Surface-water quality and suspended-sediment quantity and quality within the Big River Basin, southeastern Missouri, 2011-13
<p>Missouri was the leading producer of lead in the United States—as well as the world—for more than a century. One of the lead sources is known…
Public-domain full text preserved in the Mountain Man Mining Library. Original source: pubs.usgs.gov.
Prepared in cooperation with the U.S. Environmental Protection Agency, Region 7 Surface-Water Quality and Suspended-Sediment Quantity and Quality within the Big River Basin, Southeastern Missouri, 2011–13 Scientific Investigations Report 2015–5171 U.S. Department of the Interior U.S. Geological Survey
Front cover photographs: Left, baseflow sample collection by wading. Center, turbidity sensor equipped with a wiper to reduce sensor fouling. Right, stormflow event sampling from bridge. Back cover photographs: Top, streamflow measuring during a large stormflow event at the upstream site. Center, turbidity sensor equipped with a wiper to reduce sensor fouling. Bottom, sediment sample processing at the Missouri Water Science Center Sediment Laboratory. Background photograph: Large stormflow event during study period at the Byrnesville downstream site.
Surface-Water Quality and SuspendedSediment Quantity and Quality within the Big River Basin, Southeastern Missouri, 2011–13 By Miya N. Barr Prepared in cooperation with the U.S. Environmental Protection Agency, Region 7 Scientific Investigations Report 2015–5171 U.S. Department of the Interior U.S. Geological Survey
U.S. Department of the Interior SALLY JEWELL, Secretary U.S. Geological Survey Suzette M. Kimball, Director U.S. Geological Survey, Reston, Virginia: 2016 For more information on the USGS—the Federal source for science about the Earth, its natural and living resources, natural hazards, and the environment—visit ://www.usgs.gov or call 1–888–ASK–USGS. For an overview of USGS information products, including maps, imagery, and publications, visit ://www.usgs.gov/pubprod/. 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 information product, for the most part, is in the public domain, it also may contain copyrighted materials as noted in the text. Permission to reproduce copyrighted items must be secured from the copyright owner. Suggested citation: Barr, M.N., 2016, Surface-water quality and suspended-sediment quantity and quality within the Big River Basin, southeastern Missouri, 2011–13: U.S. Geological Survey Scientific Investigations Report 2015–5171, 39 p., ://dx.doi.org/10.3133/sir20155171. ISSN 2328-0328 (online)
Contents Abstract 1 Introduction 2 Study Background 2 Purpose and Scope 4 Description of Study Area 4 Methods 5 Field Methods 5 Laboratory Methods 7 Quality Assurance and Quality Control 7 Data Analysis and Reporting 9 Surface-Water Quality 13 Streamflow Conditions 13 Continuous Water Quality 14 Suspended-Sediment Quantity 19 Daily Suspended-Sediment Concentrations and Loads 22 Event-Based Suspended-Sediment Concentrations, Load, and Yields 24 Suspended-Sediment Quality 25 Selected Trace-Element Concentrations in Suspended Sediments 25 Event-Based Loads and Yields of Selected Trace-Element Concentrations in Suspended Sediments 31 Summary 36 References Cited 38 Figures
1. Map showing location of study sites and the Big River Basin 3
2. Graphs showing turbidity duration curves for two study sites on the Big River, Missouri, October 2011–September 2013 9
3. Graphs showing instantaneous turbidity in relation to discrete suspended-sediment concentrations used to develop a regression model for the computation of daily suspended-sediment concentrations and loads for two study sites on the Big River, October 2011–September 2013 11
4. Graphs showing daily mean streamflow in relation to discrete suspended-sediment concentrations used to develop a regression model for the computation of estimated daily suspended-sediment concentrations and loads for two study sites on the Big River, Missouri, October 2011–September 2013 12
5. Graph showing total monthly precipitation at two National Oceanic and Atmospheric Administration climate stations near the Big River Basin, Missouri, October 2011–September 2013 14
6. Graph showing daily mean streamflow computed during the study period at two study sites on the Big River, Missouri, October 2011–September 2013 15
7. Graphs showing daily mean water temperature measured at two study sites on the Big River, Missouri, October 2011–September 2013 17
8. Graphs showing daily mean turbidity measured at two study sites on the Big River, Missouri, October 2011–September 2013 18
9. Graphs showing continuous streamflow and discrete suspended-sediment concentration samples collected at two study sites on the Big River, Missouri, October 2011–September 2013 19
10. Graphs showing daily mean suspended-sediment concentrations computed at two study sites on the Big River, Missouri, October 2011–September 2013 22
11. Graphs showing daily mean suspended-sediment loads computed at two study sites on the Big River, Missouri, October 2011–September 2013 23
12. Graphs showing trace-element concentrations in two suspended-sediment size fractions collected during stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013 28
13. Graphs showing seasonal comparison of stormflow-event suspended-sediment concentrations and mass-accumulation lead and zinc concentrations within suspended sediments for two study sites on the Big River, Missouri, October 2011–September 2013 31
14. Graphs showing seasonal comparison of stormflow-event loads and yields of lead and zinc in suspended sediments at two study sites on the Big River, Missouri, October 2011–September 2013 35 Tables
1. Location information for study sites on the Big River, Missouri 5
2. Reporting limits of laboratory analyses 8
3. Summary statistics of model-calibration datasets for two study sites on the Big River, October 2011–September 2013 10
4. Streamflow statistics for two study sites on the Big River, Missouri, October 2011–September 2013 15
5. Water-quality statistics for two study sites on the Big River, Missouri, October 2011–September 2013 16
6. Concentrations and size distributions of suspended-sediment samples collected at two study sites on the Big River, Missouri, October 2011–September 2013 20
7. Daily suspended-sediment concentration and load statistics at two study sites on the Big River, Missouri, October 2011–September 2013 24
8. Suspended-sediment concentrations, loads, and yields from sampled stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013 25
9. Selected trace-element concentrations for two particle size distributions of suspended sediments collected during stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013 26
10. Streamflow, suspended-sediment concentrations and particle-size distributions, and selected total trace-element concentrations in suspended sediments collected during stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013 30
11. Selected trace element concentrations, loads, and yields in suspended sediments from sampled stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013 32
Conversion Factors [Inch/Pound to International System of Units] Multiply By To obtain Length inch (in.) centimeter (cm) inch (in.) millimeter (mm) foot (ft) meter (m) mile (mi) kilometer (km) Area acre 4,047 square meter (m2) square foot (ft2) square centimeter (cm2) square foot (ft2) square meter (m2) square inch (in2) square centimeter (cm2) square mile (mi2) square kilometer (km2) Volume cubic foot (ft3) cubic meter (m3) gallon (gal) liter (L) Flow rate foot per second (ft/s) meter per second (m/s) cubic foot per second (ft3/s) cubic meter per second (m3/s) Mass ounce, avoirdupois (oz) gram (g) pound, avoirdupois (lb) kilogram (kg) ton, short (2,000 lb) metric ton (t) ton per day (ton/d) metric ton per day ton per day (ton/d) megagram per day (Mg/d) short ton per day per square mile megagram per day per square kilometer [(Mg/d)/km2] ton per year (ton/yr) megagram per year (Mg/yr) ton per year (ton/yr) metric ton per year Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as follows: °F=(1.8×°C)+32 Temperature in degrees Fahrenheit (°F) may be converted to degrees Celsius (°C) as follows: °C=(°F-32)/1.8 Datum Vertical coordinate information is referenced to North American Vertical Datum of 1988 (NAVD 88).
Supplemental Information Concentrations of chemical constituents in water are given either in milligrams per liter (mg/L) or micrograms per liter (µg/L). Concentrations of chemical constituents in sediments are given in milligrams per kilogram (mg/kg). Water year in U.S. Geological Survey reports is the 12-month period October 1 through September 30. The water year is designated by the calendar year in which it ends and which includes 9 of the 12 months. Thus, the year ending September 30, 2013 is called the “2013 water year.” Abbreviations ADAPS automated data processing system ADCP acoustic Doppler current profiler ASTM Association of Standards and Testing Materials CWQM continuous water-quality monitor DCP data collection platform EPA U.S. Environmental Protection Agency EWI equal-width increment FISP Federal Interagency Sedimentation Project FNU formazin nephelometric units ICP-MS inductively-coupled plasma mass-spectrometry MDNR Missouri Department of Natural Resources MRL Minerals Research Laboratory NIST National Institute of Standards and Technology NOAA National Oceanographic and Atmospheric Association NWIS National Water Information System PEC probable effect concentration PVC polyvinyl chloride R 2 coefficient of determination R 2 a adjusted coefficient of determination SQG sediment quality guidelines SSC suspended-sediment concentration SSL suspended-sediment load TEC threshold effect concentration TET toxic effect threshold USFWS U.S. Fish and Wildlife Service USGS U.S. Geological Survey YSI Yellow Springs, Incorporated
Surface-Water Quality and Suspended-Sediment Quantity and Quality within the Big River Basin, Southeastern Missouri, 2011–13 By Miya N. Barr Abstract Missouri was the leading producer of lead in the United States—as well as the world—for more than a century. One of the lead sources is known as the Old Lead Belt, located in southeast Missouri. The primary ore mineral in the region is galena, which can be found both in surface deposits and underground as deep as 200 feet. More than 8.5 million tons of lead were produced from the Old Lead Belt before opera tions ceased in 1972. Although active lead mining has ended, the effects of mining activities still remain in the form of large mine waste piles on the landscape typically near tributar ies and the main stem of the Big River, which drains the Old Lead Belt. Six large mine waste piles encompassing more than 2,800 acres, exist within the Big River Basin. These six mine waste piles have been an available source of trace element-rich suspended sediments transported by natural erosional pro cesses downstream into the Big River. A study was performed by the U.S. Geological Survey in cooperation with U.S. Environmental Protection Agency, Region 7, to calculate and characterize suspended-sediment quantity and quality within the Big River basin after reclama tion of the mine waste piles ended in 2012. Streamflow and suspended sediments were quantified and sampled at two locations along a 68-mile reach of the Big River between Bonne Terre and Byrnes Mill, Missouri. The results will help regulatory agencies, such as the U.S. Environmen tal Protection Agency and U.S. Fish and Wildlife Service, determine impaired reaches and ecosystems for remedial and restoration efforts. Continuous stream stage, water temperature, and tur bidity, and discrete suspended-sediment concentration data were collected at the two sites between October 2011 and September 2013. Suspended-sediment samples were collected during various hydrologic conditions to develop a regression model between discrete suspended-sediment concentration and continuous turbidity. Suspended sediments collected during stormflow events were analyzed for concentrations of trace elements such as barium, cadmium, lead, and zinc within two sediment size fractions. Event loads and annual loads of suspended sediment and select trace elements in suspended sediments also were calculated. Suspended-sediment loads computed by the regression model increased downstream from about 201,000 tons at the upstream site to about 355,000 tons at the downstream site during the study period. Stormflow-event-based (hereinafter referred to as “event-based”) suspended-sediment loads ranged from 180 to 32,000 tons at the upstream sampling site and 390 to 53,000 tons at the downstream site along the Big River. Although only seven stormflow events at the upstream site and six at the downstream site were sampled, the event-based suspended-sediment loads accounted for nearly 30 percent of the total suspended-sediment loads computed at both sites, indicating most of the suspended sediment transported through the Big River occurs during higher streamflows. Sediment quality guidelines, known as the threshold effect concentration and the probable effect concentration, used to assess toxicity of trace-element concentrations in sediments were compared to the cadmium, lead, and zinc concentrations in suspended sediment samples collected during stormflow events. All concentrations of cadmium, lead, and zinc in event-based suspended sediment samples exceeded the threshold and probable effect concentrations. Lead and zinc concentrations in the sediment size fraction less than 0.063 millimeters also exceeded the toxic effect threshold, above which sediment is considered to be heavily polluted causing adverse effects on sediment-dwelling organ isms. Concentrations of cadmium and zinc in event-based suspended sediment samples were notably higher in samples from the upstream site than samples from the downstream site, indicating the sources of sediments enriched in these trace elements decrease in the downstream area of the watershed. The reduction in concentration of cadmium and zinc could be from dissolution of the constituents during transport or possibly a decrease in downstream source material. The lead concentration exceedance of the probable effects concentra tion as well as the threshold effects concentration indicates that lead-rich suspended sediments in the fraction less than 0.063 millimeters are readily available within the Big River Basin for transport. These sediments remain in the system from historical mining, and as the reclamation of mine waste piles in the upstream area of the watershed reduce additional sediment loadings, these fine sediments may be continually
2 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 released as the river scours the streambed and erodes stream banks causing the lead-rich suspended sediment to remain in a state of equilibrium. Barium concentrations in suspended-sediments were nearly twice as high in stormflow event samples collected at the downstream site as compared to samples from the upstream site. The source of barium in the Big River could be from Mineral Fork and Mill Creek, which flow through the historical barite (barium sulfate, also known as tiff) mining district in Washington County, and discharge into the Big River between the two study sites. Total trace-element loads and yields in suspended sedi ments were computed from the sampled events for each year in the study. The total barium loads in suspended sediments were higher for sampled events collected at the downstream site than the upstream site during both study years. Cadmium and zinc loads in suspended sediments were lower at the downstream site than the upstream site, although the decrease in total load was not substantial during the study period. Lead loads in suspended sediments were lower at the downstream site during the first study year, with a slightly higher load downstream in the second year though the increase from upstream to downstream was small. Event-based yields were higher at the upstream site, indicating that readily available sediment sources are closer to the upstream site where more mining affected areas are located. The estimates determined during large precipitation events indicate that large sources of suspended sediments with large concentrations of trace ele ments are still available for transport within the Big River. Introduction Missouri was the leading producer of lead in the world for more than a century (Missouri Department of Natural Resources, 2014). An important source of lead was the Old Lead Belt, a sub district of the Southeast Missouri Lead District, located primarily in St. Francois County in south eastern Missouri (fig. 1). Galena was the primary ore mineral and could be found on surface outcrops and as deposits that extended nearly 200 feet vertically and thousands of feet later ally. Lead mining began in southeastern Missouri in the early 1700s, at which time mines were mostly shallow pits operated by French explorers, and continued as individual small opera tions until the mid-1860s (Missouri Department of Natural Resources, 2014). In 1864, the St. Joseph Lead Company acquired 964 acres and began mining at Bonne Terre, Missouri (fig. 1). With the implementation of diamond-bit core drilling, lead deposits deep beneath the surface were discovered under much of the Big River Basin. Fifteen companies had mining operations in the Old Lead Belt by the early 1900s. Mining operations were gradually shut down during the late 1950s and early 1960s as ore deposits were depleted and mining in other parts of the State was more productive. More than 8.5 million tons of lead were produced from the Old Lead Belt before the St. Joseph Lead Company closed operations in 1972 (Missouri Department of Natural Resources, 2014). Although active lead mining ended more than 40 years ago, the effects of mining still remain within the region in the form of mining waste (tailings and chat) piles on the land surface. These piles are generally located near tributaries and the main stem of the Big River and are readily available sources of lead-rich suspended sediments that can enter into the watershed during runoff and by wind transport. The Big River is the main riverine system that drains the Old Lead Belt. The Big River runs generally south to north and is 145 miles in length from its source to the conflu ence with the Meramec River (fig. 1). Data collection during previous U.S. Geological Survey (USGS) investigations, in cooperation with Missouri Department of Natural Resources (MDNR), indicated that streambed sediments collected in the Meramec River downstream from the Big River have higher lead concentrations than those collected upstream from the Big River confluence (U.S. Geological Survey, 2015). These results indicate that lead-rich suspended sediments have been transported from the Big River in the past, but limited infor mation is currently (2015) available to determine the amount of suspended sediments associated with mining wastes that are readily available for transport by the Big River through fluvial processes and whether these suspended sediments still contain mining-related metals. Additional information was needed to quantify daily sediment loads and the concentrations of lead and other metals of concern during stormflow events within the Big River Basin. Such information can assist regulatory agencies such as the U.S. Environmental Protection Agency (EPA) and the U.S. Fish and Wildlife Service (USFWS) in determining impaired reaches and ecosystems for remedial and restoration efforts. A study was performed by the U.S. Geolog ical Survey in cooperation with U.S. Environmental Protection Agency, Region 7, to assess the amount and availability of suspended sediments within the Big River Basin after chat pile capping efforts ended in 2012 that completed a 4-year recla mation effort, and to assess the trace-element concentrations of suspended sediments transported through the basin. Study Background Although mining activities are not currently (2015) being conducted in the Big River Basin, six large chat piles consist ing of approximately 2,800 acres (Mosby and Weber, 2009) remain on the land surface upstream from the study reach that is located between two USGS streamgages (fig. 1). These large amounts of mine waste, sometimes spanning more than a mile in diameter, began with the introduction of industrial-grade mining and milling methods in the early 1900s. Materials excavated from below the land surface were crushed or ground using jig tables in the earlier periods, and after the 1920s chemical flotation techniques were used to separate lead from the host rock. The leftover host rock was discarded into chat (larger pieces from crushing) or tailings (smaller material from
Introduction 3 Figure 1. Location of study sites and the Big River Basin. JEFFERSON FRANKLIN CRAWFORD IRON MADISON ST LOUIS WASHINGTON Byrnes Mill Cedar Hill Hillsboro Potosi Bonne Terre Desloge Leadwood Irondale Bonne Terre Leadwood Desloge National Elvins Federal Federal MISSOURI ILLINOIS Big River Big River Mineral Fork Meramec River Mill Creek Cedar Creek Flat Creek Mississippi River Meramec River # # 38°30’ 38° 90°30’ 91° 20 MILES 20 KILOMETERS STUDY AREA MISSOURI St. Louis # Mine waste pile and name U.S. Geological Survey streamflow- gaging station site number Big River Basin Old Lead Belt mining district EXPLANATION Base from U.S. Geological Survey digital data 1:24,000, 1999 Universal Transverse Mercator Projection Zone 15 Byrnesville ST FRANCOIS STE GENEVIEVE
4 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 the flotation process) piles. Neither method was completely effective at separating ore from host rock. Some operations would re-mill existing chat piles to recover lead missed from basic crushing methods (Missouri Department of Natural Resources, 2014). These piles have resulted in a source of sus pended sediments available for transport by natural erosional processes such as storm event runoff and wind into the Big River. Between 2008 and 2012, the EPA conducted reclama tion efforts to cap and prevent further erosion of the piles (Mosby and Weber, 2009). Previous USGS studies have determined surface-water quality, and physical and chemical processes related to the quality of water, trace-element concentrations of suspendedsediment and streambed-sediment particles of various sizes, and loads of select trace elements in dissolved and solid phases (Smith and Schumacher, 1991, 1993); however, limited information is available on the current (2015) amount of suspended sediments transported through the Big River as well as the concentration of trace elements in suspended sedi ments. As capping efforts have been made to restrict additional sediment sources, it is important to determine the amount of suspended sediments and the trace-element concentrations of these suspended sediments that are still available for transport through the basin as well as to assess the effectiveness of the reclamation efforts over time. Sediments are fragments of parent materials that have been affected by erosional processes and are transported by, suspended in, or deposited by water or air, and also can accumulate in beds (Edwards and Glysson, 1999). The USGS defines sediments transported by water as “fluvial sediments.” Fluvial sediments are fragmented material derived from weathering of larger rocks and transported by, suspended in, or deposited by water, and can include biological and chemical precipitates and decomposed organic materials (Edwards and Glysson, 1999). Finer-grained fluvial sediments are trans ported in suspension because of the velocity and currents of the stream, whereas coarser sediments are transported closer to the streambed by rolling and skipping along the streambed. The finer, suspended particles tend to be transported at the same rate that the stream flows, whereas the coarser materi als are only transported when the system experiences higher velocities and generally are at rest most of the time (Edwards and Glysson, 1999). The supply of finer materials in suspen sion (also known as the “wash load”) usually has a greater effect on the suspended-sediment concentration (SSC) than streamflow conditions, because the rate of supply varies during and between events as well as from seasonal precipita tion and vegetation (Charlton, 2008). Increases in streamflow produce increased SSC as erosion increases, releasing finer particles from storage. The computation of daily SSC and suspended-sediment load (SSL) is useful in various applica tions such as describing variability in suspended-sediment conditions, evaluating water-resource management practices and goals, predicting reservoir capacities, evaluating waterquality criteria, understanding stream channel morphology, and comparing sediment characteristics between basins (Ras mussen and others, 2009). Turbidity is a qualitative parameter defined by an optical measurement of scattered and absorbed light as it interacts with solid particles through a fluid sample, and measurements are expressed in units based on the technology used as well as the calibration standards (ASTM International, 2011). Sus pended and dissolved organic and inorganic materials such as sediments (sands, silts, and clays), algae, microorganisms, organic acids, and dyes are common causes of turbidity in fluvial systems. Turbidity can assist with the computation of time-series SSC, monitoring land use and other human-related activities, and natural resource restoration (Anderson, 2005). Trace element concentrations are related to sediments in streams by a number of factors including grain size, surface area, and sediment composition (Horowitz and Elrick, 1987). As the proportion of grain size increases, the trace element concentrations associated with the sediments also increase. The surface area of the grains also affects the concentration of trace elements as a larger surface area allows for stronger bonding, most commonly in sediment size fractions less than 0.063 millimeters (mm). When an abundance of very finegrained sediments are present, the particles can attach together creating aggregates, which increase the mean grain size of a sample and reduces available surface area. The aggregates of sediments less than 0.063 mm can affect trace element con centrations in a sample more than larger size fractions of sands (Horowitz and Elrick, 1987). Purpose and Scope The purpose of this report is to present the results of a hydrologic investigation to characterize and calculate suspended-sediment quantity and quality being transported through the lower Big River, Mo., based on data collected October 2011 through September 2013. The report describes the techniques and methods used to collect and analyze waterquality constituents such as water temperature and turbidity; suspended-sediment concentration, load, and particle-size distribution; and selected trace-element concentration load in sediments at two sites on the main stem of the Big River dur ing stormflow events. Description of Study Area The area of study is a 68-mile reach of the Big River, between the towns of Bonne Terre and Byrnes Mill, Mo. (fig. 1; table 1). Two sites were used in this study: Big River below Bonne Terre, Mo. (USGS site number 07017610; hereinafter referred to as the upstream site) and Big River at Byrnesville, Mo. (USGS site number 07018500; hereinaf ter referred to as the downstream site). The upstream site is located less than 5 miles downstream from the Old Lead Belt, and the downstream site is located about 68 miles downstream
Methods 5 from Bonne Terre. The Big River discharges into the Meramec River, approximately 15 miles downstream from the down stream site. The Meramec River discharges into the Missis sippi River south of St. Louis, Mo. The study area is located within the Salem Plateau of the Ozark Plateaus physiographic province (Fenneman, 1938). The topography of the upper reaches of the study area is rugged with narrow, steep drainage divides and several hundred feet of relief. The lower reaches also have some steep drainage that gradually transition to large flood plains. Land-surface altitudes range from about 400 to 1,000 feet (ft) above North American Vertical Datum of 1988 (NAVD 88). Stream width is generally 50 ft in the upper reaches of the basin and can span nearly 100 ft in the lower reaches. The streambed consists mainly of coarse materials such as cobbles, gravels, and sands in the upper reaches and smaller-grained particles such as sands and silts in the lower reaches. Land use in the Big River Basin is approximately 72 percent forest, 18 percent grassland, 7 percent urbanized or developed, and 1 percent cropland (Missouri Department of Natural Resources, 2013). Some low-head mill dams are still present, which can alter and control the streamflow. Tributar ies along the Big River include Cedar Creek, Flat Creek, Mill Creek, and Mineral Fork. Mineral Fork, the largest tributary with a basin area of 189 square miles (mi2; Missouri Depart ment of Natural Resources, 2013), discharges into the Big River between the two study sites. No streamflow-gaging station (hereinafter referred to as streamgage) is located on Mineral Fork or the other tributaries in the basin. Methods In order to quantify the transport of sediment and the con centration and flux of selected metals in the Big River, stream flow and suspended sediments were quantified and sampled at the two study sites. All field and laboratory methods described in this report were consistently performed at both sites during the study period. Any variations have been documented and described fully within this study report. Field Methods The upstream site was established for this study and began operation on October 13, 2011. The site is located at the bridge on County Highway E, approximately 3 miles north of Bonne Terre (fig 1; table 1). The drainage area of the upstream site is 409 mi2. The pressure transducer orifice line and continuous water-quality monitor (CWQM) are deployed from the left stream bank (facing downstream) under the highway bridge. During base-flow conditions, streamflow measurements and suspended-sediment samples were obtained approximately 40 ft upstream from the streamgage. Measure ments and samples during high-flow conditions (hereinafter referred to as “event sampling”) were collected from the bridge deck on Highway E. The stream channel at this site location is relatively shallow and narrow (2 to 3 ft deep and 60 ft or less wide) and the streambed consists of fine sands, gravels, large cobbles, and some boulders. The downstream site has been an active streamgage from May 1922 to present (2015); is located on the left edge of the water on a privately owned bridge near Old Byrnesmill Road in Byrnesville, Mo.; and has a drainage area of 917 mi2 (fig 1; table 1). The CWQM was deployed from the left side of the stream along a rock bluff just under the bridge, next to the ori fice line. The streamgage is approximately 100 ft downstream from an old low-head mill dam. During base-flow conditions, streamflow measurements and suspended-sediment samples were obtained by wading approximately 300 ft downstream from the streamgage, which is downstream from the bridge on Old Byrnesmill Road. If wading conditions were not safe, measurements and samples were collected from the bridge on Old Byrnesmill Road, which was the location of all event sampling. The stream channel at the wading section is usu ally 3 to 5 ft deep and 80 to 100 ft wide, and the streambed consists primarily of gravels and coarse sands with some finer sands and silts. Study sites were equipped with a data collection plat form (DCP) that stored information from the non-submersible pressure transducer used for measuring stream stage and from Table 1. Location information for study sites on the Big River, Missouri. [USGS, U.S. Geological Survey; mi2, square miles; in., inches] USGS site number (fig. 1) Station name Study location description Latitude (degrees/ minutes/ seconds) Longitude (degrees/ minutes/ seconds) Drainage area (mi2) County Average monthly precipitation1 (in.) Period of stream flow record used in this study Big River below Bonne Terre, Missouri upstream 37°57′55.9″ 90°34′27.9″ St. Francois October 2011– September 2013 Big River at Byrnesville, Missouri downstream 38°23′30.2″ 90°38’16.1″ Jefferson May 1922– September 2013 1Average monthly precipitation was computed using only monthly precipitation data within the study period (October 2011 through September 2013).
6 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 the CWQM used to measure water temperature and turbidity. Stream stage, water temperature, and turbidity were recorded by the DCP every 15 minutes and transmitted hourly by way of satellite telemetry. All water-quality data were archived along with streamflow data in the USGS National Water Infor mation System (NWIS; U.S. Geological Survey, 2015). The CWQMs were deployed using polyvinyl chloride (PVC) pipe. The PVC pipe had multiple 1-inch (in.) holes drilled to allow water from the stream to flow freely around the instrument, reducing bias in data from readings of stagnant water trapped in the pipe. Instantaneous streamflow measurements were made every 4 to 6 weeks at the two sites and during event sampling to determine and maintain a stage-discharge relationship using standard USGS methods and techniques (Rantz and others, 1982; Sauer, 2002; Oberg and others, 2005; Turnipseed and Sauer, 2010; Mueller and others, 2013). Instantaneous mea surements were made with various acoustic doppler current profilers (ADCP) by wading during low-flow conditions and from the bridge decks during high flows or if the wading sec tions could not be accessed safely. Water temperature was recorded in degrees Celsius (°C) and turbidity was reported in formazin nephelometric units (FNU). Thermistors used for water temperature measurements were checked against a National Institute of Standards and Technology (NIST) calibrated thermometer and received at minimum a 3-point calibration check as documented in Wilde (2006). Many different sensors and models of turbidometers are available for field measurements. For this study, a YSI Incorporated sensor model 6136 (YSI 6136) was used to record instantaneous turbidity readings at both sites as well as the field meter to reduce any bias between sensor makes or models. The advantage of the YSI 6136 is that erroneous read ings are reduced because the sensor is equipped with a wiper. For the purpose of this study, only suspended sediments were collected. Suspended-sediment loads described in this study are considered to represent the majority of sediments transported in suspension within the Big River and are referred to as SSL. No bedload sampling was performed; therefore, any loads computed from the results are not intended to be consid ered a total sediment load. Cross-sectional (discrete) suspended-sediment samples were collected during stormflow events and almost every month during base-flow conditions. Uses of the suspendedsediment samples were two-fold—to determine a relation with turbidity data to compute a daily SSC (that is, calibrate the regression model), and to measure trace-element concentra tions of the suspended sediments being transported during a stormflow event. Stormflow events were defined in this study as a rapid and substantial change in streamflow and turbidity. The magnitude and duration of stormflow events varied and collection efforts were not always successful or possible. The monthly suspended-sediment collections at base flow were collected by wading in the stream and using a depth-inte grated and isokinetic (stream water approaching and entering the sampling nozzle have the same velocity) sampler known as a US DH-81 sampler. The US DH-81 contained a rigid, poly ethylene 1-liter (L) sampling bottle and one-quarter in. Teflon® nozzle. The equal-width increment (EWI) sampling method was used to obtain the sediment sample as described in Wilde and others (2004). The sampler was raised and lowered at a consistent transit rate at each sampling interval (vertical). If multiple 1-L bottles were used to collect at all verticals, the bottles were composited by the laboratory for analysis. If the minimum mean stream velocity was less than 1.5 feet per second (ft/s), isokinetic conditions no longer existed, and baseflow samples were collected using grab sampling methods as described in Wilde and others (2004). If stream conditions were not conducive to wading because of depth, debris, high velocities, or other conditions that made wading dangerous, suspended-sediment samples were collected from the bridge deck at each site. The sam plers were attached to a reel and cable mechanism on either a 3-wheeled base or a crane structure mounted to the front of a field vehicle. The reel was hand-operated at a constant speed based on the transit rate computed by the maximum veloc ity and depth. A one-quarter in. nozzle was used to collect all particles classified as sand-sized and smaller. Collection was performed using the same EWI techniques used for base-flow sampling except with a heavier sampler to maintain isokinetic sampling requirements (Wilde and others, 2004). Depending on the average stream depth and velocities during stormflow events, either a US DH-95 or US DH-2 sampler was used. Descriptions of each sampler and the limitations of each are described in Wilde and others (2014). Samples were collected at multiple points along the hydrograph, including the ris ing limb, the event peak, and the falling limb, when possible. Streamflow measurements also were collected as close to the peak as possible. The suspended-sediment concentration samples along the hydrograph were analyzed individually to compute the concentration flux throughout the event. Samples to be analyzed for trace-element concentrations were collected using the clean hands/dirty hands techniques as described in U.S. Geological Survey (2006). Samplers used by the USGS, as recommended by the Federal Interagency Sedimentation Project (FISP) for the col lection of suspended sediments (Davis, 2005), cover a wide range of sampling capacities and conditions and have a limita tion on the depth within the water column at which sampling can occur, based on the nozzle size and location in relation to the bottom of the sampler. The portion of depth near the streambed that cannot be reached by the nozzle is called the unsampled zone and can carry a higher concentration and coarser-sized sediment, which may or may not account for a large portion of the total suspended sediment, depending on stream velocity, depth, and turbulence through the sampled vertical (Edwards and Glysson, 1999). The concentration obtained within the measured depth is nearly equal to the concentration in the unsampled zone, as noted in Edwards and Glysson (1999), if the velocity and turbulence within the sampled depth are efficiently keeping sediments sus pended within the total depth and are greater than the forces
Methods 7 transporting sediments along the streambed in the unsampled zone. The USGS samplers used in this study had unsampled zones of 4 in. for the DH-81 and DH-2 and 4.8 in. for the DH-95 (Davis, 2005; Wilde and others, 2014). Laboratory Methods All suspended-sediment samples were processed at the USGS Missouri Water Science Center Sediment Laboratory in Rolla, Mo. Samples were delivered to the lab within 5 days of collection. Base-flow samples were analyzed for SSC in mil ligrams per liter using a filtration method as described in Guy (1969). All SSC results are available on NWISWeb at :// waterdata.usgs.gov/mo/nwis/qw. Event samples were analyzed for SSC as well as par ticle size distribution of the sands (sediments greater than 0.063 mm) and fines (sediments less than 0.063 mm) fractions. Samples collected during events were sieved using tech niques described in Guy (1969). Sieve mesh with openings of approximately 0.0625 mm and made of nylon fibers was used to reduce possible trace-element contamination from using traditional brass or stainless steel mesh sieves. All sediments in the sample were washed through the mesh sieve using deionized water. Fines which passed through the sieve were captured in a glass dish. All sediments remaining on the sieve mesh were rinsed into a separate glass dish, then both fractions were dried at 80 °C until all visible water was evaporated, followed by additional drying at 103 °C for one hour (Guy, 1969). If the mass for either sieved fraction was greater than or equal to 0.25 grams (g), the fraction could be analyzed for trace-element concentrations at the USGS Minerals Research Laboratory (MRL) in Denver, Colorado. Dried fractions for trace-element analyses were shipped to the USGS MRL in glass vials. A suite of 42 trace elements was measured by inductively-coupled plasma mass-spec trometry (ICP-MS) using an acid digestion using documented laboratory methods (Taggart, 2002). Trace elements analyzed for this study as well as sediment analyses and the reporting limits are listed in table 2. All trace-element concentrations for each size fraction were stored in NWIS and are available at ://waterdata.usgs.gov/mo/nwis/qw. Quality Assurance and Quality Control Quality assurance consists of techniques and practices used within a study to meet defined levels of quality with a known level of confidence to ensure the most accurate data possible. Such practices begin with site and equipment selec tion, sampling and maintenance frequencies and methods, personnel training and safety, and laboratory selection, all of which ensure the goals of the study are met. A USGS internal quality-assurance plan, summarized in this section, was cre ated and used to document techniques and methods specific to the study with USGS guidelines. Data transmissions to NWISWeb were reviewed daily to remove erroneous data quickly and efficiently from the sensor record. Documentation of service visits and the calibration of the CWQM were archived and describe actions taken during the visit. Sediment sampling documentation also was archived and describes sampling conditions, sample type, sampling methods, equip ment used, and other information as needed. Laboratory analysis request forms were used for both the sediment lab and the USGS MRL to document and track the samples and the analyses. Data transmitted from the DCP at each site were reviewed daily for consistency and for determining the need for event sampling. CWQMs were serviced and calibrated following USGS guidelines as described in Wagner and others (2006). Construction setup and monitor selection was deter mined during reconnaissance before the start of the project to reduce effects of fouling, low-flow conditions, and dam age from flooding or vandalism, and to guarantee safety in accessing the equipment. CWQMs were serviced on at least a monthly schedule and additionally when the data appeared erratic or anomalous. The monitors were cleaned, inspected for damage, and checked for calibration drift. If the calibra tion drift was greater than USGS criteria, the sensors were recalibrated. If the fouling or calibration drift was greater than USGS criteria as stated in Wagner and others (2006), the data were removed from the record. Efforts were made to minimize contamination from sampling equipment, surrounding structures, and vehicles dur ing event sampling. The clean hands/dirty hands technique as described in U.S. Geological Survey (2006) was used through out the sampling procedures at a site. Containers for sample compositing were kept sealed and protected in a large plastic bag to prevent contamination by airborne debris caused by wind, vehicle traffic, the sampling crane, bridge railings, and precipitation. Quality-control samples help identify and quantify bias and variability in sampling techniques such as sampling equipment, processing, shipping, and handling of the sample (U.S. Geological Survey, 2006). During the study, qualitycontrol samples were collected in the form of replicate sus pended-sediment samples and laboratory blanks. A sequential replicate sample (collected in the same order after the envi ronmental sample and composited into a second container) was performed when possible. Blank samples using deionized water processed by the USGS Missouri Water Science Center Sediment Laboratory were randomly assigned to sample ship ments during the login phase for internal laboratory validation and quality assurance. USGS Sediment Laboratories partici pate in bi-annual quality-assurance tests to document inconsis tencies within each lab and among all USGS labs for consis tency in reporting results. Quality-assurance test results for the USGS Missouri Water Science Center Sediment Laboratory are available at the USGS Branch of Quality Systems at :// bqs.usgs.gov/SLQA/. The USGS MRL created laboratory split replicates for 10 percent of a shipment to validate laboratory results for internal quality assurance and are available upon request from the USGS MRL at ://minerals.cr.usgs.gov/.
8 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Table 2. Reporting limits of laboratory analyses. [%, percent; USGS SMRL, U.S. Geological Survey Minerals Research Laboratory; mg/L, milligrams per liter; mm, millimeters; USGS SDMO, U.S. Geological Survey Missouri Sediment Laboratory; --, not available] Constituent Reporting limit Analyzing laboratory Analytical reference Aluminum, Al 0.01% USGS SMRL Taggart, 2002 Calcium, Ca 0.01% USGS SMRL Taggart, 2002 Iron, Fe 0.01% USGS SMRL Taggart, 2002 Potassium, K 0.01% USGS SMRL Taggart, 2002 Magnesium, Mg 0.01% USGS SMRL Taggart, 2002 Sodium, Na 0.01% USGS SMRL Taggart, 2002 Phosphorous, P 50 mg/L USGS SMRL Taggart, 2002 Titanium, Ti 0.01% USGS SMRL Taggart, 2002 Silver, Ag 1 mg/L USGS SMRL Taggart, 2002 Arsenic, As 1 mg/L USGS SMRL Taggart, 2002 Barium, Ba 5 mg/L USGS SMRL Taggart, 2002 Beryllium, Be 0.1 mg/L USGS SMRL Taggart, 2002 Bismuth, Bi 0.04 mg/L USGS SMRL Taggart, 2002 Cadmium, Cd 0.1 mg/L USGS SMRL Taggart, 2002 Cerium, Ce 0.05 mg/L USGS SMRL Taggart, 2002 Cobalt, Co 0.1 mg/L USGS SMRL Taggart, 2002 Chromium, Cr 1 mg/L USGS SMRL Taggart, 2002 Cesium, Cs 0.05 mg/L USGS SMRL Taggart, 2002 Copper, Cu 0.5 mg/L USGS SMRL Taggart, 2002 Gallium, Ga 0.05 mg/L USGS SMRL Taggart, 2002 Indium, In 0.02 mg/L USGS SMRL Taggart, 2002 Lanthanum, La 0.5 mg/L USGS SMRL Taggart, 2002 Lithium, Li 1 mg/L USGS SMRL Taggart, 2002 Manganese, Mn 5 mg/L USGS SMRL Taggart, 2002 Molybdenum, Mo 0.05 mg/L USGS SMRL Taggart, 2002 Niobium, Nb 0.1 mg/L USGS SMRL Taggart, 2002 Nickel, Ni 0.5 mg/L USGS SMRL Taggart, 2002 Lead, Pb 0.5 mg/L USGS SMRL Taggart, 2002 Rubidium, Rb 0.2 mg/L USGS SMRL Taggart, 2002 Sulfur, S 0.01% USGS SMRL Taggart, 2002 Antimony, Sb 0.05 mg/L USGS SMRL Taggart, 2002 Scandium, Sc 0.1 mg/L USGS SMRL Taggart, 2002 Tin, Sn 0.1 mg/L USGS SMRL Taggart, 2002 Strontium, Sr 0.5 mg/L USGS SMRL Taggart, 2002 Tellurium, Te 0.1 mg/L USGS SMRL Taggart, 2002 Thallium, Tl 0.1 mg/L USGS SMRL Taggart, 2002 Thorium, Th 0.2 mg/L USGS SMRL Taggart, 2002 Uranium, U 0.1 mg/L USGS SMRL Taggart, 2002 Vanadium, V 1.0 mg/L USGS SMRL Taggart, 2002 Tungsten, W 0.1 mg/L USGS SMRL Taggart, 2002 Yttrium, Y 0.1 mg/L USGS SMRL Taggart, 2002 Zinc, Zn 1 mg/L USGS SMRL Taggart, 2002 Suspended-sediment concentration (SSC) 0.5 mg/L USGS SDMO Guy, 1969 Suspended sediment, percent finer than 0.063 mm 0% USGS SDMO Guy, 1969
Methods 9 Data Analysis and Reporting Streamflow and water-quality constituent statistics were derived from daily mean, monthly mean, or annual mean values. The mean values were derived from the 15-minute data. Annual runoff was derived by dividing the annual mean streamflow in cubic feet per second by the drainage area in square miles, then multiplying by a conversion factor of 13.5744 to obtain the result in inches per water year. Monthly precipitation accumulations were obtained for weather stations near the study sites from the National Oceanographic and Atmospheric Administration (NOAA) and are cumulative pre cipitation measurements converted from millimeters to inches for consistency of units within the report. Daily values of continuous water temperature and tur bidity were published in the “Annual Water Year Summary Reports,” which can be accessed through NWISWeb, along with other daily, monthly, and annual statistics (U.S. Geologi cal Survey, 2015). Periods of missing data occurred because of extreme biological fouling such as algae and macroinverte brates, siltation after large events, or extreme low-flow condi tions. Turbidity values were used in a regression model for the computation of daily SSC. Because the magnitude of turbidity in fluvial systems is usually proportional to SSC (Rasmus sen and others, 2009), continuous turbidity measurements also could be used to determine if a stormflow event was large enough to transport the minimum amount of suspended sediments required for the sediment chemistry analyses at the USGS MRL (0.25 grams). A regression model for the computation of time-series SSC and SSL values was developed for each site using a dataset of discrete SSC and corresponding time-series turbid ity and (or) streamflow. The dataset was used to calibrate the model and identify outliers and trends. Outliers were reviewed for erroneous entries, sampling method problems, or labora tory errors. For robust model development, the number of data pairs as well as the distribution of data over observed hydrologic conditions during the study is important (Rasmus sen and others, 2009). The monthly base-flow sampling in combination with the multiple event values over a range of streamflows and turbidity measurements allowed for greater confidence in the model development for each site. Dura tion curves (fig. 2) show turbidity measurements at both sites were within the limits of the maximum level of accuracy (1,000 FNU) of the sensor model for all days in the study period except one. During the 2012 water year (a water year is defined as a 12-month period beginning October 1 and ending September 30, designated by the calendar year in which it ends), 27 dis crete SSC samples were collected and available for the model development at the upstream site. Non-transformed values of SSC were plotted against turbidity to determine if a statisti cally strong relationship existed (Rasmussen and others, 2009). Four outliers were detected by graphical inspection of the raw data and residual plots. The outliers occurred during storm events: two event SSC results collected on April 16, 2012; one sample on April 17, 2012; and one sample on Sep tember 8, 2012. The instantaneous turbidity values likely were biased from extreme fouling during the events rather than sediment sampling errors because when the CWQM could be accessed safely afterwards, the guard and deployment pipe were noted to be filled with gravel and mud. The four outliers were removed from the dataset. The remaining 23 data pairs were then plotted on a log-log scale to determine the strongest statistical relationship. The best-fit relation was determined to be from the power function, having an adjusted coefficient of determination (R2 a) value of 0.92 and a level of significance (p-value) of 0.0043. During the 2013 water year, 13 discrete SSC results were added to the 2012 dataset to continue the regression model for the entire study period. No outliers were found in the 2013 water year dataset. The total dataset used in the study model for the upstream site included 36 pairs of data (table 3; fig. 3). Figure 2. Turbidity duration curves for two study sites on the Big River, Missouri, October 2011–September 2013. 1,000 10,000 1,000 10,000 Turbidity, in formazin nephelometric units Big River below Bonne Terre, Missouri (07017610; upstream site) Big River at Byrnesville, Missouri (07018500; downstream site) Percent of exceedance YSI, Inc. 6136 turbidity sensor's maximum for accuracy YSI, Inc. 6136 turbidity sensor's maximum for accuracy
10 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 The final regression model for the upstream site, Big River below Bonne Terre (USGS Identifier 07017610), for the study period is as follows:
SSC 1.8239(Turb)0.984 (1) where
SSC is suspended-sediment concentration, in milligrams per liter; and
Turb is turbidity, in formazin nephelometric units, measured with a YSI model 6136. The model information used to develop equation 1 is as follows: Number of measurements 36 Residual standard error 0.239 Degrees of freedom 33 Adjusted coefficient of determination (R2 a) 0.92 The downstream site had 19 SSC samples available for the development of the 2012 water year model. As was deter mined upstream, the SSC and turbidity data plotted in log-log scale made the best-fit relation at the downstream site using the power function, with an R2 a value of 0.89 and a p-value of 0.0198. During model evaluation, three outliers were detected within the 2012 dataset—a routine base-flow collection on November 10, 2011; an event collection on March 17, 2012; and a routine base-flow collection on April 4, 2012. The SSC result for the November sampling date seemed to be biased high in relation to the turbidity results. The SSC sampling process may have inadvertently struck the sandy streambed, creating a re-suspension of bed materials. The March sample was during an event and could have been biased high during SSC collection or a possible turbidity sensor issue. The April outlier could have been because of a low-biased SSC col lection (sampling rate too fast for the velocity) or erroneous turbidity readings. These three data points did not fit the gen eral trend of the remaining points and were therefore removed from the dataset. Paired SSC and instantaneous turbidity data collected at the downstream site during water year 2013 were added to the 2012 dataset to continue the regression model for the entire study period. Twelve data points were available from 2013; however, during model calibration and evaluation, three outliers were detected: January 31, 2013; April 20, 2013; and May 28, 2013. The January and April sampling dates had reasonable SSC results but because of extreme flooding and fouling of the turbidity sensor, no turbidity data were available Table 3. Summary statistics of model-calibration datasets for two study sites on the Big River, October 2011–September 2013. [FNU, formazin nephelometric units; mg/L, milligrams per liter; ft3/s, cubic feet per second; --, not applicable] Statistical summary of datasets Summary statistic Turbidity (FNU) Model dataset Time-series dataset (17,240 hourly values) Suspended-sediment concentration (mg/L) Streamflow (ft3/s) Turbidity (FNU) Streamflow (ft3/s) Big River below Bonne Terre, Missouri (07017610; upstream site) Minimum Maximum 1,090 18,600 21,900 Mean 2,840 Median Standard deviation 5,530 1,510 Number of unavailable turbidity values 2,214 Number of data used in model Big River at Byrnesville, Missouri (07018500; downstream site) Minimum Maximum 24,600 28,280 Mean 2,023 Median Standard deviation 4,939 2,430 Number of unavailable turbidity values 1,139 Number of data used in model
Methods 11 for comparison. The May sample had erroneous turbidity data because of extreme biofouling around the deployment pipe, and data were removed from the record. The total dataset used in the study model for the downstream site included 25 pairs of data (table 3; fig. 3). The model calibration remained statis tically strong with the dataset of the full study period (seven outliers still removed), using the non-transformed SSC and turbidity data. The final regression model for the downstream site, Big River at Byrnesville (USGS Identifier 07018500) for the study period is as follows:
SSC 1.786(Turb)1.050 (2) where
SSC is suspended-sediment concentration, in milligrams per liter; and
Turb is turbidity, in formazin nephelometric units, measured with a YSI model 6136. The model information used to develop equation 1 is as follows: Number of measurements 25 Residual standard error 0.236 Degrees of freedom 23 Adjusted coefficient of determination (R2 a) 0.89 The regressions were applied to each instantaneous turbidity measurement available during the study period to compute time-series SSCs. The instantaneous turbidity measurements were stored in the NWIS subsystem ADAPS (automated data processing system), where mean daily SSC values also were computed and archived. The mean daily SSC values were then used to compute daily SSL values, also archived in ADAPS. If no turbidity data were available or if the turbidity data were deemed unusable for sediment computation, a daily mean SSC value was computed using streamflow. The relationship was determined using similar techniques as SSCturbidity relationship analyses. Daily SSC results from the SSC-turbidity relationship were plotted against daily stream flow for the period of record available. Data were transformed until a best-fit relationship was determined. A best fit for both sites during the study was determined by plotting SSC and daily streamflow on a log-log scale (fig. 4). The SSCstreamflow relationship was not as statistically accurate as the SSC-turbidity relationship, and all daily SSC values computed using streamflow were flagged as estimated. SSL values com puted from estimated SSC data also were flagged as estimated. The estimated daily SSC equation for the upstream site, Big River below Bonne Terre (USGS identifier 07017610), is as follows:
SSC 0.3341(Q)0.8148 (3) where
SSC is suspended-sediment concentration, in milligrams per liter; and
Q is daily mean streamflow, in cubic feet per second. Figure 3. Instantaneous turbidity in relation to discrete suspended-sediment concentrations used to develop a regression model for the computation of daily suspended-sediment concentrations and loads for two study sites on the Big River, October 2011–September 2013. 1,000 5,000 1,000 5,000 Discrete suspended-sediment concentration, in milligrams per liter Instantaneous turbidity, in formazin nephelometric units y 1.8239(x)0.984 R 2 a 0.92 n 36 Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) y 1.786(x)1.050 R 2 a 0.89 n 25 1,000 5,000 1,000 5,000 EXPLANATION regression equation R 2 a 0.89 n 25 y 1.786(x)1.050 R 2 a adjusted coefficient of determination n number of datasets where: y suspended-sediment concentration and x instantaneous turbidity Data pair used in model Best-fit regression model
12 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 The estimated daily SSC equation for the downstream site, Big River at Byrnesville (USGS identifier 07018500) is as follows:
SSC =0.1668(Q)0.8469 (4) where
SSC is suspended-sediment concentration, in milligrams per liter; and
Q is daily mean streamflow, in cubic feet per second. Event-based SSLs were computed using the event-aver age SSC and the sampled runoff volume. The average SSC was the arithmetic average determined from all samples col lected during the event. In some cases, only one sample could be collected because of timing of arrival, equipment issues, magnitude of event and rapid rises, and other various reasons. The sample runoff volume was the sum of the incremental streamflow in 15-minute intervals within the designated event duration divided by the event duration time. Using the stream flow data, the beginning and ending of each event was deter mined based on each event’s increase or decrease in stream flow over time. The beginning of the event was established at the 15-minute streamflow value recorded one hour before the change in streamflow doubled. The end of the event varied by site and event because each event was unique in hydrologic conditions, but was determined when the streamflow was decreasing at a constant rate of 20 cubic feet per second (ft3/s) during each 15-minute measurement for at least 2 hours. The time between the established start and ending of the event was considered the event duration in hours. Event-based SSL was computed using the following equation:
ESSL EMSSC x RV x (6.245 x 10-5) (5) where
ESSL is event-based suspended-sediment load, in pounds;
EMSSC is event-based suspended-sediment concentration, in milligrams per liter;
RV is runoff volume, in cubic feet; and 6.245 x 10-5 is a conversion from milligrams per liter to pounds per cubic foot. Sediment yield is the amount of material removed from the land surface by erosion in a given unit of time per unit area of the hydrologic basin. Sediment yield is a useful tool for comparing study sites because it removes variation in basin size and event duration by normalizing the SSL. Event-based yields were computed for this study using the following equation:
EYield ESSL/ (DA x ED) (6) where
EYield is event-based suspended-sediment yield, in pounds per square mile per hour;
ESSL is event-based suspended-sediment load, in pounds;
DA is drainage area, in square miles; and
ED is event duration, in hours. Concentrations of barium, cadmium, lead, and zinc measured in suspended-sediment samples collected during events were used to compute event-based loads and yields of these constituents. In order to properly compute the traceelement concentration within the suspended sediments, a mass Figure 4. Daily mean streamflow in relation to discrete suspended-sediment concentrations used to develop a regression model for the computation of estimated daily suspended-sediment concentrations and loads for two study sites on the Big River, Missouri, October 2011–September 2013. 1,000 10,000 100,000 1,000 5,000 Discrete suspended-sediment concentration, in milligrams per liter Daily mean streamflow, in cubic feet per second Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) 1,000 10,000 100,000 1,000 5,000 y 0.3341(x)0.8148 R 2 a 0.37 n 36 y 0.1668(x)0.8469 R 2 a 0.41 n 25 EXPLANATION regression equation R 2 a 0.41 n 25 y 0.1668(x)0.8469 R 2 a adjusted coefficient of determination n number of datasets where: y suspendedsediment concentration and x daily mean streamflow Data pair used in model Best-fit regression model
Surface-Water Quality 13 accumulation was computed using the constituent measure ments from both size fractions. The equation used for the mass-accumulation computation of the total trace-element concentration in suspended sediment collected during events is as follows: Conctotal [(ConcS x (%S/100) + (ConcF x (%F/100))] x SSC/100,000 where
Conctotal is the total trace-element concentration in suspended sediments, in milligrams per liter;
ConcS is the trace-element concentration in the sand fraction, in milligrams per kilogram;
%S is the percent of mass in total sample considered sand (particles greater than 0.063 millimeters);
ConcF is the trace-element concentration in the fines fraction, in milligrams per kilogram;
%F is the percent of mass in total sample considered fines (particles less than 0.063 millimeters);
SSC is suspended-sediment concentration of total sample, in milligrams per liter; and
100,000 is the conversion factor from milligrams per kilogram to milligrams per liter. The mass accumulation computation used the sum of mass in each size fraction adjusted by the total SSC of the sample, then converted the trace-element concentration to milligrams per liter. When only one size fraction had enough mass for analyses, the trace-element concentration of the sample was still adjusted by the SSC, but should be assumed lower than the trace-element concentration that could be in the suspended sediments. An arithmetic average of trace-element concentrations were computed from all available suspendedsediment samples collected during the designated event duration. The event-based loads and yields were computed as previously described for the event-based suspended-sediment loads and yields in equations 5 and 6, using the arithmetic average of the trace element concentrations. Surface-Water Quality Many ancillary conditions such as land use, topography, atmospheric conditions, and streamflow conditions such as extreme base flow (droughts) and stormflows (floods) can affect the overall quality of a stream. The assessment of these conditions were used to determine the surface-water quality of Big River and to compute loads and yields of suspended sediment and the loads and yields of trace elements in the suspended sediments. Streamflow Conditions During the 2012 water year, Missouri’s precipitation was less than normal at 34.67 in. compared to the long-term (approximately 100 years) State average of 41.03 in. (National Oceanic and Atmospheric Administration, 2013). In October 2011, about 40 percent of all Missouri counties had dry or drought conditions and about 60 percent were classified by the National Drought Mitigation Center (2013) as abnormally dry. From July through September 2012, all Missouri coun ties were experiencing at least moderate drought conditions. During August 2012, 35 percent of Missouri counties typically located in the northwest, southwest, and southeast corners of the State, were classified as having extreme drought conditions (National Drought Mitigation Center, University of NebraskaLincoln, 2013). Missouri’s precipitation for the 2013 water year was 43.58 in., which is above the long-term State average (National Oceanic and Atmospheric Administration, 2014b). Precipitation gages were not available at the study sites; therefore, monthly precipitation was obtained from two NOAA climate stations within the study area at DeSoto and Dittmer, Mo. (National Oceanographic and Atmospheric Administration, 2015; fig. 5). The DeSoto climate station is located in the southern part of the study area near the upstream site. The Dittmer climate station is located in the northern part of the study area near the downstream site. During the study period, monthly precipitation at both climate stations ranged from approximately 1 in. (at Dittmer in July 2012) to more than 7 in. (at DeSoto in April 2013). Monthly precipita tion was consistent between the two climate stations for most of the 2012 water year with the exception of higher monthly precipitation amounts at the Dittmer station during May, August, and September 2012 and higher monthly precipita tion at the DeSoto Station during November 2011. September 2012 had the highest monthly precipitation amounts for both climate stations during the 2012 water year. Monthly precipi tation amounts increased during the 2013 water year, particu larly from March through June, compared to the 2012 water year. Large differences in monthly precipitation between the climate stations were noted in January and July 2013, with the higher precipitation amounts recorded in the southern part of the study area at the DeSoto station. During the 2013 water year, April 2013 had the largest monthly precipitation amount recorded at the DeSoto station and March had the largest pre cipitation amount recorded at the Dittmer station. The annual mean streamflow at the upstream site dur ing the 2012 water year was approximately 189 ft3/s with an annual runoff of 6.27 in. (table 4). Annual mean streamflow and annual runoff during the 2013 water year was nearly four times higher than during the 2012 water year at 695 ft3/s and 23.06 in., respectively (table 4). At the downstream site, the annual mean streamflow during the 2012 water year was 419 ft3/s with an annual runoff of 6.22 in. The annual mean streamflow and annual runoff increased during the 2013 water year to 1,393 ft3/s and 20.63 in., respectively (table 4). (7)
14 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Streamflows were greatest in the spring months for the two study sites. At the upstream site, maximum daily streamflow was 2,390 ft3/s in March 2012 and 18,600 ft3/s in March 2013, and the minimum daily streamflow was 9.9 ft3/s in August 2012 and 42 ft3/s in September 2013 (table 4; fig. 6). Similar temporal variability also was measured at the downstream site. The maximum daily mean streamflow of 4,340 ft3/s was in March during water year 2012, and the maximum daily mean streamflow of 26,400 ft3/s was in April during the 2013 water year (table 4; fig. 6). The maximum peak streamflow during the study period was 21,900 ft3/s on April 19, 2013, at the upstream site, and the maximum peak streamflow during the study period was 28,700 ft3/s on April 20, 2013, at the downstream site. The historical maxi mum peak streamflow recorded at the downstream site was 63,600 ft3/s on September 25, 1993 (table 4). Continuous Water Quality Water temperature is an important physical property because it can assist with quality control of monitor opera tions, such as detection of biological activity, extreme stream flow conditions, and human influences (Wilde, 2006). Water temperature is also a factor in determining a stream’s physical fluid properties because viscosity is a factor in determining sediment transport efficiency (Charlton, 2008). The maximum water temperature recorded at the upstream site was 32.7 °C on July 25, 2012, and the minimum recorded water tempera ture was -0.1 °C on January 3, 2013 (table 5). At the down stream site, the maximum water temperature recorded was 30.9 °C on August 8, 2012, and the minimum recorded water temperature was 0.5 °C on January 3, 2013 (table 5). Figure 5. Total monthly precipitation at two National Oceanic and Atmospheric Administration climate stations near the Big River Basin, Missouri, October 2011–September 2013. Oct. Nov. Dec. Jan Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan Feb. Mar. Apr. May June July Aug. Sept. Total monthly precipitation, in inches DeSoto, Missouri, climate station Dittmer, Missouri, climate station EXPLANATION National Oceanic Atmospheric Administration, 2015 100 MILES KILOMETERS Big River Basin
Surface-Water Quality 15 Table 4. Streamflow statistics for two study sites on the Big River, Missouri, October 2011–September 2013. [ft3/s, cubic feet per second; ft, feet; in., inches] Water year1 Maximum daily mean streamflow (ft3/s) and date Minimum daily mean streamflow (ft3/s) and date Maximum peak streamflow (ft3/s) and date Maximum peak stage (ft) and date Annual mean streamflow (ft3/s) Annual runoff (in.) Big River below Bonne Terre, Missouri (07017610; upstream site) 2,390 March 17, August 11, 4,810 March 16, March 16, 18,600 March 18, September 14, 21,900 April 19, April 19, 2013 Big River at Byrnesville, Missouri (07018500; downstream site) 4,340 March 18, July 28, 2012 4,920 March 18, March 18, 26,400 April 20, September 15–17, 28,700 April 20, April 20, 2013 1,393 1922–2013 57,800 September 25, August 30, 63,600 September 25, September 25, 1Water year is defined as the 12-month period beginning October 1 and ending September 30 of the calendar year in which it ends (water year 2012 is the period October 1, 2011 through September 30, 2012). 2Computed with incomplete data because the streamgage was not in operation until October 13, 2011. Figure 6. Daily mean streamflow computed during the study period at two study sites on the Big River, Missouri, October 2011–September 2013. Daily mean streamflow, in cubic feet per second Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr May June July Aug. Sept. Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) 1,000 10,000 100,000
16 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Water temperatures recorded at both sites followed similar diurnal variability and were similar at the two study sites. No water temperature data were corrected or missing for the upstream site during the study period. Water temperature data for the downstream site were unavailable from June 28 to July 30, 2012, because stream levels were too low for the temperature sensor to be submerged for proper measurements. Extreme flood debris, such as organic materials, large limbs and trees, and trash, affected the water temperature measure ments during two events in water year 2013, causing data to be removed from April 19 through 24, and August 21 through 23, 2013 (fig. 7). At the upstream site, daily mean turbidity ranged from 1.4 to 520 FNU during the study period (table 5; fig. 8). The maximum value recorded during the study was 890 FNU on March 18, 2013, and the minimum value was 1.0 FNU on January 9 and 10, 2012 (table 5). Missing daily values at the upstream site were commonly caused by extreme biofouling during warmer months with lower streamflows and siltation of the instrument and its deployment pipe during high flows. Efforts were made to service and eliminate such effects but were not always successful. Other factors were instrumenta tion malfunctions and drift between calibrations. At the downstream site, turbidity measurements began on October 14, 2011. Daily mean turbidity ranged from 1.3 to 240 FNU for the study period (table 5; fig. 8). The maximum value recorded during the study was 790 FNU on April 18, 2013, and the minimum value recorded was 0 FNU on October 15, 2012 (table 5). No data were removed during water year 2012; however, during the 2013 water year, many days were removed because of siltation of the monitor and deployment pipe during higher streamflows. It was noted dur ing the 2013 water year that a large amount of trees and veg etation were removed and some construction on the property adjacent to the monitor deployment pipe had occurred. These events may have increased the monitor fouling issues during storm events because exposed soil was readily available dur ing runoff from the lack of bank stabilization. Other periods of missing record were from instrumentation malfunction. Turbidity showed a consistent temporal variability between both study sites (fig. 8). During storm events with precipitation evenly distributed throughout the basin and not localized, measured turbidity values at both sites showed similar but lagged changes with time as the runoff moved downstream. The travel time of a runoff peak was approxi mately 18 hours between sites. In some instances, a larger streamflow and greater turbidity values would be noted at the downstream site compared to the upstream site. It is possible that more localized storms in the downstream reaches of Big River caused an increase in streamflow. No streamgage exists on Mineral Fork to determine its contribution of streamflow and suspended sediment to the Big River. The low-head mill dam just upstream from the downstream site was observed to restrict streamflow during low-flow conditions and resulted in turbidity values that differed from those at the upstream site. Table 5. Water-quality statistics for two study sites on the Big River, Missouri, October 2011–September 2013. Water year1 Water temperature (degrees Celsius) Maximum daily mean and date Minimum daily mean and date Maximum recorded and date Minimum recorded and date Big River below Bonne Terre, Missouri (07017610; upstream site) July 25, 2012 January 14, 2012 July 25, 2012 January 14, 2012 July 19, 2013 January 3, 2013 July 18, 2013 January 3, 2013 Big River at Byrnesville, Missouri (07018500; downstream site) August 8, 2012 January 14, 15, 2012 August 8, 2012 January 14, 15, 2012 July 19, 2013 January 3, 2013 July 19, 2013 January 3, 2013 Water year1 Turbidity (formazin nephelometric units) Maximum daily mean and date Minimum daily mean and date Maximum recorded and date Minimum recorded and date Big River below Bonne Terre, Missouri (07017610; upstream site) September 8, 2012 November 5, 2012 March 16, 2012 January 9, 10, 2012 March 18, 2013 November 21, 24, 2013 March 18, 2013 November 20–22, 2013 Big River at Byrnesville, Missouri (07018500; downstream site) March 18, 2012 July 27, 2012 September 25, 2012 July 29, 2012 March 20, 2013 January 3–5, 7, 8, 2013 April 18, 2013 October 15, 2012 1Water year is defined as the 12-month period beginning October 1 and ending September 30 of the calendar year in which it ends (water year 2012 is the period October 1, 2011 through September 30, 2012).
Surface-Water Quality 17 Figure 7. Daily mean water temperature measured at two study sites on the Big River, Missouri, October 2011–September 2013. Daily mean water temperature, in degrees Celsius Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Gaps indicate periods in which no data are available Periods of missing data (gaps) are due to equipment issues, no-flow conditions, extreme fouling, or exceedance of correction criteria, in which case data are deleted.
18 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Daily mean turbidity, in formazin nephelometric units 1,000 1,000 Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Gaps indicate periods in which no data are available Gaps indicate periods in which no data are available Gaps indicate periods in which no data are available Periods of missing data (gaps) are due to equipment issues, no-flow conditions, extreme fouling, or exceedance of correction criteria, in which case data are deleted. Periods of missing data (gaps) are due to equipment issues, no-flow conditions, extreme fouling, or exceedance of correction criteria, in which case data are deleted. Figure 8. Daily mean turbidity measured at two study sites on the Big River, Missouri, October 2011–September 2013.
Suspended-Sediment Quantity 19 Suspended-Sediment Quantity Discrete suspended-sediment concentrations were col lected at each site at least monthly and during stormflow events (fig. 9). Samples were collected for the computation of daily SSC and SSL and were used to calibrate the regression model with the continuous turbidity data. During the 2012 water year, 11 individual base-flow and 4 sets of event SSC samples were collected at both sites (table 6); however, the November 2011 event collected at the downstream site did not have enough sediment mass for trace-element analyses and was not used for any event-based computations. During the 2013 water year, 7 individual base-flow and 3 sets of event SSC samples were collected at both sites (table 6). Base-flow SSC values ranged from 1 to 74 milligrams per liter (mg/L) at the upstream site during the study period, and event SSC values ranged from 74 to 1,091 mg/L (table 6; fig. 9). At the downstream site, base-flow SSC ranged from 4 to 127 mg/L and event SSC ranged from 27 to 871 mg/L (table 6; fig. 9). A previous study within the Ozark Plateaus (Davis and Bell, 1998) determined that median base-flow SSC ranged from 3 to 28 mg/L in streams of similar basin size and physiogra phy, with increased SSC during increased streamflow. The median base-flow SSC for the upstream and downstream site during the study period was 10 mg/L and 13 mg/L, respec tively (table 6). Figure 9. Continuous streamflow and discrete suspended-sediment concentration samples collected at two study sites on the Big River, Missouri, October 2011–September 2013. 1,000 1,000 10,000 100,000 Daily mean streamflow Suspended-sediment sample Daily mean streamflow, in cubic feet per second Suspended-sediment concentration, in milligrams per liter 1,000 10,000 100,000 1,000 10,000 10,000 Daily mean streamflow Suspended-sediment sample Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept.
20 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Table 6. Concentrations and size distributions of suspended-sediment samples collected at two study sites on the Big River, Missouri, October 2011–September 2013. [g, gram; mg/L, milligram per liter; %, percent; less than; mm, millimeter; --, not measured] Sample date and time Hydrologic event Total sample dry mass (g) Suspended-sediment concentration (mg/L) Suspended sediment (% <0.063 mm) Big River below Bonne Terre, Missouri (07017610; upstream site) 10/12/2011 16:40 base flow 11/10/2011 12:15 base flow 12/05/2011 18:00 base flow 12/15/2011 11:15 rise 12/15/2011 11:40 rise 12/15/2011 12:20 peak 12/15/2011 13:50 peak 12/15/2011 14:00 fall 12/15/2011 15:00 fall 01/10/2012 13:50 base flow 02/04/2012 16:45 base flow 03/15/2012 11:15 base flow 03/16/2012 10:50 rise 03/16/2012 15:15 rise 03/17/2012 00:01 peak 03/17/2012 16:40 fall 04/04/2012 16:15 base flow 04/16/2012 06:30 rise 04/16/2012 10:40 peak 04/17/2012 13:50 fall 05/21/2012 16:30 base flow 06/26/2012 12:45 base flow 07/31/2012 10:35 base flow 08/21/2012 15:30 base flow 09/08/2012 14:15 rise 09/08/2012 16:40 peak 09/10/2012 13:00 fall 10/16/2012 11:30 base flow 11/19/2012 12:00 base flow 12/17/2012 12:15 base flow 01/30/2013 11:20 rise 03/18/2013 11:30 peak 03/18/2013 15:50 peak 04/18/2013 13:32 rise 1,091 04/19/2013 09:14 peak 04/19/2013 11:24 fall 05/06/2013 14:30 base flow 06/26/2013 15:30 base flow 07/16/2013 10:40 base flow 08/28/2013 11:00 base flow minimum base flow maximum base flow median base flow minimum stormflow maximum stormflow median stormflow
Suspended-Sediment Quantity 21 Sample date and time Hydrologic event Total sample dry mass (g) Suspended-sediment concentration (mg/L) Suspended sediment (% <0.063 mm) Big River at Byrnesville, Missouri (07018500; downstream site) 10/13/2011 15:00 base flow 11/10/2011 15:00 base flow 11/22/2011 13:00 rise 11/22/2011 13:30 rise 11/22/2011 13:50 peak 12/05/2011 12:40 base flow 02/28/2012 10:15 base flow 03/15/2012 15:10 base flow 03/17/2012 12:40 rise 03/17/2012 20:45 peak 03/18/2012 13:20 fall 04/04/2012 12:50 base flow 04/17/2012 10:25 peak 05/21/2012 13:37 base flow 06/26/2012 15:00 base flow 06/26/2012 15:05 base flow 07/31/2012 13:45 base flow 08/21/2012 10:10 base flow 09/10/2012 10:15 fall 10/16/2012 08:50 base flow 11/30/2012 10:00 base flow 12/17/2012 13:45 base flow 01/30/2013 16:45 rise 01/31/2013 08:30 rise 03/19/2013 18:00 peak 04/18/2013 16:03 rise 04/20/2013 16:15 peak 05/28/2013 09:30 base flow 06/26/2013 11:30 base flow 07/16/2013 12:00 base flow 08/28/2013 13:40 base flow minimum base flow maximum base flow median base flow minimum stormflow maximum stormflow median stormflow Table 6. Concentrations and size distributions of suspended-sediment samples collected at two study sites on the Big River, Missouri, October 2011–September 2013.—Continued [g, gram; mg/L, milligram per liter; %, percent; less than; mm, millimeter; --, not measured]
22 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Daily Suspended-Sediment Concentrations and Loads The upstream and downstream sites had similar SSC and SSL extremes during the study, but these extremes did not always occur during the same stormflow events (figs. 10, 11). During the 2012 water year, the maximum daily mean SSC of 617 mg/L and maximum daily mean SSL of 2,920 tons at the upstream site occurred on September 8, 2012, whereas the maximum daily mean SSC of 522 mg/L and SSL of 6,120 tons at the downstream site occurred on March 18, 2012 (table 7). The maximum daily streamflow as shown in table 4 for the upstream site did not occur on the same event as the maximum daily mean SSC during the 2012 water year, but did occur on the same event during the 2013 water year (fig. 10). One pos sible explanation is that even though the streamflow was less than the water year maximum during the event on Septem ber 8, 2012, the amount of available suspended sediments in the system and the time between large events were greater, resulting in higher turbidity measurements and SSC results for that event. During the 2013 water year, the maximum daily mean SSC and streamflows both occurred on the March 18, 2013, event at the upstream site (tables 4, 7). Maximum daily mean SSC and streamflows occurred on the same events dur ing both study years for the downstream site (table 4, 7). Figure 10. Daily mean suspended-sediment concentrations computed at two study sites on the Big River, Missouri, October 2011–September 2013. Daily mean suspended-sediment concentration, in milligrams per liter 1,000 1,000 10,000 Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept.
Suspended-Sediment Quantity 23 Annual SSLs were computed to estimate the total amount of suspended sediments that passed the sites for each water year. The annual SSL increased greatly during the study period because the first water year (2012) was classified as a drought year whereas more precipitation was recorded during the 2013 water year. The upstream site had an annual SSL of 10,035 tons during the 2012 water year, which increased to 191,294 tons for the 2013 water year, and the downstream site had an annual SSL of 21,612 tons in 2012, which increased to 333,717 tons for the 2013 water year (table 7). A previous study along the Black River in southeast Missouri (Barr, 2009) also computed annual SSL for an Ozark stream to assess damage from an upstream reservoir embank ment breach. The study site, Black River below Annapolis, Mo. (USGS site number 07061600), has a drainage area of 493 mi2, which is similar to the drainage area of the Big River upstream site (409 mi2). The total annual SSL at the Black River site was 29,300 tons during the 2006 water year when sediment sources were readily available, but decreased to 17,400 tons during the 2007 water year because the readily available suspended sediments had been transported through the basin. The short-term effects of the damaged reservoir still did not cause large sediment volumes in comparison to the historical mining effects within the Old Lead Belt. Figure 11. Daily mean suspended-sediment loads computed at two study sites on the Big River, Missouri, October 2011–September 2013. 1,000 10,000 100,000 1,000 10,000 100,000 Daily mean suspended-sediment load, in tons Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept.
24 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Event-Based Suspended-Sediment Concentrations, Load, and Yields For all stormflow events sampled during the study period, an event-based average SSC and SSL were computed to help determine the mass of suspended sediments transported dur ing each sampled event. Averaged event SSCs ranged from approximately 100 mg/L to 1,400 mg/L at both study sites (table 8). Event SSLs ranged from about 180 to 32,000 tons at the upstream site and from about 390 to 53,000 tons at the downstream site (table 8). Most events lasted more than 24 hours, and the same hydrologic event was sampled at both sites, except the December 15, 2011, event, when only the upstream site was sampled (table 8). The December 2011 event was more localized in the upper portion of the watershed resulting in only a minor event by the time the event reached the downstream site. Event-specific SSLs were greater at the downstream site except for the September 2012 event, and event yields were greater at the upstream site than the down stream site during all events except April 2012 (table 8). Event loads and yields were greater during water year 2013 events, with the April 2013 event having the greatest loads and yields at both sites (table 8). It is possible that the large differences in event yields between sites are because of limited SSC col lections during events at the downstream site, and because more sediments are available in the upper part of the basin closer to the historical mining activity. It should be noted that events differed in intensity and duration between the study sites, such as April 2012 (40 hours at the upstream site and 63 hours at the downstream site), which could cause some bias in comparisons. Events at the downstream site were very slow moving and occurred over multiple days, but turbidity values did not show much variation at times, which made predictions of an event with sufficient, available suspended sediments dif ficult to determine. All event-based concentrations, loads, and yields are most likely biased low because of limited number of samples collected during the event but can be useful in approximating the minimum volume of suspended sediments transported downstream during stormflow events. Even though a limited number of events could be sam pled, using the continuous turbidity data to compute the daily SSL in conjunction with event sampling provided adequate data to determine the majority of the SSL in Big River occurs during stormflow events. Not all stormflow events that occurred during the study period could be sampled, yet the events that were sampled make up a considerable portion of the annual SSL. Because the SSCs collected during base-flow conditions during the study period are generally very low (typically less than 50 mg/L at both sites), a general base-flow SSL can be computed for each site to compare to event-based SSLs. The median of all daily median streamflow values for the upstream site during the study period (132 ft3/s) was used to represent a base streamflow condition. Using the median streamflow and a SSC of 50 mg/L, an annual base-flow SSL for the study period can be derived as 18 tons, and the total SSL of the seven events sampled at the upstream site during the study period is about 57,700 tons (table 8). When com pared to the sum of annual SSLs computed from the regres sion model for the study period (table 7), the base-flow SSL accounts for less than 0.01 percent and the sampled events account for about 29 percent of the total SSL at the upstream site. For the downstream site, the base streamflow computed Table 7. Daily suspended-sediment concentration and load statistics at two study sites on the Big River, Missouri, October 2011–September 2013. [mg/L, milligram per liter; E, estimated] Water year1 Maximum daily mean suspended-sediment concentration (mg/L) and date Minimum daily mean suspended-sediment concentration (mg/L) and date(s) Maximum daily mean suspended-sediment load (tons) and date Minimum daily mean suspended-sediment load (tons) and date Annual suspendedsediment load (tons) Number of estimated days Big River below Bonne Terre, Missouri (07017610; upstream site) September 8, many days 2,920 September 8, E 0.134 August 11, 10,035 E 1,010 March 18, 2013 E 3 August 11, 2013 E 50,500 March 18, December 9, 2012 191,294 Big River at Byrnesville, Missouri (07018500; downstream site) March 18, 2012 July 27–29, 2012 6,120 March 18, July 28, 21,612 E 927 April 20, 2013 December 22, 30, 31, 2012, January 1–9, E 66,000 April 20, 2013 January 8, 333,717 1Water year is defined as the 12-month period beginning October 1 and ending September 30 of the calendar year in which it ends (water year 2012 is the period October 1, 2011 through September 30, 2012).
Suspended-Sediment Quality 25 for the study period was 339 ft3/s. Using the same base-flow SSC of 50 mg/L, the annual base-flow SSLs were 46 tons during the study period, accounting for about 0.01 percent of the total SSL computed from the regression model (table 7). The six events sampled at the downstream site had a combined SSL of about 103,000 tons (table 8), or about 29 percent of the total SSL for the study period. Suspended-Sediment Quality Suspended sediments collected during event sampling were more abundant in the fines fraction (tables 6 and 9). The supply of materials finer than bed material that are in suspen sion (also known as the “wash load”) usually have a greater effect on the SSC than streamflow conditions, because the rate of supply varies during and between events as well as from seasonal precipitation and vegetation (Charlton, 2008). Increases in streamflow produce increased SSC as erosion increases, releasing finer particles from storage and increas ing the stream’s capacity for sediment transport. The event samples collected at the upstream site were 75 to 98 percent fines and the event samples collected at the downstream site were from 90 to 98 percent fines in all samples except one (table 6). Only one sample during the entire study, collected at the downstream site on the peak of the March 19, 2013, event, was less than 75 percent (59 percent) fines (table 6). Selected Trace-Element Concentrations in Suspended Sediments The laboratory results of event samples with adequate sample mass for analysis are listed in table 9 and include four selected trace-element concentrations—barium, cadmium, lead, and zinc. The four trace elements were selected for inclu sion in this report because they are contained in ore minerals mined in the Big River Basin and are readily available for transport from both natural-source erosion and from mine waste. In addition, cadmium, lead, and zinc are considered toxic to aquatic life at certain concentrations in bed sediments and in dissolved form (MacDonald and others, 2000). The primary minerals containing barium, cadmium, lead, and zinc are relatively insoluble and predominately are found in the solid phase (Smith and Schumacher, 1991; 1993). No samples for dissolved trace-element concentrations were collected during the study period to include in the total concentration computation of trace elements at the study sites. Smith and Schumacher (1991; 1993) collected dissolved trace element concentrations at a site along the Big River near Desloge Table 8. Suspended-sediment concentrations, loads, and yields from sampled stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013. [hr, hour; ft3, cubic foot; mg/L, milligram per liter; lb, pound; mi2, square mile; (lb/mi2)/hr, pound per square mile per hour] Event date Event duration (hr) Event number Sampled runoff volume (ft3) Suspended-sediment Average event concentration (mg/L) Load Yield (lb/mi2)/hr (lb) (ton) Big River below Bonne Terre, Missouri (07017610; upstream site) Dec. 15, 2011 50,150,000 1,110 354,000 Mar. 16–17, 2012 343,000,000 1,180 7,000,000 3,330 1,540 Apr. 16, 2012 49,110,000 352,000 Sept. 8–9, 2012 226,000,000 4,000,000 1,840 1,240 Jan. 29–31, 2013 740,000,000 13,200,000 6,590 1,670 Mar. 17–19, 2013 2,000,000,000 26,600,000 13,300 5,490 Apr. 18–19, 2013 2,210,000,000 1,380 64,600,000 32,300 9,510 Big River at Byrnesville, Missouri (07018500; downstream site) Mar. 17–18, 2012 474,000,000 1,060 8,600,000 4,300 Apr. 15–18, 2012 510,000,000 4,900,000 2,430 Sept. 9–10, 2012 244,000,000 781,000 Jan. 30–31, 2013 759,000,000 23,200,000 11,600 1,520 Mar. 17–20, 2013 4,600,000,000 62,400,000 31,200 1,840 Apr. 18–20, 2013 3,300,000,000 110,000,000 53,100 2,920
26 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Table 9. Selected trace-element concentrations for two particle size distributions of suspended sediments collected during stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013. [sand, sediment greater than 0.063 millimeters; fines, sediment less than 0.063 millimeters; mg/kg, milligram per kilogram; --, not available] Event number Sample date Sample time Trace element concentrations in suspended sediments Barium Cadmium Lead Zinc Sand (mg/kg) Fines (mg/kg) Sand (mg/kg) Fines (mg/kg) Sand (mg/kg) Fines (mg/kg) Sand (mg/kg) Fines (mg/kg) Big River below Bonne Terre, Missouri (07017610; upstream site) 12/15/2011 11:15 1,000 1,000 12/15/2011 11:40 12/15/2011 12:20 12/15/2011 13:50 12/15/2011 14:00 12/15/2011 15:00 03/16/2012 10:50 03/16/2012 15:15 1,100 03/17/2012 00:01 1,700 1,000 03/17/2012 16:40 1,700 1,100 04/16/2012 06:30 1,200 1,100 04/16/2012 10:40 09/08/2012 14:15 1,400 2,400 1,200 1,800 09/08/2012 16:40 1,000 2,300 1,600 01/30/2013 11:20 1,000 03/18/2013 11:30 1,300 03/18/2013 15:50 1,200 04/18/2013 13:32 04/19/2013 09:14 1,500 Minimum Median 1,000 1,000 Mean 1,189 Maximum 1,400 2,400 1,200 1,800 Big River at Byrnesville, Missouri (07018500; downstream site) 03/17/2012 12:40 1,200 03/17/2012 20:45 1,200 03/18/2012 13:20 1,100 1,100 04/17/2012 10:25 1,500 09/10/2012 10:15 1,300 1,800 01/30/2013 16:45 1,300 1,000 01/31/2013 08:30 1,200 1,100 03/19/2013 18:00 1,600 04/18/2013 16:03 04/20/2013 16:15 1,500 Minimum Median 1,200 1,050 Mean 1,141 1,075 Maximum 1,500 1,800 Consensus-based sediment quality guidelines (MacDonald and others, 2000) Cadmium (mg/kg) Lead (mg/kg) Zinc (mg/kg) Consensus-based threshold effect concentration (TEC) Consensus-based probable effect concentration (PEC) Toxic effect threshold (TET)
Suspended-Sediment Quality 27 (fig. 1) during base flow and stormflow events, and total recov erable trace element concentrations only were collected during base flow conditions. The available dissolved trace element concentrations show the dissolved lead and zinc concentra tions were very low, indicating a minimal portion of the total trace-element concentration would be from the dissolved phase. The full suites of trace element results for all samples are available on NWISWeb (://nwis.waterdata.usgs.gov/ mo/nwis/qwdata). Events 2, 4, 5, and 7 collected at the upstream site and events 4 and 5 collected at the downstream site had enough mass in both the sand and fines fractions for complete analyses of the total suspended sediments collected (table 9; fig. 12). Barium concentrations in the sand fraction were less than the concentrations in the fines fraction at both sites. Cadmium concentrations in sands were less than the fines fraction except for event number 5 at the upstream site (12 milligrams per kilogram [mg/kg] in sands, 10 mg/kg in fines) and event 4 at the downstream site (7 mg/kg in sands, 4 mg/kg in fines), although the concentrations were similar (table 9). Events 5 and 7 at the upstream site had similar but slightly higher con centrations of lead in the sand fraction than the fines fraction during event 5 and event 7 (table 9), and only event 4 at the downstream site had slightly higher but similar zinc concen trations in the sand fraction than the fine fraction (table 9). The greatest differences between sand and fines fraction concentrations were noted in lead and zinc concentrations for events 2 and 4 at the upstream site (fig. 12). The peak sample collected during event 2 on March 17, 2012, had lead concen trations of 750 mg/kg in the sand fraction and 1,700 mg/kg in the fines fraction, and zinc concentrations of 470 mg/kg in sand and 1,000 mg/kg in the fines (table 9, fig. 12). Event number 4 at the upstream site, collected September 8, 2012, had lead concentrations in the sand on the rising limb sample of 1,400 mg/kg and 2,400 mg/kg in the fines. The sand in the peak sample had a lead concentration of 1,000 mg/kg and the fines had a concentration of 2,300 mg/kg (table 9; fig. 12). The highest concentrations of cadmium (30 mg/kg in sand, 35 mg/kg in fines) and zinc (1,200 mg/kg in sand, 1,800 mg/kg in fines) occurred during event number 4 at the upstream site. During a previous study of surface water and sediment quality in the Old Lead Belt, trace-element concentrations in suspended sediments also were measured during events within the Big River Basin (Smith and Schumacher, 1991; 1993), but do not indicate any changes in selected concentrations between study periods. Three event samples were collected between September 1988 and February 1989 at the Big River below Desloge, Mo. (USGS site number 07017620; site 6 in Smith and Schumacher, 1993), located upstream from the study area for this report and downstream from the Desloge mine waste pile (fig. 1). Barium and cadmium concentrations at Desloge were similar to the upstream site concentrations and the lead and zinc concentrations at Desloge were higher than the upstream site concentrations. Barium concentra tions in suspended sediments ranged from 390 to 630 mg/kg at Desloge and 240 to 750 mg/kg at the upstream site, and cadmium concentrations ranged from 21 to 30 mg/kg at Desloge and 4 to 35 mg/kg at the upstream site (table 9). Lead concentrations in the three event samples at Desloge during the previous study ranged from 1,100 to 3,200 mg/kg and at the upstream site the lead concentrations ranged from 680 to 2,400 mg/kg. Zinc concentrations at Desloge ranged from 1,100 to 2,200 mg/kg, which were higher than the upstream site concentrations (470 to 1,800 mg/kg). The difference in sampling location, event intensity, changes in geomorphol ogy of the stream, or remediation efforts of mine waste piles could be reasons lead and zinc concentrations were higher at Desloge than the upstream site. Not enough data are available to determine if any long-term decreases have occurred. Sediment quality guidelines (SQG) for freshwater (Mac Donald and others, 2000) describe two qualities of environ mental effects from trace-element concentrations in suspended sediments—the threshold effect concentration (TEC) and the probable effect concentration (PEC). A TEC is defined as a concentration below which adverse effects are not expected to occur below, and a PEC is a concentration in which adverse effects are expected to occur at or above more often. Within these two sediment qualities are many levels and thresholds that help regulatory agencies assess the toxicity of sediments within a remediation study. The highest threshold in the PEC group is the toxic effect threshold (TET), which is the con centration above which sediment is considered to be heavily polluted causing adverse effects on sediment-dwelling organ isms (MacDonald and others, 2000). Both TEC and PEC are consensus-based levels of concentration to reflect the intent of each SQG type. Although these threshold and probable effects are based on streambed sediments, comparisons were made with the stormflow event sediment results because these suspended sediments were most likely at rest on the streambed during base-flow conditions and after transport will be at rest again on the streambed. Concentrations of cadmium, lead, and zinc in suspended sediment samples collected during stormflow events at both sites exceeded the consensus-based TEC and consensus-based PEC (fig. 12). All cadmium concentrations in the fine and sand fractions of suspended sediments at both sites exceeded the consensus-based TEC (0.99 mg/kg) and PEC (4.98 mg/kg). All lead and zinc concentrations in the fine fraction of all sus pended sediment from both sites greatly exceeded the con sensus-based TEC (35.8 mg/kg for lead; 121 mg/kg for zinc) and the PEC (128 mg/kg for lead; 459 mg/kg for zinc). Lead concentrations in the fine fraction of all samples at both sites also exceeded TET of 170 mg/kg. Zinc concentrations in the fine fraction for four of the five samples from the upstream site also exceeded the TET of 540 mg/kg (fig. 12). Lead and zinc concentrations in the sand fraction exceeded their respective consensus-based PECs except for the sand fraction of event number 5 collected at the downstream site. Concentrations of cadmium and zinc were notably higher in suspended sediment samples from the upstream site compared to the downstream site, especially in the fine fraction. The smaller cadmium and zinc concentrations at
28 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 the downstream site could be a result of a decrease in source material enriched in these trace elements in the downstream areas of the watershed compared to the upstream area within the Old Lead Belt. Concentrations of lead, while variable, tended to be similar in both fractions at the two sites (fig. 12). The lead concentrations greatly exceeded the PEC as well as the TET, in some samples by a factor of 10, which indi cates that lead-rich suspended sediments, especially in the fine fraction, are readily available for transport within the Big River Basin. These lead-enriched, finer sediments could remain in the system from historical mining, and as the cap ping of mine waste piles upstream from Bonne Terre continues to reduce additional sediment loadings, these fine sediments may be continually released as the river scours the streambed and erodes the stream banks causing lead-rich suspended sedi ment to remain in a state of equilibrium. Figure 12. Trace-element concentrations in two suspended-sediment size fractions collected during stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013. Milligrams per kilogram 1,000 1,200 1,400 1,600 Cadmium Barium Lead Zinc 1,000 1,200 1,400 1,600 1,800 2,000 1,000 1,500 2,000 2,500 3,000 Event number (table 9) EXPLANATION Event number Upstream site Downstream site Upstream site Downstream site Upstream site Downstream site Upstream site Downstream site Upstream site Downstream site Suspended-sediment fraction less than 0.063 millimeter (fines) Suspended-sediment fraction greater than 0.063 millimeter (sands) Consensus-based toxic effect threshold (TET; MacDonald and others, 2000) Consensus-based probable effect concentration level (PEC; MacDonald and others, 2000) Consensus-based threshold effect concentration level (TEC; MacDonald and others, 2000) Big River below Bonne Terre, Missouri (07017610) Big River at Byrnesville, Missouri (07018500) Numeric representation of stormflow events collected at each site during study period (table 9)
Suspended-Sediment Quality 29 Variation within the event sample concentrations was typ ically low, but some large variations were noted for cadmium, lead, and zinc during two events. Event 7 at the upstream site, collected April 18 and 19, 2013, showed high variation of cadmium and lead concentrations in the fines fraction of the samples (table 9). The two samples collected during this event were collected near two different streamflow peaks; the first sample occurred from localized runoff of a smaller tributary upstream from the site (categorized as a rise sample in table 6) and the second sample occurred from the main channel peak, which traveled from farther upstream (considered the actual peak for the event in table 6). The increase in cadmium and lead during the second sample could be because of more sources of these trace elements from upstream sources. During this same event, high variation among concentrations at the downstream site also were noted. The event sampling at the downstream site (event 6) occurred on the rise and near the peak of the event, during April 18 and 20, 2013. This particu lar stormflow event caused flooding conditions at the routine sampling site as the river crested, which made the normal sampling location inaccessible. The peak sample collection was performed approximately one-half mile downstream. This alternate location experienced overbank backflow conditions along the row crop and sod fields, which may account for the large increases in concentrations of selected trace elements (fig. 12). Barium concentrations were nearly twice as high in event samples collected at the downstream site as compared to samples from the upstream site (fig. 12). It is probable the source of barium in the Big River comes from the Mineral Fork and Mill Creek, which flow through the historic, open-pit barite (barium sulfate, also known as tiff) mining district in Washington County. Mineral Fork is the largest tributary to the Big River between the two study sites. In the case where both size fractions could be analyzed for trace-element concentrations, a total trace-element con centration was computed for the SSC sample. Concentrations of total trace elements (the computed total of concentrations from both the sand and fine fractions) in suspended sediments varied by site and event (table 10). Concentrations are likely affected by stream velocity, localized runoff, seasonality, and event duration and intensity. The maximum and mini mum concentrations at both sites for total barium, cadmium, lead, and zinc occurred during different events and in some instances, the maximum concentration for the upstream site was the same event for which a minimum concentration was detected at the downstream site (table 10). The maximum barium concentration at the upstream site was 0.613 mg/L during event 7, whereas the barium concentration at the downstream site during the same event (event number 6 for the downstream site) was the second-smallest concentration detected at 0.157 mg/L. The maximum lead concentration of 1.18 mg/L (the largest concentration detected during the study for both sites) occurred during event 4 (September 2012) at the upstream site; however, the minimum lead concentration detected at the downstream site during this same event on September 10, 2012 (downstream site event 3) was 0.174 mg/L. The difference in concentrations between the two sites during the same stormflow event could be the effect of limited sample collection on the stormflow hydro graphs, as well as more localized runoff in the headwaters of Big River. It is also possible that increased streamflow from Mineral Fork during the event diluted concentrations as the flows moved downstream. The maximum zinc concentration at the upstream site was the largest zinc concentration detected during the study for both sites at 0.901 mg/L, which occurred during event 4. The lowest concentration of both sites dur ing the study period was 0.074 mg/L and was detected during event 3 at the downstream site. The maximum concentrations of barium, cadmium, lead, and zinc at the downstream site all occurred during event number 4 (January 2013). The mean total trace-element concentrations in suspended sediments from stormflow events were compared between both sites during the study period. Mean concentrations of cadmium and lead were similar for both sites with mean cad mium concentrations of 0.004 mg/L at the upstream site and 0.002 mg/L at the downstream site, and mean lead concentra tions of 0.379 mg/L at the upstream site and 0.335 mg/L at the downstream site (table 10). Mean barium concentrations were higher at the downstream site (0.411 mg/L) than at the upstream site (0.184 mg/L). As noted previously, it is probable that a large source of barium enters the Big River from the Mineral Fork and Mill Creek, which drain a historical, openpit barite mining region. The mean zinc concentrations in the suspended sediments were higher upstream in the Big River than downstream, with a mean concentration of 0.291 mg/L at the upstream site, and mean concentration of 0.162 mg/L at the downstream site (table 10). Seasonal comparisons of event-based lead and zinc concentrations in suspended sediments also were used for analyses of sediment quality in the Big River, though the study period was not long enough to provide extensive data for comparisons. The seasonal comparison boxplots of SSC show that more event samples were collected in winter at the upstream site than at the downstream site (fig. 13). The large number of samples in winter at the upstream site is because the first event collected at the upstream site was a pilot sam pling event to ensure proper and consistent methods could be used during stormflow events and to determine the volume of sample needed to have adequate mass for trace-element analy ses. Spring was the most-sampled season at both sites during the study period, because stormflow events tend to occur in the wetter spring months. Median lead and zinc concentra tions collected in spring were similar at both sites, whereas median concentrations of lead and zinc were greatest in fall at the upstream site and greatest in winter at the downstream site (fig. 13). It is possible that the median concentrations in the fall were greater at the upstream site because the fall event from September 2012 was the first seasonal event in several months and the large amount of precipitation in the area of the site caused an increase in runoff. The winter event at the downstream site occurred in January 2013, after the removal
30 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Table 10. Streamflow, suspended-sediment concentrations and particle-size distributions, and selected total trace-element concentrations in suspended sediments collected during stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013. [ft3/s, cubic foot per second; mg/L, milligram per liter; %, percent; greater than; mm, millimeter; less than] Event number Sample date Sample time Stream flow (ft3/s) Suspended sediment Total trace element concentration in suspended sediments (mg/L) Concen tration (mg/L) Sand Fines Barium Cadmium Lead Zinc (% >0.063 mm) (% <0.063 mm) Big River below Bonne Terre, Missouri (07017610; upstream site) 12/15/2011 11:15 1,000 12/15/2011 11:40 12/15/2011 12:20 12/15/2011 13:50 12/15/2011 14:00 12/15/2011 15:00 03/16/2012 10:50 1,350 03/16/2012 15:15 1,370 03/17/2012 00:01 4,730 03/17/2012 16:40 1,690 04/16/2012 06:30 04/16/2012 10:40 09/8/2012 14:15 2,890 09/8/2012 16:40 3,390 01/30/2013 11:20 6,470 03/18/2013 11:30 20,100 03/18/2013 15:50 18,600 04/18/2013 13:32 11,600 1,091 04/19/2013 09:14 21,900 Median 1,370 Mean 5,300 Big River at Byrnesville, Missouri (07018500; downstream site) 03/17/2012 12:40 1,760 03/17/2012 20:45 3,380 03/18/2012 13:20 4,330 04/17/2012 10:25 2,410 09/10/2012 10:15 1,490 01/30/2013 16:45 6,050 01/31/2013 08:30 10,800 03/19/2013 18:00 27,300 04/18/2013 16:03 7,930 04/20/2013 16:15 26,700 Median 5,190 Mean 9,215 1Concentration computed from fines only. 2Concentrations computed from both sands and fines.
Suspended-Sediment Quality 31 Figure 13. Seasonal comparison of stormflow-event suspended-sediment concentrations and mass-accumulation lead and zinc concentrations within suspended sediments for two study sites on the Big River, Missouri, October 2011–September 2013. Winter Spring Summer Fall Winter Spring Summer Fall Winter Spring Summer Fall Winter Spring Summer Fall Big River at Byrnesville, Missouri (07018500; downstream site) o o o o o Total zinc concentration in suspended sediment, in milligrams per liter o o o o o 1,000 1,200 1,400 Suspended-sediment concentration, in milligrams per liter EXPLANATION Winter Spring Summer Fall Number of samples Upper adjacent 75th percentile 50th percentile (median) 25th percentile Lower adjacent Lower outside Lower detached Upper detached Upper outside December−February March–May June–August September–November Total lead concentration in suspended sediment, in milligrams per liter Big River below Bonne Terre, Missouri (07017610; upstream site) Winter Spring Summer Fall Winter Spring Summer Fall Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) Interquartile range of vegetation from the banks near the streamgage location, which could have increased the available suspended sediments during runoff. No events were sampled during summer months within the study period. Event-Based Loads and Yields of Selected Trace-Element Concentrations in Suspended Sediments Results of the four selected trace-element event-based average concentrations and the event-based load and yields are shown in table 11. Event-based loads computed at the upstream site were largest for all four constituents during April 2013 (event 7), with the lead load being the largest with 29.4 tons transported during the event. There was minimal decrease in the corresponding lead load at the downstream site, where the event-based load was 27.1 tons. The largest event-based lead load for the downstream site was 29.9 tons during event 5 in March 2013. An event-based yield also was computed to make the study sites more comparable by normalizing the drainage areas and event durations. Event-based yields were larger at the upstream site (table 11). Two events showed similar lead and zinc yields for both sites—April 2012 and January 2013. The largest event-based lead yield at the upstream site was 8.17 pounds per square mile per hour during event 7. The downstream site had smaller yields, with the largest eventbased lead yield being 1.76 pounds per square mile per hour during event 5.
32 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Table 11. Selected trace element concentrations, loads, and yields in suspended sediments from sampled stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013. [hr, hour; mg/L, milligrams per liter; lb, pound; mi2, square mile; lb/mi2/hr, pounds per square mile per hour] Event number Event date Event duration (hr) Trace element in suspended sediments Constituent Average event concentration (mg/L) Load Yield (lb/mi2/hr) (lb) (ton) Big River below Bonne Terre, Missouri (07017610; upstream site) Dec. 15, 2011 Barium Cadmium Lead Zinc Mar. 16–17, 2012 Barium 3,395 Cadmium Lead 5,944 Zinc 5,825 Apr. 16, 2012 Barium Cadmium Lead Zinc Sept. 8–9, 2012 Barium 1,885 Cadmium Lead 7,512 Zinc 5,536 Jan. 29–31, 2013 Barium 8,355 Cadmium Lead 12,930 Zinc 8,772 Mar. 17–19, 2013 Barium 14,776 Cadmium Lead 32,180 Zinc 16,737 Apr. 18–19, 2013 Barium 35,655 Cadmium Lead 58,846 Zinc 46,808 Annual total of sampled events Water year1 2012 Barium 5,619 Cadmium Lead 14,034 Zinc 11,945 Water year1 2013 Barium 58,785 Cadmium Lead 103,956 Zinc 72,317
Suspended-Sediment Quality 33 Table 11. Selected trace element concentrations, loads, and yields in suspended sediments from sampled stormflow events at two study sites on the Big River, Missouri, October 2011–September 2013.—Continued [hr, hour; mg/L, milligrams per liter; lb, pound; mi2, square mile; lb/mi2/hr, pounds per square mile per hour] Event number Event date Event duration (hr) Trace element in suspended sediments Constituent Average event concentration (mg/L) Load Yield (lb/mi2/hr) (lb) (ton) Big River at Byrnesville, Missouri (07018500; downstream site) Mar. 17–18, 2012 Barium 9,461 Cadmium Lead 7,649 Zinc 3,905 Apr. 15–18, 2012 Barium 7,158 Cadmium Lead 3,770 Zinc 2,100 Sept. 9–10, 2012 Barium Cadmium Lead 1,327 Zinc Jan. 30–31, 2013 Barium 28,357 Cadmium Lead 23,325 Zinc 10,580 Mar. 17–20, 2013 Barium 35,864 Cadmium Lead 59,814 Zinc 22,352 Apr. 18–20, 2013 Barium 75,034 Cadmium Lead 54,254 Zinc 31,576 Annual total of sampled events Water year1 2012 Barium 17,577 Cadmium Lead 12,745 Zinc 6,565 Water year1 2013 Barium 139,255 Cadmium Lead 137,393 Zinc 64,509 1Water year is defined as the 12-month period beginning October 1 and ending September 30. The water year is designated by the calendar year in which it ends and contains 9 of the 12 months.
34 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Loads computed during the study period for sampled events were higher at the downstream site and yields were higher at the upstream site. As previously noted, computed annual runoff was greater at the upstream site than the downstream site. In addition, stormflow events at the down stream site had much longer durations and lower velocities than events at the upstream site. It is possible that the lowhead dam just upstream from the sampling location at the downstream site slowed the stream velocity enough to cause sediments to fall out of suspension and be deposited on the streambed before reaching the sampling location. Other pos sible effects could be localized precipitation and fewer areas affected by mining activities in the downstream part of the Big River Basin. The additional volume of streamflow from Mineral Fork could dilute some trace-element concentrations downstream, though it is difficult to determine the amount of influence to the Big River because no streamgages are in operation on this tributary. The decreases in loads from the upstream to the downstream site also could be an artifact of the limited number of event samples that could be collected at the downstream site. Total trace-element loads and yields in suspended sedi ments were computed from the sampled events for each year in the study. Although no trace-element concentrations were collected during base-flow conditions, SSCs during base-flow typically were less than 50 mg/L and the majority of sus pended sediments are transported through the Big River during stormflow events. Given the affinity of trace elements to be transported with sediments (Horowitz and Elrick, 1987), it is likely that the majority of the selected trace elements in sus pended sediments are transported during runoff events as well. Using these assumptions, a total event-based load and yield for barium, cadmium, lead, and zinc for the events sampled were calculated for each water year (table 11). The total load and yield should be considered an estimate and a minimum repre sentation of the total trace-element loads because limited data were available for the computations and not all events during the study period were sampled or had sufficient material mass for analyses. All total loads and yields increased at each site for the four selected trace elements from the 2012 to the 2013 water years (table 11), most likely because of increased precipita tion and runoff during the 2013 water year. The total barium load in suspended sediments was higher for sampled events collected at the downstream site during both study years, with the 2013 water year having the most substantial increase in total load from 29.4 tons at the upstream site to 69.6 tons at the downstream site. Cadmium load decreased during the 2012 water year from 0.105 tons at the upstream site to 0.029 tons at the downstream site and decreased slightly downstream during the 2013 water year from 0.433 tons at the upstream site to 0.344 tons at the downstream site. Lead loads slightly decreased during the 2012 water year from 7.02 tons at the upstream site to 6.37 tons at the downstream site, and only slightly increased downstream during the 2013 water year from 52 tons at the upstream site to about 68.7 tons at the downstream site. Zinc loads were lower at the downstream site than the upstream site, although the decrease in total load was not substantial during either water year (table 11). Comparison of the total yield of the four selected trace elements based on all events in each water year showed a decrease in barium, cadmium, lead, and zinc yields from the upstream to down stream site, indicating readily-available sediment sources are closer to the upstream site (table 11). Seasonal comparisons of event-based loads and yields of lead and zinc concentrations in suspended sediments were similar between sites. The median loads and yields of lead and zinc were higher in spring at each site; however, the single winter event (event 4) at the downstream site had the highest median lead and zinc load during the study (fig. 14). Seasonal observation of the lead and zinc yields shows the single fall event (event 4) and the spring events at the upstream site have similar medians. Event-based loads and yields of lead and zinc in suspended sediments increased slightly from winter to spring at both sites and decreased greatly from spring to fall at the downstream site (fig. 14).
Suspended-Sediment Quality 35 Figure 14. Seasonal comparison of stormflow-event loads and yields of lead and zinc in suspended sediments at two study sites on the Big River, Missouri, October 2011–September 2013. o o o o o o o o o o o o Event-based lead load in suspended sediments, in tons Event-based zinc load in suspended sediments, in tons Event-based lead yield in suspended sediments, in pounds per square mile per hour Event-based zinc yield in suspended sediments, in pounds per square mile per hour Winter Spring Summer Fall Winter Spring Summer Fall Big River at Byrnesville, Missouri (07018500; downstream site) Winter Spring Summer Fall Winter Spring Summer Fall Winter Spring Summer Fall Winter Spring Summer Fall o o EXPLANATION Summer Fall Number of samples Upper adjacent 75th percentile 50th percentile (median) 25th percentile Lower adjacent Lower outside Lower detached Upper detached Upper outside December−February March–May June–August September–November Big River below Bonne Terre, Missouri (07017610; upstream site) Winter Spring Summer Fall Winter Spring Summer Fall Big River at Byrnesville, Missouri (07018500; downstream site) Big River below Bonne Terre, Missouri (07017610; upstream site) Interquartile range Winter Spring
36 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 Summary Missouri was the leading producer of lead in the United States—as well as the world—for more than a century. One of the lead sources is known as the Old Lead Belt, located in southeast Missouri. The primary ore mineral in the region is galena, which can be found both in surface deposits and underground as deep as 200 feet. More than 8.5 million tons of lead were produced from the Old Lead Belt before opera tions ceased in 1972. Although active lead mining has ended, the effects of mining activities still remain in the form of large mine waste piles on the landscape typically near tributaries and the main stem of the Big River, which drains the Old Lead Belt. Six large mine waste piles, some spanning more than a mile in diameter, exist within the Big River Basin. For the past century, the piles have been sources of lead and zinc-rich sedi ments to be transported by natural fluvial processes and wind downstream into the Big River. The U.S. Environmental Pro tection Agency (EPA) Region 7 completed 4-year reclamation efforts to cap and prevent further erosion of the piles in 2012. A study was conducted by the U.S. Geological Survey in cooperation with EPA Region 7 to assess the amount and availability of suspended sediments and to assess the traceelement concentrations of suspended sediments transported through the Big River basin after reclamation of the mine waste piles. Streamflow and suspended sediments were quanti fied and sampled at two locations in the basin: Big River below Bonne Terre, Mo. (USGS site number 07017610; also referred to as the upstream site), located less than 5 miles downstream from the Old Lead Belt, and Big River near Byrnesville, Mo. (USGS site number 07018500; also referred to as the downstream site), located about 68 miles downstream from Bonne Terre. The Big River discharges into the Meramec River, which discharges into the Mississippi River, south of St. Louis, Mo. Discrete suspended-sediment concentration (SSC) sam ples were used to develop a regression model with continu ous turbidity measurements and (or) streamflow to compute time-series SSC and suspended-sediment load (SSL) values. Discrete SSC samples were paired with corresponding turbid ity and streamflow measurements made at about the mean time of the SSC sample and plotted to identify outliers and trends. The SSC, SSL, and sediment yields were computed for hydrologic events when cross-sectional sampling and streamflow measurements were collected. Seven events were sampled at the upstream site and six events were sampled at the downstream site during the study period. Streamflows were greatest in the spring months and fol lowed similar temporal variability at both sites. The maximum peak streamflow during the study period at the upstream site was 21,900 cubic feet per second (ft3/s) on April 19, 2013, and was 28,700 ft3/s on April 20, 2013 at the downstream site. Water temperature ranged from -0.1 to about 33 degrees Celsius (oC) at both sites. Turbidity ranged from 1.0 to 890 formazin nephelometric units (FNU). Base-flow SSC values ranged from 1 to 74 milligram per liter (mg/L) at the upstream site during the study period, and event SSC values ranged from 74 to 1,091 mg/L. At the downstream site, base-flow SSC ranged from 4 to 127 mg/L and event SSC ranged from 27 to 871 mg/L. The upstream and downstream sites had similar SSC and SSL extremes during the study, but these extremes did not occur during the same events. The annual SSL increased greatly during the study period, as the first water year (2012) was classified as a drought year and more precipitation was recorded during the 2013 water year. The upstream site had an annual SSL of 10,035 tons during the 2012 water year, and 191,294 tons for the 2013 water year. The downstream site had an annual SSL of 21,612 tons during the 2012 water year, and 333,717 tons for the 2013 water year. An event-based average SSC and SSL were computed to help determine the mass of sediments transported during each sampled event. Event SSCs ranged from approximately 100 mg/L to 1,400 mg/L at both study sites and event SSLs ranged from about 180 to 32,000 tons at the upstream site, and from about 390 to 53,000 tons at the downstream site. Event SSLs were greatest at the downstream site. Event loads and yields were greater during water year 2013 events, with the April 2013 event having the greatest loads and yields at both sites. Although a limited number of events could be sampled, using the continuous turbidity data to compute the daily SSL in conjunction with event sampling provided adequate data to determine the majority of the SSL in Big River occurs dur ing stormflow events. Not all stormflow events that occurred during the study period could be sampled, yet the events that were sampled make up a considerable portion of the annual SSL. Discrete SSCs collected during base-flow conditions during the study period were generally very low (typically less than 50 mg/L). The base-flow average SSC was used with a median of daily median streamflow values to determine a base flow SSL. The base flow SSL at the upstream site was about 18 tons, or less than 0.01 percent of the SSL computed by the regression model during the study period. The event-based SSL from the seven sampled events during the study period was about 57,700 tons, accounting for about 29 percent of the SSL at the upstream site. At the downstream site, the com puted base flow SSL was about 46 tons for the study period, which accounts for about 0.01 percent of the SSL, and the SSL from the six sampled events was about 103,000 tons, account ing for about 29 percent of the SSL computed by the regres sion model during the study period. The total concentrations of barium, cadmium, lead, and zinc computed for each individual event sample were used to determine an event-based arithmetic average concentration to determine an event-based concentration, load (flux), and
Summary 37 yield of each constituent. The arithmetic average trace-element concentrations were computed from all available samples during the designated event duration. For events with enough mass in both the sand and fines fraction for analysis, it was noted that nearly all sand fraction concentrations were less than the concentrations of the fines fraction for all four trace elements. Barium concentrations were nearly twice as high in event samples collected at the downstream site as compared to samples from the upstream site. The likely source of barium in the Big River is the Mineral Fork and Mill Creek, which flow through the historical barite (barium sulfate, also known as tiff) mining district in Washington County. Cadmium and zinc concentrations were nearly two times higher at the upstream site than at the downstream site, while lead con centrations appeared similar between both sites. It is possible the suspended sediments containing cadmium, lead, and zinc have decreased concentrations downstream because slower velocities cause the sediments to fall from suspension as they are transported downstream farther from their source. Traceelement concentrations in suspended sediment sampled during stormflow events during a previous study had higher con centrations of lead and zinc than concentrations measured in suspended sediments collected during events at the upstream site. Although the previous study concentrations were higher (1,100 to 3,200 milligram per kilogram [mg/kg] of lead; 1,100 to 2,200 mg/kg of zinc) than those measured at the upstream site (680 to 1,400 mg/kg of lead; 470 to 1,200 mg/kg of zinc), there are no indications of long-term or substantial decreases as data are limited and could be affected by differences in sampling location, event intensity, changes in the geomorphol ogy of the stream, or remediation efforts of mine waste piles. Sediment quality guidelines for freshwater describe two qualities of environmental effects from trace-element concen trations in suspended sediments—the threshold effect concen tration (TEC) and the probable effect concentration (PEC)— and help regulatory agencies assess the toxicity of sediments within a remediation study. Concentrations of cadmium, lead, and zinc in suspended sediment samples collected during stormflow events at both sites exceeded the consensus-based TEC and consensus-based PEC. All lead and zinc concentra tions in the fine fraction of all suspended sediment from both sites greatly exceeded the consensus-based TEC (35.8 mg/kg for lead; 121 mg/kg for zinc) and the PEC (128 mg/kg for lead; 459 mg/kg for zinc). Lead concentrations in the fine fraction of all samples at both sites also exceeded the toxic effect threshold (TET) of 170 mg/kg, above which sediment is considered to be heavily polluted causing adverse effects on sediment-dwelling organisms. Zinc concentrations in the fine fraction in four of five sampled events from the upstream site also exceeded the TET of 540 mg/kg. Concentrations of cadmium and zinc were notably higher in suspended sediment samples from the upstream site compared to the downstream site, especially in the fine fraction. The smaller cadmium and zinc concentrations at the downstream site could be a result of a decrease in source material enriched in these trace elements in the downstream areas of the watershed compared to the upstream area within the Old Lead Belt. The lead concentrations greatly exceeded the PEC as well as the TET, in some samples by a factor of 10, which indicates that lead-rich suspended sediments, especially in the fine fraction, are readily available for transport within the Big River Basin. These lead-enriched, finer sediments could remain in the system from historical mining, and as the capping of mine waste piles upstream from Bonne Terre continues to reduce additional sediment loadings, these fine sediments may be continually released as the river scours the streambed and erodes the stream banks, causing lead-rich suspended sediment to remain in a state of equilibrium. Event-based trace-element load and yields in suspended sediments also were used for analyses of sediment quality in the Big River. The total barium load in suspended sediments was higher for sampled events collected at the downstream site during both study years. Cadmium and zinc loads in sus pended sediments were lower at the downstream site than the upstream site, although the decrease in total load was not sub stantial during the study period. Lead loads in suspended sedi ments were lower at the downstream site during the first study year, with a slightly higher load downstream in the second year though the increase from upstream to downstream was small. Event yields were higher at the upstream site. Storm events at the downstream site had much longer durations and lower velocities than events at the upstream site. All four trace element total loads and yields in suspended sediments com puted from event samples increased at each site from the 2012 to the 2013 water years, most likely because of increased pre cipitation and runoff during the 2013 water year. Total loads of barium, cadmium, and lead were higher at the downstream site than the upstream site during both water years. Comparison of the total yield of the four selected trace elements based on all events in each water year showed a decrease in barium, cad mium, lead, and zinc yields from the upstream to downstream site, indicating readily-available sediment sources are closer to the upstream site. Median lead and zinc concentrations collected in spring were similar at both sites, whereas median concentrations of lead and zinc were greatest in fall at the upstream site and greatest in winter at the downstream site. It is possible that the median concentrations in the fall were greater at the upstream site because of one large event in September 2012. The winter event at the downstream site occurred in January 2013, after the removal of vegetation from the banks near the streamgage location, which could have increased the available suspended sediments during runoff. Seasonal comparison of event-based loads computed at the upstream site were largest for all four constituents during April 2013 (event 7), with the lead load being the largest with 29.4 tons transported during the event.
38 Surface-Water Quality and Suspended-Sediment Quantity and Quality, Big River Basin, Southeastern Missouri, 2011–13 References Cited Anderson, C.W., 2005, Turbidity (version 2.1): U.S. Geologi cal Survey Techniques of Water-Resources Investigations, book 9, chap. A6, sec. 6.7, 55 p., accessed October 2014 at ://pubs.water.usgs.gov/twri9A6/. ASTM International, 2011, ASTM D7726-11, Standard guide for the use of various turbidimeter technologies for mea surement of turbidity in water: West Conshohocken, Pa., ASTM International, accessed November 2014 at :// www.astm.org/Standards/D7726.htm. Barr, M.N., 2009, Effects of the upper Taum Sauk Reservoir embankment breach on the surface-water quality and sedi ments of the East Fork Black River and the Black River, southeastern Missouri—2006-07: U.S. Geological Survey Scientific Investigations Report 2009–5111, 59 p. Charlton, R., 2008, Fundamentals of fluvial geomorphology: New York, Routledge Publishing, 234 p. Davis, B.E., 2005, A guide to the proper selection and use of federally approved sediment and water-quality samplers: U.S. Geological Survey Open-File Report 2005–1087, 20 p. Davis, J.V., and Bell, R.W., 1998, Water-quality assessment of the Ozark Plateaus study unit, Arkansas, Kansas, Mis souri, and Oklahoma—Nutrients, bacteria, organic car bon, and suspended sediment in surface water, 1993–95: U.S. Geological Survey Water-Resources Investigations Report 98–4164, 56 p. Edwards, T.K., and Glysson, G.D., 1999, Field methods for measurement of fluvial sediment: U.S. Geological Survey Techniques of Water-Resources Investigation, book 3, chap. C2, 80 p. Fenneman, N.M., 1938, Physiography of eastern United States: New York, McGraw-Hill Book Co., Inc., 689 p. Guy, H.P., 1969, Laboratory theory and methods for sediment analysis: U.S. Geological Survey Techniques of WaterResources Investigations, book 5, chap. C1, 58 p. Horowitz, A.J., and Elrick, K.A., 1987, The relation of stream sediment surface area, grainsize and composition to trace element chemistry: Applied Geochemistry, v. 2, p. 437–451. MacDonald, D.D., Ingersoll, C.G., and Berger, T.A., 2000, Development and evaluation of consensus-based sediment quality guidelines for freshwater ecosystems: Archives of Environmental Contamination and Toxicology, v. 39, no. 1, p. 20–31. Missouri Department of Natural Resources, 2013, Our Mis souri waters—Big River watershed fact sheet: accessed November 2014 at ://dnr.mo.gov/omw/index.. Missouri Department of Natural Resources, 2014, Hazardous Waste Program, Missouri lead mining history by county: accessed December 2014 at ://dnr.mo.gov/env/hwp/ sfund/lead-mo-history-more.htm. Mosby, D.E., and Weber, J.S., 2009, Final phase I damage assessment plan for southeast Missouri lead mining dis trict—Big River mine tailings superfund site, St. Francois county and Viburnum Trend sites, Reynolds, Crawford, Washington, and Iron counties: U.S. Fish and Wildlife Service, accessed December 2014 at ://dnr.mo.gov/env/ hwp/docs/semofinaldraft.pdf. Mueller, D.S., Wagner, C.R., Rehmel, M.S., Oberg, K.A., and Rainville, Francois, 2013, Measuring discharge with acous tic Dopplar current profilers from a moving boat (version 2.0, December 2013): U.S. Geological Survey Techniques and Methods, book 3, chap. A22, 95 p., accessed November 2014 at ://pubs.usgs.gov/tm/3a22/. National Drought Mitigation Center, University of NebraskaLincoln, 2013, U.S. drought monitor data archives: accessed October 2013 at ://droughtmonitor.unl.edu. National Oceanic Atmospheric Administration, 2013, 2012 Missouri climate summary: National Climatic Data Center database, accessed August 2013 at ://www..noaa. gov/cag/. National Oceanic and Atmospheric Administration, 2014a, State of the climate—National overview for annual 2013: accessed August 2014 at ://www..noaa.gov/sotc/ national/2013/13. National Oceanic Atmospheric Administration, 2014b, 2013 Missouri climate summary: National Climatic Data Center database, accessed August 2014 at ://www..noaa. gov/cag/. National Oceanic Atmospheric Administration, 2015, Monthly precipitation summaries of the Global Historical Climatol ogy Network-Daily subsets GHCND:US1MOJF0013 and GHCND:USC00232220: National Oceanic Atmospheric Administration National Climatic Data Center, accessed January 2015 at ://www..noaa.gov/cdo-web/ search?datasetid=GHCNDMS. Oberg, K.A., Morlock, S.E., and Caldwell, W.S., 2005, Quality-assurance plan for discharge measurements using acoustic Doppler current profilers: U.S. Geological Survey Scientific Investigations Report 2005–5183, 46 p. Rantz, S.E., and others, 1982, Measurement and computation of streamflow—Volume 1, Measurement of stage and dis charge: U.S. Geological Survey Water Supply Paper 2175, 284 p.
References Cited 39 Rasmussen, P.P., Gray, J.R., Glysson, G.D., and Ziegler, A.C., 2009, Guidelines and procedures for computing time-series suspended-sediment concentrations and loads from instream turbidity-sensor and streamflow data (revised 2011): U.S. Geological Survey Techniques and Methods, book 3, chap. C4, 52 p. Sauer, V.B., 2002, Standards for the analysis and process ing of surface-water data and information using electronic methods: U.S. Geological Survey Water-Resources Investi gations Report 01–4044, 91 p. Smith, B.J., and Schumacher, J.G., 1991, Hydrochemical and sediment data for the Old Lead Belt, southeastern Mis souri—1988–89: U.S. Geological Survey Water-Resources Investigations Report 91–211, 98 p. Smith, B.J., and Schumacher, J.G., 1993, Surface-water and sediment quality in the Old Lead Belt, southeastern Mis souri—1988–89: U.S. Geological Survey Water-Resources Investigations Report 93–4012, 92 p. Taggart, J.E., 2002, Analytical methods for chemical analysis of geologic and other materials: U.S. Geological Survey Open-File Report 02–223, 20 p. Turnipseed, D.P., and Sauer, V.B., 2010, Discharge measure ments at streamflow-gaging stations: U.S. Geological Sur vey Techniques and Methods, book 3, chap. A8, 87 p. U.S. Geological Survey, 2006, Collection of water samples (ver. 2.0): U.S. Geological Survey Techniques of WaterResources Investigations, book 9, chap. A4, accessed May 2012 at ://pubs.water.usgs.gov/twri9A4/. U.S. Geological Survey, 2015, National Water Information System (NWISWeb): U.S. Geological Survey database, accessed May 2013 at ://nwis.waterdata.usgs.gov/mo/ nwis. Wagner, R.J., Boulger, R.W., Jr., Oblinger, C.J., and Smith, B.A., 2006, Guidelines and standard procedures for con tinuous water-quality monitors—Station operation, record computation, and data reporting: U.S. Geological Survey Techniques and Methods, book 1, chap. D3, 51 p. plus 8 attachments, accessed May 2012 at ://pubs.water.usgs. gov/tm1d3. Wilde, F.D., Radtke, D.B., Gibs, Jacob, and Iwatsubo, R.T., eds., 2004 with updates through 2009, Processing of water samples (ver. 2.2): U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A5, April 2004, accessed May 2012 at ://pubs.water.usgs.gov/ twri9A5/. Wilde, F.D., 2006, Temperature (ver. 2): U.S. Geological Sur vey Techniques of Water-Resources Investigations, book 9, chap. A6, sec. 6.1, accessed May 2012 at ://pubs.water. usgs.gov/twri9A6/. Wilde, F.D., Sandstrom, M.W., and Skrobialowski, S.C., 2014, Selection of equipment for water sampling (ver. 3.1): U.S. Geological Survey Techniques of Water-Resources Investigations, book 9, chap. A2, April 2014 accessed May 2012 at ://pubs.water.usgs.gov/twri9A2/.
Publishing support provided by: Rolla Publishing Service Center For more information concerning this publication, contact: Director, Missouri Water Science Center U.S. Geological Survey 1400 Independence Road, MS-100 Rolla, MO 65401 (573) 308-3667 Or visit the Missouri Water Science Center Web site at: ://mo.water.usgs.gov/
Barr—Surface-Water Quality and Suspended-Sediment Quantity and Quality within the Big River Basin, Southeastern Missouri, 2011–13—SIR 2015–5171 ISSN 2328-0328 (online) ://dx.doi.org/10.3133/sir20155171