Principal locations of metal loading from flood-plain tailings, Lower Silver Creek, Utah, April 2004
Because of the historical deposition of mill tailings in flood plains, the process of determining total maximum daily loads for streams in an area like the…
Public-domain full text preserved in the Mountain Man Mining Library. Original source: pubs.usgs.gov.
Scientific Investigations Report 2007–5248 Principal Locations of Metal Loading from FloodPlain Tailings, Lower Silver Creek, Utah, April 2004 U.S. Department of the Interior U.S. Geological Survey Prepared in cooperation with the UTAH DEPARTMENT OF ENVIRONMENTAL QUALITY, DIVISION OF WATER QUALITYCover photo
Cover photo: Metal-rich, acidic water draining away from piles of tailings along the flood plain of Silver Creek, Utah. Photograph taken looking southeast in April 2004 by Briant A. Kimball.
Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 By Briant A. Kimball, Robert L. Runkel, and Katherine Walton-Day U.S. Department of the Interior U.S. Geological Survey Prepared in cooperation with the UTAH DEPARTMENT OF ENVIRONMENTAL QUALITY, DIVISION OF WATER QUALITY Scientific Investigations Report 2007–5248
U.S. Department of the Interior DIRK KEMPTHORNE, Secretary U.S. Geological Survey Mark D. Myers, Director Reston, Virginia: 2007 For additional information write to: U.S. Geological Survey Director, USGS Utah Water Science Center 2329 W. Orton Circle Salt Lake City, UT 84119-2047 Email: GS-W-UTpublic-info@usgs.gov URL: ://ut.water.usgs.gov/ For more information about the USGS and its products: Telephone: 1-888-ASK-USGS World Wide Web: ://www.usgs.gov/ Suggested citation: Kimball, B.A., Runkel, R.L., and Walton-Day, Katherine, 2007, Principal locations of metal loading from flood-plain tailings, lower Silver Creek, Utah, April 2004: U.S. Geological Survey Scientific Investigations Report 2007–5248, 33 p. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Although this report is in the public domain, permission must be secured from the individual copyright owners to reproduce any copyrighted materials contained within this report. Scientific Investigations Report 2007–5248
Contents Abstract 1 Introduction 1 Purpose and Scope 1 Description of the Study Area 1 Previous Work 3 Acknowledgments 5 Methods for Mass-Loading Approach 5 Tracer Injection and Synoptic Sampling 5 Load Calculation 6 Sample Classification 9 Discharge from Tracer Dilution 9 Chemical Variation of Synoptic Samples 11 Inflow Samples 11 Stream Samples 13 Principal Locations of Mass Loading 17 Upstream from the Study Reach 17 Upper Meadow Tailings Piles 17 Lower Meadow Tailings Piles 19 Upstream from Pivotal Promontory Access Road 19 Waste-Water Treatment Plant and Old Big 4 Mill Tailings 19 Other Sources 19 Comparison between 2002 and 2004 19 Summary and Conclusions 19 References Cited 23 Figures
1. Location of the study reach indicating upper, middle, and lower injection reaches, location of changes in stream-water chemistry (colors indicate classification by cluster analysis), and principal locations of tailings, Silver Creek, Utah, April 2004 2
2. Photographs of major sources of metal loading along the study reach, Silver Creek, Utah, April 2004 4
3. Variation of bromide concentration at transport sites with time for (A) upper, (B) middle, and (C) lower injection reaches, Silver Creek, Utah, April 2004 8
4. Variation of (A) discharge measured at U.S. Geological Survey streamflow-gaging station 10129900 with time and (B) estimated discharge and bromide concentration with distance for stream and inflow samples along the study reach, Silver Creek, Utah, April 2004 10
5. Box plots showing the range of trace-element concentration among synoptic inflow samples collected from Silver Creek, Utah, April 2004 12
6. Variation of pH with distance along the study reach for stream-water and inflow samples collected along Silver Creek, Utah, April 2004 14
7. Variation of (A) sulfate and (B) zinc concentrations with distance along the study reach for stream-water and inflow samples collected along Silver Creek, Utah, April 2004 15
8. Variation of (A) cadmium with zinc concentration and (B) mole ratio of cadmium to zinc with distance along the study reach for stream-water and inflow samples collected along Silver Creek, Utah, April 2004 16
9. (A) Variation of sulfate load with distance along the study reach and (B) change in sulfate load for individual stream segments, Silver Creek, Utah, April 2004 20
10. (A) Variation of aluminum load with distance along the study reach and (B) change in aluminum load for individual stream segments, Silver Creek, Utah, April 2004 21
11. (A) Variation of zinc load with distance along the study reach and (B) change in zinc load for individual stream segments, Silver Creek, Utah, April 2004 22
12. Zinc load at common sampling sites from studies in 2002 and 2004, Silver Creek, Utah 23 Tables
1. Details of tracer injections for three injection reaches along Silver Creek, Utah, April 2004 5
2. Bromide concentration of synoptic water samples and characteristics of the sites at which the samples were collected, Silver Creek, Utah, April 2004 25
3. Concentration of major ions in synoptic water samples collected along Silver Creek, Utah, April 2004 28
4. Concentration of trace elements in synoptic water samples collected along Silver Creek, Utah, April 2004 31
5. Method detection limits and relative standard deviation of quality-assurance samples, Silver Creek, Utah, April 2004 11
6. Median, minimum, and maximum pH value and concentration of selected consituents in groups of inflow samples collected along Silver Creek, Utah, April 2004 13
7. Summary of principal locations of mass loading for Silver Creek, Utah, April 2004 18
Conversion Factors, Datum, and Abbreviated Water-Quality Units Multiply By To obtain Length meter (m) foot (ft) kilometer (km) mile (mi) Area square kilometer (km2) acre Volume liter (L) gallon (gal) milliliter (mL) gallon (gal) microliter (μL) 0.000000264 gallon (gal) Flow rate liter per second (L/s) gallon per minute (gal/min) Mass flow milligram per second (mg/s) pound avoirdupois per day (lb/day) kilogram per day (kg/day) pound avoirdupois per day (lb/day) Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as follows: °F=(1.8×°C)+32. Horizontal coordinate information is referenced to the North American Datum of 1927 (NAD 27). Chemical concentration and water temperature are reported only in metric units. Chemical concentration is reported in milligrams per liter (mg/L), micrograms per liter (µg/L), or milli moles per liter (mM/L). Milligrams per liter is a unit expressing the mass of solute per unit vol ume (liter) of water. For concentrations less than 7,000 milligrams per liter, the numerical value is about the same as for concentrations in parts per million. Specific conductance is reported in microsiemens per centimeter at 25 degrees Celsius (µS/cm).
Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 By Briant A. Kimball, Robert L. Runkel, and Katherine Walton-Day Abstract Because of the historical deposition of mill tailings in flood plains, the process of determining total maximum daily loads for streams in an area like the Park City mining district of Utah is complicated. Understanding the locations of metal loading to Silver Creek and the relative importance of these locations is necessary to make science-based decisions. Application of tracer-injection and synopticsampling techniques provided a means to quantify and rank the many possible source areas. A mass-loading study was conducted along a 10,000-meter reach of Silver Creek, Utah, in April 2004. Mass-loading profiles based on spatially detailed discharge and chemical data indicated five principal locations of metal loading. These five locations contributed more than 60 percent of the cadmium and zinc loads to Silver Creek along the study reach and can be considered locations where remediation efforts could have the greatest effect upon improvement of water quality in Silver Creek. Introduction In heavily mined watersheds, numerous tailings and waste-rock piles may occur that can be sources of metals and acidity to streams. The challenge facing those interested in improving water quality is thus one of source determination: in a given watershed, what sources of water are most detrimental to stream-water quality and how do they compare? Source determination also is particularly important in the Total Maximum Daily Load (TMDL) process because individual sources must be identified, and their relation to the total load from all sources must be quantified. In response to the source-determination question, an approach has been developed within the U.S. Geological Survey (USGS) Toxic Substances Hydrology Program to quantify mass loading associated with various sources (Kimball and others, 2002). This approach combines the methods of tracer dilution to quantify discharge and synoptic sampling to provide spatially detailed chemical information. Given discharge and chemical data, profiles of mass loading illuminate the principal locations where sources contribute metals and acid to a stream. The purpose of this investigation was to identify the principal locations of metal mass loading to Silver Creek in Summit County, Utah (fig. 1), a tributary to the Weber River, to provide information for the Silver Creek TMDL process for the Utah Department of Environmental Quality, Division of Water Quality (UDEQ). The mass-loading approach was employed by the USGS to quantify mass loading of metals to Silver Creek along a 10,000-m study reach that is listed on Utah’s 303(d) list as being impacted by zinc and cadmium (Michael Baker Jr., Inc., 2004; Utah Department of Administrative Services, 2005). A reconnaissance mass-loading study in the southern portion of lower Silver Creek identified substantial loading of metals to Silver Creek, but the analysis only quantified the net loading; it did not give details about the location of particular sources in this portion of lower Silver Creek (Kimball and others, 2004). Almost all of these tailings occur in the flood plain of Silver Creek, and thus they are commonly called “flood-plain” tailings. Purpose and Scope The purpose of this report is to document the principal locations of metal mass loading to Silver Creek, Utah. This report (1) characterizes the chemistry of stream water and inflows along the Silver Creek study reach, (2) quantifies the metal loading along the study reach, and (3) identifies the principal locations where metal loading occurs. These results will facilitate science-based decisions about targets for remediation. Description of the Study Area This study addresses the reach of Silver Creek from the U.S. Highway 40 overpass to the Interstate 80 overpass, a reach of almost 10,000 m (fig. 1). Silver Creek originates upstream from Park City, Utah (to the southwest of the area in fig. 1), and flows into the Weber River near Wanship, Utah (to the northeast of the area in fig. 1). This has been called the southern portion of the lower Silver Creek site by UDEQ in their Innovative Assessment (Ann Tillia, Utah Department of Environmental Quality, written commun., 2005). USGS
2 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Figure 1 Figure 1. Location of the study reach indicating upper, middle, and lower injection reaches, location of changes in stream-water chemistry (colors indicate classification by cluster analysis), and principal locations of tailings, Silver Creek, Utah, April 2004.
Introduction 3 discharge-gaging station 10129900, Silver Creek near Silver Creek Junction, Utah, is located near the end of the study reach and measures flow from a drainage area of 45 km2. The flow measured at the gage includes discharge from a wastewater treatment plant (WWTP) located just upstream from the gage (fig. 1). Timing of the sampling was planned so that samples would reflect stream-water quality under snowmelt runoff conditions because Silver Creek can be ephemeral along this study reach during typical low-flow periods. Mean annual discharge at the gaging station is 82 L/s, based upon discharge records for 2002 through 2004, which were all years with drought conditions (Tibbetts and others, 2004; Wilkowske and others, 2003). Monthly mean discharge varies from a low of 44 L/s in September to a high of 206 L/s in March. April has a monthly mean discharge of 167 L/s at the stream gage; this value and that from March are a result of snowmelt runoff. Most of the discharge at the gage during low-flow months is from discharge of the WWTP. Upstream from the WWTP, which is most of the study reach, Silver Creek can be ephemeral. Diversions of Silver Creek required that the 10,000-m study reach be divided into three injection reaches for the study (fig. 1). The upper injection reach (from 0 to 1,452 m) contained a wetland area that started downstream from 525 m. Silver Creek discharges from the wetland into two branches that flow under Highway 248 through two separate culverts. The two branches converge again upstream from 1,371 m, allowing for an accounting of discharge at the end of the upper injection reach. For the stream sites between 525 m and 1,371 m, no discharge estimate was possible. The upper injection reach included two important locations for floodplain tailings. An area just downstream from the start of the study reach is locally referred to as the “flood-plain” tailings, but has been labeled “upstream tailings” in figure 1 (fig. 2A). At Richardson Flat, a tailings pond is separated from direct contact with Silver Creek by an earthen dam. During recent periods of drought, discharge at 1,452 m usually has been diverted down the valley in an irrigation ditch along the east side of the Silver Creek valley. For the purposes of this study, some of the water was allowed back into the natural channel of Silver Creek at 1,452 m to provide continuous discharge along the entire middle injection reach (1,601 to 7,259 m). Because the study occurred at the end of the snowmelt period in Silver Creek, this was a diversion into a channel that had not been dry for a substantial period of time. Thus, the diverted flow was not adsorbed by a dry alluvial channel. Much of the channel contained flow before the diversion, but continuous discharge in the natural channel was necessary to join all the ground-water inflows and to quantify loading from the ground-water discharge along the middle injection reach. In the meadow area, from 1,601 m to 7,142 m, two principal areas contain visible tailings piles; upper areas from 1,843 m to about 3,162 m; and a lower area from 5,251 m to 7,142 m. Tailings in both the upper and lower meadow areas are present in piles (mounds, berms, and hummocks) along the stream that could have been created in preparation for shipping to be reprocessed (Ann Tillia, Utah Department of Environmental Quality, written commun., 2005). Vegetation around the tailings is very scarce; the mounds are mostly bare. A typical inflow from mounds of tailings in the upper meadow area is shown in figure 2B. Ground-water discharge from the lower meadow area is shown in figure 2C. The lower injection reach (7,142 to 9,747 m) contained continuous discharge as a result of the ground-water discharge upstream in the middle injection reach. Tailings from the operation of the Old Big 4 mill, which was located near the present Pivotal Promontory access road, contribute metals to the lower tailings area. Additional tailings farther downstream also contribute metals (figs. 1 and 2D). This lower reach also receives discharge from the WWTP as well as return flow at 9,360 m from the irrigation ditch that starts at 1,452 m in the upper injection reach. Details of the ore deposits in the Park City district have been discussed by Garmoe and Erickson (1968) and Bromfield (1989). Because the study reach is affected by tailings from the ore processing, the mineralogy of the ore deposits is the most important aspect of these reports. Sphalerite (ZnS) is the principal ore mineral contributing zinc. Cadmium commonly substitutes for zinc in sphalerite; thus, this mineral is the principal source of cadmium as well. Additionally, some of the ores occurred as skarn deposits, which are hosted in carbonate rocks. Carbonate minerals, especially rhodochrosite (MnCO3), also occurred as gangue minerals in the intrusions (Rockwell and others, 1999). Thus, tailings from these ores should have abundant sphalerite and carbonate rhodochrosite. Previous Work A reconnaissance of this same study reach by Kimball and others (2004) included stream discharge and chemistry for four locations. These locations were upstream and downstream from Richardson Flat, and upstream and downstream from the WWTP. Loads of cadmium and zinc increased downstream between each of these four sampling locations. At the time of that study (Kimball and others, 2004), discharge from Silver Creek was completely diverted into an irrigation ditch, and there was no continuous flow in the natural channel. Numerous ground-water discharges from tailings in the meadow area were observed, but the amount of mass loading from the various inflows was not quantified. At the sampling site upstream from the WWTP, the ground-water discharge had combined to create continuous flow in the channel. Information from the study area has been compiled for a TMDL study (Michael Baker Jr., Inc., 2004). As with the USGS reconnaissance, however, there was little detail on sources within the meadow area. Another USGS study (Giddings and others, 2001) identified elevated metal concentrations in bed sediments of Silver Creek. The elevated
4 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 F g ure 2 Figure 2. Photographs of major sources of metal loading along the study reach, Silver Creek, Utah, April 2004. (A) Looking upstream toward injection point under the Highway 40 bridge and the “upstream” tailings; (B) Looking upstream while sampling an inflow draining from mounds of tailings in the upper meadow tailings area; (C) Looking upstream at the pond at the end of the lower meadow tailings area, upstream from Pivotal Promontory access road; and (D) Looking upstream at tailings in the flood plain downstream from historical Old Big 4 mill site.
Methods for Mass-Loading Approach 5 concentrations extended all the way from the Park City area to the mouth of Silver Creek. Acknowledgments This study was done in cooperation with the Utah Department of Environmental Quality, Division of Water Quality. Additional support was provided by the U.S. Geological Survey Toxic Substances Hydrology Program. Field assistance was provided by Andrew Burr, Jay Cederberg, Steven Gerner, James Mason, Judy Steiger, Ann Tillia, and John Whitehead. Logistical support and additional information were provided by John Whitehead and Ann Tillia of the Utah Department of Environmental Quality. Metal analyses were done at the University of Southern Mississippi under the direction of Alan Shiller; anion analyses were done by Judy Steiger at the U.S. Geological Survey Utah Water Science Center. Methods for Mass-Loading Approach A mass-loading approach to identify sources of metals combines several methods. Details of these methods are reported elsewhere (Kimball and others, 2002; Kimball and others, 2004; Kimball and others, 2006b; Kimball and others, 2006a), but some aspects are repeated here to help understand the results for Silver Creek. Data collection for the approach is based on field methods of tracer dilution (Kilpatrick and Cobb, 1985) and synoptic sampling (Bencala and McKnight, 1987). Data analysis is based on methods of calculating loads to obtain detailed longitudinal profiles of mass loading (Kimball and others, 2002; Kimball and others, 2003). Also, multivariate sample-classification methods help to interpret the detailed chemical results. Tracer Injection and Synoptic Sampling The mass-loading study began with a careful evaluation of inflows along the study reach, which was accomplished by walking the entire study reach (fig. 1). Before flow was diverted into the middle injection reach, ground-water inflows were evident, and their cumulative effect created some perennial discharge by the end of the middle injection reach. Stream sites for synoptic sampling were chosen upstream and downstream from the inflows to allow mass-balance calculations. Additional stream sites were located along the study reach at regular intervals to check for dispersed, subsurface inflow to the stream. Sampling sites for the synoptic study are referenced by the measured distance along the study reach in a downstream direction, with the injection site assigned a distance of 0 m. Inflows are referred to as left and right bank with an orientation looking downstream. Reference to a stream segment means the section of the study reach between two consecutive stream sites, and is referenced by both the upstream and downstream distances, for example the segment 1,601–1,843 m. A continuously injected chemical tracer provides a way to measure discharge that includes the hyporheic flow of the stream because it follows the water as it moves in and out of the streambed. Under ideal conditions, tracer-dilution techniques allow the detection of increases in discharge of only a few percent. Once the tracer reaches a steady concentration at each point along the stream, called the plateau condition, discharge can be calculated at any stream point from the concentration of the tracer at that point. This typical application of a tracer-injection study was adequate for the upper and lower injection reaches, but for the middle injection reach the approach was modified. Sodium bromide was selected for the injection solution because of the high pH of the stream. No geologic sources of bromide were suspected in the watershed (Nichols and Bryant, 1990). In the analysis of this experiment, bromide is assumed to be a conservative tracer. No adverse effects on organisms were observed from the injection of the tracer solution. Details of the three tracer injections are provided in table 1, and the system of pumps and controls is detailed in Kimball and others (2004). The background concentration of the tracer was much lower than the concentration of injected tracer in the stream and was mostly uniform. With these uniform background Table 1. Details of tracer injections for three injection reaches along Silver Creek, Utah, April 2004. [L/s, liters per second; mg/L, milligrams per liter] Injection reach Injection start Synoptic start Synoptic end Injection end Injection rate (L/s) Tracer- injectate concentra tion (mg/L as bromide) Background bromide concentration (mg/L) Date Time Date Time Date Time Date Time Upper 4/14/2004 12:00 4/15/2004 9:07 4/15/2004 13:25 4/15/2004 11:00 159,600 Middle 4/8/2004 9:00 4/9/2004 8:55 4/9/2004 14:12 4/9/2004 15:00 162,800 Lower 4/5/2004 15:27 4/6/2004 9:34 4/6/2004 14:00 4/6/2004 15:50 160,300
6 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 concentrations, stream discharge at any location downstream from the injection is given by:
(1) where:
QD is the stream discharge at the downstream site, in L/s,
QINJ is the injection rate (table 1), in L/s,
CINJ is the injectate concentration, in mg/L,
CD is the tracer concentration at a downstream site, in mg/L, and
CB is the naturally occurring tracer concentration, in mg/L. The amount of tracer dilution between two consecutive stream sites indicates the total inflow from surface and ground water for that segment of the study reach. Tracer dilution accounts for visible inflows, such as tributaries and springs, as well as dispersed, subsurface inflow. No separate measurement was made of tributary inflow to be able to divide the total inflow volume between surface- and ground-water components for a given stream segment. Synoptic samples were collected at numerous stream and inflow locations after the bromide concentration reached a steady-state plateau. Sampled inflows were mostly small springs and some irrigation return flows; only one welldefined tributary occurred at 9,562 m. A complete listing of sampling locations, sample information, and the chemical data are provided in tables 2, 3, and 4 (located at back of report). Samples were collected in 1.8-L HPDE bottles usually by submersing the neck of each bottle into the water near the center of flow. Samples were transported to a central processing area where 125-mL aliquots were prepared for cation and anion analyses. Onsite processing included filtration and pH measurement. Filtration was completed with in-line capsule disk filters with an effective pore size of 0.45-µm (FA samples). Some total-recoverable samples (RA) were collected to evaluate the presence of colloidal concentrations of metals. The colloidal concentration was calculated as the difference between the RA and FA concentration for those samples that included both. Both FA and RA aliquots for cation analysis were acidified to a pH of less than 2.0 with ultrapure nitric acid. Total recoverable and dissolved cation concentrations were determined from unfiltered and filtered samples, respectively, by using inductively coupled argon plasma-mass spectrometry. Cation concentrations are reported for aluminum, arsenic, barium, calcium, cadmium, cobalt, chromium, copper, iron, potassium, magnesium, manganese, molybdenum, sodium, nickel, lead, silica (as silicon), silver, strontium, uranium, vanadium, and zinc. Dissolved anion concentrations were determined from filtered, unacidified samples by ion chromatography. Anion concentrations are reported for chloride, bromide, and sulfate. E q u a t on Alkalinity (as calcium carbonate) was determined by titration from filtered, unacidified samples. Load Calculation Three specific load calculations are used to quantify the sources of loading to Silver Creek. First, the tracer injection provides estimated discharge ( Q) and synoptic sampling provides constituent concentrations (C), which are combined to determine sampled instream load:
E q u a t o n Equation 2 where:
MA is the constituent load, or mass flux, at location A, in kg/day,
CA is the concentration of the selected constituent at location A, in mg/L,
QA is the discharge at location A, in L/s, and
0.0864 is the conversion factor from mg/s to kg/ day. Sampled instream load for stream sites was calculated from the filtered concentration (FA sample) of the constituent. The longitudinal profile of sampled instream load is the basic result from the mass-loading study. The second load calculation determines the net change in mass load in one stream segment, and is used to determine if the load of a given constituent increases or decreases in the given segment. For the change in load for the segment starting at location A and ending at location B, we calculate:
Equation 3
Equation 3 where: DMB–A is the change in sampled instream load from locations A to B, in kg/day, MB is the constituent load at location B, in kg/day, and MA is defined in equation 2. Gains in constituent load (DMB–A is greater than zero) imply that there is a source that contributes to the stream between the two stream sites. Instream load also can decrease within a stream segment (DMB–A is less than zero), meaning that there was a net loss of the constituent from physical, chemical, or biological processes. Summing all the increases in load between sampling sites along the study reach (positive values of DMB–A) leads to the cumulative instream load. At the end of the study reach, the cumulative instream load is the best estimate of the total load added to the stream but is likely a minimum estimate because it only measures the net loading
Methods for Mass-Loading Approach 7 for segments and does not account for loss resulting from reaction. For those segments that include a sampled inflow, a third load calculation is possible. If stream sites A and B surround an inflow sample, location I:
Eq u a t o n
Equation 4 where: DMI is the load attributed to the inflow, I, in kg/day, is the inflow concentration, in mg/L, QB is the discharge at site B, in L/s, and QA and 0.0864 are defined in equation 2. Summing the inflow loads along the study reach produces a longitudinal profile of the cumulative inflow load. This sum can be compared to the cumulative instream load to indicate how well the sampled inflows account for the load measured in the stream. The cumulative instream and cumulative inflow profiles would be nearly equal if the sampled inflows were completely representative of the constituent concentration for all the water entering the stream, but that is rarely the case. Ground-water inflow into streams affected by mine drainage often has higher concentrations of metals than surfacewater inflows into the same stream segment. This causes the cumulative instream load to be greater than the cumulative inflow load and can indicate important areas of unsampled inflow load, which is defined as:
Equation 5Equation 5U Equation 5Equation 5Equation 5B–AEquation 5Equation 5Equation 5I Equation 5 where: DMU is the unsampled inflow load, in kg/day, and DMB–A and DMI are defined in equations 3 and 4.
Unsampled inflow can be calculated for individual stream segments even if the segment does not include a sampled inflow or for the entire study reach by comparing the cumulative instream and inflow loads. If the value is negative for the entire study reach, however, it can still be positive for some individual stream segments. Note that DMB–A includes all sources of loading within a stream segment and, in most cases, does not distinguish the quantity added by an individual source. Because there is measurement error inherent in discharge estimates, chemical analysis, and sampling, a load error equation is used to constrain the changes of sampled instream load. The load error is calculated from an equation that accounts for these potential sources of error (McKinnon, 2002):
Equation 6 + ( )( . ) Q Q A A A A Equation 6 where:
DCA is the precision of chemical analysis,
DQA is the precision of discharge calculation, and
QA, CA and 0.0864 are defined in equation 2. The value of DCA is calculated in a manner analogous to that used by Friedman and Erdman (1982) for single operator precision. The coefficient of variation (CV), representing precision, and the mean concentration are calculated for repeated analysis of a constituent in a set of standard reference samples spanning a range of concentrations. Values for CV are regressed as a power function of the mean concentrations to obtain an equation expressing analytical precision, DCA, as a function of concentration:
C a C A A b ( )
Equation 7 where:
DCA is precision for the chemical measurement at site A, in percent,
a is the coefficient from regression,
CA is the concentration of the constituent at site A, and
b is the exponent from regression. The value of DQA is based on the CV for the plateau tracer concentration at the transport sites during the period of synoptic sampling. For example, for the upper injection reach (fig. 3A), the mean bromide concentrations at transport sites T1 and T3 during synoptic sampling were 4.23 mg/L and 1.74 mg/L, respectively (site T2 was not located on the main channel, but on the returning ditch, an inflow). The value of CV for site T1 was 2.5 percent and for site T3 was 9.6 percent. Similar to the procedure for analytical precision, the values of CV for each mean are used to develop a linear regression for DQA:
Q mC b A A T
+
Equation 8 where: DQA is the discharge error at site A,
m is the slope from linear regression, C T A is the tracer concentration at site A, and
b is the intercept from linear regression. Both DCA and DQA give the percentage of CA and QA to be substituted into equation 6 to calculate load error. The load error is compared to the change in load to the next site, DMB–A. If the absolute value of DMB–A is greater than the load error, then there has been a measurable and significant change
8 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Figure 3. Variation of bromide concentration at transport sites with time for (A) upper, (B) middle, and (C) lower injection reaches, Silver Creek, Utah, April 2004.
Discharge from Tracer Dilution 9 in load. Only the values of DMB–A that are greater than the load error are included in the longitudinal profiles of sampled instream load and the cumulative instream load. Sample Classification An important objective of synoptic sampling is to recognize patterns or chemical characteristics among samples that can indicate the sources of mine drainage. Water that interacts with distinct mineral assemblages may exhibit characteristic chemical signatures that can provide distinctions among the inflow samples. Thus, groups of inflow samples are identified by their similarities. In this study, distinctions among inflow groups lead to understanding differences in drainage from the various areas where tailings occur. Groups of stream-water samples indicate where major changes occur in surface-water chemistry. Sample classification was done separately for inflow and for stream-water samples. A cluster analysis method called partitioning around medoids was used to evaluate distinctions among the inflow and stream-water samples (Kaufman and Rousseeuw, 1990). For both stream-water and inflow samples, the method uses the Euclidian distance between samples in multi-dimensional space to determine clusters or groups of samples with samples that are similar, and yet groups that are the most distinguished from each other. To emphasize the linear relations among variables, the chemical concentration of each constituent, is expressed in millimoles per liter. These values are converted to standardized variables in the analysis. Only filtered concentrations were used as input to the analysis. Discharge from Tracer Dilution Understanding the effects of flood-plain tailings on mass loading to Silver Creek is based on three critical lines of evidence. First is the estimation of discharge from the tracer dilution, second is the pattern of chemical variation of inflow and instream concentrations, and third is the longitudinal pattern of mass loading that comes from a combination of the synoptic discharge and chemical data. To estimate discharge from tracer dilution, a concentrated sodium bromide solution was slowly pumped into the stream at the upstream end of each injection reach. Details of the time, injection rate, and tracer concentration of the injectate solution for each injection reach are presented in table 1. During the periods of synoptic sampling, the tracer concentration in the middle (fig. 3B) and lower (fig. 3C) reaches appeared to attain a steady-state plateau at each transport site. During synoptic sampling for the upper injection reach (fig. 3A), however, a plateau occurred at site T1, but tracer concentrations at sites T2 and T3 appeared to be increasing. Thus, discharge estimates downstream from 525 m for the upper injection reach were not calculated. For the middle and lower injection reaches, however, where concentrations vary with downstream distance, but not with time, values of bromide concentration for each synoptic stream site can be used to estimate a discharge value by using equation 1. Smoothed bromide concentrations, using the method of Velleman and Hoagland (1981), were used in the discharge calculations, and the smoothed concentrations of the bromide tracer and estimated discharge at all of the stream sites are listed in table 2 (located at back of report). Bromide concentrations of inflow samples were variable (fig. 4B). The median bromide concentration among inflows (excluding those inflows that directly drained roads) was 0.3 mg/L (fig. 4B), which is a likely background concentration for this study reach. Twelve samples had a bromide concentration of greater than 0.5 mg/L (fig. 4B), and those samples most likely had some portion of stream water in them. Most of these samples were collected in the middle injection reach where the diversion of water could have caused some back mixing with inflows. Because higher bromide concentrations among inflow samples were likely the result of injected bromide, and not the result of natural sources of bromide, the instream bromide concentrations should remain acceptable for calculating discharge with equation 1. Discharge estimates must be viewed in the context of variation that occurred during the 10-day period of the injections (fig. 4A). Hourly-scale variation in the gagingstation record resulted from variable discharge of the WWTP, and this variation did not occur upstream from the WWTP. Daily scale variation was a result of diel variations from snowmelt. Two periods of rain occurred and discharge peaked at the gaging station at about 0:00 hours on April 8 and 0:00 hours on April 9. The period of synoptic sampling for each injection is indicated by vertical lines, and discharge at the gage varied from an average of 110 L/s during the lower injection, to 209 L/s during the middle injection, to 67 L/s during the upper injection. Discharge at the end of the middle injection reach (fig. 4B), was substantially greater than at the beginning of the lower injection reach. In a temporal context (fig. 4A), the difference is explained by the storms that occurred between the two injections. The base discharge at the gaging station was 123 L/s higher during the middle than during the lower injection and mostly accounts for the difference of 142 L/s (fig. 4A) between the two injection reaches. The comparable values of discharge at the end of the upper injection reach on April 14 and the beginning of the middle reach on April 9 should differ by much more than they do, because discharge at the end of the upper injection reach on April 9 should have been greater after the storms. Not all the discharge from the upper injection reach, however, was diverted to the natural channel for the middle injection reach, and the amount that was diverted was nearly equal to the discharge at the beginning of the middle injection reach on April 9. Thus, the temporal variations over the 10-day period can explain the discharges illustrated in figure 4B.
10 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Figure 4. Variation of (A) discharge measured at U.S. Geological Survey streamflow-gaging station 10129900 with time and (B) estimated discharge and bromide concentration with distance for stream and inflow samples along the study reach, Silver Creek, Utah, April 2004. F igure
Chemical Variation of Synoptic Samples 11 Chemical Variation of Synoptic Samples The discharge profiles of each injection reach are combined with an equally detailed profile of stream and inflow chemistry. For this 10,000-m study reach, 52 stream and 46 inflow sites were sampled to provide the desired characterization (table 2). Results of chemical determinations are listed in table 3 for major ions and table 4 for trace elements (both tables located at back of report). All samples were evaluated for charge balance and all but two samples had a balance less than 5.2 percent; the median balance was 1.97 percent. New spectroscopic technology, inductively coupled argon plasma/mass spectrometry, (ICP-MS), makes the determination of low concentrations of metals possible. Method detection limits for the analyses of the synoptic samples are listed in table 5; many detection limits were less than one part per billion. Precision for each element was determined by a modification of the method for single operator precision (Friedman and Erdmann, 1982). Statistics for calculating single operator precision were developed by running certified standards and field standard reference samples at regular intervals throughout the chemical analysis. By calculating the CV for a given concentration from these reference standards, power function equations for CV as a function of concentration were developed; coefficients and exponents for these equations are listed in table 5, and, as described in the “Methods” section, are used in the load error calculation to determine the DCA term in equation 6. Inflow Samples Metal concentrations measured for inflow samples span nine orders of magnitude, and a comparison using box plots (Velleman and Hoaglin, 1981) demonstrates this range (fig. 5). Such a large range of concentration suggests that the inflows sampled in this study most likely represent the possible range of inflow chemistry affecting Silver Creek in the study reach. A substantial percentage of the samples had cadmium, iron, manganese, strontium, and zinc concentrations that were greater than 100 µg/L. Zinc concentration in samples from 4 inflows exceeded 100,000 µg/L, and one of these exceeded 1,000,000 µg/L (fig. 5; table 4). These high concentrations indicate the potential importance of these flood-plain tailings as sources of metals to Silver Creek. Inflow samples have been classified using cluster analysis into four groups on the basis of their chemical composition. Distinctions among the groups are evident from variations in pH and concentrations of selected constituents (table 6). The groups have been arranged in an order of decreasing pH and increasing concentration (with the exception of alkalinity), and this order could represent the extent of weathering of floodplain tailings or weathering of tailings having variable content of sphalerite and other metal-rich minerals such as rhodochrosite. None of the inflow samples can be considered totally unaffected by interaction with tailings material, but the groups may represent the extent of interaction or else the effect of differing mineralogy in the tailings material. Inflow samples that have the highest values of pH (least and moderately affected groups) also have the lowest concentrations of calcium, sulfate, and zinc, but the highest concentration of alkalinity. On the other hand, samples with the lowest pH have the highest concentrations of calcium, sulfate, and zinc (substantially affected and most affected groups). Spatially, general distinctions exist among the groups of inflow samples. Inflows most affected by tailings occurred at the Table 5. Method detection limits and relative standard deviation of qualityassurance samples, Silver Creek, Utah, April 2004. [MDL, method detection limit] Constituent MDL, in micrograms per liter Coefficient of variation Coefficient Exponent Calcium Magnesium Sodium Potassium Alkalinity as CaCO3
Sulfate 1,760 Chloride Bromide Silica, as Si Aluminum Arsenic Barium Cadmium Cobalt Chromium Copper Iron Lead Lithium Manganese Molybdenum Nickel Silver Strontium Uranium Vanadium Zinc
12 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Figure 5 Figure 5. Box plots showing the range of trace-element concentration among synoptic inflow samples collected from Silver Creek, Utah, April 2004.
Chemical Variation of Synoptic Samples 13 beginning and near the end of the middle injection reach (fig. 6, orange triangles at 1,965 m and 5,928 m). Both these inflows originated directly from tailings piles (table 2). Moderately affected inflows (light blue triangles) mostly occurred from the beginning of the study reach (0 m) to near 4,403 m. Substantially affected inflows (yellow triangles) mostly occurred from 5,251 m to 8,497 m. In general, this group of substantially affected inflows not only had lower pH than the least and moderately affected groups, but also had higher concentrations of sulfate and zinc (fig. 7A and B). If the mining wastes were derived from ore deposits that had the same age of mineralization, the sphalerite might have a uniform ratio of cadmium to zinc. In a plot of cadmium with zinc, a constant ratio is represented by a line of unit slope (fig. 8A). Not all samples plot along a line of unit slope (fig. 8A). Samples from the least and the moderately affected inflows had the most variable cadmium to zinc ratio, and samples from the substantially and most-affected inflows had a relatively constant ratio. This corresponds to a spatial pattern of higher ratios occurring among inflow samples between 2,000 and 4,800 m (fig. 8B), or the area of the upper meadow tailings piles (fig. 1). Stream-water samples from this same area and also downstream to the end of the study reach generally had the same ratio and plot along the line of unit slope (fig. 8A). This result indicates that zinc and cadmium in the middle and lower injection reaches were mostly obtained from the tailings piles in those areas rather than from upstream sources. This is consistent with the substantial increases in zinc concentration among samples collected downstream from 2,000 m (fig. 7B) and has implications for remediation. Stream Samples Distinctions that occur among groups of stream-water samples have a different implication than distinctions among groups of inflow samples. As noted, distinctions among inflow sample groups could result from the degree of interaction with flood-plain tailings or the variable chemical character of tailings, both possibilities reflecting catchment sources of zinc. Distinctions among stream-water groups along the study reach in Silver Creek, however, represent changes in stream-water chemistry in response to inflows from the various sources. Consequently, the resulting classification of streamwater samples into groups represents a sequence of changes along the study reach. The locations of different groups are indicated in figure 1. Sulfate and zinc concentrations illustrate the pattern of change for stream-water samples collected along the study reach (diamond symbols for streamwater samples; fig. 7A and B). From upstream to downstream, five groups were distinguished by cluster analysis and are designated as A-E. Group A (dark blue diamonds; 0 to 1,843 m) – Sulfate
concentration at the beginning of the study reach was consistently near a median concentration of 294 mg/L. Zinc concentration progressively increased along the upper injection reach from 1,300 to almost 1,700 µg/L at 1,452 m. The increase could indicate a contribution from the “upstream” tailings (fig. 1), but the median zinc concentration of 1,590 µg/L was relatively low compared to concentrations downstream. Group B (light blue diamonds; 861 m to 1,309 m)
– The chemical character of samples from the right branch of the upper injection reach (stream-water samples collected at 861 m, 1,229 m, and 1,309 m) differed from that of the main channel, with a slightly higher sulfate concentration, but a lower zinc concentration. The difference in chemistry indicates that ground water may flow into the right branch after the stream splits, but it is of note that metal concentrations are lower as a result. Table 6. Median, minimum, and maximum pH value and concentration of selected consituents in groups of inflow samples collected along Silver Creek, Utah, April 2004. Groups are labeled by the degree to which they are affected by interaction with mining wastes. [mg/L, milligrams per liter; LD, less than detection limit; µg/L, micrograms per liter] Constituent Group Number of samples Median Minimum Maximum pH, in standard units Least Moderate Substantial Most Calcium, in mg/L Least Moderate Substantial Most Alkalinity as CaCO3, in mg/L Least Moderate Substantial Most LD LD LD Sulfate, in mg/L Least Moderate Substantial 1,083 3,250 Most 3,595 3,510 3,680 Zinc, in µg/L Least Moderate 3,380 25,500 Substantial 37,443 8,380 1,070,000
Most 200,838 132,000 270,000
14 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Figure 6 Figure 6. Variation of pH with distance along the study reach for stream-water and inflow samples collected along Silver Creek, Utah, April 2004. Group D (orange diamonds; 5,251 m to 8,862
m) – Concentrations of sulfate and zinc increased substantially a second time from the influence of the lower meadow tailings piles (fig. 7). Increases in both sulfate and zinc concentration occurred at the end of the middle injection reach, and again at the start of the lower injection reach. Particularly for zinc concentration, the increases were substantial and reflect the effect of the tailings, both in the lower meadow area (about 5,000 m to 7,142 m) and the Old Group C (yellow diamonds; 2,171 m to 4,800 m) –
Downstream from the point where water was diverted to the middle injection reach at 1,452 m, the first two stream-water samples (1,601 m and 1,843 m) were similar to the upstream stream-water samples (group A). However, there was a distinct change at 2,171 m that reflects the influence of the upper meadow tailings piles (fig. 1). Inflows from the upper meadow tailings piles caused substantial increases in both sulfate and zinc concentrations. Median concentrations between 2,174 m and 4,800 m increased to 332 mg/L sulfate and 3,730 µg/L zinc (fig. 7A and B). Big 4 mill area (7,142 m to 8,909 m). The mole ratio of the stream water for cadmium to zinc varies as a result of inflows in both these locations; first a decrease
Chemical Variation of Synoptic Samples 15 Figure 7 Figure 7. Variation of (A) sulfate and (B) zinc concentrations with distance along the study reach for stream-water and inflow samples collected along Silver Creek Utah, April 2004.
16 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 F igure
Figure 8. Variation of (A) cadmium with zinc concentration and (B) mole ratio of cadmium to zinc with distance along the study reach for stream-water and inflow samples collected along Silver Creek, Utah, April 2004.
Principal Locations of Mass Loading 17 occurred in the ratio from 5,251 m through 6,322 m, and then a steady ratio occurred near 7,571 m (fig. 8). Group E (red diamonds; 8,909 m to 9,747 m)
– Compared to the upstream group, almost all concentrations were lower as a result of dilution by the inflow of the WWTP (fig. 7), which entered Silver Creek at 8,881 m. Further dilution occurred downstream from the irrigation return flow at 9,360 m. As described, the mole ratio of cadmium to zinc in stream-water samples indicates the influence of inflows from the tailings piles in the upper and the lower meadow areas (fig. 8B). Waters from group A (dark blue diamonds) had a ratio near 0.0012, but at 2,171 m, the ratio increased to a nearly constant value of 0.0033 in response to high ratios of inflow waters. At 5,251 m, the instream ratio began to decrease in response to lower ratios of inflows from the lower meadow tailings piles, as noted above. In the lower injection reach, downstream from 7,142 m, the instream cadmium to zinc ratios in the stream-water samples were nearly constant. The initial change at 2,171 m and the subsequently constant ratio suggest the effect of tailings piles in the upper meadow area as a source of these metals. Concentrations of cadmium and zinc in stream-water of Silver Creek exceeded chronic aquatic-life standards (Utah Department of Administrative Services, 2005). All the streamwater samples exceeded the hardness-based chronic toxicity level for zinc (fig. 7B). For cadmium, water samples collected from all stream sites downstream from 1,601 m exceeded the hardness-based chronic toxicity standard. All instream concentrations of copper, lead, and nickel were less than the calculated hardness-based chronic toxicity standards. Ten locations from the lower injection reach included analysis of both the filtered and unfiltered samples (table 4). In all but the replicate sample at 9,438 m, aluminum, arsenic, copper, iron, lead, and silver were substantially in the colloidal phase. Cadmium and zinc were partly colloidal in some of the samples, but the remaining metals were mostly in the filtered phase. These metals commonly form or are sorbed to colloids in streams affected by mine drainage (Kimball and others, 1995; Nordstrom and Alpers, 1999; Smith, 1999), particularly in the pH range of these Silver Creek samples. The presence of these metals in the colloidal phase suggests they may present a chronic toxicity problem in addition to the acute toxicity. Principal Locations of Mass Loading Detailed longitudinal profiles of loading along the study reach come from the combination of the spatially detailed discharge and chemical data and indicate where the most substantial loads enter the stream. Although the three separate injection reaches were studied on different days and under different flow regimes, the combination of results from all three can be unified to present a profile for the entire stream. This combination was accomplished by calculating significant changes (using equation 5) for each stream segment within each injection reach. These significant changes were then summed incrementally along each injection reach. The resulting load at the end of the upper injection reach was then used as the starting load for the middle injection reach. Likewise, the sum of changes at the end of the middle injection reach was used as the starting load for the lower injection reach. This calculation leads to a detailed longitudinal profile of mass loading for each element that represents sums of significant changes along the entire study reach. Note that the profile calculated in this manner does not represent the absolute load. For almost all the constituents, the profile can be summarized with reference to five principal locations, summarized in table 7, that account for most of the mass loading along the study reach. Three of the locations consist of only one stream segment (1, 4, and 5), while two locations are sums of the load contributions from several stream segments (2 and 3). Photographs of some of the principal locations are shown in figure 2. Mass loading at these five principal locations is illustrated with the load profiles of sulfate, aluminum, and zinc (figs. 9, 10, and 11, respectively). Upstream from the Study Reach The first stream segment, represented by the load at 0 m, indicates the net loading from all upstream sources (fig. 2A). Metal loading has been documented at several locations upstream from the study reach (Kimball and others, 2004). These upstream sources contribute more than 10 percent of the cumulative instream loads of calcium, magnesium, sodium, potassium, sulfate, chloride, aluminum, barium, chromium, and strontium. For example, sulfate load (fig. 9A) at the upstream end of the study reach was greater than 1,300 kg/ day; this segment contributed the second largest load of any individual stream segment for sulfate (fig. 9B). Upper Meadow Tailings Piles Six stream segments, from 2,171 m to 2,757 m (fig. 2B) represent the next principal location of mass loading. This stream reach is notable for the increase in loads of several metals, including aluminum (41 percent of total load), barium (31 percent), cadmium (23 percent), copper (23 percent), iron (33 percent), lead (19 percent), nickel (29 percent), and strontium (19 percent). This stream reach had the greatest loading for aluminum (fig. 10B), but the loadings of sulfate (fig. 9B) and zinc (fig. 11B) were relatively small in this stream reach. The sampled inflow load of aluminum for this area was about twice the sampled instream load (fig. 10A). This result indicates that either the sampled inflow concentrations at the three inflows upstream from 2,171 m were higher than the concentration of aluminum that actually affected the stream load, or else there was substantial
18 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Table 7. Summary of principal locations of mass loading for Silver Creek, Utah, April 2004. less than] Constituent Area Total Others Loading upstream from the study reach Upper meadow tailings piles Lower meadow tailings piles Upstream from Pivotal Promontory access road Downstream from waste-water treatment plant; Big 4 tailings Sum of all other segments Sum of all segments Stream reach, in meters 0, start of study reach 2,171-2,337 2,337-2,519 2,519-2,637 2,637-2,757 4,403-4,800 4,800-5,251 5,251-5,624 5,624-5,950 5,950-6,093 6,093-6,322 6,332-7,142 8,862-8,909 Load, in kilograms per day Calcium 1,098 4,078 Magnesium Sodium 2,494 Potassium Sulfate 1,320 1,877 1,400 2,019 7,743 Chloride 1,630 1,000 1,190 1,063 6,179 Silicon Aluminum Arsenic Barium Cadmium Chromium Copper Iron Lead Manganese Molybdenum Nickel Strontium Zinc
Summary and Conclusions 19 precipitation of aluminum from the stream before the samples were collected at 2,171 m. The kinetics of aluminum precipitation as hydroxide phases are rapid, and at the relatively high pH of Silver Creek, rapid precipitation is likely (Broshears and others, 1996; Lydersen and others, 1991). Lower Meadow Tailings Piles Six stream segments, from 4,403 m to 6,322 m, represent the lower meadow mass loading (fig. 2C and D). This area was important for loading of several constituents, including calcium, magnesium, sulfate, aluminum, arsenic, chromium, copper, iron, nickel, and zinc. Zinc loading was particularly important, and the sum of the six stream-segment contributions resulted in the largest contribution of zinc along the entire study reach (fig. 11B). Upstream from Pivotal Promontory Access Road A single segment, from 6,332 m to 7,120 m, accounts for a substantial amount of the total mass loading (fig. 2C). This single segment contributed more than 10 percent of the total load for every constituent except aluminum, arsenic, and chromium. The pond upstream from the access road (fig. 2C) area may be a result of ground-water discharge to the stream and merits further study. Waste-Water Treatment Plant and Old Big 4 Mill Tailings Another single segment, from 8,862 m to 8,909 m, is the last principal location of mass loading to the stream (fig. 2D). The single segment that receives discharge from the WWTP also receives inflow from the right bank that drains tailings. This location differs from the other four principal locations of loading because it essentially contributed no cadmium, manganese, lead, nickel, or zinc load (table 7). Individual discharge measurements were not made on these two inflows, but chemical mass balance indicates that the metal loading that did occur came principally from the tailings while major ion loading came from the WWTP. Other Sources The sum of all other stream segments (table 7) indicates the importance of dispersed locations of mass loading. Contributions of metals from other areas of the study reach are substantial for calcium, magnesium, potassium, sulfate, silica, arsenic, cadmium, lead, and zinc. These dispersed metal loadings reflect the widespread occurrence of tailings along the study reach. Tailings are not just localized in the principal locations where loading occurred. Comparison between 2002 and 2004 Comparison of the loads between the 2002 and 2004 studies can help evaluate whether loads from 2004 were high because of the storm and snowmelt runoff. Four sampling points were common between the two studies, and the relation of zinc loads for the two studies is shown in figure 12. Although zinc load in 2004 was initially smaller than zinc load in 2002 both upstream and downstream from Richardson Flat (fig. 12, bars A and B), the 2004 load upstream from the WWTP (bar C) was substantially greater than the load in 2002. Part of the difference is a result of the diversion of flow for this study at 1,492 m. This additional water in the channel could have released the zinc from the streambed or facilitated release of greater loads from the tailings piles. However, even though the 2004 loads are much greater, the pattern of loading that indicates the principal locations of loading is still valid. Summary and Conclusions Detailed mass-loading profiles provide information to facilitate science-based decisions about targets for remediation. The significance of any particular source must be evaluated in the context of its metal loading. The study done on the southern portion of lower Silver Creek in Summit County, Utah, by the U.S. Geological Survey in cooperation with the Utah Department of Environmental Quality, Division of Water Quality, has provided discharge and chemical data to develop mass-loading profiles to indicate the principal locations where historical mill tailings are sources of metal load to the stream. Discharge was estimated by using a bromide tracer injection in three separate injection reaches. Although storms occurred between the injections, causing changes in discharge, the discharge values obtained in the separate injections were adequate to combine for mass-loading profiles. Detailed synoptic sampling provided an indication of the types of inflows affecting Silver Creek and also the major changes in stream chemical character along the study reach. These changes corresponded to the principal locations of metal loading to the stream, including (1) the beginning of the study reach, where an accounting of loading from upstream sources was possible, (2) the upper meadow tailings piles, from ground-water discharge, (3) the lower meadow tailings piles, from ground-water discharge, (4) the stream segment upstream from the Pivotal Promontory access road (6,322 m – 7,142 m), and (5) the stream segment where WWTP and additional ground-water discharge from Old Big 4 tailings occurs (8,862 m – 8,909 m). With loading data these principal sources can be appropriately compared.
20 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 F igure 9. ( A) Va riati on of sulf ate l oad w ith d istanc e al o ng the st udy r each and ( B ) cha nge i n sul fate load for i ndivi dual strea m segm e nts, Silve r Cre ek, U tah, April Figure 9. (A) Variation of sulfate load with distance along the study reach and (B) change in sulfate load for individual stream segments, Silver Creek, Utah, April 2004.Figure 9. (A) Variation of sulfate load with distance along the study reach and (B) change in sulfate load for individual stream segments, Silver Creek, Utah, April 2004. Figure 9. (A) Variation of sulfate load with distance along the study reach and (B) change in sulfate load for individual stream segments, Silver Creek, Utah, April 2004. segments, Silver Creek, Utah, April 2004.
Summary and Conclusions 21 F igure Figure 10. (A) Variation of aluminum load with distance along the study reach and (B) change in aluminum load for individual stream segments, Silver Creek, Utah, April 2004.
22 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 F igure Figure 11. (A) Variation of zinc load with distance along the study reach and (B) change in zinc load for individual stream segments, Silver Creek, Utah, April 2004.
References Cited 23 F g u r e Figure 12. Zinc load at common sampling sites from studies in 2002 and 2004, Silver Creek, Utah. References Cited Bencala, K.E., and McKnight, D.M., 1987, Identifying instream variability: Sampling iron in an acidic stream, in Averett, R.C., and McKnight, D.M., eds., Chemical Quality of Water and the Hydrologic Cycle: Chelsea, Mich., Lewis Publishers, Inc., p. 255-269. Bromfield, C.S., 1989, Gold deposits in the Park City Mining District, Utah: U.S. Geological Survey Bulletin 1857, 26 p. Broshears, R.E., Runkel, R.L., Kimball, B.A., Bencala, K.E., and McKnight, D.M., 1996, Reactive solute transport in an acidic stream: Experimental pH increase and simulation of controls on pH, aluminum, and iron: Environmental Science & Technology, v. 30, no. 10, p. 3016-3024. Friedman, L.C., and Erdmann, D.E., 1982, Quality assurance practices for the chemical and biological analyses of water and fluvial sediments: U.S. Geological Survey Techniques of Water-Resources Investigations, book 5, chap. A6, 181 p. Garmoe, W.J., and Erickson, A.J., Jr., 1968, Ore deposits of the Park City District, in Erickson, A.J., Jr., Phillips, W.R., and Garmoe, W.J., eds., Park City District, Utah: Salt Lake City, Utah, Utah Geological Society, p. 30-39. Giddings, E.M., Hornberger, M.I., and Hadley, H.K., 2001, Trace-metal concentrations in sediment and water and health of aquatic macroinvertebrate communities of streams near Park City, Summit County, Utah: U.S. Geological Sur vey Water-Resources Investigations Report 01-4213, 22 p. Kaufman, L., and Rousseeuw, P.J., 1990, Finding groups in data: An introduction to cluster analysis: New York, Wiley, 368 p. Kilpatrick, F.A., and Cobb, E.D., 1985, Measurement of dis charge using tracers: U.S. Geological Survey Techniques of Water-Resources Investigations, book 3, chap. A16, 27 p. Kimball, B.A., Callender, E., and Axtmann, E.V., 1995, Effects of colloids on metal transport in a river receiv ing acid mine drainage, upper Arkansas River, Colorado, U.S.A.: Applied Geochemistry, v. 10, p. 285-306. Kimball, B.A., Johnson, K.K., Runkel, R.L., and Steiger, J.I., 2004, Quantification of metal loading to Silver Creek through the Silver Maple Claims area, Park City, Utah, May 2002: U.S. Geological Survey Water-Resources Investiga tions Report 03-4296, 40 p. Kimball, B.A., Nordstrom, D.K., Runkel, R.L., Vincent, K.R., and Verplanck, P.L., 2006a, Questa baseline and pre-mining ground-water quality investigation. 23. Quantification of mass loading from mined and unmined areas along the Red River, New Mexico: U.S. Geological Survey Scientific Investigations Report 2006-5004, 53 p. Kimball, B.A., Runkel, R.L., and Walton-Day, K., 2003, Use of field-scale experiments and reactive solute-transport modelling to evaluate remediation alternatives in streams affected by acid mine drainage, in Jambor, J.L., Blowes, D.W., and Ritchie, A.I.M., eds., Environmental aspects of mine wastes: Vancouver, British Columbia, Mineralogical Association of Canada, p. 261-282. Kimball, B.A., Runkel, R.L., Walton-Day, K., and Bencala, K.E., 2002, Assessment of metal loads in watersheds affected by acid mine drainage by using tracer injection and synoptic sampling: Cement Creek, Colorado, USA: Applied Geochemistry, v. 17, no. 9, p. 1183-1207. Kimball, B.A., Runkel, R.L., Walton-Day, K., and Williamson, J.E., 2006b, Quantification of mass loading to Strawberry Creek near the Gilt Edge mine, South Dakota: U.S. Geo logical Survey Scientific Investigations Report 2006-5006, 41 p. Lydersen, E., Salbu, B., Paleo, A.B.S., and Muniz, I.P., 1991, Formation and dissolution kinetics of Al(OH)3 (s) in syn thetic freshwater solutions: Water Resources Research, v. 27, p. 351-357. McKinnon, T.E., 2002, Sources and seasonal variability of metal and arsenic concentrations in the surface water of the Clark Fork River Basin, Montana: University of Montana, Master of Science thesis, 115 p.
24 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Michael Baker Jr., Inc., 2004, Silver Creek Total Maximum Daily Load for dissolved zinc and cadmium: Utah Depart ment of Environmental Quality, Division of Water Quality, accessed April 3, 2005, at ://www.waterquality.utah.gov/ TMDL/Silver_Creek_TMDL.pdf Nichols, D.J., and Bryant, Bruce, 1990, Geologic map of the Salt Lake City 30’ x 60’ quadrangle, north-central Utah, and Uinta County, Wyoming: U.S. Geological Survey IMap 1944, scale 1:100,000. Nordstrom, D.K., and Alpers, C.N., 1999, Geochemistry of acid mine waters, in Plumlee, G.S., and Logsdon, M.J., eds., The environmental geochemistry of mineral deposits Part A: Processes, techniques, and health issues: Littleton, Colo., Society of Economic Geologists, p. 133-160. Rockwell, B.W., Clark, R.N., Livo, E., McDougal, R.R., Kokaly, R.F., and Vance, J.S., 1999, Preliminary materials mapping in the Park City region for the Utah USGS-EPA Imagining Spectroscopy Project using both high- and lowaltitude AVIRIS data, in Green, R.O., ed., Summaries of the 8th annual JPL airborne earth science workshop: JPL Publication, p. 365-375. Smith, K.S., 1999, Metal sorption on mineral surfaces: An overview with examples relating to mineral deposits, in Plumlee, G.S., and Logsdon, M.J., eds., The Environmen tal Geochemistry of Mineral Deposits Part A: Processes, Techniques, and Health Issues: Littleton, Colo., Society of Economic Geologists, p. 161-182. Tibbetts, J.R., Enright, M., and Wilberg, D.E., 2004, Water resources data, Utah, Water year 2003: U.S. Geological Survey Water-Data Report UT-03-1, 458 p. Utah Department of Administrative Services, 2005, Standards of quality for waters of the State: Division of Administrative Rules, accessed April 7, 2005, at ://www.rules.utah.gov/ publicat/code/r317/r317-002.htm Velleman, P.F., and Hoaglin, D.C., 1981, Applications, basics, and computing of exploratory data analysis: Boston, Mass., Duxbury Press, 354 p. Wilkowske, C.D., Allen, D.V., and Phillips, J.V., 2003, Drought conditions in Utah during 1999-2002: A historical perspective: U.S. Geological Survey Fact Sheet 037-03, 6 p.
Table 2 25 Table 2. Bromide concentration of synoptic water samples and characteristics of the sites at which the samples were collected, Silver Creek, Utah, April 2004. [Source: S, stream; LBI, left-bank inflow; RBI, right-bank inflow; Bromide: mg/L, milligrams per liter; Discharge: L/s, liters per second; NC, not calculated; NM, not measured; less than] Sample identifi- cation Distance, (meters) Source Description Northing, (meters) Easting, (meters) Sample date and time Bromide, (mg/L) Dis- charge, (L/s) Upper injection reach SQ1-0000 S T0 Upper - Injection site below U.S. Highway 40 bridge 4/15/04 11:57 SQ1-0061 LBI Discharge with iron staining from willows 4/15/04 11:53 NC SQ1-0101 S Upstream from “upstream tailings” 4/15/04 11:46 SQ1-0250 S Midway along the tailings in the left bank 4/15/04 11:41 SQ1-0428A S T1 Upper - Upstream from Richardson Flat tail ings influence 4/15/04 11:35 SQ1-0428B S T1 Upper - Upstream from Richardson Flat tail ings influence 4/15/04 11:36 SQ1-0525 S Upstream from pond area and bridge 4/15/04 11:25 SQ1-0625 LBI Pace-Homer ditch inflow; left of bridge 4/15/04 11:20 NC SQ1-0681 LBI Small ditch upstream from highway 4/15/04 11:03 NC SQ1-0682 LBI Black pipe spewing orange floc; source un known 4/15/04 10:58 NC SQ1-0731 S Downstream end of left, smaller culvert at highway 4/15/04 13:16 NC SQ1-0757 LBI Ditch downstream from highway 4/15/04 10:31 NC SQ1-0770 LBI Draining ditch on downstream side of highway 4/15/04 10:01 NC SQ1-0861 S Right channel - downstream end of larger culvert at highway 4/15/04 10:15 NC SQ1-1050 1,050 RBI Right channel - ditch from area of Richardson Flat 4/15/04 10:08 NC SQ1-1095 1,095 S Upstream end of culvert under rail trail 4/15/04 9:55 NC SQ1-1148 1,148 RBI Right channel - second ditch from area of Rich ardson Flat? 4/15/04 9:52 NC SQ1-1229 1,229 S Right channel - downstream from small pond in channel 4/15/04 9:41 NC SQ1-1235 1,235 RBI Channel draining meadow area 4/15/04 9:49 NC SQ1-1300 1,300 S Upstream from return of irr ditch 4/15/04 9:18 NC SQ1-1309 1,309 RBI T2 Upper - Right channel - returning ditch 4/15/04 13:25 NC SQ1-1371A 1,371 S At old flume in stream 4/15/04 9:13 SQ1-1371B 1,371 S At old flume in stream 4/15/04 9:14 SQ1-1452 1,452 S T3 Upper - At diversion to wetland 4/15/04 9:07 SQ1-1744 1,744 S Irrigation ditch blw culvert near wetland; 2002 sample site 4/15/04 9:26 NM NM Middle injection reach SQ2-1601 1,601 S T0 Middle - Injection site downstream from wetland culvert 4/9/04 11:55 SQ2-1843B 1,843 S T1 Middle - At fence at end of wetland 4/9/04 11:50 SQ2-1843C 1,843 S T1 Middle - At fence at end of wetland 4/9/04 11:51 SQ2-1843A 1,843 S T1 Middle - At fence at end of wetland 4/9/04 8:58 SQ2-1959 1,959 S Upstream from tailings inflow - questioned chemistry 4/9/04 11:57 SQ2-1965 1,965 RBI Pond at end of long talings pile 4/9/04 11:10 NC SQ2-2048 2,048 RBI Location of several inflows 4/9/04 11:04 NC SQ2-2118 2,118 RBI Homer Spring inflow to irrigation ditch; no input to stream 4/9/04 11:08 NM SQ2-2171 2,171 S Downstream from area of right bank inflows 4/9/04 12:00 SQ2-2337 2,337 S After braids have come back together 4/9/04 12:07 SQ2-2387 2,387 LBI Near tailings piles on right bank 4/9/04 10:50 NC
26 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Table 2. Bromide conentration of synoptic water samples and characteristics of the sites at which the samples were collected, Silver Creek, Utah, April 2004—Continued. Sample identifi- cation Distance, (meters) Source Description Northing, (meters) Easting, (meters) Sample date and time Bromide, (mg/L) Dis- charge, (L/s) Middle injection reach—Continued SQ2-2431 2,447 LBI Drains large area with tailings off to left 4/9/04 10:45 NC SQ2-2560 2,519 S Between upstream left bank inflows and down stream right bank inflows 4/9/04 12:13 SQ2-2569 2,528 RBI Drains from tailings pile 4/9/04 10:42 NC SQ2-2678 2,637 S Downstream from tailings inflow; to collect inflows 4/9/04 12:20 SQ2-2718 2,677 RBI Pond from tailings drainage 4/9/04 10:36 NC SQ2-2730 2,757 S At fence below property corner 4/9/04 12:24 SQ2-2785 2,847 S Downstream from where stream cuts through corner of property 4/9/04 12:33 SQ2-2780 2,892 RBI Direct drainage from tailings pile with Ulothrix 4/9/04 10:30 NC SQ2-2810 2,927 S At old skull in stream 4/9/04 12:28 SQ2-3027 3,144 LBI Drains flat area; no tailings piles visible 4/9/04 10:19 NC SQ2-3045 3,162 RBI Draining from tailings piles 4/9/04 10:17 NC SQ2-3254B 3,371 S T2 Middle - Upstream from old tree 4/9/04 12:41 SQ2-3254A 3,371 S T2 Middle - Upstream from old tree 4/9/04 21:54 SQ2-3379 3,496 S Downstream from area where stream is ponded 4/9/04 12:46 SQ2-3598 3,715 RBI Small pool on right bank; sample puddle 4/9/04 10:04 NC SQ2-3602 3,719 LBI Drains tailings to left of stream 4/9/04 10:01 NC SQ2-3784A 3,901 S Upstream from point where flow disperses; made a new diversion to right 4/9/04 12:55 SQ2-3784B 3,901 S Upstream from point where flow disperses; made a new diversion to right 4/9/04 12:56 SQ2-4000 4,117 RBI Inflow from natural channel; ditch from left of rail trail; water 4/9/04 9:52 NC SQ2-4050 4,167 S Location to check with discharge measurement and Br 4/9/04 13:04 SQ2-4286 4,403 S After gathering back together into channel; could be irrigation ditch 4/9/04 13:10 SQ2-4292 4,409 LBI Draining area where stream dispersed 4/9/04 9:44 NC SQ2-0054 4,517 LBI Draining wide area to left of stream 4/9/04 9:40 NC SQ2-0061 4,800 S Downstream from gathered dispersion 4/9/04 13:19 SQ2-0080 5,251 S Downstream from area where stream is ponded 4/9/04 13:32 SQ2-0096 5,493 RBI Drainage has some flow to stream; tailings in soil to right 4/9/04 9:23 NC SQ2-0100 5,624 S Downstream from possible tailings inflow 4/9/04 13:39 SQ2-0108 5,833 RBI Orange stained inflow 4/9/04 9:15 NC SQ2-0109 5,843 RBI Draining tailings 4/9/04 10:15 NC SQ2-0149 5,878 RBI Sample away from stream; water not draining to stream 4/9/04 9:12 NC SQ2-0113 5,950 S Upstream from many tailings mounds 4/9/04 13:50 SQ2-0120 6,045 RBI Draining tailings, maybe from storm, orange plume 4/9/04 9:05 SQ2-0122 6,093 S To account for inflows and separate tailings below 4/9/04 13:55 SQ2-0135 6,322 S T3 Middle - Upstream from pond above Prom ontory Road 4/9/04 14:04 SQ2-0137 6,353 RBI Orange inflow; farther right 4/9/04 8:55 NC SQ2-0005 7,259 S End of middle injection reach 4/9/04 14:12
Table 2 27 Table 2. Bromide conentration of synoptic water samples and characteristics of the sites at which the samples were collected, Silver Creek, Utah, April 2004—Continued. Sample identifi- cation Distance, (meters) Source Description Northing, (meters) Easting, (meters) Sample date and time Bromide, (mg/L) Dis- charge, (L/s) Lower injection reach SQ3-005 7,142 S T0 Lower - Injection site downstream from Promontory culvert 4/6/04 12:20 SQ3-008 7,161 S First site downstream from injection for dis charge 4/6/04 12:22 SQ3-010 7,185 RBI Draining tailings toward old “Big 4” mill site 4/6/04 11:35 NC SQ3-012 7,208 S T1 Lower - Downstream from first tailings inflow 4/6/04 12:30 SQ3-018 7,276 S To capture right bank inflows 4/6/04 12:35 SQ3-024 7,365 RBI Ponds along berm line to east 4/6/04 11:30 NC SQ3-025 7,366 LBI Draing from pond toward BFI Disposal land 4/6/04 11:25 NC SQ3-032 7,397 S To capture both inflow upstream 4/6/04 12:40 SQ3-039 7,470 S Upstream from inflow from marsh draining along fence 4/6/04 12:45 SQ3-042 7,491 LBI Draining from marsh area along much of BFI land 4/6/04 11:20 NC SQ3-048 7,571 S Downstream from inflow along fence 4/6/04 12:50 SQ3-056 7,687 S Downstream from area where stream is ponded 4/6/04 12:55 SQ3-060 7,730 LBI Draining tailings toward old mill site; pool away from stream 4/6/04 11:10 NC SQ3-066 7,825 S Near right bank talings in flood plain 4/6/04 13:00 SQ3-083 8,009 RBI Small, unconnected pools along ditch 4/6/04 11:05 SQ3-097 8,225 S Downstream from tailings inflows on both sides of stream 4/6/04 13:10 SQ3-115 8,449 RBI Pond on right bank away from stream 4/6/04 10:50 NC SQ3-121 8,497 LBI Draining in small grassy channel 4/6/04 10:45 NC SQ3-127 8,591 S Gathering of the upstream inflows 4/5/04 13:15 SQ3-131 8,701 LBI Pond by waste-water treatment plant 4/6/04 10:38 NC SQ3-140 8,862 S T2 Lower - Upstream from waste-water treat ment plant inflow 4/6/04 13:22 SQ3-141 8,881 LBI Discharge from waste-water treatment plant 4/6/04 10:22 NC SQ3-142 8,886 RBI Drains area to right including pond 4/6/04 10:11 NC SQ3-145 8,909 S Stream below gage and waste-water treatment plant inflow 4/5/04 13:34 SQ3-172 9,355 S Upstream from irrigation return flow 4/6/04 13:45 SQ3-173 9,360 RBI Return flow from irrigation ditch, through dairy farm 4/6/04 9:54 NC SQ3-178A 9,438 S Downstream from irrigation return flow 4/6/04 13:48 SQ3-178B 9,438 S Downstream from irrigation return flow 4/6/04 13:50 SQ3-186 9,562 LBI Discharge from stream on left 4/6/04 9:48 NC SQ3-189 9,598 LBI Seep inflow of very high conductance 4/6/04 9:45 NC SQ3-193 9,719 S Dowstream from high conductance seeps 4/6/04 13:58 SQ3-194 9,725 RBI Draining dairy farm 4/6/04 9:34 NC SQ3-196 9,747 S T3 Lower - Downstream from bridge to dairy 4/6/04 14:00
Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Table 3. Concentration of major ions in synoptic water samples collected along Silver Creek, Utah, April 2004. [Source: S, stream; LBI, left-bank inflow; RBI, right-bank inflow; Filter: FA, filtered acidified; RA, unfiltered acidified; Temperature: °C, degrees Celsius; pH, in standard units; mg/L, milligrams per liter; NV, no value; less than; NR, not recorded] Sample identification Distance (meters) Source Filter Temperature pH Calcium (mg/L) Magnesium (mg/L) Sodium (mg/L) Potassium (mg/L) Alkalinity as CaCO3 (mg/L) Sulfate (mg/L) Chloride (mg/L) Silica as Si (mg/L) SQ1-0000 S FA SQ1-0061 LBI FA SQ1-0101 S FA SQ1-0250 S FA SQ1-0428A S FA SQ1-0428B S FA SQ1-0525 S FA SQ1-0625 LBI FA SQ1-0681 LBI FA SQ1-0682 LBI FA SQ1-0731 S FA SQ1-0757 LBI FA SQ1-0770 LBI FA 1,880 SQ1-0861 S FA SQ1-1050 1,050 RBI FA NV SQ1-1095 1,095 S FA SQ1-1148 1,148 RBI FA SQ1-1229 1,229 S FA SQ1-1235 1,235 RBI FA SQ1-1300 1,300 S FA SQ1-1309 1,309 RBI FA SQ1-1371A 1,371 S FA SQ1-1371B 1,371 S FA SQ1-1452 1,452 S FA SQ2-1601 1,601 S FA NR SQ1-1744 1,744 S FA SQ2-1843A 1,843 S FA NR SQ2-1843B 1,843 S FA NR SQ2-1843C 1,843 S FA NR SQ2-1959 1,959 S FA NR SQ2-1965 1,965 RBI FA NR .5 3,510 SQ2-2048 2,048 RBI FA NR SQ2-2118 2,118 RBI FA NR SQ2-2171 2,171 S FA NR SQ2-2337 2,337 S FA NR SQ2-2431 2,447 LBI FA NR SQ2-2560 2,519 S FA NR SQ2-2569 2,528 RBI FA NR SQ2-2678 2,637 S FA NR SQ2-2718 2,677 RBI FA NR SQ2-2730 2,757 S FA NR SQ2-2785 2,847 S FA NR SQ2-2780 2,892 RBI FA NR SQ2-2810 2,927 S FA NR SQ2-3027 3,144 LBI FA NR SQ2-3045 3,162 RBI FA NR SQ2-3254A 3,371 S FA NR SQ2-3254B 3,371 S FA NR SQ2-3379 3,496 S FA NR SQ2-3598 3,715 RBI FA NR 1,300 SQ2-3602 3,719 LBI FA NR SQ2-3784A 3,901 S FA NR SQ2-3784B 3,901 S FA NR
Table 3 Table 3. Concentration of major ions in synoptic water samples collected along Silver Creek, Utah, April 2004—Continued. Sample identification Distance (meters) Source Filter Temperature pH Calcium (mg/L) Magnesium (mg/L) Sodium (mg/L) Potassium (mg/L) Alkalinity as CaCO3 (mg/L) Sulfate (mg/L) Chloride (mg/L) Silica as Si (mg/L) SQ2-4000 4,117 RBI FA NR SQ2-4050 4,167 S FA NR SQ2-4286 4,403 S FA NR SQ2-4292 4,409 LBI FA NR SQ2-0054 4,517 LBI FA NR SQ2-0061 4,800 S FA NR SQ2-0080 5,251 S FA NR SQ2-0096 5,493 RBI FA NR SQ2-0100 5,624 S FA NR SQ2-0108 5,833 RBI FA NR SQ2-0109 5,843 RBI FA NR 1,040 SQ2-0149 5,878 RBI FA NR 3,250 SQ2-0113 5,950 S FA NR SQ2-0120 6,045 RBI FA NR .1 3,680 SQ2-0122 6,093 S FA NR SQ2-0135 6,322 S FA NR SQ2-0137 6,353 RBI FA NR SQ2-0005 7,120 S FA NR SQ3-005 7,142 S FA SQ3-008 7,161 S FA SQ3-008 7,161 S RA SQ3-010 7,185 RBI FA 1,120 SQ3-010 7,185 RBI RA 1,120 SQ3-012 7,208 S FA SQ3-018 7,276 S FA SQ3-024 7,365 RBI FA 1,050 SQ3-025 7,366 LBI FA 1,610 SQ3-032 7,397 S FA SQ3-032 7,397 S RA SQ3-039 7,470 S FA SQ3-042 7,491 LBI FA SQ3-048 7,571 S FA SQ3-056 7,687 S FA SQ3-060 7,730 LBI FA 1,480 SQ3-060 7,730 LBI RA 1,480 SQ3-066 7,825 S FA SQ3-083 8,009 RBI FA 1,450 SQ3-097 8,225 S FA SQ3-115 8,449 RBI FA SQ3-121 8,497 LBI FA SQ3-127 8,591 S FA SQ3-127 8,591 S RA SQ3-131 8,701 LBI FA SQ3-140 8,862 S FA SQ3-141 8,881 LBI FA SQ3-142 8,886 RBI FA SQ3-145 8,909 S FA SQ3-145 8,909 S RA SQ3-172 9,355 S FA SQ3-173 9,360 RBI FA SQ3-178A 9,438 S FA SQ3-178A 9,438 S RA SQ3-178B 9,438 S FA SQ3-178B 9,438 S RA SQ3-186 9,562 LBI FA
30 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Table 3. Concentration of major ions in synoptic water samples collected along Silver Creek, Utah, April 2004—Continued. Sample identification Distance (meters) Source Filter Temperature pH Calcium (mg/L) Magnesium (mg/L) Sodium (mg/L) Potassium (mg/L) Alkalinity as CaCO3 (mg/L) Sulfate (mg/L) Chloride (mg/L) Silica as Si (mg/L) SQ3-189 9,598 LBI FA 1,400 SQ3-193 9,719 S FA SQ3-193 9,719 S RA SQ3-194 9,725 RBI FA SQ3-196 9,747 S FA SQ3-196 9,747 S RA
Table 4 31 Table 4. Concentration of trace elements in synoptic water samples collected along Silver Creek, Utah, April 2004. [Distance, in meters along the study reach; Source: S, stream; LBI, left-bank inflow; RBI, right-bank inflow; Filter: FA, 0.45-micrometer filtration; RA, unfiltered acidified; chemical concentrations reported in micrograms per liter; less than] Sample identification Dis tance Source Fil ter Alumi num Arse nic Bari um Cad mium Co balt Chro mium Cop per Iron Lead Manga nese Molyb denum Nickel Sil ver Stron tium Ura nium Vana dium Zinc Upper injection reach SQ1-0000 S FA 1,300 SQ1-0061 LBI FA 1,260 8,800 SQ1-0101 S FA 1,340 SQ1-0250 S FA 1,410 SQ1-0428A S FA 1,620 SQ1-0428B S FA 1,680 SQ1-0525 S FA 1,620 SQ1-0625 LBI FA 1,690 SQ1-0681 LBI FA 1,300 SQ1-0682 LBI FA 1,390 1,920 1,630 5,110 SQ1-0731 S FA 1,510 SQ1-0757 LBI FA 1,390 SQ1-0770 LBI FA 2,710 SQ1-0861 S FA SQ1-1050 1,050 RBI FA 5 1,100 SQ1-1095 1,095 S FA 1,560 SQ1-1148 1,148 RBI FA 2,160 SQ1-1229 1,229 S FA 1,010 SQ1-1235 1,235 RBI FA 1,310 25,500 SQ1-1300 1,300 S FA 1,950 SQ1-1309 1,309 RBI FA SQ1-1371A 1,371 S FA 1,560 SQ1-1371B 1,371 S FA 1,570 SQ1-1452 1,452 S FA 1,690 SQ2-1601 1,601 S FA 2,380 SQ1-1744 1,744 S FA 1,150 1,250 7,120 Middle injection reach SQ2-1843A 1,843 S FA 2,440 SQ2-1843B 1,843 S FA 2,380 SQ2-1843C 1,843 S FA 2,380 SQ2-1959 1,959 S FA .586 3.71 2,470 SQ2-1965 1,965 RBI FA 37,000 34,400 23,600 1,620 270,000 SQ2-2048 2,048 S FA 4,350 SQ2-2118 2,118 RBI FA SQ2-2171 2,171 S FA 3,350 SQ2-2337 2,337 S FA 3,380 SQ2-2431 2,447 LBI FA 1,020 1,590 SQ2-2560 2,519 S FA 3,360 SQ2-2569 2,528 RBI FA 1,370 11,300 SQ2-2678 2,637 S FA 3,470 SQ2-2718 2,677 RBI FA 1,270 6,110
32 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004 Table 4. Concentration of trace elements in synoptic water samples collected along Silver Creek, Utah, April 2004—Continued. Sample identification Dis tance Source Fil ter Alumi num Arse nic Bari um Cad mium Co balt Chro mium Cop per Iron Lead Manga nese Molyb denum Nickel Sil ver Stron tium Ura nium Vana dium Zinc SQ2-2730 2,757 S FA 3,580 Middle injection reach—Continued SQ2-2785 2,847 S FA 3,590 SQ2-2780 2,892 RBI FA 4,690 SQ2-2810 2,927 S FA 3,740 SQ2-3027 3,144 LBI FA 3,320 SQ2-3045 3,162 RBI FA 4,840 SQ2-3254A 3,371 S FA 4,230 SQ2-3254B 3,371 S FA 3,840 SQ2-3379 3,496 S FA 3,960 SQ2-3598 3,715 RBI FA 3,540 2,030 24,400 SQ2-3602 3,719 LBI FA 1,280 11,300 SQ2-3784A 3,901 S FA 4,150 SQ2-3784B 3,901 S FA 4,200 SQ2-4000 4,117 RBI FA 4,840 SQ2-4050 4,167 S FA 4,390 SQ2-4286 4,403 S FA 4,250 SQ2-4292 4,409 LBI FA 1,280 8,400 SQ2-0054 4,517 LBI FA 1,710 2,600 SQ2-0061 4,800 S FA 4,250 SQ2-0080 5,251 S FA 1,040 3,450 SQ2-0096 5,493 RBI FA 1,950 1,510 8,380 SQ2-0100 5,624 S FA 1,130 4,460 SQ2-0108 5,833 RBI FA 1,100 3,440 SQ2-0109 5,843 RBI FA 2,010 1,540 53,400 SQ2-0149 5,878 RBI FA 1,910 10,900 1,320 1,070,000 SQ2-0113 5,950 S FA 1,130 5,140 SQ2-0120 6,045 RBI FA 9,520 63,000 9,860 2,340 132,000 SQ2-0122 6,093 S FA 1,150 5,600 SQ2-0135 6,322 S FA 1,150 6,110 SQ2-0137 6,353 RBI FA 4,080 1,340 45,300 SQ2-0005 7,120 S FA 1,180 6,890 Lower injection reach SQ3-005 7,142 S FA 1,250 6,800 SQ3-008 7,161 S FA 1,280 6,800 SQ3-008 7,161 S RA 1,220 7,130 SQ3-010 7,185 RBI FA 1,980 2,100 30,100 SQ3-010 7,185 RBI RA 1,960 1,940 30,000 SQ3-012 7,208 S FA 1,330 6,730 SQ3-018 7,276 S FA 1,250 6,890 SQ3-024 7,365 RBI FA 2,200 1,370 26,300 SQ3-025 7,366 LBI FA 7.63 15.1 1,040 2,630 44,800 SQ3-032 7,397 S FA 1,300 7,370 SQ3-032 7,397 S RA 1,190 7,510
Table 4 33 Table 4. Concentration of trace elements in synoptic water samples collected along Silver Creek, Utah, April 2004—Continued. Sample identification Dis tance Source Fil ter Alumi num Arse nic Bari um Cad mium Co balt Chro mium Cop per Iron Lead Manga nese Molyb denum Nickel Sil ver Stron tium Ura nium Vana dium Zinc SQ3-039 7,470 S FA 1,320 7,930 Lower injection reach—Continued SQ3-042 7,491 LBI FA 1,400 8,740 SQ3-048 7,571 S FA 1,320 7,880 SQ3-056 7,687 S FA 1,280 8,420 SQ3-060 7,730 LBI FA 6,200 2,480 117,000 SQ3-060 7,730 LBI RA 6,200 2,370 117,000 SQ3-066 7,825 S FA 1,380 8,420 SQ3-083 8,009 RBI FA 3,970 1,780 25,200 SQ3-097 8,225 S FA 1,340 8,420 SQ3-115 8,449 RBI FA .236 .01 9,150 SQ3-121 8,497 LBI FA 1,540 1,800 46,000 SQ3-127 8,591 S FA 1,310 9,280 SQ3-127 8,591 S RA 1,240 9,310 SQ3-131 8,701 LBI FA 3,050 14,800 SQ3-140 8,862 S FA 1,380 9,040 SQ3-141 8,881 LBI FA SQ3-142 8,886 RBI FA 1,230 SQ3-145 8,909 S FA 1,090 4,370 SQ3-145 8,909 S RA 1,040 5,390 SQ3-172 9,355 S FA 1,120 5,100 SQ3-173 9,360 RBI FA 2,060 SQ3-178A 9,438 S FA 1,200 5,890 SQ3-178A 9,438 S RA 1,070 5,910 SQ3-178B 9,438 S FA 1,090 4,800 SQ3-178B 9,438 S RA 1,360 10,600 SQ3-186 9,562 LBI FA SQ3-189 9,598 LBI FA 4,760 2,760 SQ3-193 9,719 S FA 1,000 4,270 SQ3-193 9,719 S RA 5,260 SQ3-194 9,725 RBI FA 4,390 SQ3-196 9,747 S FA 1,100 3,900 SQ3-196 9,747 S RA 5,070
34 Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004
B.A. Kimball and others—Principal Locations of Metal Loading from Flood-Plain Tailings, Lower Silver Creek, Utah, April 2004—SIR 2007-5248