank side of meanders and deposition on point bars on the opposite side, as indicated by inclined sandstone bodies separated by mudstone-siltstone drapes. The presence of multilateral sandstone bodies provides evidence that the meandering fluvial channel migrated laterally by multiple cut-and-fill of channel cutoffs. However, lateral migration by avulsion via chute cutoff through necks of meander bends resulted in abandoned channels or channel plugs, as indicated by the inclined beds of the sandstone and siltstone-mudstone couplets with an erosional base. These sandstone bodi~s are similar to deposits of mixed-load or meandering fluvial systems described by Schumm (1977, p. 95-178) and Galloway (1985). Overbank-flood plain deposits associated with this fluvial system include sandy sediments that were transported into the flood plain by breaching through levees. Crevasse sedimentation via these breaches disrupted peat accumulation in bogs on the ·flood plain and 'resulted in formation of the thin coalcarbonaceous shale associations. However, abandoned meander belts formed by avulsion and removed from detrital influx served as elevated platforms and permitted accumulations of thick peat bog deposits, as displayed in the Ninilchik section. That is, broad (wide and long) buildup of crevasse-splay and anastomosed stream deposits formed large platforms on which peat bogs accumulated. Our interpretation of small meandering streams for the Sterling Formation at McNeil Canyon and Ninilchik differs from that of Hayes and others (1976), who suggested that the Sterling was deposited by large meandering streams. Development of point-bar sandstones with average thickness of 4 m in our study areas, compared with 9- to 14-m-thick point-bar sandstones observed elsewhere by Hayes and others (1976), clearly indicates deposition in small meandering fluvial channels. In addition, the 30-m width of channel-plug deposits and 400-m width of the channel sandstones support deposition of the Sterling Formation in a small, narrow, meandering fluvial system. In conclusion, facies analysis of the Beluga and Sterling Formations suggests deposition in narrowchanneled, anastomosing (suspended load) and small meandering (mixed-load) fluvial systems, respectively. Vertical aggradation in the anastomosing fluvial system and lateral erosion and deposition in the meandering fluvial system probably controlled thick-to-thin peat accumulation in bogs of these fluvial systems. Thick and laterally extensive peats are more likely to be preserved in deposits in vertically aggrading anastomosing streams than in laterally aggrading meandering streams. Thus, the difference in depositional processes of the fluvial systems explains the original observation of Barnes and Cobb ( 1959), that coals are thicker in the Beluga Formation than in the Sterling Formation in the Kenai Peninsula. REFERENCES CITED Barnes, F.F., and Cobb, E.H., 1959, Geology and coal resources of the Homer district, Kenai coal field, Alaska: U.S. Geological Survey Bulletin 1058-F, p. 217-260. Calderwood, K.W., and Fackler, W.C., 1972, Proposed stratigraphic nomenclature for the Kenai Group, Cook Inlet basin, Alaska: American Association of Petroleum Geologists Bulletin, v. 56, no. 4, p. 739-754. Galloway, W.E., 1985, Meandering streams-Modern and ancient, in Flores, R.M., Ethridge, F.G., Miall, A.O., Galloway, W.E., and Fouch, T.D., eds., Recognition of fluvial depositional systems and their resource potential: Society of Economic Paleontologists and Mineralogists Short Course No. 19, p. 145-166. Hartman, D.C., Pessel, G.H, and McGee, D.L., 1972, Preliminary report on stratigraphy of Kenai Group upper Cook Inlet, Alaska: Alaska Division of Geological and Geophysical Surveys Special Report 5, 4 p., 7 maps, 1 pl., scale 1:500,000. Some Facies Aspects of the Upper Part of the Kenai Group, Southern Kenai Peninsula

Hayes, J.B., Harms, J.C., and Wilson, T., Jr., 1976, Contrasts between braided and meandering stream deposits, Beluga and Sterling Formations (Tertiary), Cook Inlet, Alaska, in Miller, T.P., ed., Recent and ancient sedimentary environments in Alaska: Anchorage, Alaska Geological Society, p. Jl-J27. Hite, D.M., 1975, Some sedimentary aspects of the Kenai Group, Cook Inlet, Alaska, in Sisson, A., ed., Guide to the geology of the Kenai Peninsula, Alaska: Anchorage, Alaska Geological Society, Anchorage, p. 3-19. ---1976, Some sedimentary aspects of the Kenai Group, Cook Inlet, Alaska, in Miller, T.P., ed., Recent and ancient sedimentary environments in Alaska: Anchorage, Alaska Geological Society, p. Il-I23. Kirschner, C.E., and Lyon, C.A., 1973, Stratigraphic and tectonic development of Cook Inlet petroleum province, in Pitcher, M.G., ed., Arctic geology: American Association of Petroleum Geologists Memoir 19, p. 396-407. Magoon, L.B., 1990, Oil-source rock correlations using carbon isotope data and biological marker compounds, Cook Inlet-Alaska Peninsula, Alaska [abs.]: American Association of Petroleum Geologists Bulletin v. 74, p. 711. Magoon, L.B., Adkison, W.L., and Egbert, R.M., 1976, Map showing geology, wildcat wells, Tertiary plant· fossil localities, K-Ar age dates, and petroleum operations, Cook Inlet area, Alaska: U.S. Geological Survey Miscellaneous Geologic Investigations Map I-1019, 3 sheets, scale 1:250,000. Magoon, L.B., ed., 1986, Geologic studies of the lower Cook Inlet COST No. 1 Well, Alaska outer continental shelf: U.S. Geological Survey Bulletin 1596, 99 p. Merritt, R.D., Lueck, L.L., Rawlinson, S.E., Belowich, M.A., Goff, K.M., Clough J.C., and Reinink-Smith, L.M., 1987, Southern Kenai Peninsula (Homer district) coal-resource Geologic Studies in Alaska by the U.S. Geological Survey, 1991 assessment and mapping project: Alaska Division of Geological and Geophysical Surveys Public-Data File 87-15, 125 p. Rawlinson, S.E., 1984, Environments of deposition, paleocurrents, and provenance of Tertiary deposits along Kachemak Bay, Kenai Peninsula, Alaska: Sedimentary Geology, v. 38, p. 421-442. Reinink-Smith, L.M., 1990a, Relative frequency of Neogene volcanic events as recorded in coal partings from the Kenai lowland, Alaska: A comparison with deep-sea core data: Geological Society of America Bulletin, v. 102, p. ---1990b, Mineral assemblages of volcanic and detrital partings in Tertiary coalbeds, Kenai Peninsula, Alaska: Clays and Clay Minerals, v. 38, no. 1, p. 97-108. Schumm, S.A., 1968, Speculations concerning paleohydrologic controls of terrestrial sedimentation: Geological Society of America Bulletin, v. 79, p. 1573-1588. ---1977, The fluvial system: New York, Wiley, 338 p. Smith, D.G., 1983, Anastomosed fluvial deposits-Modern examples from western Canada, in Collinson, J.D., and Lewin, J ., eds., Modern and ancient fluvial systems: International Association of Sedimentologists Special Publication 6, p. 155-168. Smith, D.G., and Smith, N.D., 1980, Sedimentation in anastomosed river systems-Examples from alluvial valleys near Banff, Alberta: Journal of Sedimentary Petrology, v. 50, p. 157-164. Wolfe, J.A., Hopkins, D.M., and Leopold, E.B., 1966, Tertiary stratigraphy and paleobotany of the Cook Inlet region, Alaska: U.S. Geological Survey Professional Paper 398A, p. A1-A29. Reviewers: Edwin R. Landis and Richard G. Stanley

Sedimentology of the Bay of Pillars and Point Augusta Formations, Alexander Archipelago, Alaska By Susan M. Karl and Closey F. Giffen Abstract The Bay of Pillars and Point Augusta Formations are Silurian marine siliciclastic units mapped in the Alexander archipelago in the southern part of the Alexander terrane. These rock units have been correlated and used to measure right-lateral offset on the Chatham Strait fault by previous workers. Our sedimentologic studies, measured sections, and point counts of these rocks indicate that the Bay of Pillars and Point Augusta Formations are lithologically similar but not identical, and that Silurian sandstone present in the Chilkat Mountains is identical in composition to the Point Augusta Formation, supporting correlation of these rock units by previous workers. The Point Augusta Formation is herein geographically extended to include the sandstone in the Chilkat Mountains. The Bay of Pillars Formation is significantly more volcanogenic than the Point Augusta Formation and consists largely of innerfan facies deposits. The Point Augusta Formation is more calcareous, thinner bedded, and has a greater proportion of midfan facies deposits than does the Bay of Pillars Formation. Both units, nonetheless, have similar sandstone compositions that indicate magmatic-arc provenances with dissected-arc orogen sources. Distinctive red syenitic cobbles suggest a similar source for conglomerates of the two units. Both units are interpreted to represent parts of turbidite aprons formed around volcanic islands with fringing limestone reefs. Distinctive deposits of interbedded black carbonaceous, calcareous mudstone and white limestone occur in both formations and are interpreted to represent an interchannel facies. It is likely that the Bay of Pillars and Point Augusta Formations are spatially related. Consequently, estimates of offset on the Chatham Strait fault are justified; however, the lack of marker beds precludes precise measurement. The significant granitic source component in both units suggests that the Alexander volcanic arc has continental rather than oceanic basement. Sandstone compositions indicate entirely local sources and support paleontologic studies that suggest that the Alexander arc was isolated from North America in the Silurian, despite its paleobiogeographic ties to North America. INTRODUCTION Paleozoic sedimentary rocks of the Alexander terrane are important because they are only mildly deformed and their low metamorphic grade allows preservation of critical clues to the depositional and tectonic history. of the Alexander arc. There has been debate about the oceanic versus continental nature of the crust underlying the Alexander arc (Armstrong, 1985; Samson and others, 1989), about whether or not the arc was originally part of North America (Gehrels and Saleeby, 1987a; Savage, 1988; Soja, 1991), and about where the arc was before collision and accretion to North America in the Cretaceous (Hillhouse and Gramme, 1980; Haeussler and others, in press). Also, because the rocks are relatively well preserved, they have been used as a basis for correlations made in measuring 150 to 200 km of offset on the right-lateral Chatham Strait fault (Lathram, 1964; Ovenshine and Brew, 1972; Loney and others, 1975; Hudson and others, 1982). The siliciclastic rocks of the Bay of Pillars and Point Augusta Formations compose part of the basis for these correlations (fig. 1 ). The composition of the sandstones in these units also provides information about source terranes and the nature of the basement of the Alexander volcanic arc. Alexander Terrane The Alexander terrane is an allochthonous crustal block with a long history of rifting, amalgamation, and accretion. It is composed of a succession of Precambrian(?) to Quaternary, dominantly sedimentary and volcanic rocks (fig. 2). The oldest rocks consist of pelitic schist, marble, metabasite, and metarhyolite of the Wales Group (Eberlein and others, 1983); these rocks are interpreted to represent a rift assemblage (S.M. Karl, unpub. data). The Wales Group was intruded by Late Cambrian diorite to granodiorite (Gehrels and Saleeby, 1987b). These rocks are metamorphosed to greenschist and amphibolite facies and are unconformably overlain by lower grade Ordovician and Silurian basaltic and andesitic flows and breccia and by volcaniclastic turbidites of the Descon Formation (Eberlein and others, 1983). The Descon Formation gradationally overlain by the Silurian Heceta Limestone, a reef complex (Soja, 1991), and by the Silurian calcareous and volcaniclastic rocks Sedimentology of the Bay of Pillars and Point Augusta Formations, Alexander Archipelago

of the Bay of Pillars Formation. The older rocks are also intruded by Silurian to Devonian syenitic to trondjhemitic plutons. The Devonian and older rocks are depositionally overlain by Devonian limestone, shallowwater calcarenite, or polymictic conglomerate that contains abundant red syenitic cobbles. The upper Paleozoic part of the sequence consists of Devonian basaltic, andesitic, and rhyolitic volcanic and volcaniclastic rocks, representing continuing arc activity, overlain by a thick section of Mississippian carbonate rocks and Permian carbonate and clastic rocks. Major unconformities throughout the Alexander terrane underlie Upper Triassic conglomerate, limestone, and bimodal basalt and rhyolite; Jurassic to Cretaceous mafic to intermediate volcanic and volcaniclastic rocks; Paleocene polymictic conglomerate; and Eocene to Quaternary basalt, andesite, and rhyolite. Timing of amalgamation of the Alexander terrane with adjacent magmatic arc terranes is constrained by .. Pacific Ocean EXPLANATION Area that includes outcrops of Point Augusta Formation Area that includes outcrops of Bay of Pillars Formation Pennsylvanian tonalitic plutons that intruded its western contact with the Wrangellia terrane (Gardner and others, 1988) and by deposition of the Gravina belt overlap assemblage on its eastern contact with the Taku terrane (Berg and others, 1978; Brew and Karl, 1988). These amalgamated terranes were accreted to North America in the mid- to Late Cretaceous, based on the timing of deformation and metamorphism of the Jurassic and Cretaceous Gravina belt volcanic and volcaniclastic rocks. Isotopic and chemical analyses of igneous rocks from the Alexander magmatic arc provide ambiguous results with respect to the nature of the crust underlying the Alexander terrane. Interpretation of N d/Sm isotopes from a variety of sedimentary and igneous samples suggests a primitive origin (Samson and others, 1989). However, 87Sr/86Sr isotopes for Silurian plutons range from 0.7036 to 0.7212 (Armstrong, 1985) and are interpreted to in~icate assimilation of older continental crust. 50 KILOMETERS L 1 Figure 1. Location of Bay of Pillars and Point Augusta Formations. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Paleontologic studies of Silurian limestone identified a particular reef-building sponge, aphrosalpingid, that distinguishes these limestones from those of autochthonous western North America but ties them to rocks of the Nixon Fork terrane in southwestern Alaska, which are considered to be part of North America (Clough and Blodgett, 1989; Soja, 1991). Soja (1991) concludes that the Alexander terrane was isolated from North America but has biogeographic ties to the craton in the Silurian. This is supported by a recent study of Devonian faunas around the Pacific rim, which showed that the Devonian assemblage of ·the Alexander terrane is most like the Devonian assemblage of Nevada (Savage, 1988). Paleomagnetic studies also give equivocal results that do not distinguish whether the Alexander terrane was in the Northern or Southern Hemisphere in the Paleozoic (Van der Voo, 1989). Haeussler and others (in press) interpreted their data to indicate that the Alexander terrane was in the Northern. Hemisphere by the Permian, at about latitude 19° N. by the Late Triassic, suggesting that it was probably at low latitudes during the early Paleozoic. Methods of Study This study represents preliminary results of extensive field work and petrographic work conducted under the auspices of the Petersburg and Juneau mineral resource assessment projects. Data collected by numerous project geologists were compiled from several hundred stations in the Point Augusta Formation on Chichagof Island and Silurian sandstone in the Chilkat Mountains, and from more than 500 stations in the Bay of Pillars Formation on Kuiu and Prince of Wales Islands. Turbidite facies associations according to the model of Mutti and Ricchi-Lucci (1972) were recorded at all stations where applicable. Bedding thicknesses were recorded in more than 50 estimated 3- to 10-m sections in the Bay of Pillars Formation; these sections were located Eastern Chichagof Island Central Prince Annette, Gravina, of Wales and and southern Prince m.y. Kuiu Islands of Wales Islands Ordovician Cambrian EXPLANATION m Chert m Limestone

Mudstone

Conglomerate

Volcaniclastic rocks I&J Intermediate to mafic volcanic rocks

Silicic volcanic rocks Figure 2. Columns showing stratigraphic sequences in Alexander terrane and geologic time scale used by U.S. Geological Survey (1986). Sedimentology of the Bay of Pillars and Point Augusta Formations, Alexander Archipelago

ol:lo "' Table 1. Components of point-counted sandstones of the Bay of Pillars Formation

[Abbreviations: Quartz (m), monocrystalline quartz; Quartz (p), polycrystalline quartz; Quartz (c), chert; Feldspar (p), plagioclase; Feldspar (k), potassium feldspar; Biotite/Epidote (d), detrital; Foss. frag., fossil fragment; Amp, amphibole; ;f Pyr, pyroxene; Pum, pumpellyite; c, calcite, w, graywacke or chloritic pseudomatrix, w(c), chloritic matrix with patches of calcite; tr, trace. Trace components were noted in thin section but did not fall under a counted point]

Q.

:r

D7" fll & C"' ID c: !n "'

Q CJQ [ fll

loC

Sample number Quartz (m) Quartz (p) Quartz Feldspar Feldspar (c) (p) (k) Lithic Lithic Lithic Lithic (carbonate) (~udstone) (volcanic) (plutonic) Lithic (felsite) Biotite (d) Epidote Chlorite Opaques (d) Foss. Glauconite Amp frag. Pyr Pum Matrix 80SK455A 3 7 3 I6 93 25 I20 I 04 J J 1 c 80SK46 J A 18 9 14 7I 21 1 J 4 J 25 II II 8 w 80SK462A 9 3 6 77 37 230 22 40 w 80SK465A 20 I 75 53 26 2 J 7 8 2 I ' w(c) 80SK469A 22 J 6 86 38 20 192 1 8 10 3 c 80SK4 7 J A I 0 1 J 98 72 J 3 I92 16 8 20 w 82SK053A 13 4 5 96 37 26 209 1 5 4 w 82SK054A 12 17 240 73 26 70 8 w(c) 82SI$:056A 14 3 II 1 J 9 3 82 156 3 5 w(c) 82SK064A --- I 00 4 Il 125 44 5 20 65 13 16 tr w(c) 82SK065A 23 12 33 89 4 7 130 51 3 12 c 82SK071 A 1 12 20 8 331 I tr 27 c 82SK07 SA 5 I 2 46 14 55 2 7 4 4 c Table 2. Components of point-counted sandstones of the Point Augusta Formation and the Silurian sandstone of the Chilkat Mountains [First seven samples listed are from the Point Augusta Formation; last five samples are from the Silurian sandstone unit of the Chilkat Mountains. Abbreviations: s(c), siliceous matrix with patches of calcite; aU others as in table 1] Sample number Quartz (m) Quartz (p) Quartz Feldspar Feldspar (c) (p) (k) (carbonate) (mudstone) (volcanic) (plutonic) (felsite) Biotite (d) Epidote Chlorite Opaques (d) Foss. Glauconite Matrix frag. 85SK003B 19 58 1 4 3I3 4 I 11 p 85SKO I3B I24 7 1I 153 18 2 83 8 tr 2 p 85AKO 15A --- 1I7 7 6 12I 19 40 83 1 7 c 85SKO 17 A 85 5 4 92 57 3 7 112 4 1 3 c 85SKO 18A 80 5 40 44 I6 1 98 76 33 8 s(c) 85SKO 180 27 1 I35 9 210 2 7 tr 8 w 85SKO I9A 96 10 92 19 86 97 1 c 85SK029A 17 3 1 260 6 11 100 4 1 w 85SK030A 4I 5 7 I 02 4I 97 10 I 2 1 3 c 8 5S K03 2A 44 3 I 0 96 3 6 9 8 II4 2 5 c 85SK034A 26 11 15 I 03 54 1I5 64 c 85SK037 A 35 2 8 II3 38 90 173 I c

mainly on Kuiu Island because rocks from this unit on Prince of Wales Island are principally conglomerates, olistostromes (sedimentary slump deposits), limestones, and calcareous mudstones. Twenty-one detailed 25- to 50-m sections were measured in the Point Augusta Formation on Chichagof Island and in Silurian sandstone in the Chilkat Mountains. Thin sections were collected at all stations. For this study, thin sections collected by the first author were examined, and from this data set 14 samples from the Bay of Pillars Formation,. 7 samples from the Point Augusta Formation, and 5 samples from the sandstone in the Chilkat Mountains were chosen as suitable for point counting. Although a few samples contain prehnite or pumpellyite, primary clast compositions appear to be unaffected by the low metamorphic grade of these rocks (tables 1, 2). Approximately 500 points were counted on medium-grained sandstones with low matrix content by the traditional method (Dickinson, 1970); however, chert grains were combined with the sedimentary lithic parameter on ternary plots. Paleocurrents were calculated from measured unidirectional and bidirectional sedimentary structures· such as flute casts, asymmetric ripples, and groove casts. Measurements were rotated for plotting under the assumption of a horizontal fold axis because in most cases cleavage or fold axes were either not present or not measured at the outcrop where the paleocurrent indicator was measured. The regional structural style consists of broad, open folds. Measured and plotted fold axes plunge gently to moderately north or south (Karl and Hunt, 1983; S.M. Karl, unpub. data). Owing to very poor age control and pervasive faulting, unit thicknesses are not known, but both the Bay of Pillars and Point Augusta Formations have been estimated to be approximately 1,500 m thick (Muffler, 1967; Loney and others, 1975). The broad geographic distribution of the samples and sections studied covers the full lithic variation within the units, and therefore the samples and sections are assumed to represent the full stratigraphic range of the units. SEDIMENTOLOGY Bay of Pillars Formation The Silurian Bay of Pillars Formation (Muffler, 1967) is mapped on Kuiu Island, Prince of Wales Island, and smaller islands east of the Chatham Strait fault (fig. 1 ), and it is inferred to gradationally overlie the Descon Formation on Prince of Wales Island (Brew and others, 1984 ). It is depositionally overlain by the Silurian Kuiu Limestone on northern Kuiu Island (Muffler, 1967). The Bay of Pillars Formation consists dominantly of graywacke, mudstone, and calcareous mudstone, with subordinate limestone, conglomerate, and mafic to intermediate volcanic flows, breccia, and tuff (Brew and others, 1984 ). The age of the unit is based on a few graptolite collections that range from middle Llandoverian to early Ludlovian (late Early to Late Silurian) (C. Carter, in Brew and others, 1984 ). The stratigraphic position of the graptolite-bearing rocks is unknown. The graywacke of the Bay of Pillars Formation is locally volcaniclastic, calcareous, or quartzofeldspathic (Brew and others, 1984 ). Rapid lateral and vertical compositional variations, poor sorting, and angular grains indicate sediment immaturity and local sources. All clast types can be tied to local sedimentary, volcanic, or plutonic sources in coeval or adjacent units. Distinctive pink to red syenite cobbles up to 50 em in diameter are similar to nearby syenitic plutons that have yielded earliest Silurian U-Pb ages (Gehrels and Saleeby, 1987b). Meter-scale blocks of fossiliferous limestone in slump deposits within the unit are similar to the Heceta Limestone, an adjacent and coeval Silurian unit. Intercalated volcanic flows, agglomerate, and breccia indicate contemporaneous volcanic activity. The sedimentary strata of the Bay of Pillars Formation were mainly deposited as turbidites, debris flows, olistostromes, and hemipelagic sediments. The turbidites are identified by full and partial Bouma (1962) sequences; graded bedding is ubiquitous. Soft-sediment deformation of beds is common. Debris flows are poorly organized, poorly sorted, matrix-supported conglomerate and breccia, with abundant internal soft-sediment deformation. Olistostromes consist of matrix-supported meter-scale angular blocks and meter-scale soft-sediment slump structures in the matrix. There is no cleavage or evidence of tectonic deformation in these deposits. Hemipelagic sediments consist of calcareous mudstones and limestones adjacent to, truncated by, and incorporated as blocks in the turbidites, debris flows, and olistostromes. The hemipelagic mudstones and limestones are more carbonaceous than the mudstones in the turbidite beds, and they tend to be deposited as planar beds with a uniform grain size. They are generally not turbiditic but contain rare, thin turbidite beds. The mudstones are black and the interbedded limestones are very light gray, providing a dramatic striped appearance that contrasts with the massive brown sandstone turbidites; consequently, they appear to represent a separate depositional facies. They are widely interspersed with the turbidites and are generally a few meters thick at most localities. In some places these beds are slumped as units into asymmetric, irregular folds, suggesting that they were deposited on a slope. For these reasons they are interpreted as slope or slope-basin deposits cut by the turbidite channels. Measured sections and numerous field stations on Kuiu Island indicate sandstone/shale ratios consistantly Sedimentology of the Bay of Pillars and Point Augusta Formations, Alexander Archipelago

A N n 12 MEASUREMENTS B N 24 MEASUREMENTS N n 12 MEASUREMENTS Geologic Studies in Alaska by the U.S. Geological Survey, 1991 in excess of 1 and commonly as high as 50: 1. Beds range up to 12 m in thickness but average about 20 em. Meter-scale channels and amalgamated beds are common. Both thickening-upward and thinning-upward sequences in cycles 1- to 10-m thick are recorded. Thinning-upward cycles are much more common, suggesting channel migration and (or) subsidence (Normark and Piper, 1969; Nelson, 1975). Paleocurrent studies do not indicate a clearly dominant direction of flow (fig. 3). There appears to be a few more measurements that fall in the northwest quadrant of the rose diagram for the asymmetric ripples (fig. 3A) and the flute casts (fig. 3B), but the most reasonable interpretation for the data set is radial current flow. This interpretation is consistent with the interpretation of a paleogeographic setting of volcanic islands for these lower Paleozoic rocks. Turbidite facies analysis according to the model of Mutti and Ricchi-Lucci (1972) indicates dominantly inner-fan to midfan and slope facies deposits. Inner-fan facies include both matrix- and clast-supported conglomerate, locally with meter-scale crossbedding that suggests shallow-water deposition. Some conglomerates contain meter-scale angular or rounded boulders, indicating very nearby sources. Channel deposits consist dominantly of sandstone in graded or amalgamated beds of Bouma (1962) Tab and T ac units. Deposits interpreted as overbank and interchannel facies are generally more calcareous and finer grained than channel facies deposits. These beds consist of Bouma (1962) Tbe and Tee units. Centimeter-scale lenses of micritic limestone are present in many beds. The calcareous mudstone and micritic limestone may in part represent background hemipelagic sedimentation. These sediments may be related to prolific calcareous productivity or debris associated with nearby reefs represented by the adjacent and coeval Heceta limestone. Slump deposits and fine-grained black and light-gray hemipelagic calcareous and carbonaceous mudstone and limestone (described above) are interpreted to represent slope facies deposits between channel systems. These rocks of the Bay of Pillars Formation, with intercalated volcanic flows and reefs of the adjacent Heceta Formation (Brew and others, 1984 ), have been interpreted to represent a sedimentary apron around volcanic islands with fringing limestone reefs (Eberlein anq others, 1983); this study concurs with that interpretation. Sandstone compositions in the Bay of Pillars Formation vary greatly laterally and vertically on a scale of tens of meters. The three dominant types of sandstone include (1) calcareous graywacke containing carbonate Figure 3. Rose diagrams of rotated paleocurrent directions for (A) asymmetric ripples, (8) flute casts, and (0 bidirectional groove casts measured in Bay of Pillars Formation. n, number of samples.

clasts and fossil fragments, with subordinate feldspar, quartz, chert, radiolarian mudstone, chalcedony, volcanic rock fragments, and a patchy, recrystallized calcite matrix; (2) volcaniclastic graywacke containing felted intermediate to mafic volcanic rock fragments with subordinate plagioclase, monocrystalline, embayed quartz, rare felsite, epidote/clinozoisite, chlorite, clinopyroxene, amphibole, mudstone, red, green, gray or black chert, occasional fossil fragments, and a clayey to chloritic matrix; and (3) much less abundant quartzofeldspathic graywacke containing mudstone, mono- and polycrystalline quartz, plagioclase, potassium feldspar, detrital biotite, rare myrmekite, epidote/clinozoisite, and a locally calcareous or clayey matrix. Matrix of these sandstones ranges up to 35 percent of the sample. Secondary chlorite, white mica, pumpellyite, and pyrite were also identified in the various types of graywacke. Q Sandstone samples of medium grain size (0.5 mm) were point counted (table 1) and compositions were plotted on Dickinson and Suczek (1979) diagrams. Results indicate that sandstones from the Bay of Pillars Formation are feldspatholithic to lithic and relatively low in quartz content. They fall in fields of magmatic-arc provenances (fig. 4), with arc orogen sources (fig. 5). Granitic components are greatly subordinate to volcanic components (fig. 6), but indicate a dissected arc source. Point Augusta Formation The Upper(?) Silurian Point Augusta Formation (Loney and others, 1975) has been mapped on Chichagof and smaller islands. Lithologically similar, age-equivalent deposits on the Chilkat Mountains, west of the EXPLANATION !:::. Point Augusta Formation 0 Bay of Pillars Formation Recycled-orogen provenances Figure 4. Quartz (Q)-feldspar (F)-I ith ics (l) diagram with provenance fields (after Dickinson and Suczek, 1979) showing compositions of point-counted sandstones from Bay of Pillars Formation, Point Augusta Formation, and Silurian sandstone of Chilkat Mountains (included with Point Augusta Formation; see table 2). Solid symbols indicate mean values for each unit. Sedimentology of the Bay of Pillars and Point Augusta Formations, Alexander Archipelago

Chatham Strait fault (fig. 1 ), were correlated with the Point Augusta Formation by Loney and others (1975) and Brew and Ford (1985), although these rocks have not formally been included in the unit. The lower contact of the Point Augusta Formation has not been observed; the unit grades laterally and vertically to the Silurian and (or) Devonian reefal Kennel Creek Limestone (Loney and others, 1963). The Point Augusta Formation consists dominantly of graywacke, argillite, and limestone turbidites, with subordinate conglomerate and limestone. Late Silurian graptolites have been recovered from the Chilkat Mountains and used to date the Point Augusta Formation (Loney and others, 1963, 1975). Distinctive pink to red syenite clasts in Point Augusta conglomerates have been used to correlate the Point Augusta Formation with the Bay of Pillars Formation (Ovenshine and Brew, 1972; Brew and others, 1984). Qp Q Syenites on eastern Chichagof Island at Sitkoh Bay have yielded Silurian K-Ar ages (Loney and others, 1975), but some bodies intrude Upper Devonian Freshwater Bay Formation (Ford and others, 1990a). However, the older phases of the syenite suite are still inferred to be the likely source of cobbles in conglomerates of the Point Augusta Formation (Ford and others, 1990b ). It is possible that the red syenite bodies on southern Prince of Wales Island may be a source of the red syenite cobbles in both the Bay of Pillars and Point Augusta Formations .. Paleocurrents measured from this unit do not indicate any dominant flow direction (fig. 7) and support an interpretation of radial current flow. Sandstone/shale ratios in the Point Augusta Formation are high, generally in excess of 20: 1, as shown by the measured section from the type locality at Point Augusta on Chichagof Island in figure 8. Turbidite facies analysis EXPLANATION Point Augusta Formation 0 Bay of Pillars Formation Subduction-complex sources Lv Ls FigureS. Polycrystallinequartz (Qp)-volcanic lithic grains(Lv)-sedimentary lithic grains (ls) diagram (after Dickinson and Suczek, 1979) showing compositions of point-counted sandstones from Bay of Pillars Formation, Point Augusta Formation, and Silurian sandstone of Chilkat Mountains (included with Point Augusta Formation; see table 2). Solid symbols indicate mean values for each unit. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

according to the model of Mutti and Ricchi-Lucci (1972) indicate dominantly mid- to inner-fan facies depositon. Conglomerates are mostly matrix supported; depositional cycles are mostly thinning-upward cycles. Interchannel deposits are black carbonaceous and calcareous mudstones alternating on a centimeter scale with light-colored limestone (fig. 9), identical to the striped rocks described from the Bay of Pillars Formation. Sandstones in the Point Augusta Formation and the Silurian sandstones in the Chilkat Mountains are mainly calcareous graywacke containing clasts (in decreasing order of abundance) of volcanic rock fragments, mudstone and radiolarian mudstone, plagioclase, mono- and polycrystalline quartz, calcite, shelly debris, algae, chert, potassium feldspar, rare perthite, detrital biotite, epidote, chlorite, and glauconite(?) in a generally calcareous matrix (table 2). There are also some graywackes with a tuffaceous matrix that contain volcanic rock fragments, felsite, monocrystalline quartz, mudstone, calcite, chert, epidote, and chlorite. Loney and others (1975) described one locality at False Bay on Chichagof Island where calcareous and chloritic layers alternate. These beds could reflect calcareous background sedimentation interrupted by sediment from volcanic events. Secondary minerals in both types of graywacke include chlorite, white mica, prehnite, pumpellyite, and pyrite. Sandstone samples of medium grain size (0.5 mm) were point counted and plotted on ternary diagrams (after Dickinson and Suczek, 1979). Results suggest that these sandstones have magmatic-arc provenances (fig. 4) with arc orogen sources (fig. 5), similar in composition to the Bay of Pillars Formation. The Point Augusta Formation and Chilkat Mountains sandstones show a slightly higher sedimentary lithic component, a result that reflects the observed more calQm EXPLANATION t:J. Point Augusta Formation 0 Bay of Pillars Formation Increasing ratio of plutonic/volcanic components in magmatic-arc provenances P K Figure 6. Monocrystalline quartz (Qm)-plagioclase (P)-potassium feldspar (K) diagram (after Dickinson and Suczek, 1979) showingcompositionsofpoint-counted sandstones from Bay of Pillars Formation, Point Augusta Formation, and Silurian sandstone in ChilkatMountains (included with Point Augusta Formation; see table 2). Solid symbols indicate mean values for each unit. Sedimentology of the Bay of Pillars and Point Augusta Formations, Alexander Archipelago

A N n=1 B n=4 MEASUREMENT Geologic Studies in Alaska by the U.S. Geological Survey, 1991 careous composition of the graywackes relative to graywackes of the Bay of Pillars Formation. Point Augusta sandstones also have a higher proportion of plutonic components than those of the Bay of Pillars Formation. The Point Augusta sandstones and sandstones of the Chilkat Mountains are so similar in composition (table 2), and in bedding habit (figs. 8, 10) that the Point Augusta Formation is here extended to include the sandstones of the Chilkat Mountains, as suggested by the correlation of Loney and others (1975), and Brew and Ford (1985). Discussion The petrography of the Bay of Pillars and Point Augusta Formations shows that the two units are very similar in composition, but that the Point Augusta Formation has generally more quartz, a greater sedimentary lithic component, a smaller volcanic lithic component, and a greater plutonic component than the Bay of Pillars Formation (figs. 4-6). Although sedimentary sources for these units are interpreted to be local, these observations may indicate a slightly more uplifted and dissected source for the Point Augusta Formation. Several possible explanations could allow for these differences. Paleocurrent directions appear to be radial for both the Bay of Pillars and Point Augusta Formations. Thus, there is no evidence to indicate the precise spatial relation between the two units when they were originally deposited. A hypothesis to explain the slight compositional differences might be that the Point Augusta Formation sandstones were separated from the main volcanic sources to the east by carbonate reefs, whereas the Bay of Pillars Formation may have been deposited closer to volcanic sources. Although there are massive channel conglomerates in the Point Augusta Formation, overall the beds are not as thick, sandstone/shale ratios are slightly lower, and a larger proportion of outcrops are interpreted as midfan facies turbidites than those in the Bay of Pillars Formation. For these reasons, it is possible that the two units were spatially related, with the Point Augusta Formation more distal to the same source as the Bay of Pillars Formation. Alternatively, the Point Augusta Formation could have been deposited along strike in the same island chain, adjacent to extinct or dormant volc:anoes, with compositional differences reflecting local sources. Still another possibility, based on the slightly higher granitic component of the Point Augusta Formation, is that it overlaps in age with the Bay of Pillars Formation but ranges slightly younger in the Silurian, which is unrecognized owing to very sparse age control in both units. Figure7. Rosediagramsof15 rotated paleocurrentdirectionsfor (A) groove casts, (B) flute casts, and (Q asymmetric ripples measured in Point Augusta Formation. n, number of samples.

Ill a. 3'

6' 5" OQ :r

PI" "' :I a.

OQ c

6' :I

a :I a.

-6'

massive eacb-a Base of 85SK028 boudinaged ~·:.::·::·:·:j Iamey 81i!i!!',~ Top of 85SK028 18ifli Figure 8. Measured section from Point Augusta Formation at its type locality on eastern Chichagof Island, showing typical bedding thicknesses and sandstone/shale ratios; section interpreted as midfan facies using model of Mutti and Ricchi-Lucci (1972). Solid color is mudstone; dotted pattern is sandstone. Scale in meters. Letters a, b, c, d, e refer to parts of the Bouma (1962) sequence present for each bed-beds extending to "a" contain full Bouma sequences.

Although petrologic analysis of these units clearly indicates magmatic-arc provenances, the quantity of granitic material present is significant (tables 1 and 2) and suggests basement to the Silurian Alexander magmatic arc is more evolved than would be expected for oceanic crust. Therefore, it seems improbable that this Silurian volcanic arc was a primitive intraoceanic arc as suggested by Samson and others (1989). Because paleontologic analyses indicate that Silurian reef-building faunas of the Alexander volcanic arc were isolated from but related to North America (Soja, 1991), the arc must have been located at some distance from North America; however, this scenario does not preclude the possibility that the Alexander terrane originally rifted away from North America or some nearby part of Pangea. boudin 4 lenses boudin lenses boudin lens boudins boudins boudins lens CONCLUSION Both the Silurian Point Augusta and Bay of Pillars Formations were deposited as sedimentary aprons to volcanic islands with fringing carbonate reefs, but they are not lithologically identical. There is much overlap in bedding styles, sandstone compositions, and depositional environments. However, the Point Augusta Formation is more calcareous in nature, with slightly more mature or dissected arc sources than the Bay of Pillars Formation. The Silurian Chilkat Mountains sandstone is virtually identical in composition and depositional facies to the Point Augusta Formation, and the Point Augusta Formation is herein geographically extended to include these rocks. The Bay of Pillars Formation has a greater volcanic source and lens layer lens boudins lens ~r-Lr.&.rf lens lens layer lens s lenses boudin3 Top of 85SK026 boudin ..., , boudin lens nodules boudins Base of 85SK026 Figure 9. Measured section from Point Augusta Formation on northeastern Chichagof Island, showing typical bedding thicknesses for interchannel-slope facies rocks. Similar sections are also common in Bay of Pillars Formation. Solid color is mudstone; brick pattern is limestone. Arrow indicates up direction. Scale is in meters. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Ill a. §' tD

a ::r tD "'

Di" :it Dol &.

Q a·

fll ;r "'" Q 3.

6' :;,

;;" M Dol :;,

n ::r -g· K Q

·- edca-b Base of 85SK036 :~mrttl)J Top of 85SK036 Figure 10. Measured section from Silurian sandstone in southern ChilkatMountains. Solid color is mudstone, dotted pattern is sandstone, brick pattern is limestone, "V" pattern is andesite or dacite sill. Scale is in meters. Letters a, b, c, d, e refer to parts of the Bouma (1962) sequence present for each bed.

is represented by more proximal, inner-fan facies deposits than the Point Augusta Formation. Sediment compositions and facies relations clearly indicate that these units are related, but the exact lateral or vertical relationship between the units is not fully constrained. Estimates of 150-200 km of right-lateral offset on the Chatham Strait fault (Ovenshine and Brew, 1972; Hudson and others, 1982) based on correlation of Silurian clastic rocks are justified, but there is not an adequate marker for precise measurement of offset. Acknowledgments.-Special thanks go to Dave Brew for field support for turbidite studies on the Petersburg and Juneau projects. Thanks also to Dave Brew, Sue Douglass, Art Ford, Glen Himmelberg, Sue Hunt, Rich Koch, Eric Lundin, and Willie Nelson for field collaboration and assistance in measuring sections, as well as to reviewers Julie Dumoulin and Peter Haeussler. REFERENCES CITED Armstrong, R.L., 1985, Rb-Sr dating of the Bokan Mt. granite complex and its country rocks: Canadian Journal of Earth Sciences, v. 22, p. 1233-1236. Berg, H.C., Jones, D.L., and Coney, P.J., 1978, Map showing tectono-stratigraphic terranes of southeastern Alaska and adjacent areas: U.S. Geological Survey Open-File Report 78-1085, 2 sheets, scale 1:1,000,000. Bouma, A.H., 1962, Sedimentology of some flysch deposits: Amsterdam, Elsevier, 168 p. Brew, D.A., and Ford, A.B., 1985, Preliminary reconnaissance geologic map of the Juneau, Taku River, Atlin, and part of the Skagway 1:250,000 quadrangles, southeastern Alaska: U.S. Geological Survey Open-File Report 85-395, 2 sheets, 23 p. Brew, D.A., and Karl, S.M., 1988, A reexamination of contacts and other features of the Gravina belt, southeastern Alaska, in Hamilton, T.D., and Galloway, J.P., eds., Geologic studies in Alaska by the U.S. Geological Survey during 1987: U.S. Geological Survey Circular 1016, p. Brew, D.A., Ovenshine, A.T., Karl, S.M., and Hunt, S.J., 1984, Preliminary reconnaissance geologic map of the Petersburg and parts of the Port Alexander and Sumdum 1:250,000 quadrangles, southeastern Alaska: U.S. Geological Survey Open-File Report 84-405, 2 sheets, 43 p. Clough, J.G., and Blodgett, R.B., 1989, Silurian-Devonian algal reef mound complex of southwest Alaska, in Geldsetzer, H.H., Jr., James, N.P., and Tebbutt, G.E., eds., Reefs, Canada and adjacent area: Canadian Society of Petroleum Geologists, Memoir 13, p. 404-407. Dickinson, W.R., 1970, Interpreting detrital modes of graywacke and arkose: Journal of Sedimentary Petrology, v. 40, no. 2, p. 695-707. Dickinson, W.R., and Suczek, C.A., 1979, Plate tectonics and sandstone compositions: American Association of Petroleum Geologists Bulletin, v. 63, p. 2164-2182. Eberlein, G.D., Churkin, M., Jr., Carter, C., Berg, H.C., and Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Ovenshine, A.T., 1983, Geology of the Craig quadrangle, Alaska: U.S. Geological Survey Open-File Report 83-91, 28 p. Ford, A.B., Brew, D.A., and Koch, R.D., 1990, Alkalic plutonism of the Sitkoh Bay area (Chichagof Island), SE Alaska-Perplexities of age relations [abs.]: Eos (American Geophysical Union Transactions), v. 71, no. 43, p. 1699. Ford, A.B., Brew, D.A., and Loney, R.A., 1990, The Sitkoh Bay alkalic plutonic suite: Silurian or older magmatism on eastern Chichagof Island, southeastern Alaska: U.S. Geological Survey Open-File Report 90-297, 10 p. Gardner, M.C., Bergman, S.C., Cushing, G.W., MacKevett, E.M., Jr., Plafker, G., Campbell, R.B., Dodds, C.J., McCellend, W.C., and Mueller, P.A., 1988, Pennsylvanian pluton stitching of Wrangellia and the Alexander terrane, Wrangell Mountains, Alaska: Geology, v. 16, p. 967-971. Gehrels, G.E., and Saleeby, J.B., 1987a, Geologic framework, tectonic evolution, and displacement history of the Alexander terrane: Tectonics, v. 6, no. 2, p. 151-173. ---1987b, Geology of southern Prince of Wales Island, southeastern Alaska: Geological Society of America Bulletin, v. 98, p. 123-137. Haeussler, P.J., Coe, R.S., and Onstott, T.C., in press, Paleomagnetism of the Late Triassic Hound Island volcanics-Revisited: Journal of Geologic Research. Hillhouse, J.W., and Gromme, C.S., 1980, Paleomagnetism of the Triassic Hound Island volcanics, Alexander terrane, southeastern Alaska: Journal of Geophysical Research, v. 85 p. 2594-2602. Hudson, T., Plafker, G., and Dixon, K., 1982, Horizontal offset history of the Chatham Strait fault, in Coonrad, W.L., ed., The U.S. Geological Survey in Alaska: Accomplishments during 1980: U.S. Geological Survey Circular 844, p. Karl, S.M., and Hunt, S.J., 1983, Stratigraphy and turbidite facies associations in the Bay of Pillars Formation, southeastern Alaska, in New developments in the Paleozoic geology of Alaska and the Yukon: Anchorage, Alaska, Alaska Geological Society Symposium, Proceedings, p. 18. Lathram, E.H., 1964, Apparent right-lateral separation on CSF, southeastern Alaska: Geological Society of America Bulletin, v. 75, no. 3, p. 249-252. Loney, R.A., Brew, D.A., Muffler, L.J.P., and Pomeroy, J.S., 1975, Reconnaissance geology of Chichagof, Baranof, and Kruzof Islands, southeastern Alaska: U.S. Geological Survey Professional Paper 792, 105 p. Loney, R.A., Condon, W.H., and Dutro, J.T., Jr., 1963, Geology of the Freshwater Bay area, Chichagof Island, Alaska: U.S. Geological Survey Bulletin 1108-C, p. C1-C51. Muffler, L.J.P., Jr., 1967, Stratigraphy of the Keku Islets and neighboring parts of Kuiu and Kupreanof Islands, southeastern Alaska: U.S. Geological Survey Bulletin 1241-C, 52 p. Mutti, E., and Ricchi-Lucci, F., 1972, Letorbidite dell 'Appennino setentrionale-Introduzione all' analisi de facies: Societa Geologica Italiana Memoir, v. 11, p. Nelson, C.H., 1975, Turbidite fans and other base-of-slope deposits, in Dickinson, W.R., ed., Current concepts of

depositional systems with applications for petroleum geology: San Joaquin Geological Society Short Course, p Normark, W.R., and Piper, D.J.W., 1969, Deep-sea fan valleys, past and present: Geological Society of America Bulletin, v. 80, p. 1859-1866. Ovenshine, A.T., and Brew, D.A., 1972, Separation and history of the CSF, southeast Alaska, North America: International Geological Congress, 24th, Montreal, 1972 Proceedings, sec. 3, p. 245-254. Samson, S.D., McCelland, W.C., Patchett, P.J., Gehrels, G.E., and Anderson, R.G., 1989, Evidence from neodymium isotopes for mantle contributions to Phaneozoic crustal genesis in the Canadian cordillera: Nature, v. 337, no. 6209, p. 705-709. Savage, N.M., 1988, Devonian faunas and major depositional events in the southern Alexander terrane, southeastern Alaska, in McMillan, N.J., Embry, A.F., and Olan, D.J., eds., Devonian of the world: Cana~ian Society of Petroleum Geologists Memoir 14, v. 3, p. 257-264. Soja, C.M., 1991, Origin of Silurian reefs in the Alexander terrane of southeast Alaska: Palaios, v. 6, no. 2, p. 111-125. Van der Voo, R., 1989, Paleomagnetism of North America: The craton, its margins, and the Appalachian belt, in Pakiser, L.C., and Mooney, W.D., eds., Geophysical framework of the continental United States, Geological Society of America Memoir 172, p. 447-470. Reviewers: julie Dumoulin and Peter Haeussler Sedimentology of the Bay of Pillars and Point Augusta Formations, Alexander Archipelago

Depositional Environments and Some Aspects of the Fauna of Middle Ordovician Rocks of the Telsitna Formation, Northern Kuskokwim Mountains, Alaska By Elizabeth A. Measures, David M. Rohr, and Robert B. Blodgett Abstract The northern Kuskokwim Mountains of west-central Alaska contain a thick sequence of Ordovician carbonate rocks, including the type section of the Telsitna Formation, which is in the Medfra D-2 quadrangle. Silicified, fossiliferous intervals from the Middle Ordovician part of the formatio·n were sampled extensively in 1991. Two limestone intervals were measured and sampled in detail. The limestones of the two measured sections are fundamentally different: section 1 is dominated by peloid-rich packstone and contains an abundant, silicified macrofauna; section 2 is dominated by recrystallized ooid, ·ostracode grainstone. Section 1 contains skeletal wacke-mudstone, peloid grain-packstone, and coated-grain, ooid grainstone; this composition indicates deposition in normal marine, middle-shelf conditions, including some inner-shelf shoals analogous to peloidal platform sediments of the Bahamas. The macrofauna of this section and lithologically similar underlying strata are dominated by gastropods and brachiopods. Gastropods from this limestone include abundant macluritid opercula and the carrier shell Lytospira. Brachiopod genera present include Doleroides, Austinella, Strophomena, Macrocoelia, n. gen. aff. Macrocoelia, and an undetermined rhynchonellid. Conodonts and brachiopods recovered in 1991 indicate a llandeilian age for at least part of this section. Section 2 contains gastropod packstone, ooid and ostracode grainstone, mudstone, intraclast packstone, and peloid, intraclast grainstone. These lithologies .indicate deposition in a restricted marine, · shallowwater, inner-platform setting. Ooid shoals and beaches grade into a lagoon and then into tidal flats. We conclude that the Bahamian tidal flat model is a close analog to deposits of section 2. We propose that the deposits described represent formation under humid conditions similar to those found in the Bahamas today. This is the first study of carbonate petrology of Middle Ordovician strata from west-central Alaska (Nixon Fork terrane), and the results derived here will be useful in comparing and contrasting these strata with other better known Middle Ordovician stratal sections elsewhere in Alaska and the Western Cordillera of North America in order to determine displacement of these rocks. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 INTRODUCTION The northern Kuskokwim Mountains of west-central Alaska contain a thick sequence of Ordovician carbonate rocks. Approximately 120 km northwest of McGrath, in the Medfra D-2 quadrangle, is a north-trending ridge adjacent to the Telsitna River. It is informally known among geologists familiar with the region as Telsitna Ridge. Rocks on this ridge have been included by Dutro and Patton (1982) in their 2,000-m-thick Telsitna Formation, which they reported to be of Middle and Late Ordovician. age. A sparse silicified fauna is present, and collections by Dutro in 1977 and 1979 and by Blodgett in 1984 and 1985 have been published (Rohr and Blodgett, 1988; Rohr and others, 1991, 1992). In 1991, fossiliferous intervals at Telsitna Ridge were re-collected to further document the nature of the fossil assemblages. Detailed sections were measured to determine the microfacies of two adjacent, unnamed subunits ("members") of the Telsitna Formati~n. Rocks included in the Telsitna Formation comprise part of the Nixon Fork terrane of Patton ( 1978), which is characterized primarily by lower and middle Paleozoic platform carbonate rocks. To the south and east this assemblage grades into equivalent, deeper water basinal strata (Blodgett and Clough, 1985) that have been variably mapped as belonging to other terranes (Dillinger, Minchumina, and East Fork terranes). Decker and others (in press) have recognized that these "terranes" are genetically related to one another, and they have proposed the term "Farewell Terrane" to unite them as a single tectonic entity. Our primary purpose here is to document the carbonate microfacies found in Middle Ordovician strata of this region. Prior to this study, no published accounts existed on the carbonate petrology of Ordovician strata from either west-central or southwestern Alaska (see fig. 1 for distribution of Ordovician platform carbonate rocks in this region).

AGE Samples collected in 1991 for conodonts have provided age determination for several of the localities. Preliminary study of conodonts (Anita G. Harris, U.S. Geological Survey, written commun., December 1991) from several samples that were processed for silicified macrofossils indicates an early Middle Ordovician age. A sample from U.S. Geological Survey (USGS) locality 11063-CO, north of section 1, however, produced abundant representatives of Cahabagnathus sweeti (Bergstrom), indicating a very latest Llanvirnian to at least middle Llandeilian age. Because this locality is inferred to lie stratigraphically below the interval including section 1, it can be assumed that the rocks in section 1 are Llandeilian or slightly younger Middle Ordovician age. Brachiopods from section 1 and lithologically similar underlying beds (comprising an unnamed subunit of the Telsitna Formation) are mostly represented by new species, making age determination difficult; however, closely related species are present in the Siberian platform and eastern cratonal North America, and these are mostly of Llandeilian or early Caradocian age. Section 2 is from a sequence of strata that have been juxtaposed along a major fault opposite strata of section 1. However, strata of this interval are recognized farther to the north along the ridge, resting stratigraphically beneath the interval represented by section 1. Section 2 contains only a sparse, biostratigraphically nondiagnostic conodont fauna of Middle Ordovician age; the associated megafauna and its stratigraphic position beneath strata of River O MILES Q 1Qo KILOMETERS GULF OF ALASKA Figure 1. Distribution of Ordovician platform carbonate rocks in west-central and southwestern Alaska (modified from Churkin, 1973). Arrow indicates location of so-called Telsitna Ridge. 1, Northern Kuskokwim Mountains, Medfra quadrangle; 2, lone Mountain area, McGrath quadrangle; 3, White Mountain area, McGrath quadrangle;4, Holitnaand Hoholitna Rivers, Sleetmute and Taylor Mountains quadrangles. section 1 also indicate an early Middle Ordovician, probably pre-Llandeilian age for this interval. CARBONATE MICROFACIES AND DEPOSITIONAL SETTING Detailed sections were measured using a 1.5-m staff and Brunton compass along the two fossiliferous Middle Ordovician intervals on Telsitna Ridge (fig. 2). The sections are not complete because of structural complications and incomplete exposure, but they suggest relative lithologic uniformity throughout each interval. Surfaces of the limestone were weathered in such a way that few textures were observable, except on fresh-broken surfaces. Fresh surfaces from each bed were examined. Samples were collected at each lithologic change or every 1.5 m. Carbonate terminology established by Dunham (1962) is used to describe the carbonate rocks. Environmental interpretation from microfacies conforms to principles and models established by Wilson (1975), Fltigel (1982), and Scholle and others (1983). The limestones in the two sections are lithologically dissimilar. Section 1 is dominated by peloid-rich packsto~e and contains an abundant, silicified macrofauna (fig. 3). Section 2 is dominated by ooid, ostracode grainstone (fig. 4). The beds in section 1 are thicker than R26E R27E Figure 2. location of measured sections and fossil-collection sites on so-called Telsitna Ridge. Section 1 (see fig. 3) has its base at northeast end. Section 2 (see fig. 4) also has its base at northeast end. U.S. Geological Survey (USGS) locality 11 066CO is at east end of area of USGS locality 11 068-CO. Base map from Medfra D-2 quadrangle, 1958; contour interval 500ft. Depositional Environments of Middle Ordovician Rocks of the Telsitna Formation, Northern Kuskokwim Mountains

SECTION 1 Dolomite (in fauJt zone?) Peloid grain-packstone Coated grain, ooid meters grainstone Peloid grain-packstone Dolomite (in fault zone?) l.li; ~,,o;; I"' Skeletal wacke-mudstone Geologic Studies in Alaska by the U.S. Geological Survey, 1991 meters SECTION 2 Ooid grainstone & ostracode grainstone ~tropod packstone Intraclast packstone Peloid, intraclast grainstone Mudstone Ooid grainstone & ~tracode grainstone Intraclast packstone Ooid grainstone & ostracode grainstone Mudstone Ooid arainstone & ostracode grainstone Gastropod packstone & intraclast, _gastropod packstone ·Figure4. Columnar diagram of section 2 of Middle Ordovician carbonate rocks that form part of T elsitna Formation on socalled Telsitna Ridge. ¢:: Figure 3. Columnar diagram of section·1 of Middle Ordovician carbonate rocks that form part ofT elsitna Formation on so-called Telsitna Ridge.

those in section 2. Section 1 is composed of fairly homogeneous carbonate lithologies, whereas section 2 is composed of interbedded distinct lithologies that appear cyclic. Section 2 has been extensively recrystallized, but section 1 carbonates have only minor recrystallization fabrics. Section 1 The limestone in section 1 (fig. 3) can be grouped into the following three facies based upon petrographic and field observations: (1) skeletal wacke-mudstone, (2) peloid grainpackstone, and (3) coated-grain, ooid grainstone. Skeletal Wacke-Mudstone Facies The skeletal wacke-mudstone facies is characterized by a silicified fauna of brachiopods (Doleroides, Austinella, Macrocoelia, n. gen. aff. Macrocoelia, and Strophomena), straight nautiloids, and stromatoporoids. Silicified macluritacean gastropod opercula are present but uncommon. Trilobites, bryozoans, ostracodes, and cross sections of gastropods are also common but unsilicified. Thin, spaghettilike trace fossils are very common on bedding planes and pieces of float. Bedding is very irregular to nodular and 7 to 15 em thick. The matrix of the wacke-mudstone is composed of micrite and contains small gastropods, trilobites, bryozoans, ostracodes, and pelmatozoans (fig. 5A). Microarchitecture is well preserved in all but the gastropods, which have been replaced by coarse spar. The most abundant material is bioclastic debris, up to 1.0 mm long, but very thin and broken. These fragments are unidentifiable but resemble spicules. Mottling was observed in thin section and is the result of recrystallization of burrow fill to coarse calcite spar. The diverse fauna indicates normal marine, subtidal conditions (Wilson, 1975; Wilson and Jordan, 1983). The abundance of mud indicates that environmental energieswaves or currents-were not sufficiently high to remove the fine-grained carbonate. There is also a lack of high- or even moderate-energy sedimentary structures. These features and the abundant bioturbation mottling are indicative of middle-shelf conditions (Wilson and Jordan, 1983). Peloid Grain-Packstone Facies The peloid grain-packstone facies is characterized by both abundant macrofauna and peloids. :Macrofauna consists of silicified stromatoporoids (laminar and bulbous), large gastropod opercula, and straight nautiloids. Silicified brachiopods occur rarely. Silicified burrows and bryozoans are apparent at a few horizons. Broken surfaces display ostracodes, trilobites, crinoids, bryozoans, gastropods, and intraclasts. In thin section, the bioclasts display excellent preservation of microarchitecture (fig. 5E), although some bioclasts have micrite rims or coats of micrite. Peloids are rarely apparent on broken surfaces but are almost the exclusive allochem in thin section (fig. 5F). They are composed of homogeneous micrite in rounded, irregularly shaped bodies, 0.1 to 0.3 mm in diameter. Some lithologies contain poorly sorted allochems, smaller peloids, and larger skeletal grains. Rarely, the grainpackstone contains well sorted peloids, and these common! y lack bioclasts. Associated intraclasts, 3 mm to 2 em in length, are commonly composed of skeletal mudstone to wackestone, or peloid packstone to wackestone. Intraclasts are rounded and irregularly ovoid shaped. Matrix is commonly composed of micrite, and there is little evidence of recrystallization. This facies does contain abundant light-colored mottles. In thin section, these mottles contain peloids in a microspar matrix or even in poikilitic spar cement. Normal marine, subtidal conditions are indicated by the diverse fauna (Wilson, 1975). The peloids indicate shallow subtidal conditions, ingestion of sediment by infauna, and probable physical reworking of semilithified substrate. Low sedimentation rates may also be indicated by the abundance of peloids (Enos, 1983). Low sedimentation rates and the presence of boring organisms are indicated by the micritized rims on many allochems. Intermittent conditions of moderate to high energy are indicated by the well sorted texture of some units and the occurrence of coated grains. Deposition in normal marine, middle-shelf conditions is consistent with these features (Wilson and Jordan, 1983). Coated-Grain, Ooid Grainstone Facies The coated-grain, ooid grainstone facies is characterized by very abundant coated grains and ooids. The coated grains are up to 0.4 mm in length and ovoid in shape. They have a nucleus composed of unidentifiable skeletal debris. The coating is composed of micrite and is up to 0.2 mm thick. The ooids are spherical, 0.3 to 0.1 mm in diameter, and have concentric laminae (fig. 5B, C). Some of the laminae appear to have been micritized. Overall they are moderately well sorted. The facies also contains intraclasts. These clasts are up to 1 em in length and are composed of peloid pack-wackestone identical to the peloid grain-packstone facies. There is some microspar matrix as well as coarse spar cement. In the field, this facies is massively bedded, 1 to 2 m thick. It contains some silicified stromatoporoids and Murchisonia. No opercula which otherwise are common in all other facies of this section, are found associated with this facies. The limited fauna indicates environmental restriction (Wilson, 1975; Fliigel, 1982). The occurrence of ooids also indicates restricted conditions of higher than average marine salinity and also formation within 2 m of sea level (Fliigel, Depositional Environments of Middle Ordovician Rocks of the Telsitna Formation, Northern Kuskokwim Mountains

A B E F Geologic Studies in Alaska by the U.S. Geological Survey, 1991

1982). The coated grains also indicate shallow-marine, highenergy conditions. Deposition took place as shoals on the muddy inner to middle shelf. Depositional Sequence The massive nature of these three facies and their noncyclical nature seen in section 1 (fig. 3) indicate persistence of depositional conditions. Deep subtidal, middleshelf deposition was dominant, with minor deposition in shallow subtidal parts of the middle to inner shelf. Section 2 The limestones of section 2 (fig. 4) can be grouped into five distinct facies based primarily upon petrographic differences: ( 1) gastropod packstone and intraclast, gastropod packstone; (2) ooid grainstone and ostracode grainstone; (3) mudstone; (4) intraclast packstone; and (5) peloid, intraclast grainstone. Gastropod Packstone and Intraclast, Gastropod Packstone Facies The gastropod packstone and intraclast, gastropod packstone facies is characterized by abundant, large (up to 3 em in diameter), low-spired gastropods. In hand sample the gastropods appear to be the only allochem in this grain-supported lithology. They occur as casts and molds on broken surfaces. The shell material has been dissolved and replaced by large spar crystals. Other large allochems seen in hand sample include ostracodes and bivalves. Bedding varies from 15 to 30 em thick. The matrix is composed of an ooid packstone to ooid wackestone (fig. 6A). This is also the material that fills the whorls of the gastropods. The ooids are 0.1 to Figure 5. Photomicrographs (all under plane-polarized light) of samples from section 1, Telsitna Formation. Bar scale isO.S mm. A, Skeletal wacke-mudstone facies, 1 m above base of section. Skeletal wacke-mudstone showing coarse skeletal debris in micrite matrix. Debris includes pelmatozoan ossicles and trilobite fragments. 8, Coated grain, ooid grainstone facies, SO m above base of section. Coated grain, ooid grainstone showing ooids with concentric laminae and radial structure. Ooids (O.S mm in diameter) display micritized rims. C, Coated grain, ooid grainstone facies, also SO m above base. Coated grain, ooid grainstone showing peloidal interval. 0, Fault zone 87 m above base of section. Dolomite with relict ooids or peloids. f, Peloid grain-packstone facies, 77 m above base of section. Peloid grain-packstone with skeletal elements of brachiopods, ostracodes, and trilobites in micrite matrix with poorly defined peloids. F, Peloid grain-packstone facies, 41 m above base of section. Peloid grain-packstone showing irregular-shaped peloids of various sizes in microspar matrix. 0.2 mm in diameter. They are composed of single crystals of brownish calcite, but some concentric laminae have been preserved, indicating precursor allochem. The ooids are nearly perfectly circular and well sorted. The matrix also contains small ( 1 mm), disarticulated ostracodes. Fragments of trilobites, brachiopods, and bryozoans occur rarely in the matrix. The matrix is primarily cemented by spar crystals 5 mm in length. These spar crystals are poikilitic, but the enclosed ooids do not appear to be in optical continuity with the crystals. Poikilitic cement contains ooids that are not in grain contact but instead float in the spar. Boundaries of these ooids are not as well defined as ooids in nonpoikilitic cement. Most commonly the allochems are oriented randomly, but faint lamination may also be found. A variation of this lithology contains intraclasts. The intraclasts are composed of ooid packstones, the same lithology that composes the matrix of this facies, and also mudstones (microspar). Some edges of the intraclasts are corroded and bored. Peloids occur associated with the intraclasts. Peloids are irregular in shape, composed of gray microspar, and are 0.2 to 0.4 mm in diameter. The fauna is typical of very shallow marine, restricted conditions (Wilson, 1975). The ooids indicate formation in very shallow (within normal wave base), restricted, hypersaline, marine conditions (Fltigel, 1982). The intraclasts also indicate formation and subsequent deposition in very shallow marine conditions, supratidal to intertidal (Shinn, 1983 ). This association of allochems would be found in subtidal tidal channels. Intraclasts and gastropods are commonly concentrated in tidal channels (Shinn, 1983). The ooids probably formed offshore and were washed into channels by tides (Shinn, 1983). The grain-supported nature of these lithologies is also characteristic of tidal channels, as is the slightly muddy matrix (Wilson, 1975; Shinn, 1983). Tidal energy was high enough to move and concentrate large gastropods and intraclasts, and to transport subtidal sediment onshore, but not of a sufficient duration to winnow away the finer grain sizes. Ooid Grainstone and Ostracode Grainstone Facies The ooid grainstone and ostracode grainstone facies is characterized by its grain-supported nature, ubiquitous ooids, and laminated texture (fig. 6B). The ooids are similar to those of the gastropod packstone facies and are composed of single crystals of brown calcite with preserved concentric laminae. The ooids vary in size between 0.1 and 0.5 mm, and they are circular and moderately well sorted. Ooids compose from 95 to 75 percent of the lithology. Laminae vary in thickness from 1 em to 2 mm. Planar laminae and low-angle cross-laminae are both present and are primarily normally graded. UnDepositional Environments of Middle Ordovician Rocks of the Telsitna Formation, Northern Kuskokwim Mountains

A B D E F Geologic Studies in Alaska by the U.S. Geological Survey, 1991

common, inverse grading occurs as well as combined inverse to normal grading. Contacts vary from sharp and well defined to gradational, and from planar to undulatory. In hand sample, the facies is easily distinguished by the coarse spar cement. In some instances, broken surfaces have spar-crystal faces 2 to 4 em in length. Bedding is widely variable, from 8 to 60 em. Other allochems that occur in this facies include articulated and disarticulated ostracodes, trilobites, brachiopod debris, gastropods, bivalves, and peloids. Some beds contain as much as 25 percent ostracodes. Disarticulated ostracodes are commonly nested. Large macluritid gastropods rarely occur in this facies. They are poorly preserved and best seen on bedding planes as internal molds and casts. Cement is composed of large, poikilitic spar crystals. The larger spar crystals enclose poorly defined, well separated, "floating" allochems. Large spar crystals occur with the larger ooids. Smaller ooids commonly occur in microspar cement. Lamination is accentuated by the presence of spar associated with coarse basal layers and microspar associated with the finer upper layers. Bioturbation has disrupted laminae in several beds and creates structureless intervals. Small intraclasts are composed of ooid packstone. These intraclasts may be up to 1 em in length, tabular in shape, and well rounded. Hardgrounds are common in the facies and are seen as laminae with coarse spar cement and an abrupt upper surface. The hardgrounds do not display any evidence of boring or encrustations. Ooids indicate very shallow ( m) marine waters, within the influence of normal waves (Fltigel, 1982). They also commonly indicate slightly elevated salinity (Fltigel, 1982). The other allochems are also characteristic of restricted, hypersaline conditions (Wilson, 1975).

Figure 6. Photomicrographs (all under plane-polarized light) of samples from section 2, Telsitna Formation. Bar scale is 0.5 mm. A, Gastropod packstone facies, 0.3 m above base of section. Gastropod packstone showing matrix of ooid packstonegrainstone. Ooid in center (0.2 mm diameter) displays remnant concentric lamination and also shows cleavage traces of the calcite. 8, Ooid grainstone facies, 9 m above base of section. Ooid grainstone with well sorted ooids (0.08 mm in diameter) and well preserved ostracodes. Ooids are poorly defined and are floating in poikilitic spar cement. C, Mudstone facies, 28m above base of section. Skeletal wackestone showing trilobite and ostracode debris and a few recrystallized ooids. 0, Intraclast packstone facies, 2 6 m above base of section. Intraclast packstone with intraclasts of ooid ostracode packstone. Edge of intraclast in contact with matrix, which is ostracode ooid packstone. f, Intraclast packstone facies, 41 m above base of section. Intraclast packstone showing small clast composed of slightly peloidal mudstone. Bryozoans also present. F, Peloid, intraclastgrainstone facies, 29m above base of section. Peloid, intraclast grainstone with mudstone intraclasts and peloids. Matrix contains abundant ostracodes and trilobites. Coarse, clear spar cements allochems but darker microspar also present. The fine lamination and cross-lamination suggest deposition on the foreshore region of beaches (Inden and Moore, 1983) and intertidal levees, beach ridges, and tidal-flat overbank regions (Shinn, 1983). The sorting, nearly mud-free nature of the facies, nested ostracodes, and normal to inverse grading indicate continuous energy, with fluctuations in intensity, such as occur within the wave swash zone on beaches (Shinn, 1983). These features are common in beach deposits along with minor bioturbation and intraclasts (Shinn, 1983). Macluritid gastropods may have washed up on the beach after their death (Yochelson, 1975). Mudstone Facies The mudstone facies is characterized by micrite, which is unusual because the other four facies are extensively recrystallized. This facies is also unique because it contains mud-supported lithologies in a sequence dominated by grain-supported lithologies. Lithologies include mudstone and fossiliferous wackestone. In hand sample the facies is sometimes featureless but more commonly nodular. The nodular texture is composed of dark carbonate separated by light-brown crystalline carbonate. Bedding is very thin, 8 to 15 em, but may be as thick as 30 em. Allochems include articulated and disarticulated ostracodes, trilobites, brachiopod debris, gastropods, ooids, and bryozoans (fig. 6C). Wackestones may have random orientation of allochems because of bioturbation; alternatively, wackestones may contain intervals of abundant allochems alternating with intervals of few allochems. The mud-dominated nature and mud-supported lithology indicate conditions quite unlike those of the other four facies. This facies was deposited in a low-energy, subtidal environment. The allochems are a diverse mixture. Ooids were derived from the beach environment of the ooid grainstone facies and were washed into this environment. The fossils (abundant gastropods, ostracodes, bivalves) indicate restricted conditions. The nodular nature of the facies is the result of bioturbation and selective recrystallization of the burrow fill (Enos, 1983). Some of the fossils and mud of this facies were probably transported onto the tidal flats and through the tidal channels and contributed material to the rest of the facies. This facies is interpreted to represent a low-energy lagoon adjacent to tidal flats and beaches, which is a common association on inner shelves (Enos, 1983). Intraclast Packstone Facies The intraclast packstone facies is characterized by abundant intraclasts that are the main grain support. The intraclasts are composed of ooid grainstone and ostracode grainstone (fig. 6D, E) like those of the ooid Depositional Environments of Middle Ordovician Rocks of the Telsitna Formation, Northern Kuskokwim Mountains

grainstone facies. Hand samples appear laminated. Bedding is 15 to 30 em thick. The matrix contains many different types of allochems: articulated and disarticulated ostracodes, trilobites, ooids, brachiopod debris, pelmatozoan ossicles, bryozoans, and gastropods. Ostracodes compose up to 40 percent of allochems in the matrix. Trilobites and ooids are the next most abundant. The remaining allochems form only a minor percentage. Fossil material is· common! y 1 to 2 mm in size. This facies contains more biogenic material than other facies. The fossils are also coarser than in other facies. Poikilitic spar cement is very common. It is similar to the poikilitic spar found in the other facies. Finegrained micrite, not recrystallized to microspar, occurs as irregular blebs between allochems. This micrite also occurs commonly within articulated ostracode shells. Intraclasts are commonly deposited in tidal channels or on supratidal flats (Shinn, 1983). They are formed by cementation of adjacent intertidal and shallow subtidal lithologies. The intraclasts are lithologically similar to the ooid grainstone facies and probably represent submarine cementation of that facies and subsequent reworking. The matrix between the intraclasts is similar to that of the gastropod packstone facies. This facies represents a tidal-channel deposit close to a beach. Peloid, Intraclast Grainstone Facies The peloid, intraclast grainstone facies is characterized by abundant peloids and intraclasts within the grainsupported lithology (fig. 6F). Furthermore, both the peloids and intraclasts are composed of fine-grained micrite, material similar to that of mudstone facies. The intraclasts are mudstones and contain no allochems. They are tabular and well rounded. The peloids compose up to 25 percent of the allochems. They are irregular in shape and quite variable in size, 0.5 to 1.0 mm. The matrix contains some allochems. Trilobites and ostracodes make up approximately 20 percent of all allochems. In comparison to the other facies, this facies does not contain as diverse a fauna. Fossils are large, 1 to 2 mm, in comparison to fossils in the other facies. Spar and microspar both occur as cement between allochems. The spar is not as coarse as in the other facies and does not have the poikilitic texture. Microspar is not abundant but does occur associated with the smaller peloids. Lithologically this facies is similar to the gastropod packstone facies and the intraclast packstone facies, and it probably represents deposition in tidal channels. However, the lithology of the intraclasts is unlike those in the previously mentioned facies. The lithoclasts are lithologically similar to the lagoonal mudstone facies. The peloid, intraclast grainstone facies represents tidalchannel deposition near a lagoon. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Depositional Sequence The distribution of facies in section 2 (fig. 4) is interpreted as follows. The base of the section represents a tidal-channel deposit. It is overlain by a thick section of beach deposits, which are in tum overlain by a thick deposit of lagoonal sediment. The upper part of the section is composed of interbedded beach deposits, tidal-channel deposits, and lagoonal deposits. The thin-bedded nature of the entire section may indicate that the channels were broad, shallow features that migrated to form thin, blanket deposits. The thin beach deposits may have formed on the channel banks. Lack of typical tidal-flat features such as bird' s-eyes, mudcracks, stromatolites, and cryptalgal laminites may result from insufficient sampling or may indicate that deposition took place on the outermost parts of the tidal flat in the lower intertidal zone. Therefore, upper intertidal and supratidal conditions are not represented in the section measured. Sampling was insufficient to construct an idealized model of the interbedded or cyclic nature of the upper part of the section. Because the interval represented by section 2 was noted to underlie that of section 1, a deepening-upward succession is interpreted for these two unnamed subunits ("members") of the Telsitna Formation. Lack of adjacent, age-equivalent, described sections makes it impossible to form any conclusions about the nature of the carbonate platform or its paleogeography. It can be concluded that the Bahamian tidal-flat model is a close analogy to these deposits. Ooid shoals and beaches graded into a lagoon, which then graded into tidal flats. Restricted conditions limited the fauna and allowed formation of ooids. The deposits described here were probably formed under humid conditions similar to those found in the Bahamas today. PALEONTOLOGY Gastropods Gastropods are among the largest silicified fossils found at Telsitna Ridge. Both large shells (up to 15 em in diameter) and opercula (10 em high) are present. Previous collections indicated that several forms of "Maclurites" opercula are in the Ordovician section. Two of the purposes of the 1991 collecting were to determine if there was any systematic change in the opercula through the formation, and to determine if the opercula might have more biostratigraphic value than the shells to which they were once attached. At present, no biostratigraphic differentiation can be made within section 1 based upon opercula type. Two types of opercula are present: platelike to wedge shaped (fig. 7A-C, F, G),

and hom shaped (fig. 7 D, E). Both have a projecting prong to which the retractor muscle of the gastropod was attached. This is a common feature of all Maclurites opercula. In addition, all of the forms of the Telsitna Formation show evidence of a secondary muscle attachment at the opposite end of the interior surface from the prong. This attachment is in the form of small pits or small digitate projections. This secondary attachment point was previously reported only from M. logani Salter. Both types of opercula are found in the same beds. The shells to which they correspond are not known. Macluritacean opercula were also found at USGS localities 11 068-CO and 11 063-CO. The genus Maclurites is rather widespread during the Middle Ordovician. The gastropod genus Lytospira Koken was not known from the Middle Ordovician of Alaska previous to 1991. The genus is widespread during the Middle Ordovician, being reported from many localities in North America and the Baltics. Three large silicified specimens of the carrier shell were collected (fig .. 7 H-K). Linsley and Y ochelson ( 1973) have reviewed the record of fossil and living carrier shells (gastropods that attach foreign material to their shells). The oldest reported carrier shell is Lytospira Koken, 1896. L. norvegica Koken, 1925, has irregular attachment scars on the periphery and has been illustrated by Koken ( 1925) and Y ochelson ( 1963) from the Llanvirnian-Llandeilian of Norway. Lytospira of Ibexian age, illustrated by Sando ( 1957) and Flower (1968), do not have attachment scars. The Alaskan shell is an undescribed species of Lytospira. The shell is distinguished from other species by its large size, sharp crestal angulation, and spiral groove between two cords on the interior (fig. 7 H). The cords are in a position suggesting that they may have been muscle attachment features corresponding to columellar folds in conispiral gastropods. Evidence of cementation of foreign objects is not known from all species of Lytospira, but all specimens of this species appear to have it. The concave attachment scars on the exterior of the shell (fig. 7 K) are smooth, and none of the fC!reign material is preserved. The curvature of the attachment scar suggests either a relatively large gastropod fragment or one of the strophomenid brachiopods present in the same beds. Linsley and Yochelson (1973) have shown that some Devonian and Holocene carrier shells are quite particular about the types of shells that they cement, often choosing only those of the same species. Brachiopods Silicified brachiopods are found in abundance at certain horizons within section I, as well as in lithologically similar, underlying strata farther to the north along the ridge. Although the study of the recently acquired brachiopods is still preliminary, at least eight distinct species have been recognized from beds of this lithologic interval, including representatives of the followin& genera: Doleroides, Austinella, Macrocoelia, n. gen. aff. Macrocoelia, Strophomena, and an undetermined rhynchonellid. Many of these species are illustrated in figure 8. This moderate level of taxonomic diversity indicates relatively open-marine conditions for this interval. As noted earlier, the brachiopods show cl9sest affinities with faunas described from eastern cratonal North America and the Siberian platform. Brachiopods are almost wholly unknown from section 2, with only a few rare specimens of a single, undetermined orthoid being recognized from 13.4 to 13.7 m ( 44-45 ft) above the base of the section. The extreme low diversity of brachiopods here, usually considered to be stenohaline in environmental preference, coupled with the relatively high diversity of accompanying mollusks, suggest that the depositional environment of this section was subject to much more restricted, less open-marine conditions than section 1. Shallower water depth is also indicated by the dominance of such molluscan groups as gastropods and bivalves. Acknowledgments.-Field work by Rohr, Blodgett, and Measures at Telsitna Ridge during 1991 was supported by The National Geographic Society. Subsequent laboratory preparation of thin sections and study of both the macrofauna and microfauna were conducted at the National Center (Reston, Va.) of the U.S. Geological Survey. We are grateful to William Beebe of the Alaska Department of Natural Resources, Division of Forestry at McGrath for providing helicopter support. REFERENCES CITED Blodgett, R.B., and Clough, J.G., 1985, The Nixon Fork terrane-Part of an in-situ peninsular extension of the Paleozoic North American continent [abs.]: Geological Society of America Abstracts with Programs, v. 17, no. 6, p. 342. Churkin, M., Jr., 1973, Paleozoic and Precambrian rocks of - Alaska and their role in its structural evolution: U.S. Geological Survey Professional Paper 740, 64 p. Decker, John, Blodgett, R.B., Box, S.E., Bundtzen, T.K., Clough, J.G., Coonrad, W.L., Gilbert, W.G., Miller, M.L., Murphy, J.M., Robinson, .M.S., and Wallace, W.K., in press, Geology of southwestern Alaska, in Plafker, George, and Berg, H.C., eds., The Geology of Alaska: Boulder, Colo., Geological Society of America, Geology of North America, v. Gl.. Dunham, R.J ., 1962, Classification of carbonate rocks according to depositional texture, in Ham, W .E., ed., Classification of carbonate rocks: American Association of Petroleum Geologists Memoir 1, p. 108-121. Depositional Environments of Middle Ordovician Rocks of the T elsitna Formation, Northern Kuskokwim Mountains

Dutro, J.T., Jr., and Patton, W.W., Jr., 1982, New Paleozoic formations in the northern Kuskokwim Mountains, westcentral Alaska: U.S. Geological Survey Bulletin 1529-H, p. H13-H22. Enos, P., 1983. Shelf, in Scholle, P.A., Bebout, D.G., and Moore, C.H., eds., Carbonate depositional environments: American Association of Petroleum Geologists Memoir 33, p. 268-296. Flower, R.H., 1968, Part 1. Some El Paso guide fossils: New Mexico Bureau of Mines and Mineral Resources Memoir 22, p. 2-19. Fliigel, Erik, 1982, Microfacies analysis of limestones: New York, Springer-Verlag, 633 p. Inden, R.F., and Moore, C.H., 1983, Beach, in Scholle, P.A., Bebout, D.G., and Moore, C.H., eds., Carbonate depositional environments: American Association of Petroleum Geologists Memoir 33, p. 211-265. Koken, Ernst, 1896, Die Leitfossilien: Leipzig, C.H. Tarchnitz, 848 p. ---1925, Die Gastropoden des Baltischen Untersilurs: Academie Science Russie Memoire, Classe physicomathematique, ser. 8, v. 37, no. 1, 326 p. Linsley, R.M., and Yochelson, E.L., 1973 Devonian carrier shells (Euomphalidae) from North America and Germany: U.S. Geological Survey Professional Paper 824, 25 p. Patton, W.W., Jr., 1978, Juxtaposed continental and oceanicisland arc terranes in the Medfra quadrangle, west-central Alaska, in Johnson, K.M., ed., The United States Geological Survey in Alaska: Accomplishments during 1977: U.S. Geological Survey Circular 772-B, p. B38-B39. Rohr, D.M., and Blodgett, R.B., 1988, First occurrence of Helicotoma Salter (Gastropoda) from the Ordovician of Alaska: Journal of Paleontology, v. 62, p. 304-306. Rohr, D.M., Dutro, J.T., Jr., and Blodgett, R.B., 1991, Gastropods and brachiopods from the Ordovician Telsitna Formation, northern Kuskokwim Mountains, west-central Alaska [abs.]: Australia Bureau of Mineral Resources, Geology and Geophysics, Record 1991/47, p. 29. ---1992, Gastropods and brachiopods from the Ordovician Telsitna Formation, northern Kuskokwim Mountains, west-central Alaska, in Webby, B.D., and Laurie, J.R., eds., Global perspectives on Ordovician geology: Sixth International Symposium on the .Ordovician System, Proceedings, University of Sydney, Australia: Rotterdam, A.A. Balkema Press. Sando, W.J., 1957, Beekmantown Group (Lower Ordovician) of Maryland: Geological Society of America Memoir 68, 161 p. Scholle, P.A., Bebout, D.G., and Moore, C.H., eds., 1983, Carbonate depositional environments: American Association of Petroleum Geologists Memoir 33, 708 p. Shinn, E.A., 1983, Tidal flat, in Scholle, P.A., Bebout, D.G., and Moore, C.H., eds., Carbonate depositional environments: American Association of Petroleum Geologists Memoir 33, p. 171-210. Wilson, J.L., 1975, Carbonate facies in geologic history: New York, Springer-Verlag, 471 p. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Wilson, J.L., and Jordan, C., 1983, Middle shelf, in Scholle, P.A., Bebout, D.G., and Moore, C.H., eds., Carbonate depositional environments: American Association of Petroleum Geologists Memoir 33, p. 297-343. Y ochelson, E.L., 1963, The Middle Ordovician of the Oslo region, Norway, 15. Monoplacophora and Gastropoda: Norsk Geologiske Tidsskrift, v. 43, p. 133-213. --1975, Early Ordovician gastropod opercula and epicontinental seas: U.S. Geological Survey Journal of Research, v. 3, p. Reviewers: William W. Patton, Jr. and Anita G. Harris APPENDIX-FOSSIL LOCALITIES CITED IN TEXT (SEE FIG. 2 FOR LOCATION) Collections of R.B. Blodgett USGS locality 11063-CO [=91ABd4 (=91T98 of Rohr)]: Rubble zone of silicified brachiopods (predominantly Doleroides n. sp.) trending northwest across ridge crest in the NEXSW/.SW/. sec. 19, T. 18 S., R. 27 E., Medfra D2 quadrangle, lat 63°54 '37" N., long 153°40'32" W. Elevation 3,250 ft. Locality in same unnamed lithologic unit of the Telsitna Formation as section 1 but is situated stratigraphically lower. USGS locality 11066-CO [=91ABd2]: Rubble containing abun-. dant silicified brachiopods from northeast edge [3.0 to 6.1 m (10-20 ft) below ridge crest] of summit of hill in SEY.SW/.SW/. sec. 19, T. 18 S., R. 27 E., Medfra D-2 quadrangle, lat 63°54'29" N., long 153°40'29" W. Locality situated stratigraphically above USGS locality 11063-CO but beneath base of section 1. However, this and the preceding locality are lithologically part of the same interval within the Telsitna Formation. USGS locality 11071-CO: Silicified brachiopods recovered from 0 to 1.2 m (0--4 ft) above the base of section 1, SEY.NEXNEX sec. 25, T. 18 S., R. 26 E., Medfra D-2 quadrangle,lat 63°54'17" N.,long 153°41'06" W. Collections of D.M. Rohr USGS 11063-CO [=91T98 (=91ABd4 of Blodgett)]: See above. USGS 11068-CO [=91T99]: Summit of the ridge, 1,300-ft elevation, at the boundary between SW/. sec. 19 and NW/. sec. 30, T. 18 S., R. 26 E., Medfra D-2 quadrangle. The summit is a flat, circular area with a diameter of about 100 m. The shells are in medium- to thick-bedded, mollusk-brachiopod wackestone to packstone. Several types of macluritacean shells and opercula are found in the rocks as well as the brachiopods Austine/la, Doleroides, Macrocoelia, n. gen. aff. Macrocoelia, and Strophomena.

Figures 7 And 8

Figure 7. Middle Ordovician gastropods from USGS locality 11 068-CO, Telsitna Formation. A-C, F, G, Macluritacean operculum type 1: A-C, Interior, exterior, and side views, respectively, x1.7, of platelike operculum with large, projecting retractor muscle process at lower right edge and secondary muscle scar near upper right edge, USNM 460982. F, G, Interior and exterior views, x1.7, of a thicker specimen, USNM 460983. D, E, Macluritacean operculum type 2, USNM 460984, exterior and oblique interior views, x1.7. Note the large projecting muscle process and the smaller, digitate secondary muscle attachment point. H-K, Lytospira n. sp.: H, I, USNM 460731, oblique side view showing spiral groove and cords on interior surface, x1 .1, top view of shell with angular crest, x1 .1. K, Basal view of another shell, USNM 460732, x1.1, oblique basal view, x2.2, showing attachment scars and repaired break in shell near left edge of photograph. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Depositional Environments of Middle Ordovician Rocks of the Telsitna Formation, Northern Kuskokwim Mountains

Figure 8. Brachiopods from Middle Ordovician strata of the Telsitna Formation. A, 8, Doleroides n. sp. aff. D. panna (Andreeva), exterior and interior views of free pedicle valve, USNM 460985, x2, USGS locality 11 063-CO. C-F, Austine/la n. sp., ventral, dorsal, posterior, and lateral views of articulated specimen, USNM 460986, x2.6, USGS locality 11 071-CO. G, H, Strophomena sp., USGS locality 11 071-CO: G, Exterior view of a pedicle valve, USNM 460987, x2. H, Interior view of another pedicle valve, USNM 460988, x2. 1-K, n. gen. aff. Macrocoelia, n. sp.~ exterior, interior, and posteriorviewsofetched free brachial valve, USNM 460989, x2, USGS locality 11 066-CO. L, Macrocoelia n. sp., interiorviewofbrachial valve(on right hand of specimen, some remnants of attached portions of the pedicle valve are visible), USNM 460990, x2, USGS locality 11071-CO. 200 . Geologic Studies in Alaska by the U.S. Geological Survey, 1991

G H Depositional Environments of Middle Ordovician Rocks of the Telsitna Formation, Northern Kuskokwim Mountains

Cenozoic Uplift History of the Mount McKinley Area in the Central Alaska Range Based on Fission-Track Dating By George Plafker, Charles W. Naeser, Robert A. Zimmermann, John S. Lull, and Travis Hudson Abstract Fission-track dating of apatite and zircon from the Paleocene McKinley pluton at Mount McKinley, Mount Dan Beard, and Mount Huntington provides new data on uplift rates of this part of the western Alaska Range. Zircon ages of 52 to 39 Ma from the three mountains indicate rapid middle to late Eocene cooling-inferred to result from uplift and erosion. The apatite ages indicate relative stability during middle to late Miocene time. Rapid uplift and cooling of Mount McKinley began in the early Pliocene (about 4.2 Ma). Apatite ages (9.7-3.9 Ma) indicate that Mount McKinley has uplifted at an average rate of roughly 1.3 mm/ yr for the last 4.2 m.y. Apatite and zircon ages from the same elevations on Mounts Huntington and Dan Beard are significantly older than those on Mount McKinley. This age difference indicates that Mount McKinley has been elevated by about 1,800 m relative to Mounts Huntington and Dan Beard during the late Cenozoic. This differential movement could be due to either regional tilting or to differential uplift along unidentified intervening faults. INTRODUCTION The rugged central Alaska Range has some of the greatest local topographic relief in the world and includes Mt. McKinley, the highest peak in North America (6,193 m, 20,320 ft). A suite of 13 granitoid rock samples was collected from Mounts McKinley, Dan Beard (3,127 m, 10,260 ft), and Huntington (3,731 m, 12,240 ft) by Geoff Radford and Bill Kitson during a climbing expedition in May 1981 (fig. 1 ). The samples were taken from . a large vertical elevation range on the three mountains in order to obtain information on uplift/ cooling rates by fission-track studies of apatite and zircon. Ten of the hand-specimen-size samples yielded zircon and apatite suitable for dating; the three highest samples collected on Mount McKinley at elevations of 5,305 m, 5,854 m, and 6,128 m did not contain adequate mineral assemblages for fission-track dating (samples A, B, C, fig. 1). Fission-track samples from Mount McKinley were collected from the northern part of the McKinley pluton; Geologic Studies in Alaska by the U.S. Geological Survey, 1991 those from Mount Huntington and Mount Dan Beard are in the southern and eastern parts of the pluton, respectively (fig. 2). As described by Reed and Nelson (1977) and Lanphere and Reed (1985) mainly on the basis of float samples, the McKinley pluton is dominantly medium- to coarse-grained, hypidiomorphic-granular biotite granite. Petrographic data for the 7 samples collected from this pluton on Mount McKinley show compositional variation from leucocratic, tourmaline-bearing granite near the top of the mountain (samples A, B, C) to muscovite granite, biotite-muscovite granite, and biotite-muscovite granodiorite below (samples D, E, F, G, table 1). Ruth pluton, immediately to the east, is a large granitoid mass of weakly foliated, medium- to coarse-grained biotite with rare hornblende, coarse-grained biotite granite and granodiorite, and medium- to coarse-grained biotite and biotitemuscovite granite (Reed and Nelson, 1977). Except for the presence of reported rare hornblende in the Ruth pluton, it is essentially identical in composition with the McKinley pluton, and the two bodies likely merge at depth to form a continuous batholith. Both the McKinley and Ruth plutons are considered to be part of the early Tertiary McKinley sequence of granitoid intrusions in the Alaska Range that have a late Paleocene emplacement age (average 57.3 Ma) based on K-Ar dating of nine samples of biotite and muscovite (Lanphere and Reed, 1985). FISSION-TRACK DATA Results of fission-track dating of zircons and apatites· from the granitic rocks are presented in table 2. The analytical techniques used in this study are similar to those described in Bryant and Naeser (1991). Significance of Fission-Track Ages in Interpreting Uplift Histories The fission-track dating method is based on determination of the ratio of fission tracks spontaneously formed

in a mineral by decay of radioactive elements in the mineral to fission-tracks induced in the mineral by exposing it to a calibrated radioactive source. For a fission-track age to be geologically significant, the fission tracks must be retained once they are formed and be stable at ambient surface temperatures. Heating can cause partial to complete fading of the spontaneous tracks. Fission tracks are stable in most nonopaque minerals (opaque minerals do not retain tracks) at temperatures of 50°C or less. If a mineral is heated above a critical temperature and held there for a time, the fission tracks will begin to disappear (anneal). If the mineral is held at that temperature for a sufficiently long time, the tracks will completely disappear. The process of annealing is a time-temperature function (Fleischer and others, 1965); that is, a short time at a high temperature will have the same effect on the fission tracks as a long time at low temperature. As long as a mineral is held at or above its annealing temperature, it will not retain fission tracks. When this mineral is cooled to temperatures below which total annealing can take place, the mineral will again begin to accumulate tracks. Therefore, if an apatite or zircon is emplaced at depth in the crust where CONTOUR INTERVAL 2,000 FEET the temperature is sufficient to cause total track annealing, it would have a zero apparent age. If it is then uplifted and cooled, it will yield a fission-track age that is related to the uplift event (Naeser, 1979a). There is, however, a zone between total track loss and total track retention called the partial-annealing zone (PAZ) (Naeser, 1979a). In the case of apatite, this zone is about 30°C to 50°C wide (Naeser, 1981). If a mineral resides in this zone for any length of time, the fission tracks are only partially retained and the age calculated will be an intermediate age with no numerical significance in a strict geological context. There are several techniques that can be employed to test an apatite sample in order to determine whether or not the age from that sample is an intermediate "mixed" age or not. The best method is to determine the length distribution of confined fission tracks (Gleadow and others, 1986). In some studies, such as this one, the track densities are too low for lengths to be determined. This is in contrast to relatively slow uplift, where the blocking temperature for zircon could be around 160°C. The annealing temperature for fission tracks in apatite under conditions of geologic heating is fairly well KILOMETERS Figure 1. Topographic map showing locations of sam pies A through N from Mount McKinley, Mount Huntington, and Mount Dan Beard. Stippled pattern indicates area of exposed bedrock. Cenozoic Uplift History of the Mount McKinley Area in the Central Alaska Range Based on Fission-Track Dating

63° MAP AREA 152" 151" EXPLANATION Unconsolidated deposits (Quaternary) Granodiorite of McGonagall (Tm) and Foraker (Tf) plutons (Oligocene) Granite and quartz monzonite of the Mount McKinley sequence of Reed and Lanphere (1972) (early Tertiary). Includes McKinley and Ruth plutons (Eocene) Clastic sedimentary rocks, volcanic rocks, and minor coal of the Cantwell Formation (Paleocene) Kahiltna terrane ii Mainly slate, argillite, and graywacke (Jurassic and Cretaceous) Mystic terrane Undifferentiated flysch, volcanogenic flysch, wildflysch, pillow basalt, chert, and ultramafic rocks; shelf- and slope-deposited clastic rocks and reef limestone; and subaerial clastic sedimentary rocks (late and middle Paleozoic) Argillite, slate, sandstone, schist, and limestone (Paleozoic) Contact-Dashed where approximate; dotted where concealed Fault-Sawteeth on upper plate; dotted where concealed Figure 2. Generalized map showing geologic setting of McKinley pluton, which underlies Mount McKinley, Mount Huntington, and Mount Dan Beard (modified from Reed and Lanphere, 1972; Reed and Lanphere, 1974, fig. 2; Reed and Nelson, 1977). Location shown on figure 5. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 63°

Table 1. Summary of lithologic and petrographic data for granitoids from Mount McKinley Sample Elevation Lithology (fig. 1) A B D E F G 6,128 m (20,100 ft) 5,854 m (19,200 ft) 5,305 m (17,400 ft) 4,907 m (16,100 ft) 4,054 m (13,300 ft) 3,444 m (11.300 ft) 3,048 m (10,000 ft) Granite Granite Granite Granite Granodiorite Granodiorite Granite Granularity, texture, and structure Medium- to coarsegrained, seriate; hypidiomorphic; massive. Medium-grained, equigranular to seriate; hypidiomorphic; massive. Fine- to coarsegrained, seriate to porphyritic with K-feldspar phenocrysts; hypidiomorphic; massive. Medium- to coarsegrained. seriate; hypiodiomorphic; massive. Fine -to mediumgrained, seriate; hypidiomorphic; massive Medium- to coarsegrained, equigranular; hypidiomorphic; massive Medium- to coarsegrained; equigranular; hypidiomorphic; massive. Essential Plagioclase,

K-feldspar. biotite muscovite Pla&ioclase,

K-feldspar, muscovite K-feldspar, pla&ioclase, Q!!W, muscovite Quartz, pla&ioclase, K-feldspar. biotite, muscovite

pla&ioclase, K-feldspar, biotite, musc.ovite Plagioclase, quartz, K-feldspar, biotite Quartz, pla&ioclase, K-feldspar, biotite 1 Mineralogy Accessory Tourmaline, zircon, garnet Secondary Tourmaline, Sericite 3garnet, 3topaz, 4cassiterite,:t:a patite. :t:zircon (?) Tourmaline, Sericite garnet Apatite, zircon, 3garnet Apatite, zircon, 3garnet Apatite, zircon, tourmaline, epidote Apatite, zircon Sericite, chlorite, white mica, finegrained epidote Sericite, chlorite, calcite Chlorite, sericite, fine-grained epidote Sericite, epidote, chlorite Plagioclase Remarks An2o-An1o: patchy normal and oscillatory zoning. An15-Anw; patchy zoning, complex twining. An2o-An1o: patchy zoning. Some bent twin lamellae. An2o-An1o: zoning not apparent. Altered. An3o-An2o: oscillatory zoning. An2o-An1o; normal and oscillatory zoning. An2o-An1o; fuzzy oscillatory zoning. Partly altered. Tourmaline relatively abundant Zircon with pleochroic haloes in biotite. Tourmaline abundant(23%), brown, color zoned. No biotite or hornblende. K-feldspar is perthitic. Tourmaline Biotite mostly replaced by chlorite and white mica. Plagioclase largely replaced by sericite and e idote. Minor chlorite after biotite; sericite and calcite after plagioclase. Chlorite after biotite; sericite and epidote after plagioclase. Chlorite after biotite; sericite and epidote after plagioclase. Kfeldspar strongly erthitic. 1 Visual estimate listed in order of decreasing abundance (underline indicates approximately equal abundance). 2 Color index (C.I.); visual estimate. 3 Identified in mineral separates. 4 Identified by X-ray analysis. known (Naeser, 1979b). With slow cooling it will begin to retain tracks at temperatures of about 1 00°C. If the cooling is rapid, as in an active tectonic region, a temperature of about 120°C can be used. The annealing of fission tracks in zircon, under geological conditions, is not very well known. Limited data suggest that temperatures between about 160°C (C.W. Naeser, unpub. data) and about 230°C (Hurford, 1986) will completely anneal fission tracks in zircon over periods of time in excess of several million years. The lower value of 160°C was determined from a metamorphic complex in the Yukon-Tanana upland between the Alaska Cenozoic Uplift History of the Mount McKinley Area in the Central Alaska Range Based on Fission-Track Dating

Table 2. Fission-track data for Mounts McKinley, Huntington, and Dan Beard Sample Mineral Grains Observed Track Densities 1 Neutron 2 Age± 95 Elevation (fig. 1) (Z Zircon) Counted Fossil Induced Fluence percent Cl (A-Apatite) (1 o6 tracks/cm2) (#counted) (1 015 n/cm2) (Ma) MOUNT MCKINLEY G -Z 9.09 (1,852) A .363 ( 37) 9.50 (1,848) A .0333 ( 33) E Z 10.55 (2,980) A .106 ( 79) 0 Z 11.22 (1,927) A .0629 ( 34) 6.61 (1,346) 1.06 (1,082) 6.78 (1,317) .907 ( 898) 6.33 (1,787) 2.13 (1,578) 7.72 (1,320) .744 ( 402) 38.7 :t 4.8* 3.9 :t 1.3 40.9 :t 3.4 4.2 :t 1.4 45.5 :t 6.7* 5.8 :t 1.4 40.9 :t 4.8* 9.7 :t 3.5 3,048 m 10,000 ft 3,444 m 11,300 ft 4,054 m 13,300 ft. 4,907 m 16,100 ft MOUNT HUNTINGTON K Z 8.21 (1,179) A .956 ( 71) L Z 8.27 (1,130) A .166 ( 123) M- --- Z 7.44 (1,378) 11.20 (1 ,03 7) A .234 ( 116) 5.12 ( 734) 1.76 (1,309) 5.32 ( 727) 2.27 (1 ,685) 4.66 ( 863) 6.87 ( 636) 2.66 (1,317) 46.7 :t 4.7 7.0 :t 1.7 45.2 :t 5.5 9.4 :t 1.8 46.5 :t 4.4 45.2 :t 5.1* 11.4 :t 2.3 2,560 m 8,400 ft 2,957 m 9,700 ft 3,200 m 10,500 ft 3,414 m 11,200 ft MOUNT DAN BEARD H Z 6.15 ( 939) A .229 ( 243) 9.87 (1,600) A .244 ( 242) 9.71 (1,348) A .205 ( 124) 4.30( 656) 2.38 (2,519) 5.59 ( 906) 2.34 (2,325) 5.11 ( 710) 1.84 (1,113) 41.7 :t 4.5 12.5 :t 1.7 51.4 :t 4.7 11.8 :t 1.7 51.8 :t 7.8* 14.3 :t 2.8 2,438 m 8,000 ft 2,743 m 9,000 ft 2,987 m 9,800 ft 1 Samples irradiated in three different batches (superscripts 1,2, and 3). For irradiation #1, 2,143 tracts were counted in the standard detectors and the uncertainty in neutron tluence is 2.2 percent. For irradiation #2, 2,489 tracks were counted in the two standard detectors and a significant neutron gradient was observed across the sample; neutron tluences are interpolated between the standard values and the uncertainties and are approximately 2,8 percent. For irradiation #3, 2,899 tracks were counted, no gradient was observed, and the uncertainty in neutron tluence is 1.9 percent. 2 Age calculation formula is after Fleischer and others (1965) using the following constants: A.o 1.55 X w-10 yr-1; A f 7.03 X w- 17 yr-1 ;u2351238 0.00725; a- 580.2 barns. Asterisk indicates that sample age failed the x2 test (5 percent level). Range and the Fairbanks area that cooled very slowly (Naeser, 1981), while the higher value of Hurford (1986) is based on data from the Alps of Switzerland where the cooling rates are much higher. Zircons in the McKinley area probably record cooling from about 230°C because uplift rates are comparable to those encountered in the Alps. Zircon Fission-Track Data The zircon fission-track ages record rapid postemplacement cooling during the Eocene (fig. 3). All of these ages have large error bars that range from 7.8 Ma Geologic Studies in Alaska by the U.S. Geological Survey, 1991 to 3.4 Ma. On Mount McKinley, ages are about 46 to 39 Ma for four samples through a vertical range of 1 ,859 m; on Mt. Huntington, ages are 47 to 45 Ma for four samples through a vertical range of 640 m; and on Mt. Dan Beard, the two highest samples are 52 to 51 Ma through a vertical range of 244 m. A possible exception is on Mount Dan Beard, where samples I (about 51 Ma) and H (about 42 Ma), with vertical separation of 305 m, suggest slowing of the cooling rate during the late Eocene. Cooling is assumed to result from postemplacement uplift and erosion of the pluton. Concordancy of most of the ages from a single mountain indicates that the uplift was probably very rapid, but this cooling occurred at slightly different times on each of

these mountains. The large error bars for all the zircon samples preclude meaningful determination of uplift rates except possibly on Mt. Dan Beard, where it may be as low as 0.03 mm/yr for the interval between samples H and I (fig. 3). Apatite Fission-Track Data The apatite fission-track data provide information on the Neogene uplift history of Mounts McKinley, Huntington, and Dan Beard. We assume that cooling is a direct function of uplift and erosion. A plot of apatite fission-track age versus elevation for the three mountains is shown in figure 4. Straight lines are shown through all data points except I on Mt. Dan Beard, which we regard as probably anomalous. Alternative regression curves can be fit through the data points, but this was not done because the small number of samples from each mountain and the large error bars do not merit a more sophisticated analysis. The data from Mount McKinley show a change in slope occurring at about 4.2 Ma, with apparent uplift rates of 0.2 to 0.4 rnmlyr prior to 4.2 Ma. The elevation of this change in slope on Mt. McKinley is 3,444 m (11,300 ft). We interpret this to indicate that the I D 0 o ..

f- f- E

12 fI fIF samples above 3,750 m on Mt. McKinley were in the partial-annealing zone prior to the most recent uplift and the samples from below 3,750 m were totally annealed prior to this latest period of uplift and cooling. Similarly, we interpret the apatite ages from Mounts Huntington and Dan Beard to be from an uplifted partial-annealing zone with apparent uplift rates of 0.1 to 0.3 mm/yr. However, if the apatite ages from above 3,750 m on Mount McKinley and all of the apatite ages from Mounts Huntington and Dan Beard represent uplifted partial-annealing zones, these ages can not be used to determine uplift/cooling rates. An abrupt increase in the uplift rate on Mount McKinley occurred during the early Pliocene. The straight-line uplift rate defined by samples F and G is about 1.3 mm/yr since about 4.2 Ma and they-intercept on the graph (fig. 4) would be at -2.1 km. Thus, for an apatite closure temperature of 120°C at a depth of 5.1 km below ground surface (difference in elevation between elevation of sample G and the y intercept), the indicated geothermal gradient is about 25°Cikrn. Alternative regression curves can be fit through the Mt. McKinley apatite data that tend to shift the curve toward they-axis, thereby decreasing the average uplift rate and increasing the geothermal gradient. For example, a bestfit line between points F and G and the graph origin 0 0 o .. -

N 11

1--o--1 ::E z 10 g

r1 1 f- J a h I

// 1 H . 1 D 1 3.0 s

MT. McKINLEY MT. HUNTINGTON MT. DAN BEARD

ZIRCON FISSION - TRACK AGE, in Ma Figure]. Zircon fission-trackages versus elevation for Mounts McKinley, Huntington, and Dan Beard (locations of samples on fig. 1; fission-track data in table 2). Vertical lines are error bars for each sample. Cenozoic Uplift History of the Mount McKinley Area in the Central Alaska Range Based on Fission-Track Dating

yields an apparent uplift rate of 0.80 mrnlyr since about 4 Ma, with a geothermal gradient of 40°C/k.m. Although the geothermal gradient in this area is unknown, data from geologically comparable regions in the western Cordillera suggest it is likely to be less than 40°C/k.m and that the corresponding average uplift rate of 0.80 mrnlyr is likely to be an approximate minimum rate for Mt. McKinley. The recent uplift rate of about 1.3 mrnlyr for Mt. McKinley, although high, is considerably slower than the 10 mrnlyr rate determined for the tectonically active Nanga Parbat area in the Himalaya Mountains (Zeitler and others, 1982) or known Holocene uplift rates of up to 1 0+ mrnlyr in some areas along the Gulf of Alaska coast (Plafker and Rubin, 1978). Apatite fission-track ages comparable to those on Mount McKinley are found approximately 1,800 m lower on Mount Huntington and even lower on Mount Dan Beard. Although there is more scatter in the zircon fission-track data, they show the same general relationship. Thus, the combined apatite and zircon data indicate that Mount McKinley has undergone differential uplift of at least 1,800 m in the Neogene relative to Mounts Hun~ington and Dan Beard. This difference suggests either regional southeastward tilting or the presence of an unidentified fault between Mount McKinley and Mounts Huntington and Dan Beard, with the Mount McKinley side relatively uplifted. TECTONIC SETTING Reconnaissance geologic mapping in the study area by Reed and Lanphere (1974), Reed and Nelson (1977), and Jones and others (1983) shows that the McKinley pluton lies in a structurally complex southwest-trending zone of thrust faults that splay off the Denali fault system within and east of the study area (figs. 2, 5). Neotectonic Data The long Denali fault system marks the northern boundary of the Saint Elias and Wrangell blocks with the remainder of Alaska (fig. 5). Seismicity is low along most of the fault, and there is no evidence of historic slip. In the segment between the Wrangell block and North America, however, the fault is dominantly dextral strike slip with geologically determined Holocene slip rates of 9 to 20 mrnlyr with an overall decrease from east to west along. the northern margin of the Wrangell block (Plafker and others, 1977). The slip rate is close D N

" 1 /

K (.o.t ;' ,R/ tl! APATITE MT. McKINLEY MT. HUNTINGTON MT. DAN BEARD AGE, FISSION - TRACK in Ma Figure 4. Neogene uplift rates of Mounts McKinley, Huntington, and Dan Beard based on apatite fission-track dating (locations of samples on fig. 1; fission-track data in table 2). Vertical lines are error bars for each sample. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

to 9 mrnlyr along the Denali fault 30 km east of the map area; there is no evidence for Holocene dextral displacements west of the point where the Shellabarger fault splays off the Denali fault just north of the McKinley pluton (fig. 2). A kinematic interpretation of Holocene PacificNorth American plate interaction in southern Alaska has been made by Lahr and Plafker ( 1980) and Plafker and others (in press) based on integrated slip rates from fault, seismologic, and plate-motion data (fig. 5). This analysis suggests that the Wrangell tectonic block is rotating counterclockwise· about an axis near the southwestern end of the block. The Denali fault is the northern boundary of the block, and a diffuse zone of seismicity and inferred faulting (dot pattern on fig. 5) is inferred to bound the block on the northwest. This boundary, as defined by seismicity, extends southwest through the central Alaska Range and through Cook Inlet and Shelikof Strait to merge with the Aleutian megathrust southwest of Kodiak Island (Lahr and Plafker, 1980). As a consequence of the block rotation, dextral slip on the Denali fault is inferred to be taken up on the northwestern boundary of the block in the central and western Alaska Range by contractional folding and Figure 5. South-central Alaska showing relationship of Mount McKinley area to principal tectonic features, including Wrangell block, Dena I i fault, and main Dena I i fault splays in map area and southeast of map area. Open arrows with numerals in circles indicate direction and amount of motion of Pacific plate (6.3 em/ yr) and Wrangell block (0.5 cm/yr) relative to North America; paired facing open arrows and numerals indicate direction and rate of convergence across southeastern (Aleutian megathrust) and western (dot pattern indicating diffuse zone of seismicity) boundaries of Wrangell block; paired arrows along Denali fault and numeral indicate sense and amount of relative displacementon northwestern boundary of Wrangell block, and paired arrows along the Fairweather/Queen Charlotte faults and numeral indicate sense and amount of relative displacement between Pacific plate and Saint Elias block. Modified from Lahr and Plafker (1980) and Plafker and others (in press). faulting. The present stress regime probably dates from about 3.9 to 3.4 Ma, when there was a change to a more northerly relative motion between the Pacific and North American plates as recorded in oceanic magnetic anomalies (Harbert and Cox, 1989). Stratigraphic Record of Late Cenozoic Uplift Late Cenozoic uplift of part of the Alaska Range is suggested by the stratigraphic record in the Cook Inlet basin and in local basins on the north side of the Alaska Range. In the upper Cook Inlet region, Oligocene and Miocene units (mainly the Hemlock Conglomerate and Tyonek Formation) consist of fluvial, deltaic, and estuarine sandstone, siltstone, and coal deposited in cyclic fining-upward sequences (Kirschner and Lyons, 1973). Sandstone in these middle Tertiary units has abundant quartz and a distinctive heavy mineral assemblage that together indicate a high-grade metamorphic provenance of the type that occurs north of the Alaska Range. In contrast, the overlying middle Miocene to Pliocene sequence (Beluga and Sterling Formations) includes over 3,000 m of thick-bedded sandstone and conglomerate in which the predominant heavy-mineral fraction is hornblende derived from erosion of Alaska Range plutonic rocks and hypersthene from volcanic rocks of the Alaska Peninsula segment of the Aleutian magmatic arc. Kirschner and Lyon (1973) interpreted these data as indicating that the middle Tertiary sequence was deposited by an ancestral major river system that drained much of interior Alaska (comparable to the present Yukon and Tanana Rivers), and that this system was cut off as a result of uplift of the Alaska Range in late Miocene time. The change in sandstone petrology within the Cook Inlet region also approximately coincides with a sedimentary change in the Nenana basin north of the central Alaska Range. In this basin, early and middle Tertiary fluvial and lacustrine coal-bearing strata are unconformably overlain by as much as 640 m of conglomerate and conglomeratic sandstone containing minor lenticular interbeds of shale and lignite of the Nenana Gravel (Wahrhaftig and others, in press). The gravel was derived from the rising Alaska Range and was deposited in large alluvial fans along the north flank of the range (Kirschner, in press; Wahrhaftig and others, in press). The age of the Nenana Gravel is bracketed by isotopic dates from ash layers below and above the unit of 8.3 Ma and 2.75 Ma, respectively, and recent preliminary studies of the contained pollen assemblages suggest that it is of early Pliocene age (T. Agar, written comm., 1991 ). Thus, the formation is at least partly contempoCenozoic Uplift History of the Mount McKinley Area in the Central Alaska Range Based on Fission-Track Dating

raneous with the Beluga and Sterling Formations of the Cook Inlet basin (Wahrhaftig and others, in press). Assuming that gravel deposition began during or shortly after uplift of the Alaska Range, the early Pliocene age (between 5.2 and 3.4 Ma) is compatible with fission-track data suggesting onset of rapid uplift of Mt. McKinley at about 4.2 Ma. Furthermore, these data yield an approximate uplift rate of 1.2 to 1.8 mm/yr, assuming 6 km of differential uplift between the base of the Nenana Gravel and the present height of Mt. McKinley, and an early Pliocene age for the gravel. The calculated uplift rates bracket the rate of 1.3 mm/yr obtained for Mt. McKinley. However, these rates can only be considered crude approximations because they do not incorporate a probable time lag between onset of uplift and deposition of the gravel (which would tend to lower the calculated uplift rates), or the amount of erosion off the top of Mt. McKinley (which would tend to increase the rates). POSSIBLE CAUSE OF THE LATE CENOZOIC UPLIFT OF MOUNT MCKINLEY Stratigraphic data cited above suggest that prior to late Cenozoic time the central and western Alaska Range was not a major topographic high. The combined fission-track and stratigraphic data suggest to us that the late Cenozoic uplift was largely tectonic in origin. In the study area, the geometric relationships between the Denali fault and the northeast-trending structures that appear to splay off it on the south side would result in a contractional horizontal component of 3.1 mm/yr and a vertical component of about 1.8 mm/yr if all the strike-slip motion were taken up by slip on a single thrust fault dipping at an average angle of 30° (fig. 6). This is more than enough to account for the 1.3 mm/yr average late Cenozoic uplift rate of Mount McKinley indicated by fission-track dating. Thus~ the uplift inferred from fission-track studies could result from oblique underthrusting along one or more northeast-trending faults that splay off the Denali fault within the western Alaska Range. The abrupt increase in uplift/cooling rates for the Mount McKinley region at about 4.2 Ma could be related to a change in relative Pacific-North American plate motion between about 3.2 and 3.9 Ma that involved clockwise rotation of about 15° in the eastern Aleutian arc (Harbert and Cox, 1989; DeMets and others, 1990). The inferred rotation would result in an increased component of compression across the Denali fault in the central Alaska Range. An alternative possibility, for which there is no direct evidence, is that the accelerated uplift/ cooling at Mt. McKinley resulted from a change in the degree of coupling between the Pacific plate and the Yakutat block relative to the Wrangell block. Earthquake data suggest that active faults other than the Denali fault may be present in the Mt. McKinley area (Lahr and Plafker, 1980). However, geologic mapping in the rugged and extensively glaciated central Alaska Range has been reconnaissance in nature, and few faults other than the Denali fault have been recognized. Figure 6. Diagram iII ustrati ng how strike-slip displacement averaging 9 mm/yr on Dena I i fau It could produce 3.1 mm/yr contraction and 1.8 mm/yr uplift on a thrust splay dipping 30° northwest that intersects Denali fault at an angle of 20°. Geologic Studies in Alaska by the U.S. Geological Survey, 1991

One major thrust fault identified by Reed and Nelson (1977) and Jones and others (1983), the Shellabarger fault, extends northeast along the Kahiltna and Peters Glaciers to an acute intersection with the Denali fault (fig. 2). The Shellabarger fault has been interpreted as a terrane boundary that juxtaposes a basement complex of predominantly Paleozoic rocks on the northwest (Mystic terrane of Jones and others, 1983) against a sequence of Mesozoic flysch that comprises the Kahiltna terrane on the southeast (Reed and Nelson, 1977; Jones and others, 1983). Numerous faults and folds with comparable orientation occur east of the map area in the Healy quadrangle (Csejtey and others, 1992). Some of these faults have had late Cenozoic displacements and could represent active splays off the Denali fault (indicated schematically on fig. 5 by a dashed fault line). SUMMARY AND CONCLUSIONS Fission-track dating provides new data on the uplift history of mountains within the highest part of the Alaska Range. This area is uniquely suited for determination of uplift rates because local topographic relief is among the highest in the world, there is ample stratigraphic evidence for Neogene uplift, and the highest parts of the range are underlain by the McKinley pluton, which is made of rocks suitable for fission-track dating. The most important implications are the following: 1. The zircon data indicate an episode of rapid cooling and presumed uplift of Mounts McKinley, Huntington, and Dan Beard in Eocene time following late Paleocene emplacement of the McKinley pluton. 2. On Mount McKinley, the uplift rate increased dramatically to about 1.3 mrnlyr from 4.2 Ma to the present; data are not available to determine whether comparable uplift occurred at Mounts Huntington and Dan Beard. 3. The cause of the dramatic change in uplift rate on Mount McKinley is uncertain, although it most likely reflects an increase in contractional deformation within the western Alaska Range fold-fault system, possibly resulting from late Cenozoic changes in the Pacific-North American relative plate motion. 4. The apatite and zircon data require that Mount McKinley was elevated about 1,800 m (6,000 ft) relative to Mounts Huntington and Dan Beard during the late Cenozoic. This differential uplift may reflect either regional southeastward tilting or the presence of an unidentified up-to-the-northwest fault between Mount McKinley and Mounts Huntington and Dan Beard. Acknowledgments.-This study was made possible by the scientific interest and mountaineering skill of geologists Geoff Radford and Bill Kitson, who collected the samples described in this paper. We especially acknowledge B.L. Reed's detailed and thoughtful technical review of the manuscript. REFERENCES CITED Bryant, Bruce, and Naeser, C.W., 1991, Implications of lowtemperature cooling history on a transect across the Colorado Plateau-Basin and Range boundary, west central Arizona: Journal of Geophysical Research, v. 96, no. B7, p. 12,375-12,388. Csejtey, Bela, Jr., Mullen, M.W., Cox, D.P., and Stricker, G.D., 1992, Geology and geochronology of the Healy · quadrangle, south-central Alaska: U.S. Geological Survey Miscellaneous Investigations Series Map, 1-1961, 63 p., 2 pis., scale 1:250,000. DeMets, D.C., Gordon, R.G., Argus, D.F., and Stein, S.A., 1990, Current plate motions: Geophysical Journal International, v. 101, no. 2, p. 425-478. Fleischer, R.L., Price, P.B., and Walker, R.M., 1965, Effects of temperature, pressure, and ionization on the formation and stability of fission tracks in minerals and glasses: Journal of Geophysical Research, v. 70, p. 1497-1502. Harbert, William, and Cox, Allan, 1989, Late Neogene motion of the Pacific plate: Journal of Geophysical Research, v. 94,no. B3,p. 3052-3064. Hurford, A.J ., 1986, Cooling and uplift patterns in the Lepontine Alps south-central Switzerland and an age of vertical movement on the Insubric fault line: Contributions to Mineralogy and Petrology, v. 92, p. 413-427. Gleadow, A.J.W., Duddy, l.R., Green, P.F., and Lovering, J.F., 1986, Confined fission track lengths in apatite: A diagnostic tool for thermal analysis: Contributions to Mineralogy and Petrology, v. 94, p. 405-415. Jones, D.L., Silberling, N.J., and Coney, P.J., 1983, Tectonostratigraphic map and interpretive bedrock geologic map of the Mount McKinley region, Alaska: U.S. Geological Survey Open-File Report 83-11, s sheets, scale 1:250,000 .. Kirschner, C.E., in press, Interior basins of Alaska, in Plafker, G., and Berg, H.C., eds., Geology of Alaska: Boulder, Colo., Geological Society of America, Geology of North America, v. G 1. Kirschner, C.E., and Lyon, C.A., 1973, Stratigraphic and tectonic development of Cook Inlet Petroleum Province, in Pitcher, M.G., ed., Arctic geology: American Association of Petroleum Geologists Memoir 19, p. 396-407. Lahr, J.C., and Plafker, George, 1980, Holocene Pacific-North America plate interaction in southern Alaska: Implications for the Yakataga seismic gap: Geology,·v. 8, p. 483-486. Lanphere, M.A. and Reed, B.L., 1985, The McKinley sequence of granitic rocks: A key element in the accretionary history of southern Alaska: Journal of Geophysical Research, v. 90, no. B13, p. 11,413-11,430. Naeser, C.W., 1979a, Thermal history of sedimentary basins: Fission-track dating of subsurface rocks: Society of Economic Paleontologists and Mineralogists Special Publication 26, p. 109-112. ---1979b, Fission-track dating and geologic annealing of fission tracks, in Jaeger, E., and Hunziker, J.C., eds., LeeCenozoic Uplift History of the Mount McKinley Area in the Central Alaska Range Based on Fission-Track Dating

tures in Isotope Geology: New York, Springer-Verlag, p. ---1981, The fading of fission tracks in the geologic environment-data from deep drill holes: Nuclear Tracks, v. 5, p. 248-250. Plafker, George, and Rubin, Meyer, 1978, Uplift history and earthquake recurrence as deduced from marine terraces on Middleton Island, Alaska, in Proceedings of Conference VI, Methodology for identifying seismic gaps and soon-tobreak gaps: U.S. Geological Survey Open-File Report 78-943, p. 687-721. Plafker, George, Hudson, Travis, and Richter, D.H., 1977, Preliminary observations on late Cenozoic displacements along the Totschunda and Denali fault systems, in Blean, K.M., ed., The United States Geological Survey in Alaska: Accomplishments during 1976: U.S. Geological Survey Circular 751-B, p. B67-B69. Plafker, George, Gilpin, L.M., and Lahr, J.C., in press, Neotectonic map of Alaska, in Plafker, G., and Berg, H.C., eds., Geology of Alaska: Boulder, Colo., Geological Society of America, Geology of North America, scale 1 :2,500,000, 1 sheet and text. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Reed, B.L., and Lanphere, M.A., 1972, Generalized geologic map of the Alaska-Aleutian Range batholith showing potassium-argon ages of the plutonic rocks: U.S. Geological Survey Miscellaneous Field Studies Map MF-372, scale 1:1 ,000,000, 2 sheets .. ---1974, Offset plutons and history of movement along the McKinley segment of the Denali fault system, Alaska: Geological Society of America Bulletin, v. 85, p. 18831892. Reed, B.L., and Nelson, S.W., 1977, Geologic map of the Talkeetna quadrangle, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-870-A, scale 1:250,000. Zeitler, P.K., Johnson, N.M., Naeser, C.W., and Tahirkheli, Rashid A.K., 1982, Fission-track evidence for Quaternary uplift of the Nanga Parbat region, Pakistan: Nature, v. 298, p. 255-257. Wahrhaftig, Clyde, Bartsch-Winkler, S., and Stricker, G.D., in press, Coal in Alaska, in Plafker, G., and Berg, H.C., eds., Geology of Alaska: Boulder, Colo., Geological Society of America, Geology of North America, v. G 1. Reviewers: Marvin A. Lanphere and Bruce L. Reed

Isotopic Variations in Calcite Veins from the Kandik Region of East-Central Alaska By Kevin L. Shelton, Michael B. Underwood, Deborah Bergfeld, and David G. Howell Abstract The Kandik River and Tatonduk belts of east-central Alaska are juxtaposed along the Glenn Creek fault zone. Measurements of oxygen and carbon isotopic compositions of vein-filling calcite in each belt, together with calculations of b180water values, indicate that the parent fluids in the Kandik River section were enriched in 180 relative to fluids that moved through the Tatonduk stratigraphy. The Kandik .River fluid reservoir equilibrated with a large volume of metasedimentary and volcanic rocks at elevated temperatures (up to 300°C) and low water-to-rock ratios (that is, the veins are broadly synmetamorphic). Tatonduk veins, in contrast, precipitated at lower temperatures (generally below 150°C) from less evolved meteoric waters; these fluids most likely attained their isotopic compositions under conditions of moderate to high water-to-rock ratios and modest amounts of interaction with intrabasinal carbonate, siliciclastic, and volcanic rocks. The exchange of fluids across the Glenn Creek fault zone (in either direction) appears to have been very limited, and the isotopic compositions provide additional supporting evidence for the emplacement of a hot Kandik River hanging wall over a cooler Tatonduk footwall. Conversely, veins from basaltic rocks assigned to the Woodchopper Canyon terrane yield isotopic values similar to those of the neighboring Kandik River belt. Evidently, both of these units were affected by the same high-temperature fluids, with vein precipitation occurring after the Woodchopper Canyon terrane and Kandik River belt had been amalgamated. INTRODUCTION The Kandik region of east-central Alaska (fig. 1) has been the site of a multidisciplinary study of the effects of thrust faulting on thermal structure. As outlined by Underwood and others (this volume), the study area contains two major tectono-stratigraphic units, herein referred to as the Kandik River "belt," which is dominated by Mesozoic deep-marine strata, and the Tatonduk "belt," which is mostly Proterozoic to Permian in age (see also Churkin and others, 1982; Howell and Wiley, 1987; Dover, 1990; Howell and others, 1992). Pelitic rocks of the Kandik River belt (Glenn Shale and Biederman Argillite) typically have well-developed pressure-solution cleavage and (or) pencil structure, plus relatively high ranks of organic metamorphism; conversely, the Tatonduk belt is mildly deformed and much lower in thermal maturity (Laughland and others, 1990). In essence, the Tatonduk stratigraphy is a relatively straightforward example of a rifted continental margin that experienced thermal subsidence during the Paleozoic (Payne and Allison, 1981; Howell and Wiley, 1987; Dover, 1990). The western portion of the study area contains diverse sequences of rock that Churkin and others (1982) assigned to the Woodchopper Canyon, Slaven Dome, and Takoma Bluff terranes. Rock units within these smaller terranes include the Step Conglomerate (Permian); sequences of Paleozoic argillite, chert, limestone, and dolomite (units Pza and Pzl of Brabb and Churkin, 1969); the Woodchopper Volcanics (Devonian); and argillite, dolomite, and volcanic rocks of Proterozoic(?) age. Stratigraphic and structural relations are uncertain between the Mesozoic siliciclastic rocks of the Kandik River belt and Devonian to Permian strata that crop out in the Step Mountains region (fig. 1 ). Fault contacts separate the Kandik River belt from the Woodchopper Canyon, Slaven Dome, and Takoma Bluff terranes; Churkin and others (1982) referred to this complicated region as the Eureka suture zone. Some of the Paleozoic strata within the suture zone may have formed the depositional basement of the Kandik River belt (Howell and others, 1992). The Glenn Creek fault zone marks the structural boundary between the Kandik River and Tatonduk belts (fig. 1 ). Past field investigations of the region have resulted in conflicting interpretations of the geometry and sense of slip on the Glenn Creek fault (Brabb and Churkin, 1969; Howell and Wiley, 1987; Dover and Miyaoka, 1988; Laughland and others, 1990). Recent measurements of kinematic indicators, however, provide evidence that the fault is a southeast-verging thrust, with younger rocks of the Kandik River belt thrust over the older Tatonduk rocks (Howell and others, Isotopic Variations in Calcite Veins from the Kandik Region

1992). This orogenic event evidently occurred during late Early Cretaceous time. In-depth analyses of vitrinite reflectance and illite crystallinity (Underwood and others, this volume) uphold interpretations derived from an earlier reconnaissance-level investigation of regional thermal maturity (Laughland and others, 1990). These data are important because they provide independent confirmation of the widespread occurrence of low-grade metamorphic rocks in the hanging wall of the Glenn Creek fault; the footwall sequences, in contrast, are much lower in thermal maturity. For example, the average value of mean random vitrinite reflectance (percent Rm) is 3.8 percent for the Kandik River belt, whereas a comparable average percent Rm value for the Tatonduk belt is only 1.2 percent (Underwood and others, this volume). Using the --Contact correlation between percent Rm and paleotemperature established by Barker (1988), these data correspond to average temperature estimates of about 285°C (Kandik River) and 165°C (Tatonduk). Likewise, values of illitecrystallinity index for the Kandik River belt typically fall within the boundaries of the zone of anchimetamorphism (transition into lowermost greenschist facies). In contrast, the Tatonduk belt displays a great deal of scatter and inconsistency in illite crystallinity due to the effects of mixing among authigenic clay minerals and detrital populations of illite and muscovite inherited from a variety of source rocks (Underwood and others, this volume). A complete understanding of the regional thermal history and its relation to the history of rock deformation cannot be determined without considering the role of KANDIK RIVER BELT Indian Grave t\1ountain Fault, dashed where approximate Strike-slip fault, arrows indicate direction of relative movement 15 MILES

15 Kl LOMETERS

Thrust fault, sawteeth on upper plate +Syncline -t- Anticline Figure 1. Geologic index map of east-central Alaska showing Kandik River belt (no pattern) and Tatonduk belt (horizontal lined pattern). Thehachured linetothesoutheastoftheGienn Creek fault depicts the limit of a paleothermal anomaly detected in the Tatonduk footwall. Also shown are unit TKs of Brabb and Churkin (1969) and the Woodchopper Canyon (WC), Slaven Dome (SD), and Tacoma Bluff (TB) terranes of Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Churkin and others (1982). Rocks southwest of Tintina fault zone are assigned to Yukon-Tanana composite terrane (Coney and Jones, 1985). Modified from Brabb and Churkin (1969) and Foster (1976). For relevant fossil control and alternative interpretations of the structural geology, see Dover and Miyaoka (1988) and Miyaoka (1990). ·

fluid migration. Perhaps the best evidence for the fluidmigration history comes from the mineral phases that were precipitated as vein fillings. Veins containing calcite, quartz, and quartz-calcite intergrowths are ubiquitous in the cleaved and complexly folded sequences of the Kandik River belt. Although we made a concerted effort to sample and measure both common and unusual vein orientations (fig. 2), most of the veins are aligned at high angle to bedding and (or) cleavage, locally filling conjugate sets of extension fractures. On the other hand, calcite veins in the Tatonduk belt occur sporadically, and there is little sense of systematic orientation (fig. 2). In this paper we present initial results and tentative interpretations of isotopic analyses of calcite veins. Research in progress will establish the isotopic signature of quartz veins and microthermometric data for fluid inclusions. From these data, we expect to be able to estimate the temperatures of vein precipitation, the fluid compositions and salinities, the isotopic character of parent fluids (from which the veins precipitated), the degree of fluidrock interaction, and the ultimate source or sources of the fluids. METHODS Previous studies have shown the utility of stable isotopes in elucidating the origin and history of hydrotherN s Figure 2. Schmidt equal-area projection (lower hemisphere) of poles to veins for calcite veins in Tatonduk belt (open circles) and composite Kandik River belt and Woodchopper Canyon terrane (closed squares). Statistical distribution of data for Kandik River belt is biased in that a concerted effort was made to sample and measure unusual vein orientations. Most veins are aligned at high angle to bedding and (or) cleavage. mal fluids and fluid-rock interactions in tectonically disturbed terranes (Magaritz and Taylor, 1976; Dietrich and others, 1983; Wickham and Taylor, 1987; Rye and Bradbury, 1988; Burkhard and Kerrich, 1988; Bebout and Barton, 1989; Nesbitt and Muehlenbachs, 1989). In this study we measured the carbon and oxygen isotopic compositions of vein-filling calcites. Techniques of mineral extraction and analysis followed those of McCrea ( 1950), and the data are reported in conventional b notation as per mil deviations relative to the Pee Dee Belemnite (PDB) standard for C and the Vienna standard mean ocean water (SMOW) standard for 0. The standard error for each analysis is approximately ±0.1 ·per mil, using the University of Missouri's automated Finnigan MAT Delta E mass spectrometer. Specimens of calcite from calcite and calcite-quartz veins were extracted from a wide variety of lithologies and stratigraphic positions (table 1; fig. 3). Most of the Kandik River specimens (20) came from fractured sandstone beds in the Biederman Argillite (Cretaceous); ·in addition, we analyzed three samples from the Glenn Shale. Within the Woodchopper Canyon terrane (of Churkin and others, 1982), we selected one sample from a shale interval and four veins cutting through the Devonian basaltic rocks. We collected a total of 10 samples from the following units within the Tatonduk belt: shale and basalt of the Tindir Group (Proterozoic and Cambrian); Adams Argillite (Cambrian); argillite of the Road River Formation (Ordovician and Silurian); turbidite sequences of the Nation River Formation (Devonian); limestone and shale interbeds of the Calico Bluff Formation (Mississippian and Pennsylvanian); and the Tahkandit Limestone (Permian). All formational assignments (table 1) conform to the .maps of Brabb and Churkin (1969) and Foster (1976). RESULTS If all of the samples from the study area are considered collectively, the o180 values of calcite veins range from 4.4 to 21.3 per mil; oBc values range from -7.6 to +7.6 per mil (table 1). A comparison of the isotopic . data from the Tatonduk belt, the Kandik River belt, and the Woodchopper Canyon terrane indicates several noteworthy features (fig. 4 ). First, although the three data sets show some overlap of b Be values, veins from the Tatonduk belt are enriched in Be (average 0.1 per mil) relative to those of the Kandik River belt and Woodchopper Canyon terrane (average -3.5 per mil). Most of the overlap is associated with a single site in the Nation River Formation (MJ90-K59), where values of vitrinite reflectance are somewhat higher than normal for this particular formation (Underwood and others, this volume). Second, with one exception, calcite in veins Isotopic Variations in Calcite Veins from the Kandik Region

Table 1. Isotopic data for calcite veins, Kandik region, east-central Alaska [Values given in per mil] Tatonduk belt Sample Formation lithology1 f113C mj90-k29b Tindir shale +0.6 mj90-k77c Tindir basalt mj90-k77d Tindir basalt um90-k34b Tindir lms/shale +7.6 mj90-k32c Road River shale mj90-k33c Adams argillite +2.9 mj90-k59f Nation River ss/shale mj90-k59g Nation River ss/shale mj90-k64c Calico Bluff lms/shale mj90-k73b Tahkandit limestone +1.8 1 Abbreviations: lms, limestone; ss, sandstone. from northwest of the Glenn Creek fault show no consistent deviation in isotopic values as a function of the hostrock lithology or age; in other words, values from veins in basaltic rocks of the Woodchopper Volcanics (Devonian) are virtually identical to those from the Biederman Argillite (Cretaceous) and the Glenn Shale (TriassicCretaceous). The one exception (sample DH91-K22c) comes from a thin shale interval of the Woodchopper Volcanics (fig. 3); this specimen yielded the lowest [)180 value (4.4 per mil) and ·the highest b13C value (-0.2 per mil). Third, calcite from veins in argillites and basalts of the Tatonduk belt display consistently lower b 180 values than veins in Tatonduk limestone, sandstone, and (or) interbedded shale units (table 1 ). Thus, calcite veins from the Kandik River belt and the Woodchopper Canyon terrane display a much narrower isotopic range than veinfilling calcite in the Tatonduk belt. If reasonable inferences are made regarding temperature conditions at the time of vein emplacement, then we can demonstrate that the pore waters that deposited calcites in the two geologic domains were isotopically distinct. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Kandik River belt/Woodchopper Canyon terrane Sample Formation Lithology1 f113C mj90-k80b Biederman ss/argillite mj90-k82e Biederman ss/ argillite mj90.:k82f Biederman ss/ argillite mj90-k82g Biederman ss/argillite mj90-k82h Biederman ss/ argillite mj90-k83g Biederman sst argillite mj90-k84d Biederman ss/argillite mj90-k84e Biederman ss/argillite mj90-k85c Biederman ss/argillite mj90-k85d Biederman sst argillite mj90-k87c Biederman ss/argillite mj90-k88e Biederman ss/argillite mj90-k88f Biederman ss/argillite mj90-k88g Biederman ss/argillite mj90-k90g Biederman ss/argillite mj90-k90h Biederman ss/argillite mj90-k90j Biederman sst argillite mj90-k91d Biederman ss/argillite mj90-k91e Biederman sst argillite dh91-k24d Biederman ss/ argillite um90-k24b Glenn Shale shale um90-k25b Glenn Shale shale um90-k25f Glenn Shale shale dh91-k20 Woodchopper basalt dh91-k22c Woodchopper shale --0;2 dh91-k23a Woodchopper basalt dh91-k23b Woodchopper basalt dh91-k25b Woodchopper basalt INTERPRETATION Oxygen Isotopes Based on measurements of vitrinite reflectance (Underwood and others, this volume), it is certain that the average and the range of maximum paleotemperatures were quite different in the Kandik River and Tatonduk belts. Most of the calcite veins from the Kandik River belt have b180 values within the narrow range of 17 to 19 per mil (fig. 4 ). In comparison, Magaritz and Taylor (1976) showed that carbonate veins in a wide variety of rocks from the Franciscan Complex in coastal California produce b180 values of carbonates that concentrate between 14 and 16 per mil, and these data demonstrate a slight enrichment in 180 with respect to the whole-rock values of the host lithologies. If we assume that the Kandik River calcite veins were precipitated at temperatures approximately equal to the average belt maxima, then it is possible to calculate b180 values

for the parent waters responsible for the vein fillings (O'Neil and others, 1969; Friedman and O'Neil, 1977; O'Neil, 1986). The average paleotemperature for the entire Kandik River belt is approximately 285°C; if the paleotemperature calculation is restricted to the cleaved rocks within the Biederman Argillite (from which most veins were collected), then the average value rises to almost 300°C. A comparable average value for the Tatonduk belt is only 165°C; however, this statistic may be inappropriate for the purposes of calculating values of b180water because it includes a disproportionate concentration of data points from an aureole of elevated thermal maturity located to the southeast of the surface trace of the Glenn Creek fault (fig. 1). We believe that this anomaly was caused by syntectonic conductive heat transfer from hanging wall to footwall (see Laughland and others, 1990). A skewed population of data shifts the mean paleotemperature to a higher value than might be expected given equal spatial distribution of samples throughout the belt. If the footwall anomaly is eliminated from the calculation, then the Tatonduk mean drops to roughly 130°C. We believe that a value of -· ldh91-lc24l 150°C is a reasonable compromise that takes into account the possibility of slightly elevated paleotemperatures near the thrust. Calculated b180water values (using an approximate temperature of 300°C) for the Woodchopper Canyon and Kandik River veins (excluding sample DH91-K22<;) range from 8.2 to 14.7 per mil, with a mean of 12.4 per mil (fig. 5). A b180water value near 12 per mil is consistent with a metamorphic parent fluid (Sheppard, 1986);· in other words, the fluids must have equilibrated isotopically with a large volume of metamorphic or sedimentary rocks at elevated temperatures and low · water-to-rock ratios (Valley, 1986). Similar calculations of b180water values in the Catalina Schist of California yielded average values of about 13 per mil for both greenschist-facies and blueschist-facies units (Bebout and Barton, 1989). We suggest that dehydration reactions within pelitic and basaltic rocks during lowermost greenschist-facies metamorphism (Valley, 1986) provided the source for fluids moving through the Kandik River belt and the Woodchopper Canyon terrane. The extreme scatter of b180 values in Tatonduk calcite veins indicates that the fluids were possibly affected imj90-lc29 I "

lum90-lcH l

30 Kl LOMETERS Figure 3. Sample localities for calcite veins in Tatonduk belt, Kandik River belt, and Woodchopper Canyon terrane (of Churkin and others, 1982). See figure 1 for a complete geologic base. Isotopic Variations in Calcite Veins from the Kandik Region

by several types of rocks, including marine carbonate rocks, which typically yield relatively high b180 values (25±2 per mil, Keith and Weber, 1964). Using a temperature of 150°C, calculated b180water values for the Tatonduk belt are -6.5 to +8.6 per mil, with an average of 2. 7 per mil (fig. 5). These estimates are significantly depleted in 180 relative to the b180water values estimated for the geologic domain to the west, such that almost no overlap exists between the two data sets (fig. 5). Assuming a constant temperature of calcite-vein formation, the Tatonduk results are consistent with precipitation dominantly from less evolved meteoric waters whose isotopic compositions were attained through interaction with carbonate, siliciclastic, and basaltic rocks under moderate to high water-to-rock ratio conditions (Sheppard, 1986). Because similar b180water values are associated with host formations containing interbeds of limestone and shale, chert and shale, and sandstone and shale, we suggest that an open system of vein-forming fluids probably crossed many of the formational boundaries within the Tatonduk stratigraphy. The lowest b180 values come from mudrock and chert-shale units (Adams Argillite, Road River Formation), where low formation permeabilities evidently inhibited significant isotopic exchange between fluids and the host rock. Thus, the 180depleted signature of the source meteoric-water reservoir was dominantly preserved in these strata. Also, at temperatures below 150°C, exchange reactions are controlled by kinetics (O'Neil, 1986); equilibration with the silicic host rocks under these circumstances is unlikely. The one anomalous sample in the Woodchopper Canyon terrane, extracted from the shale member of the D Tatonduk belt Kandik River belt m Woodchopper Canyon terrane J D

oo a:

w lo oo 0 z () (') <(p

b180, IN PER MIL (SMOW) Figure 4. Values of o13C [relative to th~ Pee Dee Belemnite (PDB) standard] and o180 [relative to Standard Mean Ocean Water (SMOW)] for vein-filling calcite extracted from rocks of Tatonduk belt, Kandik River belt, and Woodchopper Canyon terrane, east-central Alaska. See table 1 for values and identification of host lithologies. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Woodchopper Volcanics, probably precipitated during a later stage, lower temperature emplacement event. This sample has a b13C value of -0.2 per mil, which is very close to the average value for the Tatonduk belt. In addition, if a lower paleotemperature of 150°C is used to calculate the b180water value for this specimen, it yields a value of -8.3 per mil, which likewise indicates a relatively unexchanged meteoric water. This vein probably precipitated late in the paragenetic history. Carbon Isotopes The b 13C values for calcite veins in the Kandik River belt and Woodchopper Canyon terrane (average= -3.4 per mil) are compatible with a fluid source that incorporated oxidized carbon from a mixed source (marine limestone with minor input from degradation of organic carbon). The enrichment of 13C in calcite veins from the Tatonduk belt (average b13C 0.1 per mil) may be due to lower temperatures of calcite-vein formation (Emrich and others, 1970). If we again assume average calcite depositional temperatures of 150°C and 300°C for the Tatonduk belt and the composite Kandik River/Woodchopper Canyon domain, respectively, then it is possible to calculate the b13C values of dissolved C02 in equilibrium with calcite (Friedman and O'Neil, 1977). Calculated average b13C values of dissolved C02 are as follows: Kandik River/ Woodchopper Canyon, -1.3 per mil; Tatonduk belt, -1.4 per mil. These values indicate that calcite-depositing fluids in each belt contained dissolved C02 of similar CIJ 1--z llJ

llJ a: 15

Cij

llJ

u.. a: llJ en 5 Calcite-depositing waters D Tatonduk belt (-150°C} Kandik River belt ( -300°C} +5 +10 b180water' IN PER MIL (SMOW) +15 Figure 5. Calculated values of o180water for calcite veins in Tatonduk belt (open bars) and composite Kandik River belt and Woodchopper Canyon terrane (black bars) of east-central Alaska. Calculations follow technique outlined by O'Neil and others (1969) and Friedman and O'Neil (1977).

isotopic composition. Furthermore, these calculations support the idea that calcite veins were deposited at significantly different temperatures in the two belts. The highest o Be value of 7.6 per mil occurs in a calcite vein extracted from a shale unit of the Tindir Group of the Tatonduk belt. Although this carbonisotope signature could have been affected by the presence of 13C-enriched interbedded carbonate rocks, another possible cause is the presence of methane. Methane concentrates 12C relative to oxidized carbon species and thus causes the coexisting C02 in a mixedcarbon fluid to become enriched in 13C; precipitation of calcite from the dissolved C02 component in this type of fluid would cause an increase in o13C values. In contrast, oxidation of methane will result in carbonates depleted in Be (for example, Hudson, 1977; Ritger and others, 1987). There is no evidence to indicate that any calcite in veins of the Kandik study area was precipitated from a 13Cdepleted fluid reservoir containing abundant hydrocarbons or their oxidized equivalents. Tectonic Evolution The distinct isotopic compositions of calcite-depositing fluids in the Kandik River belt and the Tatonduk belt allow us to place additional constraints on models of regional tectonic evolution (for example, Dover, 1990; Laughland and others, 1990; Howell and others, 1992). Tatonduk veins produce a wide scatter in both oBc and o180 values, whereas the Kandik River data cluster tightly, with average values of about o180 +18 per mil and o13C -4 per mil. There is almost no overlap between the two data sets, particularly if one calculates the o 180 values of the parent waters from which the veins precipitated (fig. 5). Thus, calcite veins in the Tatonduk belt did not share a common emplacement history with veins of the Kandik River belt; these two belts, moreover, did not share common fluid reservoirs or common pathways of fluid migration. All but one of the Woodchopper and Kandik River samples show the effects of considerable isotopic exchange within a rock-dominated hydrothermal system. Fluids within this geologic domain probably were derived from dehydration reactions of pelitic rocks and basalt under conditions of subgreenschist facies metamorphism (Valley, 1986). Calcite veins in Woodchopper basaltic rocks (Devonian), moreover, are virtually identical to those in the Biederman Argillite (Cretaceous); this evidence strongly supports the idea that the two tectono-stratigraphic units were linked structurally prior to the peak heating and fluid-migration events. Most of the Biederman veins are aligned at a high angle to pressure-solution cleavage; precipitation occurred mostly within extension fractures in the sandstone interbeds, rather than within the pelitic rocks. Because of this, we believe that the vein emplacement was broadly synchronous with the formation of cleavage; furthermore, fluid movement and vein precipitation were byproducts of regional metamorphism, not the cause. If fluid migration had occurred long after low-temperature metamorphism, then we would expect to see evidence of fracture propagation through both sandstone beds and cleaved pelites. We see no evidence for preferential concentration of veins along pathways of enhanced permeability, such as fault zones. The absolute age of Kandik River vein precipitation and its relative timing with respect to veins of the Tatonduk belt remain debatable. In the Tatonduk belt, fluid flow and local vein formation definitely occurred at lower temperatures, with various degrees of isotopic exchange between fluids and host rocks. The Tatonduk veins could have precipitated long before the two belts were juxtaposed along the Glenn Creek fault zone; alternatively, some or all of the veins may have formed late in the diagenetic history, after the hanging wall of the thrust system had cooled. We favor the former interpretation, for the following reasons. First, except for the narrow thermal anomaly beneath the thrust fault (fig. 1 ), vitrinite reflectance within the Tatonduk belt was notreset to higher values during the Cretaceous orogenic event (Laughland and others, 1990; Underwood and others, this volume). All of the veining was not necessarily synchronous with peak heating, but because we find only local evidence for a syntectonic thermal overprint, burial temperatures within most of the Tatonduk belt probably reached peak values prior to thrusting. Veins in the Nation River Formation at site MJ90-K59 (fig. 1) may represent an exception. Second, there is no evidence to suggest that hot fluids escaped the Kandik River belt and penetrated downdip into the Tatonduk belt. In fact; we did not recognize obvious increases in the spatial density of veining near the fault zone in either the hanging wall or the footwall. More importantly, there are no isotopic anomalies associated with veins in rocks on either side of the Glenn Creek fault, and there is no detectable gradient in isotopic values with distance from the fault. One or more of these occurrences might be expected if the fault had served as a major pathway of syntectonic fluid movement. Instead, isotopic compositions are different on either side of the fault zone. Third, if widespread fluid migration had occurred after thrusting, then one might expect some updip penetration of a later stage, lower temperature plumbing system from the underlying Tadonduk strata into the overlying Kandik River section. Instead, we note that veins in basalt of the Tindir Group just to the southeast of the Glenn Creek fault are isotopically distinct with respect to veins in the overlying Glenn Shale and Biederman Argillite. The one unusual sample (DH91Isotopic Variations in Calcite Veins from the Kandik Region

K22c; b180 4.4 per mil) from shale of the Woodchopper Canyon terrane probably did precipitate very late. in the thermal history as meteoric fluids migrated through the section. CONCLUSIONS The isotopic data from calcite veins amplify previous interpretations of the regional thermal structure of the Kandik region, particularly with regard to the emplacement of a hot Kandik River hanging wall over cooler footwall sequences of the Tatonduk belt. Fluid reservoirs in these two belts were not connected during the principal phases of vein precipitation, and most of the veins were formed before the belts were juxtaposed along the Glenn Creek fault zone. Conversely, similarities in isotopic data support the idea that the Woodchopper Canyon terrane and the Kandik River belt had been amalgamated prior to vein precipitation. These results should be regarded as preliminary, but they demonstrate the utility of stable-isotope geochemistry in deciphering the geodynamic and hydrogeologic histories of neighboring tectono-stratigraphic domains. Acknowledgments.-Mark Johnsson, Tom Brocculeri, and Lu Haufu assisted in the field. Financial support to the University of Missouri was generously supplied by ARCO Alaska, Inc. We thank G. Van Kooten and his ARCO colleagues for their scientific cooperation and logistical aid. Superintendent Don Chase granted permission to sample in the Yukon-Charley Rivers National Preserve. Acknowledgment is also made to the Donors of the Petroleum Research Fund, administered by the American Chemical Society, for partial support of this research (grant #22773-AC2 to Underwood). Editorial suggestions by D. Bradley and C. Dusel-Bacon helped improve the manuscript. REFERENCES CITED Barker, C.E., 1988, Geothermics of petroleum systems: Implications of the stabilization of kerogen maturation after a geologically brief heating duration at peak temperature, in Magoon, L.B., ed., Petroleum systems of the United States: U.S. Geological Survey Bulletin 1870, p. 26-29. Bebout, G.E., and Barton, M.D., 1989, Fluid flow and metasomatism in a subduction zone hydrothermal system: Catalina Schist terrane, California: Geology, v. 17, p. Brabb, E.E., and Churkin, M., Jr., 1969, Geologic map of the Charlie River quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Geologic Investigations Map I-973, scale 1: 250,000. Burkhard, M., and Kerrich, R., 1988, Fluid regimes in the deformation of the Helvetic nappes, Switzerland, as inferred Geologic Studies in Alaska by the U.S. Geological Survey, 1991 from stable isotope data: Contributions to Mineralogy and Petrology, v. 99, p. 416-429. Churkin, M., Jr., Foster, H.L., Chapman, R.H., and Weber, F.R., 1982, Terranes and suture zones in east-central Alaska: Journal of Geophysical Research, v. 87, p. 37183730. Coney, P.J., and Jones, D.L., 1985, Accretion tectonics and crustal structure in Alaska: Tectonophysics, v. 119, p. Dietrich, D., McKenzie, J.A., and Song, H., 1983, Origin of calcite in syntectonic veins as determined from carbonisotope ratios: Geology, v. 11, p. 547-551. Dover, J.H., 1990, Geology of east-central Alaska: U.S. Geological Survey Open-File Report 90-289, 66 pp. Dover, J.H., and Miyaoka, R.T., 1988, Reinterpreted geologic map and fossil data, Charley River quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2004, 2 sheets, scale 1:250,000. Emrich, K., Ehhalt, D.H., and Vogel, J.C., 1970, Carbon isotope fractionation during the precipitation of calcium carbonate: Earth and Planetary Science Letters, v. 8, p. Foster, H.L., 1976, Geologic map of the Eagle quadrangle, Alaska: U.S. Geological Survey Miscellaneous Geologic Investigations Map I-922, 1 sheet, scale 1 :250,000. Friedman, 1., and O'Neil, J.R., 1977, Compilation of stable isotope fractionation factors of geochemical interest: U.S. Geological Survey Professional Paper 440-KK, p. 1-12. Howell, D.G., Johnsson, M.J., Underwood, M.B., Lu, Haufu, and Hillhouse, J.W., 1992, Tectonic evolution of the Kandik region, east-central Alaska: Preliminary interpretations, in Bradley, D.C. and Ford, A.B., eds., Geologic studies in Alaska by the U.S. Geological Survey During 1990: U.S. Geological Survey Bulletin 1999, p. 127-140. Howell, D.G., and Wiley, T.J., 1987, Crustal evolution of northern Alaska inferred from sedimentological and structural relations in the Kandik area: Tectonics, v. 6, p. Hudson, J.D., 1977, Stable isotopes and limestone lithification: Journal of the Geological Society of London , v. 133, p. Keith, M.L., and Weber, J.N., 1964, Carbon and oxygen isotopic composition of selected limestones and fossils: Geochimica et Cosmochimica Acta, v. 28, p. 1787-1816. Laughland, M.M., Underwood, M.B., and Wiley, T.J., 1990, Thermal maturity, tectonostratigraphic terranes, and regional tectonic history: An example from the Kandik area, east-central Alaska, in Nuccio, V.F., and Barker, C.E., eds., Applications of thermal maturity studies to energy exploration: Society of Economic Paleontologists and Mineralogists, Rocky Mountain Section, Special Publication, p. 97 -Ill. Magaritz, M., and Taylor, H.P., Jr., 1976, Oxygen, hydrogen and carbon isotope studies of the Franciscan Formation, Coast Ranges, California: Geochimica et Cosmochimica Acta, v. 40, p. 215-234. McCrea, J.M., 1950, The isotope chemistry of carbonates and a paleotemperature scale: Journal of Chemical Physics, v. 18, p. 849-857. Miyaoka, R.T., 1990, Fossil locality map and fossil data for the

southeastern Charley River quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2007, 1 sheet, scale 1:63,360. Nesbitt, B.E., and Muehlenbachs, K., 1989, Origin and movement of fluids during deformation and metamorphism in the Canadian Cordillera: Science, v. 245, p. 733-736. O'Neil, J.R., 1986, Theoretical and experimental aspects of isotopic fractionation, in Valley, J.W., Taylor, H.P., Jr., and O'Neil, J.R., eds., Stable isotopes in high temperature geological processes: Reviews in Mineralogy, v. 16, p. 1-40. O'Neil, J.R., Clayton, R.N., and Mayeda, T.K., 1969, Oxygen isotope fractionation in divalent metal carbonates: Journal of Chemical Physics, v. 51, p. 5547-5558. Payne, M.W., and Allison, C.W., 1981, Paleozoic continentalmargin sedimentation in east-central Alaska: Geology, v. 9, p. 274-279. Ritger, S., Carson, B., and Suess, E., 1987, Methane-derived authigenic carbonates formed by subduction-induced porewater expulsion along the Oregon/Washington margin: Geological Society of America Bulletin, v. 98, p. 147156. Rye, D.M., and Bradbury, H.J., 1988, Fluid flow in the crust: An example from a Pyrenean thrust ramp: American Journal of Science, v. 288, p. 197-235. Sheppard, S.M.F., 1986, Characterization and isotopic variations in natural waters, in Valley, J.W., Taylor, H.P. ,Jr., and O'Neil, J.R., eds., Stable isotopes in high temperature geological processes: Reviews in Mineralogy, v. 16, p. Valley, J.W., 1986, Stable isotope geochemistry of metamorphic rocks, in Valley, J.W., Taylor, H.P., Jr., and O'Neil, J .R., eds., Stable isotopes in high temperature geological processes: Reviews in Mineralogy, v. 16, p. 445-490. Wickham, S.M., and Taylor, H.P., Jr., 1987, Stable isotope constraints on the origin and depth of penetration of hydrothermal fluids associated with Hercynian regional metamorphism and crustal anatexis in the Pyrenees: Contributions to Mineralogy and Petrology, v. 95, p. 255-268. Reviews: William Carothers and Robert G. Bohannon Isotopic Variations in Calcite Veins from the Kandik Region

Statistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, Kandik Region of East-Central Alaska By Michael B. Underwood, Thomas Brocculeri, Deborah Bergfeld, David G. Howell, and Mark Pawlewicz Abstract Measurements of mean random vitrinite reflectance (Rm) and illite crystallinity index (CI) show that the Kandik River belt of east-central Alaska reached higher levels of thermal maturity than its neighbor, the Tatonduk belt. A statistical correlation between all available pairs of Rm and Cl data yields a best-fit curve that passes through the domain of the anchizone (that is, the transition into lowermost greenschist-facies metamorphism). The equation to this curve is: Rm 0.31 - 6.79[1og(CI)], with the Cl parameter measured in units of A 0 28. Individual data points from both belts display a considerable amount of scatter on either side of the correlation curve, and the correlation coefficient is only r 0.60. We attribute the relatively poor statistical correlation to the effects of several internal and external variables: (1) contamination of the authigenic illite signal by detrital illite/muscovite; (2) differences in both the precursor clay-mineral assemblages and the bulk geochemistry of the diverse host rocks; (3) variations in the rate and (or) physical mechanism of heating; (4) fluctuations in the hydrogeochemistry and hydrogeology at the time of peak heating; and (5) differences in tectonic history, particularly the effects of deformation fabrics such as slaty cleavage and pencil structure. INTRODUCTION The Kandik region of east-central Alaska (fig. 1) contains a wide variety of geologic units that are parautochthonous with respect to the North American continent (Brabb and Churkin, 1969; Foster, 1976; Payne and Allison, 1981; Howell and Wiley, 1987; Dov~r and Miyaoka, 1988). This part of Alaska has been described and interpreted within the conceptual framework of terrane analysis (Churkin and others, 1982; Coney and Jones, 1985; Howell and Wiley, 1987; Laughland and others, 1990; Howell and others, 1992). Conversely, other geologists regard the Kandik region as a thrustfaulted continental-margin succession of North American Geologic Studies in Alaska by the U.S. Geological Survey, 1991 affinity (Dover, 1990; D. Bradley, 1992, written commun.). Regardless of which concept is favored, there are two major fault-bounded domains of strata to consider within the Kandik region, herein referred to as the Tatonduk "belt" and the Kandik River "belt" (following the suggestion of D. Bradley, 1992, written commun.). The structural boundary between the Kandik River and Tatonduk belts is the younger-over-older Glenn Creek fault zone (fig. I), which is part of a complicated system of southeast-verging thrust faults (Dover and Miyaoka, 1988; Howell and others, 1992). In this paper, we make no inferences regarding the absolute distance of tectonic transport for either belt with respect to the Proterozoic edge of North America. The Tatonduk belt (fig. 2) contains a Middle Proterozoic to Lower Cambrian basement complex (Tindir Group) composed largely of basalt, carbonate rocks, and shale (Young, 1982; Allison, 1988). These strata are overlain by a Paleozoic sequence of sedimentary rocks that records the rifting and progressive subsidence of the North American continental margin (Payne and Allison, 1981; Churkin and others, 1982; Howell and Wiley, 1987; Howell and others, 1992). The Kandik River belt (fig. 3), in contrast, is dominated by deep-marine deposits of Mesozoic age (Churkin and others, 1982). The principal rock units include the Glenn Shale (Triassic to Lower Cretaceous) and the Lower Cretaceous Kandik Group (Keenan Quartzite, Biederman Argillite, and Kathul Graywacke). Fossil localities for these and all other units in the Kandik study area have been summarized by Dover and Miyaoka (1988) and Miyaoka (1990). Pre-Mesozoic strata occur in two general localities west of the Glenn Creek fault zone: in the core of the Step Mountains antiform, and west of the Mardow Creek fault (fig. 1). Three Paleozoic formations have been mapped in the Step Mountains: (1) conglomerate and fossiliferous limestone lenses of the Permian Step Conglomerate (Brabb, 1969); (2) undated mudrocks correlated by Brabb (1969) with the Ford Lake Shale (Upper

Devonian and Mississippian); and (3) poorly dated chertpebble conglomerates, assigned to the Devonian Nation River Formation by Brabb and Churkin ( 1967). West of the Mardow Creek fault, within the so-called Eureka suture zone of Churkin and others (1982), the pre-Mesozoic units include ( 1) the Devonian Woodchopper Volcanics [named the Woodchopper Canyon terrane by Churkin and others (1982)]; (2) poorly dated interbeds of argillite and chert,· sandstone and conglomerate, and limestone and dolomite [mostly units Pza and Pzl of Brabb and Churkin (1969)]-these strata were assigned to the Slaven Dome terrane by Churkin and others (1982); and (3) the Permian Step Conglomerate, plus Proterozoic(?) deposits of undivided sedimentary rocks, dolomite, and volcanic rocks [the Takoma Bluff terrane of Churkin and. others (1982)]. The map of Dover and Miyaoka ( 1988) shows widespread occurrences of Devonian conglomerates (Nation River Formation rather than Step Conglomerate) within the Takoma Bluff terrane, but fossil control for these rocks is poor. The youngest rocks within the study area are assigned to the unit TKs of Brabb and Churkin ( 1969). These strata are Albian to early Tertiary in age (Dover and Miyaoka, 1988; Miyaoka, 1990). Unit TKs unconformably overlies both the Tatonduk and Kandik River belts and consists of poorly indurated interbeds of conglomerate, sandstone, and mudrock, plus local lenses of coal and plant fragments. · One of the principal goals of our research in the Kandik region is to establish a clearly defined tectonic history, including the thermal evolution of all pertinent rock units. Preliminary accomplishments within this framework included reconnaissance-level measurements of vitrinite reflectance and illite "crystallinity" KANDIK RIVER BELT Indian Grave Mountain EXPLANATION 15 MILES

15 Kl LOMETERS c::: :1

Limit of paleothermal anomaly --Contact Fault, dashed where approximate Strike-slip fault, arrows indicate direction of relative movement Thrust fault, sawteeth on upper plate +Syncline -;-- Anticline Figure 1. Geologic index map of east-central Alaska showing Kandik River belt (no pattern) and Tatonduk belt (horizontal lined pattern). Hachured line southeast of Glenn Creek fault represents detected limit of a footwall thermal aureole. Also shown are unit TKs of Brabb and Churkin (1969), and the Woodchopper Canyon (WC), Slaven Dome (SD), and the Tacoma Bluff (TB) terranes of Churkin and others (1982). Rocks southwest of Tintina fault zone are assigned to Yukon-Tanana composite terrane (Coney and Jones, 1985). Modified from Brabb and Churkin (1969) and Foster (1976). For relevant fossil control and alternative interpretations of structural geology, see Dover and Miyaoka (1988) and Miyao'ka (1990). Statistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, Kandik Region

(Underwood and others, 1989; Laughland and others, 1990). Those studies showed that most of the rocks within the Kandik River belt attained significantly higher ranks of thermal maturity than those of the Tatonduk belt. We are now expanding on the analytical techniques and initial interpretations by including isotopic assessments of fluid-rock interactions (for example, Shelton and others, this volume) and micro thermometric data from fluid inclusions. Many techniques are available to help document peak burial temperatures and pressures (P). Most traditional mineralogic studies suffer from a lack of precision with respect to P-T estimates because paragenetic sequences respond to a complex array of variables, and most of the common index minerals are stable over wide ranges of burial conditions (for example, Kisch, 1983; Liou and others, 1987; Frey, 1987). One widely used method for characterizing specific levels of diagenesis and incipient metamorphism is the measurement of. illite Devonian Cambrian Proterozoic and Glenn Shale Tahkandit Limestone Calico Bluff Formation Ford Lake Shale Nation River Formation McCann Hill Chert Road River Formation Hillard Limestone Adams Argillite Funnel Creek Limestone Bituminous limestone t Sandstone, shale redbeds Basalt Thin-bedded dolomite Limestone and shale Sandstone Dolomite Sandstone and shale Figure 2. Schematic stratigraphic section for Tatonduk belt [otherwise known as Tatonduk terrane of Churkin and others (1982)1. Modified from Howell and others (1992). Geologic Studies in Alaska by the U.S. Geological Survey, 1991 "crystallinity" by X-ray diffraction (XRD). This procedure is rapid, simple, inexpensive, applicable over a wide variety of geologic conditions, and useful for documenting relatively subtle differences among individual samples. XRD data are especially helpful when combined with analyses of organic metamorphism, as measured, for example, by vitrinite reflectance [for general descriptions of this technique, see Dow (1977); Bostick (1979); Barker and Pawlewicz (1986)]. Tertiary Proterozoic(?) Devonian Paleozoic Unit TKs of Brabb and Kandik River belt Kathul Graywacke

Biederman Argillite

a -o Glenn Shale Takoma Bluff terrane Step Conglomerate Undivided siliciclastic, volcanic, and carbonate rocks (age uncertain) Woodchopper Canyon terrane Basalt, shale, limestone Slaven Dome terrane Units Pza and Pzl of Brabb and Churkin {1969} Argillite, chert, limestone, and dolomite Figure 3. Schematic stratigraphic section for Kandik River belt and rocks of Eureka suture zone. Following the terminology of Churkin and others (1982), Precambrian rocks and Step Conglomerate would be assigned to Takoma Bluff terrane, Woodchopper Volcanics would be assigned to Woodchopper Canyon terrane, and units Pza and Pzl of Brabb and Churkin (1969) would be assigned to Slaven Dome terrane. Modified from Howell and others (1992).

In this paper, we report on the responses of both organic and inorganic matter to conditions of diagenesis and low-temperature metamorphism within the Kandik region of east-central Alaska (fig. 1). So that meaningful comparisons can be made among many rock units with different burial histories, we have restricted the discussion to two types of data: mean random vitrinite reflectance (Rm), and illite "crystallinity" (CI). This paper establishes a statistical correlation between R and CI m that is specific to the Kandik study area and compares the Kandik curve to trends established in other orogenic belts. Discussions of spatial variations in thermal maturity, and how they relate in detail to the stratigraphy and geodynamic evolution of the Kandik region, will be presented at a later date. BASIC PRINCIPLES Illite Crystallinity Many studies of clay mineralogy have demonstrated that detrital and authigenic smectites are replaced gradually by illite during sediment diagenesis; in addition, a mixed-layer illite-smectite phase, which is the intermediate product of illitization, displays improved degrees of "ordering" as diagenesis proceeds, together with progressive increases in the ratio of illite to smectite (Dunoyer de Segonzac, 1970; Perry and Hower, 1970; Reynolds and Hower, 1970; Kisch, 1983; Moore and Reynolds, 1989). Weaver (1960) was the first to relate systematic variations in the shape of the 10 A (001) illite X-ray diffraction peak to the lattice reorganization of illite. In reality, however, this process is quite complicated, in that it that involves an increase in the size of crystallites, changes in chemical composition, and a progression toward greater regularity of the structural layers (Kisch, 1983). Frey (1970) demonstrated that increases in illite "crystallinity" can be accompanied by changes in several rock and mineral properties, including the intensity ratio of the illite basal reflections [/(002)//(001)], the color of the host rock, the mean bulk density, and microscopic rock texture due to reactions between clastic quartz grains and clay matrix. Hunziker and others ( 1986) showed that four chemical modifications occur as illite is transformed into well-crystalline K-mica: (1) expandable layers decrease in relative abundance and eventually disappear; (2) the 1Md illite polymorph (disordered monoclinic) is replaced by the 2M1 form (two octahedral layers with an overall monoclinic symmetry); (3) the total layer charge and the potassium content in illite interlayer positions both increase; and ( 4) the chemical variability of individual illite/mica grains decreases. Other authors have pointed out that the term illite "crystallinity" is not perfectly appropriate to describe this complicated structural and chemical reorganization (Kisch, 1983; Frey, 1987), but it will be used in theremainder of this paper without quotation marks. Many workers have used illite crystallinity data as part of detailed studies of regional and local structural evolution (for example, Frey and others, 1980; Kemp and others, 1985; Roberts and Merriman, 1985; A wan and Woodcock, 1991; Hesse and Dalton, 1991). Some caution is warranted in this regard, however, particularly if XRD data serve as the only indicator of paleotemperature. One might assume that crystallinity values respond in a linear fashion to increases in temperature or burial depth, as suggested by some borehole data (Yang and Hesse, 1991). Nevertheless, the inorganic reactions are complicated by the effects of many additional variables, including the duration or rate of heating, fluid pressure, fluid composition (K+ must be available), rates of fluid migration, tectonic stress, original composition of the host sediment (inhibiting effects of Na+ and Mg2+), the content of organic matter, and the chemical make-up of illite and (or) mixed-layer precursors (Frey, 1987). In addition, data from regions of intermediate-level diagenesis can be notoriously unreliable because of a "contamination" effect caused by mixtures of authigenic illite, illite-smectite mixed-layer phases, and higher grade detrital micas eroded from metamorphic terrains. As reviewed by Blenkinsop (1988), there are several procedures to quantify the shape of the 10-A illite peak. To be consistent with most other studies of this type, we have employed the Kubler index (otherwise known as crystallinity index, or Cl), which is defined as the width of the 10-A peak at one-half the peak height (Kubler; 1968). Blenkinsop (1988) considered the Kubler crystallinity index to be marginally superior to the other indices at all grades of diagenesis and metamorphism. The crystallinity index decreases as the diagenetic/metamorphic grade increases, and values can be calculated very precisely using digital XRD data. The original units of crystallinity index were millimeters, but most subsequent workers have converted to angular units of ll 0 28 to minimize the effects of variable machine settings (for example, Kisch, 1980). Robinson and others (1990) completed additional studies of error and precision and concluded that geologic interpretations of CI gradients and anomalies should be based on differences of at least 0.1ll 0 28. The most reliable application, therefore, is simply to define broad zones or stages of advanced diagenesis and lowtemperature metamorphism. A five-fold subdivision was introduced by Weaver (1960), but most workers follow the system of Kubler ( 1968), who defined the zones as diagenesis, anchimetamorphism (transition into lowermost greenschist-facies metamorphism), and epimetamorphism (lowermost greenschist facies). PinStatistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, kandik Region

pointing the zonal boundaries has been problematic because of operational inconsistencies in sample preparation, analytical equipment, and analytical technique [see Kisch (1987, 1990), Kisch and Frey (1987), and Robinson and others (1990) for comprehensive discussions of these problems]. With these problems in mind, Blenkinsop ( 1988) advocated the following values of CI for the two principal boundaries of thermal alteration: diagenesis-to-anchizone 0.42A 0 28 and anchizone-toepizone 0.25A Vitrinite Reflectance and Paleotemperature The goal of correlating CI zones with other indicators of low-grade metamorphism (both inorganic and organic) has been discussed in detail by Kisch (1987). Most noteworthy, for the purposes of our study, is the rt:;lation between CI values and coal rank, which is typically quantified in terms of vitrinite reflectance. To establish a universal CI-Rm correlation is troublesome, however, because the organic and inorganic systems quite clearly respond to different (though overlapping) sets of external variables. Unlike illite crystallinity, for example, vitrinite reflectance increases primarily in response to higher burial temperatures, thereby serving as a more direct indicator of maximum heating (Barker, 1989, 1991). Time-dependent models of organic metamorphism (Hood and others, 1975; Bostick and others, 1978; Waples, 1980; Middleton, 1982; Ritter, 1984; Wood, 1988; Hunt and others, 1991) are useful for studies of first-cycle sedimentary basins. However, serious problems are inherent in the application of these models if one's goal is to calculate maximum paleotemperatures with any degree of accuracy, particularly within uplifted orogenic sequences where the tectonic and thermal histories can be very complicated and poorly constrained. In addition to the guesswork involved in the selection of an appropriate value of activation energy for a given rock unit (Antia, 1986; Wood, 1988), large uncertainties usually exist in the inferred reaction rates for time-temperature integrals (lssler, 1984; Ritter, 1984), not to mention in the choices of effective heating time. Conversely, many empirical and theoretical studies over the past decade have demonstrated that vitrinite reflectance is affected most by maximum temperature and almost imperceptibly by the duration of heating, at least under geologic situations where time is measured in millions of years (Wright, 1980; Gretener and Curtis, 1982; Suggate, 1982; Barker, 1983, 1989, 1991; Price, 1983; Barker and Pawlewicz, 1986). These workers, in other words, have argued that the time required for stabilization of kerogen maturation, at a given burial temperature, is usually on the order of 1 m.y. or less. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Several statistical correlations now exist between Rm and absolute paleotemperature, with the temperature scales established through direct borehole measurements (Barker, 1983, 1988; Price, 1983; Barker and Pawlewicz, 1986). The Barker (1988) correlation yields the lowest temperature estimates for a given value of Rm as compared with other time-independent methods. Recent models based on chemical kinetics (Burnham and Sweeney, 1989; Sweeney and Burnham, 1990) match the curve of Barker (1988) quite closely, provided the inferred heating time is liplited to around 1 m.y. Furthermore, Barker and Goldstein (1990) arrived at a close match between homogenization temperatures of fluid inclusions (which are not affected by heating duration) and estimates of paleotemperature derived from the Barker and Pawlewicz (1986) regression analysis. The Barker ( 1988) equation is based on the same set of data as Barker and Pawlewicz (1986), but the statistical model is more defensible. Thus, for the purposes of providing a straightforward estimate of peak burial temperature, we prefer the equation of Barker (1988). The relevant equation is: T (°C) 148 + 104[ln (Rm)], and the error associated with paleotemperature estimates is roughly ±30°C. No universal agreement has been reached on the anchizone boundaries in terms of mean vitrinite reflectance or paleotemperature values. The boundaries suggested by various workers are diagenesis-to-anchizone 2.5 percent Rm to 3.1 percent Rm [see numerous studies summarized by Kisch (1987)], and anchizone-to-epizone 3.7 percent Rm to 5.5 percent Rm (Frey and others, 1980; Kisch, 1980; Ogunyomi and others, 1980; Duba and Williams-Jones, 1983). The lower and upper temperature limits of the anchizone have been approximated only vaguely at 200°C and 300°C, respectively (Kisch, 1987). METHODS Specimens of shale and their low-grade metamorphic equivalents were collected from fresh outcrop exposures. Virtually all of the mudrock-bearing formations within the Kandik region were sampled (table 1). The standard U.S. Geological Survey and University of Missouri techniques for preparation and optical measurements of vitrinite reflectance have been described at length elsewhere (Barker and Pawlewicz, 1986; Underwood and others, 1989) and are not repeated here. We report the data as mean values based on measurements (in oil) of at least 25 individual randomly oriented particles per specimen. Although a few coal specimens were analyzed during the investigation (particularly common in unit TKs ), all of the reported data come from dispersed organic particles that were concentrated from mudstones, shales, and argillites.

Illite Sample Preparation The rocks were ground to a fine powder using a mortar and pestle; the powder then was washed in distilled water and disaggregated further using an ultrasonic cell disrupter. A pinch of sodium phosphates and sodium carbonates was added to prevent flocculation of suspended clay minerals, and the . <2-Jlm size fraction was segregated by centrifugation (770 rpm for 3.3 min.; then 1,500 rpm for 15 min.). Oriented aggregates of the clay-sized material were placed on glass slides using the vacuum-filtration and peel technique (Moore and Reynolds, 1989). Prior to XRD analysis, the slides were placed in a chamber containing ethylene glycol for 12 h at approximately 60°C to saturate any existing expandable clay phases; this step was necessary to prevent seasonal changes in atmospheric humidity from affecting the results. Diffractometer Settings and Digital Data Processing All XRD measurements were completed at the University of Missouri using a Scintag PAD V microprocessor-controlled diffractometer, interfaced with a Microvax 2000 microcomputer. X-ray scans were run under the following machine settings: voltage 30 kV; amperage 20 rnA; radiation CuKa; filters none; receiving slits 0.3° (close to detector); scatter slits 2° (close to beam emission); scan continuous with 0.03° chopper increment; time constant 1.8 second/step; scan rate 1 °28/min; rate meter automatic control. The (101) quartz peak (3.342 A) was used as a standard for internal calibration of 28 peak positions. The scan for illite (and other clay minerals) was run from 2°28 to 15°28 with the sample holders spun. The resulting digital output was processed first for a background correction and Ka2 stripping, using Scintag software. The illite (001) peaks then were identified and deconvoluted using an interactive graphics subroutine. The deconvolution program is designed to fit XRD peaks to models, based on a Split Pearson Vll profile shape (Gaussian-Lorentzian hybrid). Automatic computations yield peak positions and d-spacings (in °28 and A values), peak intensities (counts per second), integrated peak areas (total counts), and full peak width at half maximum (A 0 28). The analytical sensitivity for measurements of peak width using the Scintag PAD V diffractometer is 0.001°28. However, prior tests of internal reproducibility (Underwood and others, 1993) resulted in standard deviations about the mean ranging from 0.009°28 to 0.015°28. We conclude, therefore, that the data are meaningful only to the nearest 0.01°28. RESULTS All of the sample numbers, formational designations, values of illite crystallinity index, plus corresponding data from measurements of vitrinite reflectance are listed in table 1. All formational assignments conform to the maps of Brabb and Churkin ( 1969) and Foster (1976). Many of the specimens did not yield Rm data, either because they are too old to have incorporated debris from terrestrial plants or because the organic matter is too lean to produce reliable results. Pre-Devonian Strata Results from the older rock units within the Tatonduk belt (that is, stratigraphically below the Nation River Formation) are shown in figure 4. The principal epiclastic units within this part of the stratigraphic column (fig. 2) include interbeds of sandstone and shale or dolomite and shale within the Tindir Group, mudrocks of the Adams Argillite, and interbeds of chert, siliceous shale, and graptolitic shale of the Road River Formation. These 0 data define an extremely wide range of apparent burial condipons, extending from the upper limit of the ancliizone (0.25A 028) to the zone of early diagenesis (0.80A 0 28). Because these specimens do not contain appropriate organic constituents, we have no means of establishing comparable ranks of organic metamorphism or assessing the reliability of CI values as indicators of in situ burial conditions. Nation River Formation The Nation River Formation is an important Devonian unit of inferred submarine-fan deposits (sandstone turbidites, conglomerate, and shale). Vitrinite is fairly abundant in the mudrocks, which supports the idea of a continental provenance for the sediment. With one exception, CI values from the Nation River Formation all fall within the confines of the diagenetic zone (fig. 5). Similarly, ranks of organic metamorphism are generally low (below 1.0 percent Rm). Consistent exceptions to the low organic rank occur at two localities: (1) within the Step Mountains region, where Rm 2.5 percent to 3.7 percent; and (2) within a relatively narrow zone (exposed along the Nation River) immediately to the southeast of the Glenn Creek fault, where Rm values are between 2.0 percent and 2.8 percent (see fig. 1 for general localities). Two samples of the Ford Lake Shale at Step Mountains also produced anomalously high values of Rm (table 1). Structural and tectono-stratigraphic interpretations of the Step Mountains region are controversial. This feature could represent a simple faulted anticline (Brabb and Statistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, Kandik Region

Table 1. Values of mean random vitrinite reflectance (Rm) and illite crystallinity (CI), Kandik region, east-central Alaska Sample Formation R,. Sample Formation R,. Pre-Devonian strata tb90-k50b Nation River tb90-k51a Nation River tb90-k52a Nation River mj90-k26a Tindir tb90-k53c Nation River mj90-k26b Tindir tb90-k59c Nation River mj90-k27a Tindir 786-3-2d Nation River mj90-k28 Tindir 786-3-3a Nation River mj90-k29a Tindir 786-4-2c Nation River mj90-k75a Tindir 786-29-1 Nation River mj90-k76a Tindir 85jcr-15a Nation River um90-k34a Tindir 85jcr-24 Nation River um90-k35b Tindir um90-k37c Tindir um90-k39b Tindir Glenn Shale 786-7-2 Tindir 85jcr-67 Tindir 886-1-1 Funnel Creek mj90-k17a Glenn mj90-k33a Adams rnj90-k41a Glenn mj90-lC63a Adams rnj90-k4lb Glenn um90-k33 Adams rnj90-k42a Glenn um90-k41a Adams mj90-k43b Glenn tb90-k56b Adams mu90-k9a Glenn tb90-k57b Adams um90-k17 Glenn tb90-k58a Adams um90-k18 Glenn um90-k44a Hillard um90-k20 Glenn um90-k46 Hillard um90-k21 Glenn um90-k47b Hillard um90-k24a Glenn mj90-k15a Road River um90-k25c Glenn um90-k50a Road River um90-k26b Glenn um90-k51a Road River um90-k28a Glenn um90-k52a Road River um90-k30a Glenn tb90-k54b Road River um90-k31 a Glenn tb90-k55a Road River um90-k32c Glenn 85jcr-25b Road River um90-k78f Glenn um90-k78h Glenn tb90-k18 Glenn Nation River Formation 786-7 -3a Glenn 786-7 -3b Glenn mj90-k5 Nation River 786-23-1a Glenn 786-23-lb Glenn mj90-k6 Nation River 786-31-3a Glenn mj90-k31b Nation River 786-31-3b Glenn rnj90-k34a Nation River 786-31-3c Glenn rnj90-K35c Nation River rnj90-k59a Nation River mj90-k68a Nation River Kandik Group rnj90-k69d Nation River rnj90-k70a Nation River rnj90-k71a Nation River mu90-kl Oa Keenan rnj90-k72a Nation River mj90-k18b Biederman rnj90-k74a Nation River mj90-k23a Biederman um90-k54c Nation River mj90-k37a Biederman tb90-k49c Nation River mj90-k38a Biederman Geologic Studies in Alaska by the U.S. Geological Survey, 1991

Table 1. Values of mean random vitrinite reflectance (Rm) and illite crystallinity (CI), Kandik region, east-central AlaskaContinued Sample Formation Rm Kandik Group mj90-k39a Biederman mj90-k40a Biederman mj90-k54 Biederman mj90-k56a Biederman mj90-k79c Biederman mj90-k79g Biederman mj90-k80c Biederman mj90-k81 b Biederman mj90-k82b Biederman mj90-k83d Biederman mj90-k84b Biederman mj90-k85b Biederman mj90-k86b Biederman mj90-k87b Biederman mj90-k89c Biederman mj90-k90d Biederman mj90-k9lb Biederman . mj90-k92a Biederman mj90-k93a Biederman um90-k27 a Biederman 786-8-2b Biederman 786-12-ld Biederman 786-13-1 Biederman 786-14-1 Biederman 786-14-3 Biederman 786-14-4 Biederman 786-23-2 Biederman 786-23-3 Biederman 786-26-1 b Biederman 786-26-3a Biederman 786-27-la Biederman 786-27-1 b Biederman 786-27-4a Biederman 786-27 -4b Biederman mj90-k20a Kathul mj90-k21b Kathul mj90-k22a Kathul Churkin, 1969; Dover and Miyaoka, 1988), or a window into the Tatonduk belt, with the Paleozoic footwall strata exposed by erosion through a folded Glenn Creek fault (Laughland and others, 1990). Alternatively, Howell and others (1992) have speculated that the Devonian and Permian strata at Step Mountains could be part of a separate nappe composed of basement rocks that crop out to the west in the Eureka suture zone [that is, the Takoma Bluff and (or) Slaven Dome terranes of Churkin and others, Sample Formation Rm 786-24-lb Kathul 786-24-2b Kathul 786-24-3a Kathul 786-24-3b Kathul 786-25-8a Kathul 786-25-8b Kathul 786-25-10 Kathul 85jcr-14 Kathul Miscellaneous mj90-k50 Pza mj90-k5la Pza mj90-k51 b Pza mj90-k5lc Pza mj90-k94a Pza mj90-k98b Pza mj90-k99 Pza mj90-k4 Ford Lake rnj90-kll Ford Lake rnj90-k62a Ford Lake mu90-k2 Ford Lake mu90-k3 Ford Lake mu90-k4 Ford Lake mu90-k8a Ford Lake mu90-klla Ford Lake 85jcr-ll Ford Lake mj90-k64a Calico Bluff mj90-k67b Calico Bluff um90-k16 Step rnj90-k36a TKs rnj90-k58b TKs rnj90-k65a TKs mj90-k66a TKs mj90-k95b TKs um90-k22j TKs tb90-k17 TKs 1982]. The existing thermal-maturity data fail to resolve these conflicting hypotheses, as any one of the scenarios would explain the generally higher values of thermal maturity in the core of the structure. The second group of anomalous Rm data from the Nation River Formation confirms the existence of athermal aureole created by displacement along the Glenn Creek fault (fig. 1 ). Laughland and others ( 1990) first suggested that this anomaly was caused by heat conducStatistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, Kandik Region

tion as a hot hanging wall (Kandik River belt) was uplifted and thrust over the cooler footwall (Tatonduk belt). The southeastward termination of the paleothermal anomaly may be related to wedging out of the hanging wall (Laughland and others, 1990) or to a transition from a thrust ramp or associated fault-bend fold in the hanging wall to a flat in the hanging wall. The values of illite crystallinity within the aureole (0.57 A 0 28 to 0.46/l 0 28) do not deviate with respect to the Nation River average of 0.52A 0 28. Thus, even though thrusting evidently reset Rm to higher values, the temperature increase was insufficient [in magnitude and (or) duration] to change CI beyond the levels inherited from detrital source rocks and (or) earlier stages of diagenesis. Glenn Shale CI values for the Glenn Shale vary considerably (fig. 6). This stratigraphic unit has been mapped without ambiguity within both the Kandik River belt and the Tatonduk belt (figs. 2 and 3; Churkin and others, 1982; Howell and others, 1992). One group of samples yielding higher CI values (sites 786-23 and 786-31) was obtained from a poorly mapped region on the Canadian side of the international border (see Underwood and others, 1989); these rocks are provisionally included in the Tatonduk belt, which is generally lower in thermal maturity. Based upon one datum (site um90-k30), CI values associated with shale outcrops on the flanks of the Michigan Creek anticline (see fig. 1 for location) are also consistent with low thermal maturity, as are specimens of the Nation River Formation obtained from the fold's core (sites 786-3 and 786-4). 111111111111111111111111111111111 I

w

Pre-Devonlan Units r.o I"': Anchizone n 31

1 .· u 0 ~.H W.; ILLITE CRYSTALLINITY INDEX (A026) Figure 4. Illite crystallinity (CI) histogram for pre-Devonian strata of Tatonduk belt. Rock units consist of Tindir Group, Funnel Creek Limestone, Adams Argillite, and Road River Formation. See table 1 for complete listing of data. n, number of samples. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Within the Kandik River belt, one anomalous zone of relatively low thermal maturity (sites um90-k17 to um90k21) is associated with an unusual east-west trending syncline (cored by the Glenn Shale and Keenan Quartzite); this structure is located to the west of and is truncated by the Mardow Creek fault (fig. 1; Brabb and Churkin, 1969). Most of the remaining data, particularly in the vicinity of . the Glenn Creek fault zone, fall within the confines of the anchizone. These samples of Glenn Shale display excellent slaty cleavage or pencil structure, formed by the intersection of cleavage and fissility. One fault-zone specimen (mj90-k17a), a semi-schist from Indian Grave Mountain (fig. 1), falls within the zone of epimetamorphism. In general, the inorganic and organic data are consistent, in that most anchizone samples yield correspondingly high Rm values of 3.4 percent to 4.8 percent (table 1). Kandik Group Data from the Kandik Group (which forms the bulk of the Kandik River belt) display the most consistent cluster of any stratigraphic unit within the study area (fig. 7). Most of these samples were obtained from the Biederman Argillite; all of the interbeds of mudrock that we sampled from Biederman Argillite display excellent slaty cleavage, and sandstone interbeds typically contain extension fractures filled with quartz and calcite veins. The ubiquitous development of cleavage is consistent with the fact that most of the Biederman data fall within the boundaries of the anchizone; a few data points are s 7 6 M Nation River Formation ,

w Anchizone

n =25 :5 3 1 1 1:11 2 l l··"171i'.I-·"'~M . 0o.2 ILLITE CRYSTALLINITY INDEX {L\026) Figure 5. Illite crystallinity (CI) histogram for Nation River Formation (Devonian). See table 1 for complete listing of Cl data and corresponding values of vitrinite reflectance. Included in this compilation are two data from Step Mountains (sites mj90-k5 and mj90-k6). All other samples were collected southeast of Glenn Creekfaultzone. With one exception (site mj90-k59), all samples yielding Rm values above 1.0 percent were collected within inferred footwall aureole of the Glenn Creek fault (see fig. 1 ). n, number of samples.

within the uppermost zone of diagenesis (fig. 7). The corresponding values of vitrinite reflectance range from 1.6 percent to 5.1 percent, and most of the samples within the anchizone are above 3.0 percent Rm(table 1). Miscellaneous Two additional rock units deserve specific mention, even though data are relatively sparse. Unit Pza of Brabb and Churkin (1969) occurs to the west of the Mardow Creek fault (that is, within the Slaven Dome terrane of Churkin and others, 1982). The original stratigraphic position of these poorly dated argillites remains uncertain, as all contacts with the Glenn Shale and the Kandik Group have been mapped as faults (Brabb and Churkin, 1969). CI values for unit Pza are uniformly low, ranging between 0.40A 0 28 and 0.22A 028 (table 1); Rm values range from 3.2 percent to 4.2 percent. These results are consistent with the phyllitic appearance of hand specimens and the hypothesis that assigns unit Pza to the depositional basement of the Kandik River belt (Howell and others, 1992). CI values from unit TKs of Brabb and Churkin (1969) are extremely erratic and inconsistent with respect to levels of organic metamorphism. These poorly indurated and mildly deformed alluvial and fluvial deposits unconformably overlie rocks within both the Tatonduk and Kandik River belts (fig. 1). Coal specimens (unpublished U.S. Geological Survey and University of Missouri data) and dispersed organic matter from this unit consistently fall below 0.6 percent Rm, yet values of CI for the mudstones range from 0 .26A 0 28 to 0.68A 0 28 (table 1). Clearly, most of the illite and wellcrystalline K-mica incorporated into these mudstones is z Glenn 5 5 t t .. m s h a 1 e o Anchizone W 4 t-· .. ·t·· ~r·t i

() n 27 ILLITE CRYSTALLINITY INDEX (A028) Figure 6. lllitecrystallinity{CI) histogram for Glenn Shale (Triassic to Lower Cretaceous). Results from both Kandik River belt and Tatonduk belt are included. See table 1 for complete listing of Cl data and corresponding values of vitrinite reflectance. n, number of samples. detrital in origin, rather than a product of diagenetic alteration of the host rocks. DISCUSSION The following discussion places the available data from the Kandik region within the context of comparable studies completed elsewhere, particularly as they apply to calibrations of the anchizone. A cross plot of all pairs of CI and Rm data is shown in figure 8. In this diagram, we have adopted the following boundaries for the anchizone: CI boundaries set at 0.42A 0 28 and 0.25A 0 28; Rm boundaries set at 2.6 percent and 4.5 percent. These Rm limits agree with most of the data sets cited by Kisch ( 1987), but they are more specifically dictated by the results of several studies summarized by Underwood and others (1991). Using the correlation between Rm and paleotemperature established by Barker (1988), the temperature boundaries for the anchizone are 245°C and 305°C. These temperatures seem reasonable in light of the widespread acceptance of the 300°C isotherm as an approximate lower limit for greenschist-facies metamorphism (for example, Ernst, 1974). We believe that a logarithmic regression is appropriate because of the well-established exponential relationship between Rm and temperature, with or without the hypothesized linear response to time or heating rate (Dow, 1977; Bostick, 1979; Barker and Pawlewicz, 1986). The precise numerical relationship between CI and paleotemperature is impossible to isolate, however, because of the many variables involved in illitic reactions. Nevertheless, diagenetic progressions in illitesmectite mixed-layer phases typically follow linear depth 10 1'1'1"'1'

!z8 ()6 w

...J4

ILLITE CRYSTALLINITY INDEX (A 026) Figure7.lllitecrystallinity(CI)histogramforKandikGroup{Lower Cretaceous). Rock units consist of Keenan Quartzite, Biederman Argillite, and Kathul Graywacke. See table 1 for complete listing of Cl data and corresponding values of vitrinite reflectance. n, number of samples. Statistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, Kandik Region

trends (Perry and Hower, 1970; Freed and Peacor, 1989), and borehole gradients in CI are usually linear (for example, Yang and Hesse, 1991). Consequently, at least for the purpose of regression analysis, we assume that CI values change as a linear function of temperature, whereas Rm values change exponentially with temperature. The equation to a best-fit curve for the Kandik data is Rm 0.31 - 6. 79[log(CI)]. This result closely matches similar curves established for two study areas within the Franciscan Complex of California (King Range terrane and Sur-Obispo terrane) and one site within the Shimanto Belt of Japan (fig. 9). It is important to point out, however, that a great deal of scatter exists in individual data points from the Kandik study area (fig. 8). As a measure of this scatter, the correlation coefficient for the Kandik curve is only r 0.60. If both CI and Rm had responded in kind to the same set of external and (or) internal variables, then one might expect the correlation to be much tighter. As statistics for comparison, values of r for the Sur-Obispo terrane and the Shimanto data are 0.89 and 0.84, respectively (Underwood and others, 1991 ). Geologic reasons for the scatter of data in the Kandik study area need to be assessed. Sample preparation methods, optical equipment, and operator technique will affect values of vitrinite reflectance, as will statistical manipulations of the resulting data (Dembicki, 1984). Beyond these possible influences, sev0

eral geologic and geochemical parameters appear to have influenced our results, particularly with respect to the CI data. One obvious reason for the poor statistical correlation is the fact that many of the rock units were exposed to only moderate degrees of diagenesis, and those mudstones contain mixed populations of illite/mica and illite-smectite mixed-layer clays. This is particularly obvious for samples within unit TKs, but low-rank sequences within most of the Tatonduk belt probably contain mixtures of detrital muscovite eroded from a broad spectrum of metamorphic source rocks. In support of this conclusion, petrographic analyses of sandstones from the Nation River Formation show that the detritus includes significant percentages of low-grade metasedimentary rock fragments (Brocculeri and others, this volume). Because of this recycling phenomenon, absolute values of CI within the field of diagenesis are not reliable as indicators of in situ thermal history. At the very least, we believe that CI data from any samples yielding Rm values less than 1.0 percent must be viewed as suspect (fig. 8). Furthermore, because the pre-Devonian units within the Tatonduk belt cannot be evaluated for vitrinite reflectance (fig. 4), we would caution against overinterpretation of those CI data unless some other measure of thermal maturity can provide independent confirmation.

A second factor to consider is the effect of claymineral precursors and bulk geochemistry of the host

0 0 - ,13? Anchizone CO a 8? o? a? C? D(flaa :YDetritai Figure B. Scatter diagram and illitecrystallinity(CI)-vitrinite reflectance(Rm)correlation curve for Kandik River and Tatonduk belts of east-central Alaska. Cl boundaries for anchizone conform to designations of Blenkinsop (1988). Rm boundaries of 2.6 percent and 4.5 percent correspond to paleotemperatures of approximately 245°C and 305°C, based on the correlation of Barker (1988). Data from unitPza of Brabb and Churkin (1969) have been combined with those of Kandik River belt. Samples from Step Mountains include Nation River Formation, Ford Lake Shale, and Glenn Shale. Inferred footwall aureole associated with Glenn Creek fault is based on data from seven sites within Nation River Formation. Data points with question marks refer to sample sites in Canada, where belt designations are uncertain (see Underwood and others, 1989). Cl values are considered unreliable below equivalent Rm values of about 1.0 percent because of probable contamination by detrital illite/muscovite. Unit TKs of Brabb and Churkin (1969). Geologic Studies in Alaska by the U.S. Geological Survey, 1991

rock on illite diagenesis. Although we have no chemical data to provide quantitative constraints on this argument, it is obvious that the Tadonduk and Kandik River belts both contain a wide variety of rock types spanning long periods of geologic time. For example, interbeds between the mudrocks vary from chert, to dolomite and limestone, to sandstone and chert-pebble conglomerate; the mudrocks themselves differ considerably in color, texture, content of organic matter (see Underwood and others, 1989), carbonate content, and so on. Maintaining uniform geochemical conditions under these circumstances is not realistic. In comparison, the Franciscan and Shimanto study areas (fig. 9) are representative of relatively limited time intervals, consistent arc-related sediment sources, and uniform environments of deep-marine deposition. By increasing the variability of the host-rock lithologies, the scatter in CI data is bound to expand. Other published studies have led to the suggestion that unusual heating events can affect the relationship between Rm and CI. In particular, CI values tend to be suppressed with respect to Rm under conditions of rapid heating, such as the emplacement of igneous intrusions (Pevear and others, 1980; Smart and Clayton, 1985; Kisch, 1987). This is because inorganic reactions are relatively sluggish compared with changes in the organic constituents. Within the King Range terrane of California, a similar lag in CI has been documented in an apparent response to localized hy-

z w (.) a: w e:. w (.) z

(.) w J LL w a: w t:: z

z c( w Anchizone drothermal discharge and mineralization around a vent site (Laughland, 1991 ). Another factor in hydrothermal systems is the chemical effect of hot fluids (Duba and Williams-Jones, 1983). The retardation of several different clay-mineral reactions is unequivocal in Holocene waterdominated geothermal systems, and this effect has been attributed to disequilibrium conditions triggered by rapid heating (Barker and others, 1986). Thus, the effects of both rapid conductive heating and advective heat transfer, by either focused or diffuse hydrothermal systems, need to be considered during interpretations of data from east-central Alaska, particularly for high-rank strata within the Kandik River belt. Preliminary assessments of fluid-rock interactions within our study area show that the Tatonduk and Kandik River belts were affected by different systems of fluid (Shelton and others, this volume). For example, syntectonic quartz and calcite veins are widespread within the cleaved rocks of the Kandik Group and the Glenn Shale; our studies show that the calcite veins precipitated from metamorphic fluids that reached isotopic equilibration under conditions of low water-to-rock ratios and relatively high temperature. In contrast, calcite veins in the Tatonduk belt are quite rare, and those that do occur precipitated from low-temperature waters with isotopic signatures much closer to meteoric values (Shelton and others, this volume). Based on these data, - · - Ouachita Sur-Obispo Shimanto King Range -oKANDIK Rm 0.31 - 6.79 [log(CI)] r 0.60 ILLITE CRYSTALLINITY INDEX (L\029) Figure 9. Comparison among Kandik illite crystallinity - vitrinite reflectance curve and equivalent correlations for other orogenic belts. Boundaries for anchizone are same as those defined in figure 8. Sources of data and statistical correlations are as follows: Ouachita Mountains, Arkansas and Oklahoma (Guthrie and others, 1986; Houseknecht and others, 1987); Shimanto Belt, Japan (Underwood and others, in press); King Range terrane, Franciscan Complex, California (Laughland, 1991; Underwood and others, 1991 ); Sur-Obispo terrane, Franciscan Complex, California (laugh land, 1991; Underwood and others, 1991 ). Rm, mean random vitrinite reflectance; r, correlation coefficient. Statistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, Kandik Region

it is clear that the two terranes experienced fundamentally different histories of fluid migration; some of the scatter in the Rm -CI correlation, therefore, probably was caused by responses of different illite populations to the variations in pore-fluid chemistry, as well as to rates of fluid migration during the respective episodes of peak heating. A final possibility is the effect of tectonic deformation. Stress regimes across the entire study area varied drastically at the time of peak heating, from environments of passive burial by sedimentation at one end of the spectrum to zones of complex folding, pressure solution, and thrust imbrication at the other. Studies completed elsewhere show that CI values can decrease within shear zones and toward the hinges of small folds, presumably in response to increases in effective stress (Nyk, 1985; Aldahan and Morad, 1986). Similarly, Kreutzberger and Peacor ( 1988) demonstrated that CI can be affected by pressure solution, even under isothermal conditions. Merriman and Roberts ( 1985), Kemp and others ( 1985), and Mitra and Yonkee (1985) noted obvious connections between values of illite crystallinity and zones of weak or strong cleavage, and Kisch (1991) has compiled an extensive data set showing how CI data vary with different types of cleavage. On the other hand, Robinson and Bevins (1986) were unable to recognize a regular association of well-cleaved rocks and more advanced stages of illite crystallinity, so the link between cleavage and CI is not universal. One key appears to be the timing of peak heating with respect to the development of cleavage. The compilation of data shown on figure 8 does not discriminate perfectly between zones with well-developed folds and slaty cleavage and zones without any penetrative structural fabrics. In general, however, there is a clear connection between higher ranks of thermal maturity and cleavage. Virtually all of the high-rank samples that we selected from the Biederman Argillite, for example, display excellent slaty cleavage and (or) pencil structure, as do most samples of the Glenn Shale from the Kandik River belt. Many of the Rm values, particularly for rocks of the Biederman Argillite, are actually higher than expected for given values of CI (that is, data points plot above the correlation curve). The intensity and geometry of cleavage appear to be fairly uniform across the Kandik River belt, although we have not attempted to document these structural fabrics in great detail. Unfortunately, for the purposes of statistical analysis, high-rank rocks (>3.0 percent Rm) are not present in the Tatonduk belt, and low-rank rocks ( <2.0 percent Rm) are quite rare in the Kandik River belt. Therefore, our ability to make meaningful comparisons between cleaved and uncleaved rocks of the same starting composition, and over a consistent gradient in paleotemperature, is limited. Perhaps the best evidence to refute a universal link between cleavage, Geologic Studies in Alaska by the U.S. Geological Survey, 1991 paleotemperature, and lower CI values comes from the anomalous zone of higher thermal maturity within the Nation River Formation (just to the southeast of the Glenn Creek fault). These specimens yield elevated values of vitrinite reflectance (due to conductive heat transfer across the thrust), but there is no demonstrable change in CI relative to the formational average, and no cleavage. We believe that additional studies must be completed at a variety of scales before the effects of deformation fabrics on both CI and Rm can be assessed properly, but this variable undoubtedly contributed to the inconsistent results displayed in figure 8. CONCLUSIONS Values of CI and Rm collectively show that the Tatonduk belt of east-central Alaska has been exposed to significantly lower levels of thermal maturity than its neighbor, the Kandik River belt. A best-fit statistical correlation between all available pairs of Rm and CI values is consistent with limits established elsewhere for the anchizone. The equation to this curve is: Rm 0.31 - 6.79[log(CI)]. However, the amount of scatter of individual data points is unusually large, as shown by the value of the correlation coefficient (r 0.60). The relatively poor match between these popular indicators of thermal maturity probably was influenced by several factors: (1) widespread "contamination" of the illite signature due to mixing among detrital populations, authigenic illite, and illite-smectite mixed-layer phases; (2) changes in both the precursor clay-mineral assemblages and the bulk geochemistry of the host rocks; (3) pronounced fluctuations in both the chemical composition of pore fluids and the rates of fluid flow during peak heating events; (4) variations in both the rates and the physical causes of peak heating; and (5) differences in both the timing and the degree of structural disruption with respect to peak heating, particularly the presence or absence of cogenetic slaty cleavage and pencil structure. The technique of illite crystallinity works best where variations in geologic and geochemical parameters can be reduced to a minimum, and this ideal situation did not develop during the tectono-thermal evolution of eastcentral Alaska. Acknowledgments.-Mark Johnsson and Lu Haufu assisted in the field. Eric Hathon provided guidance during the XRD analyses. Superintendent Don Chase granted permission to sample in the Yukon-Charley Rivers National Preserve. Financial support to the University of Missouri was generously supplied by ARCO Alaska, Inc. We thank Gerry Van Kooten and his ARCO ·colleagues for their scientific cooperation and logistical aid. Acknowledgment is also made to the Donors of the Petroleum Research Fund, administered by

the American Chemical Society, for partial support of this research (Grant #22773-AC2 to Underwood). Dwight Bradley and Cynthia Dusel-Bacon provided many helpful editorial suggestions. REFERENCES CITED Aldahan, A.A., and Morad, S., 1986, Mineralogy and chemistry of diagenetic clay minerals in Proterozoic sandstones from Sweden: American Journal of Science, v. 286, p. Allison, C.W., 1988, Paleontology of Late Proterozoic and Early Cambrian rocks of east-central Alaska: U.S. Geological Survey Professional Paper 1449, 50 p., 18 plates. Antia, D.D.J., 1986, Kinetic method for modeling vitrinite reflectance: Geology, v. 14, p. 606-608. Awan, M.A., and Woodcock, N.H., 1991, A white mica crystallinity study of the Berwyn Hills, North Wales: Journal of Metamorphic Geology, v. 9, p. 765-773. Barker, C.E., 1983, The influence of time on metamorphism of sedimentary organic matter in selected geothermal systems, western North America: Geology, v. 1.1, p. ---1988, Geothermics of petroleum systems: Implications of the stabilization of kerogen thermal maturation after a geologically brief heating duration at peak temperature, in Magoon, L.B., ed., Petroleum systems of the United States: U.S. Geological Survey Bulletin 1870, p. 26-29. ---1989, Temperature and time in the thermal maturation of sedimentary organic matter, in Naeser, N.D., and McCulloh, T.H., eds., Thermal history of sedimentary basins, methods and case histories: New York, Springer-Verlag, p. 73-98. ---1991, Implications for organic maturation studies of evidence for a geologically rapid increase and stabilization of vitrinite reflectance at peak temperature: Cerro Prieto geothermal system, Mexico: American Association of Petroleum Geologists Bulletin, v. 75, p. 1852-1863. Barker, C.E., Crysdale, B.L., and Pawlewicz, M.J., 1986, The relationship between vitrinite reflectance, metamorphic grade, and temperature in the Cerro Prieto, Salton Sea, and East Mesa geothermal systems, Salton Trough, United States and Mexico, in Mumpton, F.A., ed., Studies in diagenesis: U.S. Geological Survey Bulletin 1578, p. 83-95. Barker, C.E., and Goldstein, R.H., 1990, Fluid-inclusion technique for determining maximum temperature in calcite and its comparison to the vitrinite reflectance geothermometer: Geology, v. 18, p. 1003-1006. Barker, C.E., and Pawlewicz, M.J., 1986, The correlation of vitrinite reflectance with maximum temperature in humic organic matter, in Buntebarth, G., and Stegena, L., eds., Paleogeothermics: New York, Springer-Verlag, p. 79-93. Blenkinsop, T.G., 1988, Definition of low-grade metamorphic zones using illite crystallinity: Journal of Metamorphic Geology, v. 6, p. 623-636. Bostick, N.H., 1979, Microscopic measurement of the level of catagenesis of solid organic matter in sedimentary rocks to aid exploration for petroleum and to determine former burial temperatures-A review, in Scholle, P.A., and Schluger, P.R., eds., Aspects of diagenesis: Society of Economic Paleontologists and Mineralogists Special Publication 26, p. 17-43. Bostick, N.H., Cashman, S., McCulloh, T.H., and Wadell, C.T., 1978, Gradients of vitrinite reflectance and present temperature in the Los Angeles and Ventura basins, California, in Oltz, D., ed., Symposium in geochemistry: Low temperature metamorphism of kerogen and clay minerals: Society of Economic Paleontologists and Mineralogists, Pacific Section, Special Publication, p. 65-96. Brabb, E.E., 1969, Six new Paleozoic and Mesozoic formations in east-central Alaska: U.S. Geological Survey Bulletin 1274-I, p. 1-26. Brabb, E.E., and Churkin, M., Jr., 1967, Stratigraphic evidence for the Late Devonian age of the Nation River Formation, east-central Alaska: U.S. Geological Survey Professional Paper 575-D, p. D4-D15. ---1969, Geologic map of the Charlie River Quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Geological Investigations Map I-973, scale 1: 250,000. Burnham, A.K., and Sweeney, J.J., 1989, A chemical kinetic model of vitrinite maturation and reflectance: Geochimica et Cosmochimica Acta, v. 53, p. 2649-2657. Churkin, M., Jr., Foster, H.L., Chapman, R.H., and Weber, F.R., 1982, Terranes and suture zones in east-central Alaska: Journal of Geophysical Research, v. 87, p. 37183730. Coney, P.J., and Jones, D.L., 1985, Accretion tectonics and crustal structure in Alaska: Tectonophysics, v. 119, p. Dembicki, H., Jr., 1984, An interlaboratory comparison of source rock data: Geochimica et Cosmochimica Acta, v. 48, p. 2641-2649. Dover, J.H., 1990, Geology of east central Alaska: U.S. Geological Survey Open-File Report 90-289, 66 p. Dover, J.H., and Miyaoka, R.T., 1988, Reinterpreted geologic map and fossil data, Charley River Quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2004, 2 sheets, scale 1:250,000. Dow, W.G., 1977, Kerogen studies and geological interpretations: Journal of Geochemical Exploration, v. 7, p. 79-87. Duba, D., and Williams-Jones, A.E., 1983, The application of illite crystallinity, organic matter reflectance, and isotopic techniques to mineral exploration: A case study in southwestern Gaspe, Quebec: Economic Geology, v. 78, p. 1350-1363. Dun oyer de Segonzac, G., 1970, The transformation of clay minerals during diagenesis and low-grade metamorphism: A review: Sedimentology, v. 15, p. 281-346. Ernst, W.G., 1974, Metamorphism and ancient continental margins, in Burk, C.A., and Drake, C.L., eds., The geology of continental margins: New York, Springer-Verlag, p. 907919. Foster, H.L., 1976, Geologic map of the Eagle Quadrangle, Alaska: U.S. Geological Survey Miscellaneous Geological Investigations Map I-922, 1:250,000, 1 sheet. Freed, R.L., and Peacor, D.R., 1989, Geopressured shale and sealing effect of smectite to illite transition: American Association of Petroleum Geologists Bulletin, v. 73, p. Statistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, Kandik Region

1223-1232. Frey, M., 1970, The step from diagenesis to metamorphism in pelitic rocks during Alpine orogenesis: Sedimentology, v. 15, p. 261-279. ---1987, Very low-grade metamorphism of clastic sedimentary rocks, in Frey, M., ed., Low-temperature metamorphism, New York, Chapman and Hall, p. 9-58. Frey, M., Teichmuller, M., Teichmuller, R., Mullis, I., Kunzi, B., Breitschmid, A., Gruner, U., and Schwizer, B., 1980, Very low-grade metamorphism in external parts of the Central Alps: Illite crystallinity, coal rank and fluid inclusion data: Eclogae Geolgicae Helveticae, v. 73, p. Gretener, P.E., and Curtis, C.D., 1982, Role of temperature and time on organic metamorphism: American Association of Petroleum Geologists Bulletin, v. 66, p. 1124-1129. Guthrie, I.M., Houseknecht, D.W., and Johns, W.D., 1986, Relationships among vitrinite reflectance, illite crystallinity, and organic geochemistry in Carboniferous strata, Ouachita Mountains, Oklahoma and Arkansas: American Association of Petroleum Geologists Bulletin, v. 70, p. 26-33. Hesse, R., and Dalton, E., 1991, Diagenetic and low-grade metamorphic terranes of Gaspe Peninsula related to the geologic structure of the Taconian and Acadian orogenic belts, Quebec Appalachians: Journal of Metamorphic Geology, v. 9, p. 775-790. Hood, A., Gutjahr, C.C.C., and Heacock, R.L., 1975, Organic metamorphism and the generation of petroleum: American Association of Petroleum Geologists Bulletin, v. 59, p. 986-996. Houseknecht, D.W., Johns, W.D., and Guthrie, I.M., 1987, Relationships among vitrinite reflectance, illite crystallinity, and organic geochemistry in Carboniferous strata, Ouachita Mountains, Oklahoma and Arkansas: Reply: American Association of Petroleum Geologists Bulletin, v. 71, p. 347. Howell, D.G., Iohnsson, M.J., Underwood, M.B., Lu Haufu, and Hillhouse, J.W., 1992, Tectonic evolution of the Kandik region, east central Alaska: Preliminary interpretations, in Bradley, D.C., and Ford, A.B., eds., Geologic studies in Alaska by the U.S. Geological Survey, 1990: U.S. Geological Survey Bulletin 1999, p. 127-140. Howell, D.G., and Wiley, T.J., 1987, Crustal evolution of northern Alaska inferred from sedimentological and. structural relations in the Kandik area: Tectonics, v. 6, p. Hunt, I.M., Lewan, M.D., and Hennet, R.I-C., 1991, Modeling oil generation with time-temperature index graphs based on the Arrhenius equation: American Association of Petroleum Geologists Bulletin, v. 75, p. 795-807. Hunziker, I.C., Frey, M., Clauer, N., Dallmeyer, R.D., Friedrichsen, H., Flehmig, W., Hochstrasser, K., Roggwiler, P., and Schwander, H., 1986, The evolution of illite to muscovite: Mineralogical and isotopic data from the Glarus Alps, Switzerland: Contributions to Mineralogy and Petrology, v. 92, p. 157-180. Issler, D.R., 1984, Calculation of organic maturation levels for offshore eastern Canada-Implications for general application of Lopatins method: Canadian Journal of Earth Science, v. 21, p. 283-304. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Kemp, A.E.S., Oliver, G.H.J., and Baldwin, J.R., 1985, Lowgrade metamorphism and accretion tectonics: Southern Uplands terrane, Scotland: Mineralogical Magazine, v. 49, p. 335-344. Kisch, H.J ., 1980, Illite crystallinity and coal rank associated with lowest-grade metamorphism of the Taveyanne graywacke in the Helvetic zone of the Swiss Alps: Eclogae Geolgicae Helveticae, v. 73, p. 753-777. ---1983, Mineralogy and petrology of burial diagenesis (burial metamorphism) and incipient metamorphism in clastic rocks, in Larsen, G., and Chilingar, G.V., eds., Diagenesis in sediments and sedimentary rocks, part 2, Developments in sedimentology, v. 25B: Amsterdam, Elsevier, p. 289-493. ---1987, Correlation between indicators of very low-grade metamorphism, in Frey, M., ed., Low-temperature metamorphism: New York, Chapman and Hall, p. 227-300. ---1990, Calibration of the anchizone: A critical comparison of illite 'crystallinity' scales used for definition: Journal of Metamorphic Geology, v. 8, p. 31-46. ---1991, Development of slaty cleavage and degree of very-low grade metamorphism: A review: Journal of Metamorphic Geology, v. 9, p. 735-750. Kisch, H.J ., and Frey, M., 1987, Appendix: Effect of sample preparation on the measured lOA peak width of illite (illite 'crystallinity'), in Frey, M., ed., Low temperature metamorphism: New York, Chapman and Hall, p. 301304. Kreutzberger, M.E., and Peacor, D.R., 1988, Behavior of illite and chlorite during pressure solution of shaly limestone of the Kalkberg Formation, Catskill, New York: Journal of Structural Geology, v. 10, p. 803-811. Kubler, B., 1968, Evaluation quantitative du metamorphisme par Ia cristallinite de l'illite: Bulletin Centre de Recherches de Pau- SNPA, v. 2, p. 385-397. Laughland, M.M., 1991, Organic metamorphism and thermal history of selected portions of the Franciscan accretionary complex of coastal California: Columbia, Mo., University of Missouri, Ph.D. dissertation, 318 pp. Laughland, M.M., Underwood, M.B., and Wiley, T.I., 1990, Thermal maturity, tectonostratigraphic terranes, and regional tectonic history: An example from the Kandik area, east-central Alaska, in Nuccio, V.F., and Barker, C.E., eds., Applications of thermal maturity studies to energy exploration: Society of Economic Paleontologists and Mineralogists, Rocky Mountain Section, Special Publication, p. 97-111. Liou, J.G., Maruyama, S., and Cho, M., 1987, Very low-grade metamorphism of volcanic and volcaniclastic rocks - mineral assemblages and mineral facies, in Frey, M., ed., Low temperature metamorphism: New York, Chapman and Hall, p. 59-113. Merriman, R.J., and Roberts, B., 1985, A survey of white mica crystallinity and polytypes in peletic rocks of Snowdonia and Llyn, North Wales: Mineralogical Magazine, v. 49, p. Middleton, M.F., 1982, Tectonic history from vitrinite reflectance: Geophysical Journal of the Royal Astronomical Society, v. 68, p. 121-132. Mitra, G., and Yonkee, W.A., 1985, Relationship of spaced

cleavage to folds and thrusts in the Idaho-Utah-Wyoming thrust belt: Journal of Structural Geology, v. 7, p. 361373. Miyaoka, R.T., 1990, Fossil locality map and fossil data for the southeastern Charley River quadrangle, east-central Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2007, 1:100,000, p. 1-46. Moore, D.M., and Reynolds, R.C., Jr., 1989, X-ray diffraction and the identification and analysis of clay minerals: New York, Oxford University Press, 332 pp. Nyk, R., 1985, Illite crystallinity in Devonian slates of the Meggen mine (Rhenish Massif): Neues Jahrbuch fur Mineralogie Monatshefte, v. 1985, p. 268-276. Ogunyomi, 0., Hesse, R., and Heroux, Y., 1980, Pre-orogenic and synorogenic diagenesis and anchi-metamorphism in Lower Paleozoic continental margin sequences of the northern Appalachians in and around Quebec City, Canada: Bulletin of Canadian Petroleum Geology, v. 28, p. 559-577. Payne, M.W., and Allison, C.W., 1981, Paleozoic continentalmargin sedimentation in east-central Alaska: Geology, v. 9, p. 274-279. Perry, E., and Hower, J., 1970, Burial diagenesis in Gulf Coast pelitic sediments: Clays and Clay Minerals, v. 18, p. 165177. Pevear, D.R., Williams, V.E., and Mustoe, G.E., 1980, Kaolinite, smectite, and K-rectorite in bentonites: Relation to coal rank at Tulameen, British Columbia: Clays and Clay Minerals, v. 28, p. 241-254. Price, L.C., 1983, Geologic time as a parameter in organic metamorphism and vitrinite reflectance as an absolute paleogeothermometer: Journal of Petroleum Geology, v. 6, p. 5-38. Reynolds, R.C., Jr., and Hower, J., 1970, The nature of interlayering in mixed-layer illite-montmorillonites: Clays and Clay Minerals, v. 18, p. 25-36. Ritter, U., 1984, The influence of time and temperature on vitrinite reflectance: Organic Geochemistry, v. 6, p. 473-480. Roberts, B., and Merriman, R.J., 1985, The distinction between Caledonian burial and regional metamorphism in metapelites from North Wales: An analysis of isocryst patterns: Journal of the Geological Society of London, v. 142, p. 615-624. Robinson, D., and Bevins, R.E., 1986, Incipient metamorphism in the Lower Palaeozoic marginal basin of Wales: Journal of Metamorphic Geology, v. 4, p. 101-113. Robinson, D., Warr, L.N., and Bevins, R.E., 1990, The illite 'crystallinity' technique: A critical appraisal of its precision: Journal of Metamorphic Geology, v. 8, p. 333-344. Smart, G., and Clayton, T., 1985, The progressive illitization of interstratified illite-smectite from Carboniferous sediments of northern England and its relationship to organic maturity indicators: Clay Minerals, v. 20, p. 455-466. Suggate, R.P., 1982, Low-rank sequences and scales of organic metamorphism: Journal of Petroleum Geology, v. 4, p. Sweeney, J.J., and Burnham, A.K., 1990, Evaluation of a simple model of vitrinite reflectance based on chemical kinetics: American Association of Petroleum Geologists Bulletin, v. 74, p. 1559-1570. Underwood, M.B., Laughland, M.M., and Kang, S.M., 1993, A comparison between organic and inorganic indicators of diagenesis and metamorphism, Upper Shimanto Group, Muroto Peninsula, Shikoku, Japan, in Underwood, M.B., ed., Thermal evolution of the Tertiary Shimanto Belt, southwest Japan: An example of ridge-trench interaction: Geological Society of America Special Paper 273 [in press]. Underwood, M.B., Laughland, M., Shelton, K., Solomon, R., Kang, S.M., Orr, R., Brocculeri, T., Bergfeld, D., and Pawlewicz, M., 1991, Correlations among paleotemperature indicators within orogenic belts: Examples from pelitic rocks of the Franciscan Complex (California), the Shimanto Belt (Japan), and the Kandik Basin (Alaska) [abs.]: Eos (American Geophysical Union, Transactions), v. 72, p. 549. Underwood, M.B., Laughland, M.M., Wiley, T.J., and Howell, D.G., 1989, Thermal maturity and organic geochemistry of the Kandik Basin region, east-central Alaska: U.S. Geological Survey Open-File Report 89-353, 41 p. Waples, D.W., 1980, Time and temperature in petroleum exploration: Application of Lopatin's method to petroleum exploration: American Association of Petroleum Geologists Bulletin, v. 64, p. 916-926. Weaver, C.E., 1960, Possible uses of clay minerals in search for oil: American Association of Petroleum Geologists Bulletin, v. 44, p. 1505-1518. Wood, D.A., 1988, Relationships between thermal maturity indices calculated using Arrhenius equation and Lopatin method; Implication for petroleum exploration: American Association of Petroleum Geologists Bulletin, v. 72, p. Wright, N.J.R., 1980, Time, temperature, and organic maturation: The evolution of rank within a sedimentary pile: Journal of Petroleum Geology, v. 2, p. 411-425. Yang, C., and Hesse, R., 1991, Clay minerals as indicators of diagenetic and anchimetamorphic grade in an overthrust belt, External Domain of southern Canadian Appalachians: Clay Minerals, v. 26, p. 211-231. Young, G.M., 1982, The late Proterozoic Tindir Group, eastcentral Alaska: Evolution of a continental margin: Geological Society of America Bulletin, v. 93, p. 759783. Reviewers: Mark johnsson and Zenon Valin Statistical Comparison Between Illite Crystallinity and Vitrinite Reflectance, Kandik Region

GEOLOGIC FRAMEWORK STUDIES-GEOLOGIC NOTE The Arctic Alaska Superterrane By Thomas E. Moore INTRODUCTION A recent major synthesis by Moore and others (1992, in press) on the geology of northern Alaska utilizes the tectonostratigraphic terrane classification of Jones and others (1987). Moore and others (1992, in press) extensively revised the terrane boundaries and stratigraphic definitions for northern Alaska shown by Jones and others (1987) .in order to incorporate recent geologic mapping and observations. However, a debate· occurred among the coauthors of Moore and others (1992, in press) concerning the use of the subterrane nomenclature used by Jones and others (1987). Some authors (and reviewers) expressed the view that subdivision of the Arctic Alaska terrane into subterranes emphasized the differences between tectonic units and downplayed evidence that the various parts of the Arctic Alaska terrane originated as a single paleogeographically connected body. Other authors argued that the terrane and subterrane units mapped by Jones and others coincide with independently mapped tectonic units of northern Alaska and hence represent a useful classification of constituent lithotectonic units. In the first view, the division of the Arctic Alaska terrane into subterranes is regarded as spurious and even misleading to an understanding of the largely known geologic history of the region. In the second view, the sub terrane units provide a nomenclature that facilitates discussion and understanding of mappable bodies of rocks with distinctive stratigraphic aspects that have demonstrably undergone significant displacement relative to neighboring units, without prejudicing the discussion by using a genetic classification. In this article I comment on some problems of the subterrane nomenclature as applied in northern Alaska and compare current usage there with the nomenclature developed for other situations in the northern Cordillera. I conclude that (1) tectonostratigraphic subdivisions of the Arctic Alaska terrane are a useful tool for tectonic investigations in northern Alaska; (2) the subdivisions of the Arctic Alaska terrane should be designated as terranes rather than subterranes, and (3) the entire entity, which has been displaced relative to North America, should be designated the Arctic Alaska superterrane. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 TERRANE TERMINOLOGY USED IN THIS PAPER The definition of terranes as "fault-bounded geologic packages of regional extent characterized by a geologic history which differs from that of neighboring terranes" of Howell and others (1985) is used in this article. The definition of Howell and others (1985) is preferred to the earlier one of Coney and others ( 1980), which has been construed by some to mean that tectonostratigraphic units are not terranes if they may be explained or interpreted as facies equivalents of one another or of cratonal North America. Howell and others (1985) also defined a composite terrane as consisting of "two or more distinct parts that became amalgamated and subsequently shared a common geologic history prior to their accretion." The terms "subterrane" and "superterrane," although in widespread use (for example, Jones and others, 1987; Csejtey and others, 1982), were not discussed in print until Coney (1989) wrote that "composite terranes***sometimes have been referred to as superterrane[s], which [are] made up of two or more 'subterranes'." I utilize herein Coney's interpretation of a superterrane as an aggregate of subordinate terranes but regard a superterrane as a designation for combinations of terranes grouped on the basis of interpretation of similar kindred or affinity as well as those that share the same geologic history after an amalgamation event. LITHOTECTONIC UNITS (TERRANES) OF NORTHERN ALASKA The Arctic Alaska terrane (Jones and others, 1987) is one of the largest lithotectonic units in Alaska. It underlies about 20 percent of the state and consists of Proterozoic to Cretaceous rocks of continental affinity (Jones and others, 1986, 1987; Plafker, 1990). The southern part of the Arctic Alaska terrane is exposed in the Brooks Range, a Mesozoic fold-thrust belt, whereas its northern part occurs under the North Slope and Beaufort Sea continental shelf, which are the foreland for the orogen. The Arctic Alaska terrane is flanked to the north by Cretaceous oceanic crust of the Canada basin and terminates to the east at high angle against cratonic

North America under Cenozoic sedimentary rocks of the MacKenzie delta. To the south, it is structurally overlain by oceanic rocks of the Angayucham terrane. Grantz and others (1991) estimated from potential field data a thickness of 30 to 45 km for the terrane under the North Slope and Brooks Range, respectively. These estimated thicknesses have recently been confirmed by seismicrefraction studies of the Trans-Alaska Crustal Transect (TACT) program of the U.S. Geological Survey (Levander and others, 1991 ). The stratigraphy of the Arctic Alaska terrane under the North Slope consists of three major rock sequences: structurally and stratigraphically complex pre-Mississippian rocks (the Franklinian sequence of Lerand, 1973), the Mississippian to Lower Cretaceous Ellesmerian sequence, and the Lower Cretaceous and younger Brookian sequence. The Ellesmerian sequence consists of northerly derived quartzose clastic and carbonate sedimentary rocks, whereas the Brookian sequence consists of lithic clastic material shed northward from the Brookian orogen (Lerand, 1973). Because the Arctic Alaska terrane is bordered to the north by the oceanic Canada basin of Cretaceous age, the northern quartz-rich source region for Paleozoic and lower Mesozoic rocks demonstrates that the Arctic Alaska terrane has been displaced relative to a once-contiguous continental area that lay to the north (present coordinates). Stratigraphic studies of the upper Paleozoic and lower Mesozoic section in the Brooks Range have shown that while many of the same stratigraphic units exposed there are present under the North Slope, thrust slices in the frontal part of Brooks Range consist of coeval but contrasting facies that were juxtaposed against each other and the North Slope sequence by extensive thrust faulting in Jurassic and Cretaceous time (Tailleur and Brosge, 1970; Martin, 1970; Mayfield and others, 1988). Estimates of shortening cannot be precisely constrained because of the reconnaissance nature of the mapping, but reconstructed facies suggest about 500 km of shortening (Oldow and others, 1987; Mayfield and others, 1988). Nonetheless, evidence of stratigraphic links between these thrust slices contrast~d with the apparent absence of such links between terranes elsewhere in the Cordillera. Accordingly, Jones and others (1987) recognized five continental tectonostratigraphic units in northern Alaska which they designated as subterranes (North Slope, Endicott Mountains, De Long Mountains, Coldfoot, and Hammond) of the Arctic Alaska terrane. Most of these units in the western Brooks Range were adapted from tectonostratigraphic units defined independently by Martin (1970) and Mayfield and others (1988). In addition, Jones and others (1987) identified three other continental lithotectonic units in northern Alaska (Venetie, Sheenjek, Kagvik) as separate terranes on the basis of their own work and the controversial investigation of Churkin and others (1979). Moore and Mull (1989) and Moore and others (1992, in press) interpreted these terranes to be the same as some of the subterranes of the Arctic Alaska terrane and defined the Slate Creek subterrane as consisting of the Venetie and southern part of the Coldfoot subterranes. A terrane map for northern Alaska and a tectonostratigraphic correlation diagram are shown in figures 1 and 2. These diagrams, modified from Moore and Mull (1989) and Moore and others (1992, in press), are revised to show the subterranes of the Arctic Alaska terrane as separate terranes and the Arctic Alaska terrane as a superterrane as proposed below. in the remainder of this paper, the discussion will be according to the nomenclature shown in these figures, except where noted. The largest terrane of the Arctic Alaska superterrane is the North Slope terrane. It consists of deformed, stratigraphically disrupted, lower Paleozoic rocks, some of which may be correlative with coeval North American rocks, and the extensive covering succession of upper Paleozoic and Mesozoic shelfal strata of the Ellesmerian sequence. This terrane is the benchmark to which all of the others of the Arctic Alaska superterrane are compared. The Endicott Mountains and De Long Mountains terranes consist of thrust slices composed of allochthonous upper Paleozoic and Mesozoic shelfal strata that are assigned to many of the same stratigraphic units as the North Slope terrane but are stratigraphically and lithologically distinct from one another and from the North Slope terrane. The Hammond and Coldfoot terranes encompass the metamorphic rocks of the southern Brooks Range and consist of lower Paleozoic rocks mostly older than those of the Endicott Mountains and De Long Mountains terranes but coeval with the older rocks of the North Slope terrane. The southernmost terrane of the Arctic Alaska terrane is the Slate Creek terrane, which is a phyllitic unit that separates the continental rocks of the Arctic Alaska superterrane from the structurally overlying oceanic rocks of the Angayucham terrane. Of particular significance for the terrane nomenclature in northern Alaska are the stratigraphies of the Coldfoot and Hammond terranes of the southern Brooks Range. These units are metamorphosed, highly deformed, and, as far as known, consist mostly of preMississippian rocks; in contrast, the North Slope, Endicott Mountains, and De Long Mountains terranes contain sedimentary upper Paleozoic and early Mesozoic successions. The Coldfoot terrane consists of penetratively deformed quartzose schist, calc-schist, metavolcanic rocks, and marble, all metamorphosed (first to blueschist and subsequently to greenschist facies) in Mesozoic time (Dusel-Bacon and others, 1989). Moore and others ( 1992, in press) divided the Coldfoot terrane into two regional quartz-rich schist units distinThe Arctic Alaska Superterrane

guished by the presence or absence of calc-schist and marble. The subjacent calc-schist unit locally yields Silurian to Devonian conodonts, but the protolith age of the noncalcareous schist unit is uncertain. The original stratigraphic relations of these units to each other are unknown, and their boundary is shown as a fault on most maps. Both units are intruded by metagranitic rocks of Proterozoic and Devonian age. The extensive Hammond terrane consists of a variety of possibly unrelated units that includes at least two distinctive lower Paleozoic carbonate platform sequences (Dumoulin and Harris, 1987 a, b), Devonian mixed clastic and carbonate units, volcaniclastic rocks of presumed Devonian age, a belt of granitic batholiths of Devonian age, amphibolite-facies lower Paleozoic metamorphic rocks, and Proterozoic metagranitic rocks. The carbonate platform sequences, juxtaposed in the Baird Mountains in the southwestern Brooks Range, overlap in age but differ in the amount of time represented; where rocks of the sequences are coeval, they display contrasting facies. Dumoulin and Harris ( 1987b) suggested that one carbonate sequence is correlative with rocks of the Seward Peninsula, whereas the other sequence appears to extend eastward throughout much of the southern Brooks Range. Amphibolite and quartz-rich schists of Proterozoic protolith age have also been identified in structural highs in the Hammond terrane, and in the Schwatka Mountains, an upper Paleozoic sequence lithologically similar to the Ellesmerian sequence is present (Mull and Tailleur, 1977). All rocks of the Hammond terrane are partially to completely metamorphosed, structurally imbricated, and isoclinally folded so that original stratigraphic relationships are ambiguous. The Hammond terrane may eventually be found to comprise two or more tectonostratigraphic units that are yet to be recognized (for example, the two carbonate sequences of Dumoulin and Harris, 1987 a, b). The continental affinity of protoliths, the structural position beneath the oceanic Angayucham terrane, and the absence of included ophiolitic belts are evidence for incorporation of the Hammond and Coldfoot terranes in the Arctic Alaska superterrane. However, the correlation North Slope terrane beneath Cretaceous and Tertiary foredeep sedimentary rocks EXPLANATION Arctic Alaska superterrane I :." : ·j North Slope terrane c;;:;:;J Endicott Mountains terrane 1888881 De Long Mountains terrane

Sheenjek River terrane

Hammond terrane

Coldfoot terrane Venetie terrane 111.1 Slate Creek terrane Angayucham terrane

150 Miles Kilometers Terrane boundary Northern Brooks Range mountain front Bathymetric contour, in meters Figure 1. Distribution of constituent terranes of Arctic Alaska superterrane in northern Alaska according to terrane nomenclature proposed in this paper (terrane boundaries from Moore and others, 1992). Geologic Studies in Alaska by the U.S. Geological Survey, 1991

of the Coldfoot terrane with the North Slope terrane is tenuous, since they share no major stratigraphic units. Likewise, the Hammond terrane shares few stratigraphic units with the North Slope terrane, although correlation of the Hammond with the North Slope terrane is plausible because of the presence of Ellesmerian rocks in the Schwatka Mountains. More detailed mapping in that small area, however, could prove, in time, that those rocks occur in a fenster much like the Ellesmerian sequence at the Mt. Doonerak fenster (figure 1) and thus may not be part of the Hammond terrane. The Coldfoot, Hammond, and North Slope terranes all contain Devonian plutonic rocks that may be interpreted as a magmatic belt that links the terranes, but restoration of the terranes by undoing Mesozoic shortening would increase the distance between the plutonic belts. This consideration led Hubbard and others ( 1987) to conclude that two Devonian plutonic belts of different origins are present in northern Alaska. Because the Devonian and older stratigraphy for the North Slope terrane is complex and possibly partly of accretionary origin, linkages between the terranes are plausible but must be considered to be conjectural. · The designation of the Hammond and Coldfoot as subterranes of the Arctic Alaska terrane as shown by Jones and others (1987) and Moore and others (1992, in press) clearly implies that these units have an integral stratigraphic and paleogeographic relationship to the remainder of the Arctic Alaska terrane. This implication is deceiving, however, because the stratigraphy and structure of these regions are complex and have been studied only at reconnaissance scale, and because the nature of their premetamorphic relationship has not been demonstrated. For example, the paleogeographic relationship of the contrasting carbonate platform sequences of the Hammond terrane described by Dumoulin and Harris (1987a, b) has not been investigated in detail and might indicate substantial amounts of shortening that is not explained by current stratigraphic models. Designation of both carbonate sequences as part of a single subterrane of a larger terrane implies that the observed stratigraphic differences and absence of firm tectonic correlation are a nuance of little importance to tectonic models. I suggest that the tectonic significance of these carbonate units, as well as the stratigraphic complexity and lithological diversity of the Hammond and Coldfoot terranes, are not presently understood and are a subject for careful consideration. Classification of the Hammond and Coldfoot as terranes, as I propose here, rather than as subterranes highlights rather than obscures this uncertainty. ARCTIC ALASKA SUPERTERRANE EXPLANATION mJ Foreland and successor basin

Conglomerate

Carbonate rocks

Shale WjJ Metasedimentary rocks [i] Uncertain stratigraphic relationship

Syntectonic sedimentary rocks Will Sandstone

Chert (g) Volcanic rocks Phyllite D Non-deposition m Metagranitic rocks Slate Creek and Venetie terranes Coldfoot Hammond De Lon_g Endicott terrane terrane Mountams Mountians and terrane Sheenjek River terranes North Slope terrane sequence Pre-Miss. rocks Ma 0 Tertiary Earl( Cre. Jurassic Triassic Permian Penn. Miss. Dev. Silurian Ord. Cam b. Prot. Figure 2. Terrane correlation diagram for terranes of Arctic Alaska superterrane according to terrane nomenclature proposed in this paper (modified from Moore and others, 1992). Geologic time scale used in this figure is from Palmer (1983). The Arctic Alaska Superterrane

Similarly, the relationship of the Slate Creek terrane, which consists of Devonian quartz-rich turbidites, phyllonite, and melange, to the other terranes of the Arctic Alaska superterrane is unclear. This unit has been interpreted as representing a deep-marine equivalent of part of the Endicott Mountains terrane (Murphy and Patton, 1988), but recent work suggests that it may be a tectonic unit produced by deformation associated with down-to-the-south extensional deformation in mid-Cretaceous time (Miller and Hudson, 1991; Moore and others, 1991). Patton and Box (1989) classified the Slate Creek as a thrust panel of the Angayucham terrane, whereas Moore and others (1992, in press) interpreted it as a subterrane of the Arctic Alaska terrane. Because there is uncertainty about its nature and correlation, the Slate Creek should be regarded as a separate terrane so that all models for its origin may be considered. PROPOSED NEW TERRANE TERMINOLOGY FOR NORTHERN ALASKA On the basis of the foregoing discussion, I conclude that the subterrane nomenclature of Jones and others ( 1987) as used in northern Alaska is misleading because it implies a genetic relationship between and within tectonostratigraphic units that, at least in the southern Brooks Range, cannot be firmly established on the basis of critical examination of observed field relations. The stratigraphies of the Endicott Mountains terrane, De Long Mountains terrane, and stratigraphically higher part of the North Slope terrane, in contrast, are structurally and stratigraphically less complicated and are more thoroughly investigated than the Hammond, Coldfoot, and Slate Creek terranes. Although these terranes of the northern Brooks Range display many distinctive stratigraphic and lithologic differences, they have been interpreted to be stratigraphically related and to represent different parts of an upper Paleozoic and Lower Mesozoic continental margin that was collapsed in Jurassic and Early Cretaceous time (for example, Oldow and others, 1987; Mayfield and others, 1988; Moore and Mull, 1989; Grantz and others, 1991; Moore and others, 1992, in press). This is the basis for their designation as subterranes of the Arctic Alaska terrane as shown by Jones and others (1987), Moore and Mull (1989), and Moore and others (1992, in press). However, if these allochthonous units are to be regarded as subterranes on the basis of their inferred origin, how well documented by geological data must the interpretation of their common origin be? Because the answer to this question is debatable, not easily measured by any standard, and could be revised in the light of new data, I suggest that the genesis of these units should not be considered at all for the purpose of their classification, but instead, classi242 Geologic Studies in Alaska by the U.S. Geological Survey, 1991 fication should be based on map and stratigraphic data alone. This descriptive approach results in a terminology that is based on observed geologic data rather than on a specific paleogeographic model. Thus, because the tectonostratigraphic units of the northern Brooks Range have distinctive stratigraphies and have been displaced a significant but undetermined distance relative to each other (for example, Martin, 1970; Mayfield and others, 1988), I suggest that they also should be regarded as separate terranes rather than subterranes. If, for the purpose of a tectonic synthesis or other regional investigation, there is a need to discuss combinations of terranes by their inferred origin, what nomenclature should then be used? Elsewhere in the northern Cordillera, groups of terranes have been discussed as composite terranes or superterranes. For example, Jones and others (1987) divided southern Alaska south of the Denali fault into about 10 terranes. Based on geologic data, Csejtey and others (1982) argued that two of the largest terranes in this area, the Peninsular and Wrangellia terranes, were geologically connected by Middle Jurassic time, on this basis, they combined the terranes into the Talkeetna superterrane. Later work showed that the Wrangellia and Alexander terranes have been contiguous since Pennsylvanian time (Gardner and others, 1988), and most workers now regard the Peninsular, Wrangellia, and Alexander terranes as the PeninsularAlexander-Wrangellia (PAW) superterrane (Plafker and others, 1989). This superterrane composes about 20 percent of Alaska. Similarly, Monger and Berg (1987) divided the Canadian Cordillera into 12 terranes (including the Alexander and Wrangellia terranes). Monger and others (1982) concluded on the basis of stratigraphic evidence that most of these terranes can be combined into two composite terranes, which they designated composite terrane I and composite terrane IT (the PeninsularAlexander-Wrangellia superterrane). Recently, Plafker ( 1990) has grouped the terranes of the northern Cordillera into eight composite terranes in order to discuss his tectonic model for the development of the region. These cases show that recent classifications have retained previously defined terranes as building blocks that can be combined in various ways as the genetic relations between terranes are reinterpreted or as new data are reported. This nomenclatural system retains mapped terranes as separate entities and presents hypotheses of their inferred paleogeographic origin as combinations of terranes by designating them as composite terranes or superterranes. This nomenclature has been utilized by a number of workers (for example, Oldow and others, 1989; Nokleberg and others, 1992) for the purpose of grouping terranes for regional analysis where simplifying assumptions are useful for discussions of the larger tectonic features. It is also consistent with Coney's (1989) view that subterranes are terranes that compose composite terranes

and superterranes. Thus, for purposes of consistency and comparability with terminology in use elsewhere in the northern Cordillera, I propose that the subterranes of the Arctic Alaska terrane of Jones and others (1987) should be regarded as independent terranes and that their Arctic Alaska terrane should be regarded as a superterrane or composite terrane. CONCLUSION The nomenclature of natural systems should not presuppose a particular genetic model, regardless of the degree of acceptance of the model by knowledgeable scientists, because it discourages formulation of alternative hypotheses. In the case of the terrane nomenclature of northern Alaska, the classification of its mapped constituent tectonostratigraphic units as subterranes on the basis of the prevailing interpretation of their common origin constitutes such a presupposition. Despite the conclusion of many workers, including myself, that at least some of the tectonostratigraphic units of northern Alaska originated as part of the same continental margin, the present structural position is that of displaced and fault-bounded stratigraphically distinct bodies of rock. Thus, they should be classified as independent terranes whose interrelationships in the future can be reevaluated free of the specific model implied by the nomenclature established by Jones and others (1987). For consideration as a single entity the combination of terranes in northern Alaska that are interpreted as fragments of a continental margin and for the purpose of comparison of this entity with groups of terranes of common origin established elsewhere in the northern Cordillera, I propose that the combined terranes be designated the Arctic Alaska superterrane. REFERENCES CITED Churkin, M., Jr., Nokleberg, W.J., and Huie, C., 1979, Collision-deformed Paleozoic continental margin, western Brooks Range, Alaska: Geology, v. 7, no. 8, p. 379-383. Coney, P.J., 1989, Structural aspects of suspect terranes and accretionary tectonics in western North America: Journal of Structural Geology, v. 11, p. 107-125. Coney, P.J., Jones, D.L., and Monger, J.W.H., 1980, Cordilleran suspect terranes: Nature, v. 288, p. 329-333. Csejtey, Bela, Jr., Cox, D.P., and Evarts, R.C., 1982, The Cenozoic Denali fault system and the Cretaceous accretionary development of southern Alaska: Journal of Geophysical Research, v. 87, p. 3741-3754. Dumoulin, J.A., and Harris, A.G., 1987a, Lower Paleozoic carbonate rocks of the Baird Mountains quadrangle, western Brooks Range, Alaska, in Tailleur, I.L., and Weimer, Paul, eds., Alaskan North Slope geology: Bakersfield, Calif., Pacific Section, Society of Economic Paleontologists and Mineralogists, v. 50, p. 311-336. ---1987b, Cambrian through Devonian carbonate rocks of the Baird Mountains, western Brooks Range, Alaska [abs.]: Geological Society of America Abstr~ts with Programs,v. 19,no.6,p. 373-374. Dusel-Bacon, Cynthia, Brosge, W.P., Till, A.B., Doyle, E.O., Mayfield, C.F., Reiser, H.N., and Miller, T.P., 1989, Distribution, facies, ages, and proposed tectonic associations of regionally metamorphosed rocks in northern Alaska: U.S. Geological Survey Professional Paper 1497-A, 44 p., 2 sheets, scale 1 : 1 ,000,000. Gardner, M.C., Bergman, S.C., MacKevett, E.M., Jr., Plafker, George, Campbell, R.C., Cushing, G.W., Dodds, C.J., and McClelland, W.D., 1988, Middle Pennsylvanian pluton stitching of Wrangellia and the Alexander terrane, Wrangell Mountains, Alaska: Geology, v. 16, p. 967-971. Grantz, Arthur, Moore, T.E., and Roeske, Sarah, 1991, Gulf of Alaska to Arctic Ocean: Boulder, Colo., Geological Society of America, Centennial Continent/Ocean Transect #15, 72 p., scale 1 :500,000, 3 sheets. Howell, D.G., Jones, D.L., and Schermer, E.R., 1985, Tectonostratigraphic terranes of the circum-Pacific region, in Howell, D.G., ed., Tectonostratigraphic terranes of the circum-Pacific region: Houston, Texas, Circum-Pacific Council for Energy and Mineral Resources, Earth Science Series, v. 1, p. 3-30. Hubbard, R.J., Edrich, S.P., and Rattey, R.P., 1987, Geologic evolution and hydrocarbon habitat of the Arctic Alaska microplate, in Tailleur, I.L., and Weimer, Paul, eds., Alaskan North Slope geology: Bakersfield, Calif., Pacific Section, Society of Economic Paleontologists and Mineralogists, v. 50, p. 797-830. Jones, D.L., Silberling, N.J., and Coney, P.J., 1986, Collision tectonics in the Cordillera of western North America: Examples from Alaska, in Coward, M.P., and Ries, A.C., eds., Collision tectonics: Geological Society of London Special Publication 19, p. 367-387. ---1987, Lithotectonic terrane map of Alaska (west of the 14lst meridian): U.S. Geological Survey Miscellaneous Field Studies Map MF-1874-A, 1 sheet, scale 1:2,500,000. Lerand, Monti, 1973, Beaufort Sea, in McCrossam, R.G., ed., The future petroleum provinces of Canada-Their geology and potential: Canadian Society of Petroleum Geology Memoir 1, p. 315-386. Levander, A.R., Wissinger, E.S., Fuis, G.S., and Lutter, W.J., 1991, The 1990 Brooks Range seismic experiment: Near vertical incidence reflection images [abs.]: Eos (American Geophysical Union Transactions), v. 72, no. 44 suppl., p. Martin, A.J., 1970, Structure and tectonic history of the western Brooks Range,. De Long Mountains, and Lisburne Hills, northern Alaska: Geological Society of America Bulletin, v. 81, p. 3605-3622. Mayfield, C.F., Tailleur, I.L., and Ellersieck, Inyo, 1988, Stratigraphy, structure, and palinspastic synthesis of the western Brooks Range, northwestern Alaska, in Gryc, George, ed., Geology and exploration of the National Petroleum Reserve in Alaska, 1974 to 1982: U.S. The Arctic Alaska Superterrane

Geological Survey Professional Paper 1399, p. 143-186. Miller, E.L., and Hudson, T .L., 1991, Mid-Cretaceous extensional fragmentation of a Jurassic-Early Cretaceous compressional orogen, Alaska: Tectonics, v. 10, p. 781796. Monger, J.W.H., and Berg, H.C., 1987, Lithotectonic terrane map of western Canada and southeastern Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF1874-B, 1 sheet, scale 1:2,500,000, 12 p. Monger, J.W.H., Price, R.A., and Tempelman-Kluit, D.J., 1982, Tectonic accretion and the origin of the two major metamorphic and plutonic welts in the Canadian Cordillera: Geology, v. 10, p. 70-75. Moore, T.E., and Mull, Gil, 1989, Geology of the Brooks Range and North Slope, in Schmidt, R.A.M., Nokleberg, W.J., and Page, R.A., Alaska geological and geophysical transect, Valdez to Coldfoot, Alaska: American Geophysical Union, Field Trip Guidebook T104, p. 107-131. Moore, T.E., Nokleberg, W.J., Jones, D.L., Till, A.B., and Wallace, W.K., 1991, Contrasting structural levels of the Brooks Range orogen along the Trans-Alaska Crustal Transect (TACT) [abs.]: Eos (American Geophysical Union Transactions), v. 72, no. 44 suppl., p. 295. Moore, T.E., Wallace, W.K., Bird, K.J., Karl, S.M., Mull, C.G., and Dillon, J.T., 1992, Stratigraphy, structure, and geologic synthesis of northern Alaska: U.S. Geological Survey Open-File Report 92-330, 283 p., 1 pl. ---in press, Geology of northern Alaska, in Plafker, George, and Berg, H.C., eds., The geology of Alaska: Boulder, Colo., Geological Society of America, The geology of North America, v. Gl. Mull, C.G., and Tailleur, I.L., 1977, Sadlerochit(?) Group in the Schwatka Mountains, south-central Brooks Range, in Blean, K.M., ed., The United States Geological Survey in Alaska: Accomplishments during 1976: U.S. Geological Survey Circular 751-B, p. B27-B29. Murphy, J.M., and Patton, W.W., Jr., 1988, Geologic setting and petrography of the phyllite and metagraywacke thrust panel, north-central Alaska, in Galloway, J.P., and Hamilton, T.D., eds., Geologic studies in Alaska by the U.S. Geological Survey during 1987: U.S. Geological Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Survey Circular 1016, p. 1 04-108. Nokleberg, W.I., and sixteen others, 1992, Circum-North Pacific tectono-stratigraphic terrane map [abs.]: Submitted to International Geologic Congress, Kyoto, Japan. Oldow, J.S., Bally, A.W., Ave Lallemant, H.G., and Leeman, W.P., 1989, Phanerozoic evolution of the North American Cordillera; United States and Canada, in Bally, A.W., and Palmer, A.R., eds., The geology of North America-An overview: Boulder, Colo., Geological Society of America, Geology of North America, v. A, p. 139-232. Oldow, J.S., Seidensticker, C.M., Phelps, J.C., Julian, F.E., Gottschalk, R.R., Boler, K.W., Handschy, J.W., and Ave Lallemant, H.G., 1987, Balanced cross sections through the central Brooks Range and North Slope, Arctic Alaska: American Association of Petroleum Geologists Publication, 19 p., 8 pis., scale 1:200,000. Palmer, A.R., 1983, The Decade of North American Geology 1983 time scale: Geology, v. 11, p. 503-504. Patton, W.W., Jr., and Box, S.E., 1989, Tectonic setting of the Yukon-Koyukuk basin and its borderlands, western Alaska: Journal of Geophysical Research, v. 94, no. Bll, p. 15,807-15,820. Plafker, George, 1990, Regional geology and tectonic evolution of Alaska and adjacent parts of the northeast Pacific Ocean margin: Pacific Rim Congress 90, Proceedings: Australasian Institute of Mining and Metallurgy, Queensland, Australia, p. 841-853. Plafker, George, Nokleberg, W.J., and Lull, J.S., 1989, Bedrock geology and tectonic evolution of the Wrangellia, Peninsular, and Chugach terranes along the Trans-Alaska crustal transect in the Chugach Mountains and southern Copper River basin, Alaska: Journal of Geophysical Research, v. 94, no. B4, p. 4255-4295. Tailleur, I.L, and Brosge, W .P ., 1970, Tectonic history of northern Alaska, in Adkison, W.L., and Brosge, M.M., eds., Geological seminar on the North Slope of Alaska, Proceedings: American Association of Petroleum Geologists Pacific Section Meeting, Los Angeles, Calif., p. E1-E19. Reviewers: Warren J. Nokleberg and Arthur Grantz

BIBLIOGRAPHIES U.S. Geological Survey Reports on Alaska Released in 1991 Compiled by Ellen R. White [Some reports dated 1990 did not become available until 1991; they are included in this listing.] Also listed is the 1992 release (Bulletin 1999) because it covers research done in 1990. ABBREVIATIONS B1950 Goldfarb, R.J., Nash, J.T., and Stoeser, J.W., eds., 1990, Geochemical studies in Alaska by the U.S. Geological Survey, 1989: U.S. Geological Survey Bulletin 1950, Chaps. A-F, variously paged. [Chapters are not available separately but are indexed in this bibliography.] B1999 Bradley, D.C., and Ford, A.B., eds., 1992, Geologic studies in Alaska by the U.S. Geological Survey, 1990: U.S. Geological Survey Bulletin 1999, 244 p. Good, E.E., Slack, J.F., and Kotra, R.K., 1991, USGS research on mineral resources-1991, program and abstracts: U.S. Geological Survey Circular 1062, 99 p. [Annual V.E. McKelvey Forum on Mineral and Energy Resources, 7th, Reno, Nevada, 1991.] Casadevall, T.J., ed., 1991, First International Symposium on Volcanic Ash and Aviation Safety, Program and Abstracts, Seattle, Washington, July 8-12, 1991: U.S. Geological Survey Circular 1065, 58 p. OF90-680 Jacobson, M.L., compiler, 1990, National Earthquake Hazards Reduction Program; Summaries of technical reports, v. XXXI: U.S. Geological Survey Open-File Report 90-680, 603 p. dF91-352 Jacobson, M.L., compiler, 1991, National Earthquake Hazards Reduction Program; Summaries of technical reports, v. XXXII: U.S. Geological Survey Open-File Report 91-352, 707 p. OF91-631 Carlson, P.R., ed., 1991, Sediment of Prince William Sound, beach to deep fjord floor, a year after the EXXON Valdez oil spill: U.S. Geological Survey Open-File Report 91-631, 101 p. Abers, G.A., 1990, Seismic monitoring of .the Shumagin seismic gap, Alaska, in OF90-680, p. 1-3. --1991, Seismic monitoring of the Shumagin seismic gap, Alaska, in OF91-352, p. 1-4. Abers, G.A., and Jacob, K.H., 1991, Analysis of seismic data from the Shumagin seismic gap, Alaska, in OF91-352, p. Allen, M.S., 1990, Gold anomalies and newly identified gold occurrences in the Lime Hills quadrangle, Alaska, and their association with the Hartman sequence plutons: B1950, Chap. F., p. F1-F16. Arbogast, B.F., Erickson, B.M., Gray, J.E., and McNeal, J.M., 1991, Analytical results and sample locality map of moss, moss-sediment, and willow samples from the Iditarod quadrangle, Alaska: U.S. Geological Survey Open-File Report 91-380-A (paper version) 101 p., scale 1 :250,000; 91-380-B (5 114 inch diskette version, text in ASCII file format, IBM compatible computer required, using MS DOS.) Barker, J .C., 1991, Investigation of rare-earth and associated elements, Zane Hills pluton, northwestern Alaska: U.S. Geological Survey Open-File Report 36-91, 39 p. including foldout maps. Barnes, D.F., 1991, Map showing geologic interpretation of aeromagnetic data for the Chugach National Forest, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-1645-H, 8 p., scale 1:250,000. Barnes, D.F., and Kelley, J.S., 1991, Applications of gravity data to studies of framework geology, evaluation of mineral deposits, and mineral prospecting in northwestern Alaska [abs.]: C1062, p. 3-4. Beavan, John, 1990, Crustal deformation measurements in the Shumagin seismic gap, Alaska, in OF90-680, p. 176-180. ---1991, Crustal deformation measurements in the Shumagin seismic gap, Alaska, in OF91-352, p. 229-231. Beget, J.E., Swanson, S.E., and Stone, David, 1991, Frequency and regional extent of ash eruptions from Alaskan volcanoes [abs.]: C1065, p. 13. Bering Sea EEZ-Scan Scientific Staff, 1991, Atlas of the U.S. Exlusive Economic Zone, Bering Sea: U.S. Geological U.S. Geological Survey Reports on Alaska Released in 1991

Survey Miscellaneous Investigations Series Map I-2053, 147 p. [l0x22 inches, in color.] Bird, K.J., 1991, Geology, play descriptions, and petroleum resources of the Alaskan North Slope (petroleum provinces 58-60): U.S. Geological Survey Open-File Report 88-450Y, 52p. Blodgett, R.B., Clough, J.G., Harris, A.G., and Robinson, M.S., 1992, The Mount Copleston Limestone, a new Lower Devonian formation in the Shublik Mountains, northeastern Brooks Range, Alaska: B 1999, p. 3-7. Box, S.E., and Elder, W.P., 1992, Depositional and biostratigraphic framework of the Upper Cretaceous Kuskokwim Group, southwestern Alaska: B1999, p. 8-16. Boyd, T .M., 1990, Analysis of the 1957 Andreanof Islands earthquake, in OF90-680, p. 54-60. ---1991, Analysis of the 1957 Andreanof Islands earthquake, in OF91-352, p. 76-85. Bradley, D.C., and Ford, A.B., 1992, Introduction: B1999, p. Bradley, D.C., and Kusky, T.M., 1992, Deformation history of the McHugh Complex, Seldovia quadrangle, south-central Alaska: B1999, p. 17-32. Brew, D.A., Drew, L.J., Schmidt, J.M., Root, D.H., and Huber, D.F., 1991, Assessment of undiscovered mineral resources, Tongass National Forest, southeastern Alaska [abs.]: C1062, p. 6. ---1991, Undiscovered locatable mineral resources of the Tongass National Forest and adjacent lands, southeastern Alaska: U.S. Geological Survey Open-File Report 91-10, 370 p., 12 pis., scale 1:250,000. Brew, D.A., and Drinkwater, J.L., 1991, Tongass Timber Reform Act Wilderness areas, supplement to U.S. Geological Survey Open-File Report 91-10 (Undiscovered locatable mineral resources of the Tongass National Forest and adjacent lands, southeastern Alaska): U.S. Geological Survey Open-File Report 91-343, 35 p., foldout map. Bronston, M.A., 1990, A view of sea-floor mapping priorities in Alaska from the mining industry: C1052, p. 86-91. Cady, J.W., 1990, Aeromagnetic map of Alaska from lat 65°- 680 N., long 141°-162° W.: Color-shaded relief: U.S. Geological Survey Geophysical Investigations Map GP992, 2 sheets, scale 1:500,000. Carlson, P.R., 1991, Conclusions and recommendations: 1989 Prince William Sound oil spill, the following year, in OF91-631, p. 99-101. Carlson, P.R., Barnes, P.W., Hayden, Bran, and Carkin, B.A., 1991, Morphology of bottom sediment of Prince William Sound along the oil spill trajectory, in OF91-631, p. 1-30. Carter, L.D., and Hillhouse, J.W., 1992, Age of the late Cenozoic Bigbendian marine transgression of the Alaskan Arctic coastal plain: Significance for permafrost history and paleoclimate: B1999, p. 44-51. Cathrall, J.B., and Antweiler, J.C., 1992, Occurrence of platinum-group elements in some gold-mining districts of Alaska: B1999, p. 33-43. Cathrall, J.B., Carlson, R.R., Antweiler, J.C., and Mosier, E.L., 1991, Platinum group elements in native gold, alluvium concentrates, and mineralized rock concentrates from some gold mining districts of Alaska: U.S. Geological Survey Open-File Report 91-348, 36 p., 1 sheet. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Clendenen, W.S., Sliter, W.V., and Byrne, Tim, 1992, Tectonic implications of the Albatross sedimentary sequence, Sitkinak Island, Alaska: B1999, p. 52-70. Coel, R.J., Crock, J.G., and Kyle, J.R., 1991, Biogeochemical studies of gold in a placer deposit, Livengood, Alaska: U.S. Geological Survey Open-File Report 91-142, 51 p. Combellick, R.A., and Reger, R.D., 1991, Investigation of peat stratigraphy in estuarine flats near Anchorage, Alaska, as a means of detemiining recurrence intervals of major earthquakes, in OF91-352, p. 162-163. Cooper, A.K., Marlow, M.S., Geist, E.L., and Smith, G .L., 1991, Multichannel seismic-reflection profiles collected in 1980 from the southern Bering Sea, Alaska: U.S. Geological Survey Open-File Report 91-317, 5 p. Cronin, T.M., Briggs, W:M., Jr., Brouwers, E.M., Whatley, R.C., Wood, Adrian, and Cotton, M.A., 1991, Modem arctic Podocopid ostracode database: U.S. Geological Survey Open-File Report 91-385, 51 p. Crowe, D.E., Shanks, W.C., III, and Valley, J.W., 1991, Lasermicroprobe studies of sulfur isotopes in stockwork and massive sulfide ores, Rua Cove Mine, south-central Alaska [abs.]: C1062, p. 13-14. Csejtey, Bela, Jr., 1992, Discrepancies between geologic evidence and rotational models-Talkeetna Mountains and adjacent areas of south-central Alaska: B 1999, p. 71-80. Curtis, S.M., Ellersieck, Inyo, Mayfield, C.F., and Tailleur, I.L., 1990, Reconnaissance geologic map of the De Long Mountains A-1 and B-1 quadrangles and part of the C-1 quadrangle, Alaska: U.S. Geological Survey Miscellaneous Investigations Series Map I-1930, 2 sheets, scale 1:63,360, color. [This map supersedes USGS Open-File Report 83-185.] Dean, K.G., and Whiting, Larry, 1991, Analysis of satellite images of Redoubt Volcano plumes [abs.]: Cl065, p. 17. Detterman, R.L., 1990, Stratigraphic correlation and interpretation of exploratory wells, Alaska Peninsula: U.S. Geological Survey Open-File Report 90-279, 56 p., 2 pis. [correlation charts.] Dickinson, K.A., and Skipp, G.L., 1992, Clay mineral depositional facies and uranium resource potential in part of the Tertiary Kenai Group, Kenai Peninsula, Alaska: B1999, p. 81-99. Dusel-Bacon, Cynthia, 1991, Metamorphic history of Alaska: U.S. Geological Survey Open-File Report 91-556, 48 p., 2 sheets, scale 1 :2,500,000. Dusel-Bacon, Cynthia, Brew, D.A., and Douglass, S.L., 1991, Metamorphic facies map of southeastern Alaska-Distribution, facies, and ages of regionally metamorphosed rocks: U.S. Geological Survey Open-File Report 91-29, 46 p., 2 sheets, scale 1:1,000,000. Elder, W.P., and Miller, J.W., 1991, Maps showing fossil localities and checklists of Jurassic and Cretaceous macrofauna of western Alaska: U.S. Geological Survey Open-File Report 91-629, 71 p., 7 pis., scale 1:500,000. EIJersieck, Inyo, Curtis, S.M., Mayfield, C.F., and Tailleur, I.L., 1990, Reconnaissance geologic map of the De Long Mountains A-2 and B-2 quadrangles and part of the C-2 quadrangle, Alaska: U.S. Geological Survey Miscellaneous Investigations Series Map I-1931, 2 sheets, scale 1:63,360, color. [This map supersedes USGS Open-File

Report 83-184.] Feist, Monique, and Brouwers, Elisabeth, 1991, A new Tolypella from the Ocean Point dinosaur locality, North Slope, Alaska, and the Late Cretaceous to Paleocene Nitelloid charophytes: U.S. Geological Survey Bulletin 1990-F, 7 p., 1 pl. Ferrians, O.J., Jr., 1991, Bibliography of Quaternary geology, Copper River basin and adjacent areas, south-central Alaska: U.S. Geological Survey Open-File Report 91107-A, 20 p. (paper version); 91-107-B (IBM PC, XT, AT, or compatible diskette version). Frederiksen, N.O., 1990, Pollen zonation and correlation of Maastrichtian marine beds and associated strata, Ocean Point dinosaur locality, North Slope, Alaska: U.S. Geological Survey Bulletin 1990, 31 p., 7 pls. Frost, G.M., and Stanley, R.G., 1991, Compiled geologic and Bouguer gravity map of the Nenana basin area, central Alaska: U.S. Geological Survey Open-File Report 91562, 30 p., 2 pls., scale 1:250,000. Frost, T .P ., 1990, Geology and geochemistry of mineralization in the Bethel quadrangle, southwestern Alaska: B 1950, Chap. C, p. C1-C9. Frost, T.P., and Box, S.E.,. 1991, Lithologic and tectonic controls on mercury mineralization in the Bethel 1 ° x 3 ° quadrangle, southwestern Alaska [abs.]: C1062, p. 29-31. Gaccetta, J.D., and Church, S.E., 1989, Lead isotope data base for sulfide occurrences from Alaska, December 1989: U.S. Geological Survey Open-File Report 89-688, 60 p. Galloway, J.P., Huebner, Mark, Lipkin, Robert, and Dijkmans, J.W.A., 1992, Early Holocene calcretes from the subarctic active Nogahabara Sand Dune field, northern Alaska: B1999, p. 100-111. Gehrels, G .E., 1991, Geologic map of Long Island and southem and central Dall Island, southeastern Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF2146, scale 1:63,360. Goldfarb, R.J., Bailey, E.A., Folger, P.F., and Schmidt, J.M., 1991, The use of heavy-mineral concentrate data to show geochemical favorability for zinc-lead-silver and copper- (cobalt) mineral occurrences in the Baird Mountains quadrangle, northwest Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2151, scale 1:250,000. Goldfarb, R.J., Gray, J.E., Pickthom, W.J., Gent, C.A., and Cieutat, B.A., 1990, Stable isotope systematics of epithermal mercury-antimony mineralization, southwestern Alaska: B1950, Chap. E., p. E1-E9. Goldfarb, R.J ., and Pickthom, W .J ., 1991, Synorogenic auriferous fluids of the Juneau gold belt, southeast Alaska-Stable-isotope evidence for a deep crustal origin [abs.]: C1962, p. 32-33. Gough, L.P., Severson, R.C., Harms, T.F., Papp, C.S.E., and Shacklette, H.T., 1991, Biogeochemistry of selected plant materials, Alaska: U.S. Geological Survey Open-File Report 91-292, 30 p. Gray, J.E., Detra, D.E., Goldfarb, R.J., and Slaugther, K.E., 1991, Geochemical exploration criteria for epithermal cinnabar and stibnite deposits, southwestern Alaska [abs.]: C1062, p. 34-35. Gray, J.E., Frost, T.P., Goldfarb, R.J., and Detra, D.E., 1990, Gold associated with cinnabar- and stibnite-bearing deposits and mineral occurrences in the Kuskokwim River region, southwestern Alaska: B1950, Chap. D, p. D1-D6. Grybeck, D.J., 1991, Tapping the potential mineral resources of Alaska, in United States Geological Survey Yearbook 1990, p. 45-47. Grybeck, D.J., Nokleberg, W.J., and Bundtzen, T.K., 1991, Comparative metallogeny of the Soviet Far East and Alaska [abs.]: C1062, p. 36. Hill, P.L., 1991, Bibliographies and location maps of publications on aeromagnetic and aeroradiometric surveys for Hawaii and Alaska: U.S. Geological Survey Open-File Report 91-370-E, 35 p. [revised 3-1-91.] Hobbs, P.V., Radke, L.F., and Coffman, D.J., 1991, Airborne lidar detection and in situ measurements of ash emissions from the 1990 volcanic eruptions of Mount Redoubt [abs.]: C1065, p. 24. Boblitt, R.P., 1991, Lightning detection and location as a remote ash-cloud monitor at Redoubt Volcano, Alaska [abs.]: Ct065, p. 24. Hopkins, D.M., Gray, J.E., Hageman, P.L., McDougal, C.M., and Slaughter, K.E., 1991, Gold, mercury, tellurium, and thallium data and sample locality map of stream-sediment samples from the lditarod quadrangle, Alaska: U.S. Geological Survey Open-File Report 91-283-A, 37 p., 1 sheet, scale 1:250,000 (paper version); 91-283-B (5.25 inch, 360K diskette version). Hopkins, D.M., Gray, J.E., and Slaughter, K.E., 1991, Lowlevel gold determinations by use of flow injection analysis-atomic absorption spectrophotometry-An application to precious-metal-resource assessment in the lditarod 1° x 3° quadrangle, southwestern Alaska [abs.]: C1062, p. 39. Howell, D.G., Bird, K.J., Huafu, Lu, and Johnsson, M.J., 1992, Tectonics and petroleum potential of the Brooks Range fold and thrust belt-A progress report: B1999, p. 112126. Howell, D.G., Johnsson, M.J., Underwood, M.B., Huafu, Lu, and Hillhouse, J.W., 1992, Tectonic evolution of the Kandik region, east-central Alaska: Preliminary interpretations: B1999, p. 127-140. Karl, S.M., 1992, Arc and extensional basin geochemical and tectonic affinities for the Maiyumerak basalts in the westem Brooks Range: B1999, p. 141-155. Karl, S.M., Goldfarb, R.J., Kelley, K.D., Sutphin, D.M., Finn, C.A., Ford,A.B., and Brew, D.A., 1991, Mineral-resource potential of the Sitka 1 ° x 3 ° quadrangle, southeastern Alaska [abs.]: C1062, p. 45-46. Kelley, J.S., 1990, Generalized geologic map of the Chandler Lake quadrangle, north-central Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2144-A, 19 p., scale 1 :250,000. Kelley, K.D., 1990, Interpretation of geochemical data from AdmiraJty Island, Alaska-Evidence for volcanogenic massive sulfide mineralization: B1950, Chap. A, p. A1A9. Kilburn, J.E., Box, S.E., Goldfarb, R.J., and Gray, J.E., 1992, Geochemically anomalous areas in the eastern Goodnews Bay 1° by 3 ° quadrangle, southwest Alaska: B 1999, p. U.S. Geological Survey Reports on Alaska Released in 1991

Kilburn, J.E., Box, S.E., Goldfarb, R.J., Gray, J.E., and Jones, J.L., 1991, Mineral-resource assessment of the Goodnews 1° x 3° quadrangle and parts of the Hagemeister Island and Nushagak Bay quadrangles, southwestern Alaska [abs.]: Cl062, p. 46. Kisslinger, Carl, Hill, Julie, and Kindel, Bruce, 1991, Central Aleutians Islands seismic network, in OF91-352, p. 2427. Kisslinger, Carl, Kubicheck, Sharon, Kindel, Bruce, and Hill, Julie, 1990, Central Aleutians Islands seismic network, in OF90-680, p. 15-17. Kvenvolden, K.A., Rapp, J.B., and Hostettler, F.D., 1991, Tracking hydrocarbons from Prince William Sound, Alaska-About one year after the EXXON Valdez oil spill, in OF91-631, p. 69-98. Lahr, J.C., Stephens, C.D., Page, R.A., and Fogleman, K.A., 1990, Alaska seismic studies, in OF90-680, p. 18-23. ---1991, Alaska seismic studies, in OF91-352, p. 28-34. Lipscomb, S.W., 1991, Streamflow and sediment transport characteristics of the lower Campbell Creek basin, Anchorage, Alaska, 1986-88: U.S. Geological Survey Water-Resources Investigations Report 91-4074, 38 p. Lisowski, M., Savage, J.C., Prescott, W.H., King, N.E., and Svarc, J.L., 1991, Crustal strain, in OF91-352, p. 295304. [l:las section on Alaska.] Lynch, J .S ., 1991, Mount Redoubt: Tracing volcanic ash plumes from space [abs.]: C1065, p. 30. Madden, D.J., 1991, Geochemical maps showing distribution of anomalously abundant elements in stream-sediment and glacial-moraine samples from the Anchorage 1 ° x 3 ° quadrangle, southern Alaska: U.S. Geological Survey Miscellaneous Investigations Series Map 1-1976, scale 1:250,000. ---1991, Geochemical maps showing distribution of anomalously abundant elements in the nonmagnetic, heavy-mineral-concentrate fraction of stream sediment from the Anchorage 1 ° x 3 ° quadrangle, southern Alaska: U.S. Geological Survey Miscellaneous Investigations Series Map 1-1977, scale 1:250,000. Madden-McGuire, D.J., and Winkler, G.R., 1991, Areas of mineral-resource favorability (with emphasis on gold and . chromite) in the Anchorage 1 ° x 3 ° quadrangle, southern Alaska [abs.]: Cl062, p. 50-51. Mann, D.M., and Fisher, M.A., 1991, Multichannel seismic-reflection data collected in 1980 in Norton Sound, Alaska: U.S. Geological Survey Open-File Report 91-254, 5 p., 1 pl., scale 1:250,000. Marincovich, Louie, Jr., and Moriya, Shigehiro, 1992, Early middle Miocene mollusks and benthic foraminifers from Kodiak Island, Alaska: B1999, p. 163-169. Mayfield, C.F., Curtis, S.M., Ellersieck, Inyo, and Tailleur, I.L., 1990, Reconnaissance geologic map of the De Long Mountains A-3 and B-3 quadrangles and parts of the A-4 and B-4 quadrangles, Alaska: U.S. Geological Survey Miscellaneous Investigations Series Map 1-1929, 2 sheets, color, scale 1:63,360. [This map supersedes USGS Open-File Report 83-183.] McDanal, S.K., Arbogast, B.F., and Cathrall, J.B., 1991, Analytical results and sample locality map of stream-sediment, heavy-mineral-concentrate, pebble, and rock samples from Geologic Studies in Alaska by the U.S. Geological Survey, 1991 the Craig Study Area; Craig, Dixon Entrance, Ketchikan, and Prince Rupert quadrangles, Alaska: U.S. Geological Survey Open-File Report 91-36-A, 122p., 2 sheets, scale 1:250,000 (paper copy), 91-36-B (DS/HB IBM compatible diskette version). McGimsey, R.G., Richter, D.H., DuBois, G.D., and Miller, T.P., 1992, A postulated new source for the White River Ash, Alaska: B1999, p. 212-218. McLean, Hugh, and Stanley, R.G ., 1992, Reconnaissance sandstone petrology and provenance of the Cantwell Formation, central Alaska: B1999, p. 170-179. Miller, T.P., 1991, Redoubt Volcano, Alaska, in United States Geological Survey Yearbook Fiscal Year 1990, p. 12-15. Miller, T.P., and Davies, J.N., 1991", The 1989-90 eruption of Redoubt Volcano: Chronology, character and effects [abs.]: C1065, p. 33. Mullen, M.W., and Grantz, Arthur, 1991, Bathymetric map of southern North wind Ridge and vicinity, Arctic Ocean: U.S. Geological Survey Open-File Report 91-136, scale 1:250,000. Murray, T.L., Bauer, C.l., and Paskievitch, J.F., 1991, Using a personal computer to obtain predicted plume trajectories during the 1989-1990 eruption of Redoubt Volcano, Alaska [abs.]: C1065, p. 34. Nelson, R.E., and Carter, L.D., 1992, Preliminary interpretation of vegetation and paleoclimate in northern Alaska during the late Pliocene Colvillian marine transgression: B1999, p. 219-222. Nelson, S.W., and Blome, C.D., 1991, Preliminary geochemistry of volcanic rocks from the McHugh Complex and Kachemak terrane, southern Alaska: U.S. Geological Survey Open-File Report 91-134, 14 p. Nokleberg, W.J., Lange, I.M., Roback, R.C., Yeend, Warren, and Silva, S.R., 1991, Map showing locations of metalliferous lode and placer mineral occurrences, mineral deposits, prospects, and mines, Mount Hayes quadrangle, eastern Alaska Range, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-1996-C, 42 p., scale 1:250,000. Nokleberg, W.J., Patton, W.W., Jr., and Hearn, P.P., 1991, The Soviet Far East and Alaska, in United States Geological Survey Yearbook 1990, p. 72-73. Palmer, I.F., Jr., 1990, Mapping requirements for planning the Outer Continental Shelf mining program Norton Sound, Alaska, lease sale: Cl052, p. 97-105. Phillips, R.L., Grantz, Arthur, Mullen, M.W., and White, J.M., 1991, Preliminary lithostratigraphy of piston cores from the Beaufort Sea continental slope off northeastern Alaska: U.S. Geological Survey Open-File Report 91-34, 2 sheets. Quinterno, P.J., 1991, Benthic foraminifera from Prince William Sound, Alaska-About one year after the EXXON Valdez oil spill, in OF91-631, p. 31-68. Riehle, J.R., Church, S.E., and Magoon, L.B., 1991, Resource assessment of the Mount Katmai 1° x 3° quadrangle and adjacent parts of the Naknek and Afognak quadrangles, Alaska Peninsula [abs.]: Cl062, p. 65-66. Roberts, S.B., 1991, Subsurface cross section sbowing coal beds in the Sagavanirktok Formation, vicinity ofPrudhoe Bay, east-central North Slope, Alaska: U.S. Geological

Survey Coal Investigations Map C-139-A. Roberts, S.B., Stricker, G.D., and Affolter, R.H., 1991, Stratigraphy and chemical analysis of coal beds in the Upper Cretaceous and Tertiary Sagavanirktok Formation, eastcentral North Slope, Alaska: U.S. Geological Survey Coal Investigations Map C-139-B. ---1992, Reevaluation of coal resources in the Late Cretaceous-Tertiary Sagavanirktok Formation, North Slope, Alaska: B1999, p. 196-203. Roeske, S.M., Pavlis, T.L., Snee, L.W., and Sisson, V.B., 1992, 40 Ar/39 Ar isotopic ages from the combined Wrangellia-Alexander terrane along the Border Ranges fault system in the eastern Chugach Mountains and Glacier Bay, Alaska: B1999, p. 180-195. Rowan, E.L., Bailey, E.A., and Goldfarb, R.J ., 1990, Geochemical orientation study for identification of metallic mineral resources in the Sitka quadrangle, southeastern Alaska: B1950, Chap. B, p. B1-B12. Sampson, J.A., Labson, V.F., and Long, C.L., 1992, Electrical resistivity cross sections in east-central Alaska: B 1999, p. Sass, J.H., Lachenbruch, A.H., and Williams, C.F., 1991, Heat flow and tectonic studies, in OF91-352, p. 494-498. [Section on North Slope and on Colville Basin, Alaska.] Schlatter, T.W., and Benjamin, S.G., 1991, A mesoscale data assimilation system adapted for trajectory calculations over Alaska [abs.]: C1065, p. 38-39. Schneider, D.J., and Rose, W.l., 1991, Utility of AVHRR sensor for remote sensing of Alaskan eruption clouds [abs.]: Cl065, p. 38. Schneider, J.L., 1990, 1990 annual report on Alaska's mineral resources: U.S. Geological Survey Circular 1056, 67 p. ---1991, 1991 Annual report on Alaska's mineral resources: U.S. Geological Survey Circular 1072, 69 p. Scott, W.E., and McGimsey, R.G., 1991, Mass, distribution, grain size, and origin of 1989-1990 tephra-fall deposits of Redoubt Volcano, Alaska [abs.]: C1065, p. 39. Seitz, H.R., 1991, Hydrologic conditions at Anaktuvuk Pass, Alaska, 1989: U.S. Geological Survey Open-File Report 90-591, 17 p. Severson, R.C., and Gough, L.P., 1990, Geochemical studies of plants and soils in the Beluga coal field, Alaska: U.S. Geological Survey Circular 1033, p. 159-160. Shasby, M.B., 1991, Alaska disaster-Natural and manmade, in United States Geological Survey Yearbook 1990, p. Snyder, E.R., 1990, Activities of the Alaska district, Water Resources Division, U.S. Geological Survey, 1990: U.S. Geological Survey Open-File Report 90-157, 21 p. ---1991, Location maps and list of U.S. Geological Survey reports on water resources in Alaska, 1950 to 1990: U.S. Geological Survey Open-File Report 91-60, 44 p. Stanley, R.G., Flores, R.M., and Wiley, T.J., 1992, Fluvial facies architecture in the Tertiary Usibelli Group of Suntrana, central Alaska: B 1999, p. 204-211. Stover, C.W., and Brewer, L.R., 1991, Earthquake descriptions: Alaska, in Stover, C.W., and Brewer, L.R., United States earthquakes, 1985: U.S. Geological Survey Bulletin 1954, p. 9-19. [See also: Summary of United States earthquakes: Alaska, p. 76-99.] Tanaka, H.L., 1991, Development of a prediction scheme for the volcanic ash fall from Redoubt Volcano [abs.]: Cl065, p. 44-45. Theis, C.V., 1991, Short papers on water supplies and engineering geology, Alaska Highway, 1943-1944: U.S. Geological Survey Open-File Report 91-80, 61 p. Trabant, D.C., Krimmel, R.M., and Post, Austin, 1991, A preliminary forecast of the advance of Hubbard Glacier and its influence on Russell Fiord, Alaska: U.S. Geological Survey Water-Resources Investigations Report 90-4172, 34 p. Tripp, R.B., and Madden, D.J., 1991, Mineralogical maps showing distribution of ore-related minerals in the nonmagnetic, heavy-mineral-concentrate fraction of stream sediment from the Anchorage 1° x 3 ° quadrangle, southern Alaska: U.S. Geological Survey Miscellaneous Investigations Series Map I-1975, 1 sheet, scale 1:250,000. Valin, Z.C., Bader, J.W., Barnes, D.P., Fisher, M.A., and Stanley, R.G., 1991, Simple Bouguer gravity anomaly maps of the Nenana basin area, Alaska: U.S. Geological Survey Open-File Report 91-33, 5 sheets. [Quads. included: Tanana, Kantishna River, Mt. McKinley, Livengood, Fairbanks, and Healy.] White, E.R., 1992, Reports about Alaska in non-USGS publications released in 1990 that include USGS authors: B1999, p. 236-242. ---1992, U.S. Geological Survey reports on Alaska released in 1990: B1999, p. 231-235. Wilson, F.H., Detterman, R.L., and Harris, E.E., 1991, Generalized geologic map of the Port Moller, Stepovak Bay, and Simeonof Island quadrangles, Alaska Peninsula, Alaska: U.S. Geological Survey Miscellaneous Field Studies Map MF-2155-A, seal~ 1:250,000. Y eend, Warren, 1991, Gold placers of the Circle district, Alaska-Past, present, and future: U.S. Geological Survey Bulletin 1943, 42 p., 1 pl., scale 1:63,360. ---1992, Gold placers, gold source, and high terrace gravels in the Fortymile River area, Alaska: B 1999, p. Yehle, L.A., Schmoll, H.R., and Dobrovolny, Ernest, 1991, Geologic map of the Anchorage B-8 SW quadrangle, Alaska: U.S. Geological Survey Open-File Report 91143, 30 p., 2 sheets, scale 1:25,000. U.S. Geological Survey Reports on Alaska Released in 1991

Reports About Alaska in Non-USGS Publications Released in 1991 That Include USGS Authors Compiled by Ellen R. White [Some reports dated 1989 and 1990 did not become available until 1991; they are included in this listing. USGS authors are marked with asterisks ABBREVIATIONS Eos GSA2 GSA5 Eos (American Geophysical Union, Transactions), v. 72, no. 44, suppl. Geological Society of America Abstracts with Programs, v. 23, no. 2. Geological Society of America Abstracts with Programs, v. 23, no. 5. Wood Wood, C.A., and Kienle, Jiirgen, eds., 1990, Volcanoes of North America: New York, Cambridge University Press, 354 p. Workshop Mesoscale Modeling, Circumpolar Climate Change, Arctic Science Conference, 42nd, Arctic Workshop, 21st, University of Alaska, Fairbanks, Alaska, 1991, Abstracts: University of Alaska Museum, Alaska Quaternary Center Occasional Paper 4, 97p. *Aleinikoff, J.N., *Moore, T.E., *Karl, S.M., and Dillon, J.T., 1991, Pb isotopic ratios in Late Proterozoic and Devonian granitic rocks of the Brooks Range, Alaska [abs.]: Eos, p. *Barnes, D.F., 1991, Small or undectable [undetectable] gravity changes accompany vertical crustal movements in northern southeast Alaska and adjacent Canada [abs.]: Eos, p. 111. *Barnes, D.F., *Nokleberg, W.J., and *Brocher, T.M., 1991, Gravitational and seismic evidence for Tertiary structural basins in the Alaska Range and Tanana lowland [abs.]: GSA2, p. 5. Beaudoin, B.C., *Fuis, G.S., *Mooney, W.D., *Nokleberg, W.J., *Lutter, W.J., and Christensen, N.I., 1991, Thin, low-velocity crust beneath the Yukon-Tanana terrane, east-central Alaska [abs.]: GSA2, p. 5. *Bird, K.J., 1991, North Slope of Alaska, in Gluskoter, H.J., Rice, D.D., and Taylor, R.B., eds., Economic geology, U.S.: Boulder, Colo., Geological Society of America, Geology of North America, v. P-2, Chap. 28, p. 447-462, pl. 6A. Boyd, T.M., *Engdahl, E.R., and *Spence, W., 1991, Aleutian earthquake catalog: 1957 through 1989 [abs.]: Eos, p. *Brew, D.A., 1990, Behm Canal and Rudyerd Bay, southeastern Alaska, in Wood, p. 95-96. ---1990, Duncan Canal, southeastern Alaska, in Wood, p. Geologic Studies in Alaska by the U.S. Geological Survey, 1991 ---1990, Origin and distribution of granitic rocks in the Coast plutonic-metamorphic complex, northern CanadianAlaskan Cordillera, southeastern Alaska, U.S.A. [abs.], in Chappell, B.W., ed., Hutton Symposium on Granites and Related Rocks, 2nd, Canberra [Australia], 1991, Abstracts: Canberra, Australia, Bureau of Mineral Resources, Geology and Geophysics, p. 14. ---1990, Tlevak Strait and Suemez Island, southeastern Alaska, in Wood, p. 95. ---1991, Geology, tectonics, and metallogeny of southeastern Alaska and adjacent parts of the Pacific Ocean rim [abs.]: GSA5, p. 218. *Brew, D.A., *Drew, L.J., *Schmidt, J.M., *Root, D.H., and *Huber, D.F., 1991, Undiscovered locatable mineral resources of the Tongass National Forest and adjacent areas, southeastern Alaska [abs.]: Alaska Miners Association, Juneau Branch, Conference, Juneau, Alaska, 1991, Abstracts of Professional Papers, p. 45-46. *Brew, D.A., *Hammarstrom, J.M., Himmelberg, G.R., *Wooden, J.L., *Loney, R.A., and *Karl, S.M., 1991, Crawfish Inlet pluton, Baranof Island, southeastern Alaska-a north-tilted Eocene body or an untilted enigma? [abs.]: GSA2, p. 8. *Brew, D.A., *Karl, S.M., *Barnes, D.F., *Jachens, R.C., *Ford, A.B., and Horner, Robert, 1991, A northern Cordilleran ocean-continent transect: Sitka Sound, Alaska, to Atlin Lake, British Columbia: Canadian Journal of Earth Sciences, v. 28, no. 6, p. 840-853.

*Brewer, Max, 1991, Research projects on the National Petroleum Reserve [abs.], in Geiselman, Joy, and Mitchell, K.L., eds., Federal Arctic Research Information Workshop, Anchorage, Alaska, 1991, Proceedings: U.S. Department of the Interior, Minerals Management Service, OCS Study MMS 91-0053, p. 73-74. *Brocher, T.M., *Fisher, M.A., *Geist, E.L., *Moses, M.J., and *Hart, P.E., 1990, Images of subducting oceanic lithosphere and a mid-crustal decollement beneath southern Alaska [abs.]: Terra Abstracts, v. 2, p. 192. *Brocher, T.M., *Fisher, M.A., *Luzitano, Robert, *Fuis, G.S., *Labson, V.F., *Stanley, W.D., and Christensen, N.I., 1991, Crustal structure and evolution of the Alaska Range, Alaska [abs.]: GSA2, p. 8. *Brocher, T.M., *Moses, M.J., *Fisher, M.A., *Stephens, C.D., and *Geist, E.L., 1991, Images of the plate boundary beneath southern Alaska, in Meissner, Rolf, Brown, Larry, Diirbaum, H.-I., Franke, Wolfgang, Fuchs, Karl, and Seifert, Friedrich, eds., Continental lithosphere: Deep seismic reflections: Washington, D.C., American Geophysical Union, Geodynamics Series v. 22, p. 241-246. *Brocher, T.M., *Nokleberg, W.J., Christensen, N.I., *Lutter, W.J., *Geist, E.L., and *Fisher, M.A., 1991, Seismic reflection/refraction mapping of faulting and regional dips in the eastern Alaska Range: Journal of Geophysical Research, v. 96, no. B6, p. 10,233-10,249. *Bruns, T.R., *Fisher, M.A., *Geist, E.L., and *Brocher, T.M., 1990, Deep crustal structure across the Yakutat terranePrince William terrane subduction suture, northern Gulf of Alaska, from multichannel seismic reflection data [abs.]: Terra Abstracts, v. 2, p. 192-193. Burns, L.E., Pessel, G.H., Little, T.A., Pavlis, T.L., Newberry, R.J., *Winkler, G.R., and Decker, John, 1991, Geology of the northern Chugach Mountains, southcentral Alaska: Alaska Division of Geological and Geophysical Surveys Professional Report 94, 63 p., 2 sheets, scale 1:63,360. *Cacchione, D.A., and *Drake, D.E., 1991, Bottom and nearbottom sediment dynamics in Norton Sound, Alaska in U.S. Department of Commerce and U.S. Department of the Interior, OCSEAP, Final reports of principal investigators v. 74: U.S. Department of the Interior, Minerals Management Service, OCS Study MMS 91-0085, p. 77143. *Cady, J.W., 1991, Angayucham and Tozitna geophysical domains-Geophysical and geochemical ties between parts of the Angayucham and Tozitna terranes, northern Alaska [abs.]: Eos, p. 296. *Carlson, P.R., 1990, GLORIA imagery provides clues to Quaternary sedimentary history in the Gulf of Alaska [abs.], in American Association for the Advancement of Science, Arctic Division, Circumpolar Perspectives, Arctic Science Conference, Anchorage, Alaska, 1990, Proceedings, p. 46. *Carlson, P.R., *Bruns, T.R., and *Fisher, M.A., 1990, Development of slope valleys in the glacimarine environment of a complex subduction zone, northern Gulf of Alaska, in Dowdeswell, J .A., and Scourse, J.D., eds., Glacimarine environments: Processes and sediments: Geological Society of London Special Publication 53, p. 139-153. *Carlson, P.R., *Stevenson, A.J., *Mann, D.M., *Bruns, T.R., and Dobson, M., 1991, From glaciers to deep-sea fans: Quaternary sedimentation in Gulf of Alaska [abs.]: GSA5, p. 385. *Carter, L.D., and *Whelan, J.F., 1991, Isotopic evidence for restricted Arctic sea ice during a late Pleistocene warm period: Implications for sea ice during future climatic warming at high latitudes [abs.]: GSA5, p. 237. *Chouet, B.A., *Page, R.A., Davies, J.N., and Power, J.A., 1991, Forecasting eruptions at Redoubt Volcano, Alaska [abs.]: GSA2, p. 13. ---1991, Forecasting eruptions at Redoubt Volcano, Alaska [abs.]: Seismological Research Letters, v. 62, no. 1, p. 25. *Clifton, H.E., 1990, Variation in sequence development in transgressive-regressive successions on the western margin of North America [abs.] in International Sedimentological Congress, 13th, Nottingham, England, 1990, Abstracts of Papers, p. 96. [Narrow Cape Formation on southern Kodiak Island is one of the four examples.] Crampin, Stuart, [Comment on] and, *Brocher, T.M., and Christensen, N.I., [Reply on] "Seismic anisotropy due to preferred mineral orientation observed in shallow crustal rocks in southern Alaska": Geology, v. 19, no. 8, 1991, p. [Original article appeared in Geology, v. 18, 1990, p. 737-740.] *Deming, D., *Sass, J.H., and *Lachenbruch, A.H., 1991, Heat flow and subsurface temperature as evidence for basinscale groundwater flow, N. Slope of Alaska [abs.]: Eos, p. 504. *Dorava, J.M., 1991, Generalized stream channel evolution resulting from the 1989-90 eruptions of Redoubt Volcano, Alaska [abs.]: Eos, p. 214. *Dumoulin, J.A., and *Harris, A.G., 1991, Lower and Middle Paleozoic metacarbonate rocks of the Snowden Mountain area, central Brooks Range, northern Alaska [abs.]: Eos, p. 300. *Dusel-Bacon, Cynthia, and Hansen, V .L., 1991, High-pressure, medium-temperature early Mesozoic metamorphism and deformation within the Yukon-Tanana composite terrane, eastern Alaska [abs.]: GSA2, p. 20. Elias, S.A., Short, S.K., *Waythomas, C.F., and Ten Brink, N. W ., 1991, Late Quaternary paleoenvironments of the north Alaska Range [abs.], in Workshop, p. 62. *Foster, H.L., 1990, Prindle, eastern Alaska, in Wood, p. 108109. *Frost, G.M., and *Stanley, R.G., 1991, Preliminary geologic and Bouguer gravity map of the Nenana basin area, central Alaska [abs.]: GSA2, p. 26. *Frost, T.P., and *Box, S.E., 1991, Depth controls on magmatism-related gold and mercury mineralization, Bethel quadrangle, southwestern Alaska [abs.]: GSA2, p. *Fuis, G.S., *Ambos, E.L., *Mooney, W.D., Christensen, N.I., and *Geist, Eric, 1991, Crustal structure of accreted terranes in southern Alaska, Chugach Mountains and Copper River basin, from seismic refraction results: Journal of Geophysical Research, v. 96, no. B3, p. 4187-4227. *Fuis, G.S., and Clowes, R.M., 1991, A comparison of deep continental-margin structure along transects in southern Alaska, southern Vancouver Island, and central California [abs.]: GSA2, p. 42. *Fuis, G.S., *Lutter, [*Moore, T.E.], W.J., Levander, A.R., and Reports About Alaska in Non-USGS Publications Released in 1991 That Include USGS Authors

Wissinger, E.S., 1991, A preliminary seismic-velocity model for the uppermost crust of the Brooks Range, arctic Alaska [abs.]: Eos, p. 296. *Fuis, G.S., and *Plafker, George, 1991, Evolution of deep structure along the Trans-Alaska Crustal Transect, Chugach Mountains and Copper River basin, southern Alaska: Journal of Geophysical Research, v. 96, no. B3, p. 4229-4253. *Galloway, J.P., 1991, Development of an Alaskan radiocarbon data base as a subset of the international radiocarbon data base (IRDB) [abs.]: GSA2, p. 28. Gehrels, G.E., McClelland, W.C., Samson, S.D., Patchett, P.J., and *Brew, D.A., 1991, U-Pb geochronology of Late Cretaceous and early Tertiary plutons in the northern Coast Mountains batholith: Canadian Journal of Earth Sciences, v. 28, no. 6, p. 899-911. *Geist, E.L., *Scholl, D.W., and *Vallier, T.L., 1991, Collision of the Aleutian island arc with Kamchatka [abs.]: Eos, p. *Goldfarb, R.J., 1989, Genesis of lode gold deposits of the southern Alaskan Cordillera [abs.]: Fluid Inclusion Research, v. 22, p. 128. [Abstract of Ph.D. thesis, Boulder, Colo., Univ. of Colorado.] *Goldfarb, R.J., *Leach, D.L., *Rose, S.C., and *Landis, G.P., 1989, Fluid inclusion geochemistry of gold-bearing quartz veins of the Juneau gold belt, southeastern Alaska: Implications for ore genesis [abs.]: Fluid Inclusion Research, v. 22, p. 128. *Goldfarb, R.J., Newberry, R.J., *Pickthorn, W.J., and *Gent, C.A., 1991, Oxygen, hydrogen, and sulfur isotope studies in the Juneau gold belt, southeastern Alaska: Constraints on the origin of hydrothermal fluids: Economic Geology, v. 86, no. 1, p. 66-80. *Goldfarb, R.J., *Snee, L.W., Miller, L.D., and Newberry, R.J., 1991, New evidence for widespread fluid flow and gold deposition in southeast Alaska at 55 Ma [abs.]: GSA2, p. ---1991, Rapid dewatering of the crust deduced from ages of mesothermal gold deposits: Nature, v. 354, no. 6351, Nov. 28, p. 296-298. ---1991, Timing of gold deposition within the Juneau gold belt [abs.]: Alaska Miners Association, Juneau Branch, Conference, Juneau, Alaska, 1991, Abstracts of Professional Papers, p. 12-13. *Grantz, Arthur, and *Moore, T.E., 1991, Crustal model of northern Alaska from the Brooks Range to the Arctic Ocean basin [abs.]: GSA2, p. 30. ---1991, A prediction of the crustal structure of the Brooks Range, Alaska, based on regional geologic and gravity data [abs.]: Eos, p. 299. *Grantz, Arthur, *Moore, T.E., and Roeske, Sarah, 1991, Gulf of Alaska to Arctic Ocean: Boulder, Colo., Geological Society of America, Centennial Continent/Ocean Transect #15, 72 p., 3 pis., scale 1:500,000. *Gray, J.E., *Goldfarb, R.J., *Detra, D.E., and *Slaughter, K.E., 1991, Geochemistry and exploration criteria for epithermal cinnabar and stibnite vein deposits in the Kuskokwim River region, southwestern Alaska: Journal of Geochemical Exploration, v. 41, no. 3, p. 363-386. *Griscom, Andrew, and *Sauer, P.E., 1991, Tectonics of the Geologic Studies in Alaska by the U.S. Geological Survey, 1991 Aleutian subduction fault in the northern Gulf of Alaska from magnetic data [abs.]: GSA2, p. 31. *Hamilton, T.D., 1991, The last interglaciation and the Old Crow tephra: Data from 10 Alaskan sites [abs.], in Workshop, p. 68-69. *Hamilton, T.D., Westgate, J.A., and Beget, J.E., 1991, The Old Crow tephra: A stratigraphic marker for the last interglaciation in Alaska and the Yukon Territory? [abs.]: GSA5, p. 62. *Heinrichs, T.A., *Mayo, L.R., Echelmeyer, K.E., and Harrison, W.D., 1991, Black Rapids Glacier, Alaska-Unexpected behavior during the quiescent phase of a surge-type glacier [abs.]: Eos, p. 158. *Hildreth, Wes, 1990, Griggs, Alaska Peninsula, in Wood, p. ---1990, Katmai, Alaska Peninsula, in Wood, p. 71-72. ---1990, The Katmai eruption of 1912: a comparison with the Minoan eruption of Santorini, in Hardy, D.A., Keller, J., Galanopoulos, V.P., Flemming, N.C., and Druitt, T.H., eds., Thera and the Aegean world III: London, England, Thera Foundation, Earth, v. 2, p. 455-462. [Proceedings of the ·Third International Congress, Santorini, Greece, 1989.] ---1990, Mageik, Alaska Peninsula, in Wood, p. 67-68. ---1990, Novarupta, Falling Mountain, and Cerberus, Alaska Peninsula, in Wood, p. 70-71. ---1990, Trident, Alaska Peninsula, in Wood, p. 68-69. ---1991, The timing of caldera collapse at Mount Katmai in response to magma withdrawal toward Novarupta: Geophysical Research Letters, v. 18, no. 8, p. 1541-1544. Himmelberg, G.R., *Brew, D.A., and *Ford, A.B., 1991, Development of inverted metamorphic isograds in the western metamorphic belt, Juneau, Alaska: Journal of Metamorphic Geology, v. 9, no. 2, p. 165-180. *Hinkley, T.K., *Fitzpatrick, J.J., *Landis, G.P., *Rye, R.O., and Holdsworth, G., 1991, High-resolution paleoclimate reconstruction from Alaskan ice-core records [abs.]: GSA5, p. 352. *Horton, R.J., *Karl, S.M., *Griscom, Andrew, *Taylor, C.D., and *Bond, K.R., 1991, Annette Islands Reserve Mineral Assesment Project, southeast Alaska, in Manydeeds, S.A., and Smith, B.D., eds., Mineral frontiers on Indian lands, Annual Northwest Mining Convention, 97th, Spokane, Wash., 1991: U.S. Bureau of Indian Affairs, Division of Energy and Mineral Resources, p. 22-31, Introduction on p. 19-21. *Howell, D.G., Fehri, N., and *Bird, K.J., 1991, Thin- versus thick-skinned thrusting in the central Brooks Range orogen, Alaska-Constraints based on surface geology [abs.]: Eos, p. 295. *Johnson, M.J., *Howell, D.G., and *Bird, K.J., 1991, Use of vitrinite reflectance to constrain regional structural patterns: An example from the North Slope of Alaska [abs.]: Eos, p. 549-550. *Jolly, A.D., *Page, R.A., *Stephens, C.D., *Lahr, J.C., Power, J .A., and Cruse, G .R., 1991, Seismicity in the vicinity of Mt. Spurr volcano, south-central Alaska, based on revised velocity model [abs.]: Eos, p. 567. Kamata, Hiroki, Johnston, D.A., and *Waitt, R.B., 1991, Stratigraphy, chronology, and character of the 1976 pyro-

clastic eruption of Augustine Volcano, Alaska: Bulletin of Volcanology, v. 53, no. 6, p. 407-419. *Karl, S.M., 1991, Regional geology of the Chichagof mining district, southeastern Alaska [abs.]: Alaska Miners Association, Juneau Branch, Conference, Juneau, Alaska, 1991, Abstracts of Professional Papers, p. 12-13. *Keith, T.E.C., 1991, Argillic alteration in the Novarupta vent region, Katmai National Park, Alaska: Geophysical Research Letters, v. 18, no. 8, p. 1549-1552. ---1991, Fossil and active fumaroles in the 1912 eruptive deposits, Valley of Ten Thousand Smokes, Alaska: Journal of Volcanology, v. 45, no. 3-4, p. 227-254. *Keith, T.E.C., and *Ingebritsen, S.E., 1991, Advective flux of solutes and heat from the Valley of Ten Thousand Smokes, Katmai National Park, Alaska [abs.]: Eos, p. *Kempema, E.W., *Reimnitz, Erk, and *Hunter, R.E., 1990, Flume studies and field observations of the interactions of frazil ice and anchor ice with sediment, in U.S. Department of Commerce, and U.S. Department of the Interior, Outer Continental Shelf Environmental Assessment Program, Final Reports of Principal Investigators, v. 72, p. ["This is a report of flume experiments Observations of frazil and anchor ice from the Alaska!l Beaufort Sea also are presented." This report has also been published as U.S. Geological Survey OpenFile Report 86-515.] Kissling, Eduard, and *Lahr, J.C., 1991, Tomographic image of the Pacific slab under southern Alaska: Eclogae Geologicae Helvetiae, v. 84, no. 2, p. 297-315. *Kleinman, J.W., and *Iwatsubo, E.Y., 1991, A geodetic network in the Novarupta area, Katmai National Park, Alaska: Geophysical Research Letters, v. 18, no. 8, p. 1517-1519. *Kvenvolden, K.A., *Collett, T.S., and *Williams, R.S., 1991, Methane in permafrost ice near Fairbanks, Alaska [abs.]: Eos (American Geophysical Union, Transactions), v. 72, no. 17, suppl., p. 67.

*Kvenvolden, K.A., and *Lorenson, T.D., 1991, Varying amounts of methane in shallow permafrost cores from Alaska [abs.]: Eos, p. 162. [Three cores taken on the University of Alaska, Fairbanks, campus.] *Kvenvolden, Keith, *Lorenson, T.D., and *Collett, T.S., 1991, Arctic shelf gas hydrate as a possible source of methane [abs.]: GSA5, p. 238. *Labson, V.F., Rodriguez, B.D., Sampson, J.A., and Bisdorf, R.J ., 1991, Deep crustal geoelectric structure in the vicinity of the Tintina fault zone, north-central Alaska [abs.]: GSA2, p. 43. *Labson, V.F., *Sampson, J.A., and *Heran, ·w.D., 1991, Magnetotelluric electrical resistivity profile across the Brooks Range, Alaska [abs.]: Eos, p. 300. *Landis, G.P., and *Hofstra, A.H., 1991, Fluid inclusion gas chemistry as a potential minerals exploration tool: Case studies from Creede, CO, Jerritt Canyon, NV, Coeur d'Alene district, ID and MT, southern Alaska mesothermal veins, and mid-continent MVT's: Journal of Geochemical Exploration, v. 42, no. 1, p. 25-59. Lange, I.M., *Nokleberg, W.J., and Newkirk, S.R., 1991, Primary and secondary textures in multiply deformed and metamorphosed Devonian age massive sulfide deposits, Yukon-Tanana terrane, Alaska [abs.]: Geological Society of America Abstracts with Programs, v. 23, no. 4, p. 40. Levander, A.R., Wissinger, E.S., *Fuis, G.S., and *Lutter, W.J., 1991, The 1990 Brooks Range seismic experiment: Near vertical incidence reflection images [abs.]: Eos, p. Levander, A.R., Wissinger, E.S., Henrys, S.A., and *Fuis, G.S., 1991, Results from the 1988 and 1990 Brooks Range seismic experiments, arctic Alaska [abs.]: Eos (American Geophysical Union, Transactions), v. 72, no. 17, suppl., p. *Light, T.D., *Brew, D.A., and *Ashley, R.P., 1989, The Alaska-Juneau and Treadwell lode gold systems, southeastern Alaska: U.S. Geological Survey Bulletin 1857, p. D27-D36 [Comment]: Fluid Inclusion Research, v: 22, p. Lowell, R.P., and *Keith, T.E.C., 1991, Chemical and thermal constraints on models of thermal springs, Valley of Ten Thousand Smokes, Alaska: Geophysical Research Letters, v. 18, no. 8, p. 1553-1556. *Lutter, W.J., *Fuis, G.S., [*Moore, T.E.], Levander, A.R., and Wissinger, E., 1991, A simultaneous inversion of seismic travel-time data for the velocity and interface position for the 1990 Brooks Range experiment [abs.]: Eos, p. 300. *Major, J.J., 1991, Destructive geomorphic processes on volcanoes [abs.]: Eos, p. 228. [Redoubt Volcano is an example.] *Marincovich, Louie, Jr., and *Powell, C.L., II, [Comment on] and McNeil, D.H., and Miller, K.G., [Reply] "High-latitude application of 87Sr/86Sr: Correlation of Nuwok beds on North Slope, Alaska, to standard Oligocene chronostratigraphy": Geology, v. 19, no. 5, 1991, p. 537539. [Original article appeared in Geology, 1990, v. 18, no. 5, p. 415-418.] McNutt, S.R., *Miller, T.P., and Taber, J.J., 1991, Geological and seismological evidence of increased explosivity during the 1986 eruptions of Pavlov Volcano, Alaska: Bulletin of Volcanology, v. 53, no. 3, p. 86-98. Meen, J.K., *Snee, L.W., Ross, Kent, and Elthon, Don, 1991, Age and geologic relations of the Hall Cove Complex, Duke Island [abs.]: GSA5, p. 389. *Miller, M.L., *Bradshaw, J.Y., Kimbrough, D.L., *Stern, T.W., and Bundtzen, T.K., 1991, Isotopic evidence for Early Proterozoic age of the Idona Complex, west-central Alaska: Journal of Geology, v. 99, no. 2, p. 209-223. *Miller, T.P., 1990, Aniakchak, Alaska Peninsula, in Wood, p. ---1990, Black Peak, Alaska Peninsula, in Wood, p. 5859. ---1990, Dutton, Alaska Peninsula, in Wood, p. 51-52. ---1990, Emmons and Hague, Alaska Peninsula, in Wood, p. 52-53. ---1990, Fisher, eastern Aleutian Islands, in Wood, p. 4648. ---1990, Iliamna, Cook Inlet, Alaska, in Wood, p. 80-81. ---1990, Pavlof and Pavlof Sister, Alaska Peninsula, in Wood, p. 53-54. ---1990, Ugashik and Peulik, Alaska Peninsula, in Wood, p. 63-64. Reports About Alaska in Non-USGS Publications Released in 1991 That Include USGS Authors

*Miller, T.P., and Davies, J.N., 1991, Volcanic hazards and earthquake potential of the North Pacific [abs.]: GSA5, p. 218. [Aleutian-Alaskan volcanic arc-trench system.] Moll, S.H., *Light, T.D., and *Bie, S.W., 1991, Digital methods for lode gold exploration in central Alaska [abs.]: GSA5, p. 414-415. *Moll-Stalcup, E.J ., 1990, Kookooligit, Bering Sea, Alaska, in Wood, p. 104-105. ---1990, St. Michael, western Alaska, in Wood, p. 103104. ---1990, Yukon Delta, western Alaska, in Wood, p. 99102. [Ungulungwak Hill-Ingrichuak Hill, p. 100; Ingakslugwat Hills, p. 1 00~ 101; Nushkolik Mountain, p. 101; Ingrisarak Mountain, p. 102; Nelson Island, p. 102.] *Moll-Stalcup, E.J., and *Arth, J.G., 1991, Isotopic and chemical constraints on the petrogenesis of Blackburn Hills volcanic field, western Alaska: Geochimica et Cosmochimica Acta, v. 55, no. 12, p. 3753-3776. *Molnia, B.F., *Post, Austin, *Carlson, P.R., and *Trabant, D.C., 1991, Evolution and morphology of proglacial Vitus Lake, Bering Glacier, Alaska [abs.]: Eos, p. 158-159. *Molnia, B.F., *Trabant, D.C., *Post, Austin, and *FrankMolnia, D.G., 1990, The potential for an irreversible calving retreat of Bering Glacier, Alaska [abs.], in American Association for the Advancement of Science, Arctic Division, Circumpolar Perspectives, Arctic Science Conference, Anchorage, Alaska, 1990, Proceedings, p. 44. Moore, J.C., Diebold, J., *Brocher, T.M., *Fisher, M.A., *Geist, E.L., *Moses, M.J., Talwani, M., Ewing, J.l., Davies, J., Stone, D., Sample, J., and von Hoene, R., 1990, Comparison of EDGE and TACT images of subducting oceanic lithosphere and crustal reflectivity beneath southern Alaska [abs.]: Terra Abstracts, v. 2, p. 202. Moore, J.C., Diebold, John, *Fisher, M.A., Sample, J., *Brocher, T., Talwani, M., Ewing, John, von Huene, Roland, Rowe, C., Stone, D., *Stevens, Chris, and Sawyer, Dale, 1991, EDGE deep seismic reflection transect of the eastern Aleutian arc-trench layered lower crust reveals underplating and continental growth: Geology, v. 19, no. 5, p. 420-424. *Moore, T.E., *Nokleberg, W.J., Jones, D.L., *Till, A.B., and Wallace, W.K., 1991, Contrasting structural levels of the Brooks Range orogen along the Trans-Alaskan Crustal Transect (TACT) [abs.]: Eos, p. 295. *Moore, T.E., *Plafker, George, and *Weber, F.R., 1991, Evidence bearing on reconstruction of crystalline and sedimentary terranes in central Alaska [abs.]: GSA2, p. *Moore, T.E., Wallace, W.K., and Jones, D.L., 1991, TACT geologic studies in the Brooks Range: Preliminary results and implications for crustal structure [abs.]: GSA2, p. 80. Motyka, R.J., *Trabant, D.C., and Noll, Rick, 1990, Preliminary analysis of the Taku Glacier advance [abs.], in American Association for the Advancement of Science, Arctic Division, Circumpolar Perspectives, Arctic Science Conference, Anchorage, Alaska, 1990, Proceedings, p. 13. Murphy, J.M., and *Patton, W.W., Jr., 1991, Apatite fission track studies on granitic plutons of the Yukon-Koyukuk basin and Ruby geanticline, Alaska [abs.]: Eos, p. 505. Murphy, J.M., O'Sullivan, P.B., *Nelson, S., Miller, E.L., and Geologic Studies in Alaska by the U.S. Geological Survey, 1991 *Moore, T .E., 1991, Apatite fission track thermochronology of the southern Brooks Range, Alaska: Evidence of episodic Tertiary uplift, segmentation and high-angle fault reactivation [abs.]: Eos, p. 299. Newberry, R.J., and *Brew, D.A., 1991, Stratabound sulfide deposits of uncertain origins in metamorphic rocks vs. Black box data: 3 Alaskan examples [abs.]: Alaska Miners Association, Juneau Branch, Conference, Juneau, Alaska, 1991, Abstracts of Professional Papers, p. 10-11. [Groundhog Basin prospect, 20 km east of Wrangell; Liberty Bell deposit, 25 km northeast of Healy; Dream prospect, 20 km northwest of Kensington.] *Nokleberg, W.J., *Foster, H.L., *Lanphere, M.A., *Aleinikoff, J.N., and Pavlis, T.L., 1991, Structure and tectonics of the Yukon-Tanana, Wickersham, Seventymile, and Stikinia terranes along the Trans-Alaska-CrustalTransect (TACT), east-central Alaska [abs.]: GSA2, p. *Nokleberg, W.J., and *Fisher, M.A., eds., 1989, Alaskan geological and geophysical transect, Valdez to Coldfoot, Alaska, June 24-July 5, 1989, in Sedimentation and tectonics of western North America, v. 1: Washington, D.C., American Geophysical Union, 28th International Congress Field Trip Series T104, p. T104:1-T104:131. [For detailed indexing of T104 see USGS Bulletin 1946, p. 116-121.] Oldow, J.S., Ave Lallemant, H.G., Gottschalk, R.R., and *Snee, L.W., 1991, Timing and kinematics of Cretaceous contraction and extension in the southern Brooks Range, Alaska [abs.]: Eos, p. 295. ---1991, Timing and kinematics of Cretaceous contraction and extension in the ·southern Brooks Range, Alaska [abs.]: GSA2, p. 85. *Page, R.A., Biswas, N.N., *Lahr, J.C., and Pulpan, Hans, 1991, Seismicity of continental Alaska, in Slemmons, D.B., Engdahl, E.R., Zoback, M.D., and Blackwell, D.D., eds., Neotectonics of North America: Boulder, Colo., Geological Society of America, DNAG series, Decade Map volume to accompany the neotectonic maps, part of the continent-scale maps of North America, Chap. 4, p. *Page, R.A., *Stephens, C.D., *Fogleman, K.A., *Brocher, T.M., *Fisher, M.A., and *Puis, G.S., 1991, Seismicity and current tectonics of south-central Alaska [abs.]: GSA2, p. 86. Papike, J.J., *Keith, T.E.C., Spilde, M.N., Shearer, C.K., Galbreath, K.C., and Laul, J.C., 1991, Major and trace element mass flux in fumarolic deposits, Valley of Ten Thousand Smokes, Alaska, rhyolite-rich protolith: Geophysical Research Letters, v. 18, no. 8, p. 1545-1548. Papike, J.J., Spilde, M.N., and *Keith, T.E.C., 1991, Chemical mass flux in fumarolic deposits, Valley of Ten Thousand Smokes, Alaska: Influence of assumed protolith composition on enrichment/depletion systematics [abs.]: Eos, p. Patrick, B.E., Dinklage, W.S., and *Till, A.B., 1991, Metamorphism and progressive deformation in the Walker Lake region of the southern Brooks Range, Alaska [abs.]: GSA2, p. 87. *Patton, W.W., Jr., 1991, Deep crustal composition of the

Yukon~Koyukuk basin, Alaska [abs.]: GSA2, p. 88. Phillips, R.L., *Grantz, Arthur, *Mullen, M.W., *Chase, T., and *Young, J ., 1990, Preliminary stratigraphy and sediment processes on southeastern Northwind Ridge, Arctic Ocean [abs.], in Sediments 1990, International Sedimentological Congress, 13th, Nottingham, England, 1990, Abstracts of Posters, p. 176. *P1afker, George, 1990, Regional vertical tectonic displacement of shorelines in south-central Alaska during and between great earthquakes: Northwest Science, v. 64, no. 5, p. 250-258. *Plafker, George, *Lajoie, K.R., and *Rubin, Meyer, 1991, New data on the recurrence interval and seismic cycle for great 1964-type earthquakes in the Copper River delta, Alaska [abs.]: Seismological Research Letters, v. 62, no. 1, p. 38. *Reimnitz, Erk, and *Kempema, E.W., 1990, Field observations on slush ice generated during freezeup in Arctic . coastal waters, in U.S. Department of Commerce, and U.S. Department of the Interior, Outer Continental Shelf Environmental Assessment Program, Final Reports of Principal Investigators, v. 72, p. 331-355. *Reimnitz, Erk, *Kempema, E.W., and *Barnes, P.W., 1990, Anchor ice and bottom-freezing in high-latitude marine sedimentary environments: Observations from th~ Alaskan Beaufort.Sea, in U.S. Department of Commerce, and U.S. Department of the Interior, Outer Continental Shelf Environmental Assessment Program, Final Reports of Principal Investigators, v. 72, p. 257-281. [This report has also been published as U.S. Geological Survey OpenFile Report 86-298.] *Richter, D.H., 1990, Capital, Wrangell Mountains, Alaska, in Wood, p. 89-90. ---1990, Drum, Wrangell Mountains, Alaska, in Wood, p. ---1990, Eastern Wrangell, Wrangell Mountains, Alaska, in Wood, p. 91-92. ---1990, Sanford, Wrangell Mountains, Alaska, in Wood, p. 87. ---1990, Tanada, Wrangell Mountains, Alaska, in Wood, p. 90-91. *Riehle, J .R., 1990, Chiginagak, Alaska Peninsula, in Wood, p. 61-62. ---1990, Edgecombe, southeastern Alaska, in Wood, p. ' ---1990, Hayes, Cook Inlet, Alaska, in Wood, p. 84-85. ---1990, Kialagvik, southeast Alaska [should be "Alaska Peninsula"], in Wood, p. 62-63. ---1990, Yantami, Alaska Peninsula, in Wood, p. 60-61. Roeske, S.M., Murphy, J.M., and *Snee, L.W., 1991, Isotopic constraints on the metamorphic and cooling history of high Pnow T and moderate Plhigh T metamorphic rocks and E. Cretaceous granites in the SW Ruby terrane, central Alaska [abs.]: Eos, p. 440. Roeske, S.M., *Snee, L.W., and Pavlis, T.L., 1991, Strike-slip and accretion events along the southern Alaska plate margin in the Cretaceous and early Tertiary [abs.]: GSA5, p. Roeske, S.M., *Walter, Marianne, and *Aleinikoff, J.N., 1991, Cretaceous deformation of granitic rocks in the southwest Ruby terrane, central Alaska [abs.]: GSA2, p. 93. *Saltus, R.W., Stone, D.B., Kienle, J., and Goodliffe, A.M., 1991, New gravity data at Katmai National Park, Alaska, suggest a magma body analogue to that at the GeysersClear Lake region, California [abs.]: Eos, p. 429. *Savage, J.C., and *Lisowski, M., 1991, Strain accumulation along the Denali fault at the Nenana River and Delta River crossings, Alaska: Journal of Geophysical Research, v. 96, no. B9, p. 14,481-14,492. *Savage, J.C., and *Plafker, George, 1991, Tide gage measurements of uplift along the south coast of Alaska: Journal of Geophysical Research, v. 96, no. B3, p. 4325-4335. *Scholl, D.W., and *Stevenson, A.J., 1991, Exploring the idea that early Tertiary evolution of the Alaska orocline and the Aleutian-Bering Sea region is a manifestion of Kulaplate-driven Cordilleran tectonism and escape tectonics? [abs.]: GSA5, p. 435. ---1991, Tectonic evolution of the Pacific's Alaska-Bering Sea rim in terms of large-scale plate-boundary driven transgressive deformation [abs.], in Toronto 1991, Geological Association of Canada, Mineralogical Association of Canada and Society of Economic Geologists, Joint Annual Meeting, Toronto, Canada, 1991, Program with Abstracts, p. Al12. *Schoonmaker, J.W., Jr., *Jones, J.E., and *Molnia, B.F., 1989, Preliminary results of glacier studies from digital. radar data, in Agenda for the 90's, ASPRS/ACSM [American Society for Photogrammetry and Remote Sensing, and American Congress on Surveying and Mapping] Annual Convention, Baltimore, Md., 1989, Technical Papers, v. 3, p. 1-9. *Stanley, W.D., *Nokleberg, W.J., and *Labson, V.F., 1991, Flysch belts and collisional processes: Eastcentral Alaska and Alpine-Carpathian regions [abs.]: GSA2, p. 100. *Starratt, S.W., 1991, Late Quaternary paleoceanography of the Navarin basin region, Bering Sea: Evidence from diatom floras and radiolarian faunas [abs.]: GSA2, p. 100. *Stricker, G.D., 1991, Economic Alaskan coal deposits, in Gluskoter, H.J., Rice, D.O., and Taylor, R.B., eds., Economic geology, U.S.: Boulder, Colo., Geological Society of America, Geology of North America, v. P-2, Chap. 37, p. 591-602, pl. 7. Swanson, S.E., Beget, J.E., and *McGimsey, R.G., 1991, Compositional equivalence of tephra and lava groundmass glasses in the 1989-90 eruption of Mount Redoubt, Alaska: Implications for eruption monitoring [abs.]: GSA5, p. 396. Taber, J.J., Billington, S., and *Engdahl, E.R., 1991, Seismicity of the Aleutian arc, in Slemmons, D.B., Engdahl, E.R., Zoback, M.D., and Blackwell, D.D., eds., Neotectonics North America: Boulder, Colo., Geological Society of America, DNAG series, Decade Map volume to accompany the neotectonic maps, part of the continent-scale maps of North America, Chap. 3, p. 29-46. *Thompson, Ken, 1991, [USGS] Water Resources Divsion Arctic research [abs.], in Geiselman, Joy, and Mitchell, K.L., eds., Federal Arctic Research Information Workshop, Anchorage, Alaska, 1991, Proceedings: U.S. Department of the Interior, Minerals Management Service, OCS Study MMS 91-0053, p. 74-75. Reports About Alaska in Non-USGS Publications Released in 1991 That Include USGS Authors

*Till, A.B., and *Moore, T.E., 1991, Tectonic relations of the schist belt, southern Brooks Range, Alaska [abs.]: Eos, p. *Till, A.B., and Patrick, B.E., 1991, Ar-Ar evidence for a 110105 MA amphibolite-facies overprint on blueschist in the south-central Brooks Range, Alaska [abs.]: GSA5, p. 436. *Trabant, D.C., *Molnia, B.F., and *Post, Austin, 1991, Bering Glacier, Alaska-Bed configuration and potential for calving retreat [abs.]: Eos, p. 159. Underwood, M.B., Laughland, M., Shelton, K., Solomon, R., Kang, S.M., Orr, R., Brocculeri, T., Bergfeld, D., and *Pawlewicz, M., 1991, Correlations among paleotemperature indicators within orogenic belts: Examples from pelitic rocks of the Franciscan Complex (California) and the Kandik Basin, Alaska [abs.]: Eos, p. *Vallier, T .L., 1990, Koniuji, central Aleutian Islands, in Wood, p. 28-29. *Waitt, R.B., 1991, Repeated failures of summit domes of Augustine Volcano delivering tsunami-generating debris avalanches to Cook Inlet, Alaska [abs.]: Eos, p. 602. *Waitt, R.B., and Beget, J.E., 1991, Tsunami hazard from debris avalanches off Augustine Volcano, Alaska [abs.]: Eos, p. 227-228. Walker, H.J., and *Brewer, M.C., 1991, The Colville River delta: Hydrologic characteristics [abs.], in Workshop, p. Walker, H.J., and *Brewer, M.C., 1991, The Colville River delta: Morphology [abs.], in Workshop, p. 27. Walker, H.J., and *Brewer, M.C., 1991, When Colville River water meets the sea [abs.], in Workshop p. 29. *Ward, P.L., *Pitt, A.M., and *Endo, Eliot, 1991, Seismic evidence for magma in the vicinity of Mt. Katmai, Alaska: Geophysical Research Letters, v. 18, no. 8, p. 1537-1540. Warme, J.E., Gardner, M.H., Ethridge, F.G., Houston, W.S., and *Flores, R.M., 1991, Tertiary non-marine stratigraphic section along Shelikof Straits, Katmai National Park, Alaska [abs.]: GSA2, p. 107. *Waythomas, C.F. and Kaufman, D.S., [Comment on] and Bigelow, N., Beget, J.E., and Powers, W.R., [Reply on], and Beget, J.E., Bigelow, Nancy, and Powers, Roger, Geologic Studies in Alaska by the U.S. Geological Survey, 1991 1991, [Reply to Comment on] "Latest Pleistocene increase in wind intensity recorded in eolian sediments from central Alaska": Quaternary Research, v. 36, no. 3, p. [Original article appeared in Quaternary Research, v. 34, 1990, p. 160-168.] *White, Willis, 1991, Activities of the Branch of Alaska Geology [abs.], in Geiselman, Joy, and Mitchell, K.L., eds., Federal Arctic Research Information Workshop, Anchorage, Alaska, 1991, Proceedings: U.S. Department of the Interior, Minerals Management Service, Alaska OCS Region OCS Study MMS 91-0053, p. 71. *Wilson, F.H., 1990, Kupreanof (Stepovak Bay) Alaska Peninsula, in Wood, p. 55-56. ---1991, Geology of the Alaska Peninsula, southwestern Alaska, and the Alaska Peninsula terrane [abs.]: GSA5, p. *Wiltshire, D.A., and *Molnia, B.F., 1990, Arctic data interactive: A hypermedia system [abs.], in American Association for the Advancement of Science, Arctic Division, Circumpolar Perspectives, Arctic Science Conference, Anchorage, Alaska, 1990, Proceedings, p. 48. Wissinger, E.S., Levander, A.R., *Lutter, W.J., and *Puis, G .S ., 1991, The 1990 Brooks Range seismic experiment: Near vertical incidence reflection data processing [abs.]: Eos, p. 299. *Wong, F.L., 1990, St. George, Bering Sea, Alaska, in Wood, p. 97-98. ---1990, St. Paul, Bering Sea, Alaska, in Wood, p. 96-97. *Yount, M.E., 1990, Dana, Alaska Peninsula, in Wood, p. 55. --1990, Redoubt, Cook Inlet, Alaska, in Wood, p. 81-82. ---1990, Veniaminof, Alaska PeJ?.insula, in Wood, p. 5658. Zoback, M.D., and *Zoback, M.L., 1991, Tectonic stress field of North America and relative plate motions, in Slemmons, D.B., Engdahl, E.R., Zoback, M.D., and Blackwell, D.D., eds., Neotectonics of North America: Boulder, Colo., Geological Society of America, DNAG series, Decade Map volume to accompany the neotectonic maps, part of the continent-scale maps of North America, Chap. 19, p. 339-366.

SELECTED SERIES OF U.S. GEOLOGICAL SURVEY PUBLICATIONS Periodicals Earthquakes & Volcanoes (issued bimonthly). Preliminary Determination of Epicenters (issued monthly). Technical Books and Reports Professional Papers are mainly comprehensive scientific reports of wide and lasting interest and importance to professional scientists and engineers. Included are reports on the results of resource studies and of topographic, hydrologic, and geologic investigations. They also include collections of related papers addressing different aspects of a single scientific topic. Bulletins contain significant data and interpretatioi)S that are of lasting scientific intere~t but are generally more limited in scope or geographic coverage than Professional Papers. They include the results of resource studies and of geologic and topographic investigations; as well as collections of short papers related to a specific topic. Water-Supply Papers are comprehensive reports that present significant interpretive results of hydrologic investigations of wide interest to professional geologists, hydroiogists, and engineers. The series covers investigations in all phases of hydrology, including hydrogeology, availability of water, quality of water, and use of water. Circulars present administrative information or important scientific information of wide popular interest in a format designed for distribution at no cost to the public. Information is usually of short-term interest. Water-Resources Investigations Reports are papers of an interpretive nature made available to the public outside the formal USGS publications series. Copies are reproduced on request unlike formal USGS publications, and they are also available for public inspection at depositories indicated in USGS catalogs. Open-File Reports include unpublished manuscript reports, maps, and other material that are made available for public consultation at depositories. They are a nonpermanent form of publication that may be cited in other publications as sources of information. Maps Geologic Quadrangle Maps are multicolor geologic maps on topographic bases in 7 1/2-or 15-minute quadrangle formats (scales mainly 1:24,000 or 1 :62,500) showing bedrock, surficial, or engineering geology. Maps generally include brief texts; some maps include structure and columnar sections only. Geophysical Investigations Maps are on topographic or planimetric bases at various scales; they show results of surveys using geophysical techniques, such as gravity, magnetic, seismic, or radioactivity, which reflect subsurface structures that are of economic or geologic significance. Many maps include correlations with the geology. Miscellaneous Investigations Series Maps are on planimetric or topographic bases of regular and irregular areas at various scales; they present a wide variety of format and subject matter. The series also includes 7 1/2-minute quadrangle photo geologic maps on planimetric bases which show geology as interpreted from aerial photographs. Series also includes maps of Mars and the Moon. Coal Investigations Maps are geologic maps on topographic or planimetric bases at various scales showing bedrock or surficial geology, stratigraphy, and structural relations in certain coal-resource areas. 011 and Gas Investigations Charts show stratigraphic information for certain oil and gas fields and other areas having petroleum potential. Miscellaneous Field Studies Maps are multicolor or black-andwhite maps on topographic or planimetric bases on quadrangle or irregular areas at various scales. Pre-1971 maps show bedrock geology in relation to specific mining or mineral-deposit problems; post-1971 maps are primarily black-and-white maps on various subjects such as environmental studies or wilderness mineral investigations. Hydrologic Investigations Atlases are multicolored or black-andwhite maps on topographic or planimetri<... bases presenting a wide range of geohydrologic data of both regular and irregular areas; principal scale is 1:24,000 and regional studies are at 1:250,000 scale or smaller. Catalogs Permanent catalogs, as well as some others, giving comprehensive listings of U.S. Geological Survey publications are available under the conditions indicated below from the U.S. Geological Survey, Books and Open-File Reports Section, Federal Center, Box 25425, Denver, CO 80225. (See latest Price and Availability List.) "Publications of the Geological Survey, 1879-1961" may be purchased by mail and over the counter in paperback book form and as a set of microfiche. "Publications of the Geological Survey, 1962-1970" may be purchased by mail and over the counter in paperback book form and as a set of microfiche. "Publications of the U.S. Geological Survey, 1971-1981" may be purchased by mail and over the counter in paperback book form (two volumes, publications listing and index) and as a set of microfiche. Supplementsfor 1982,1983, 1984, 1985,1986, and for subsequent years since the last permanent catalog may be purchased by mail and over the counter in paperback book form. State catalogs, "List of U.S. Geological Survey Geologic and Water-SupplyReportsandMapsFor(State),"maybepurchasedbymail and over the counter in paperback booklet form only. "Price and Availability List of U.S. Geological Survey Publications," issued annually, is available free of charge in paperback booklet form only. Selected copies of a monthly catalog "New Publications of the U.S. Geological Survey" available free of charge by mail or may be obtained over the counter in paperback booklet form only. Those wishing a free subscription to the monthly catalog "New Publications of the U.S. Geological Survey" should write to the U.S. Geological Survey, 582 National Center, Reston, VA 22092. Note.--Prices of Government publications listed in older catalogs, announcements, and publications may be incorrect. Therefore, the prices charged may differ from the prices in catalogs, announcements, and publications.