U.S. Geological Survey
Water-Resources Investigations Report 02-4048

Prepared in cooperation with the
KANSAS DEPARTMENT OF HEALTH AND ENVIRONMENT

Download this report as a PDF file (2.09 MB)

Download a free copy of Acrobat Reader

Sediment Deposition and Occurrence of Selected Nutrients and Other Chemical Constituents in Bottom Sediment, Tuttle Creek Lake, Northeast Kansas, 1962-99

By Kyle E. Juracek and David P. Mau

Tuttle Creek Lake

CONTENTS

Abstract
Introduction

Description of Tuttle Creek Lake Basin
Acknowledgments

Methods

Bathymetric Survey
Estimation of Bottom-Sediment Volume, Mass, and Mean Annual Net Deposition and Yield
Sediment-Core Collection, Processing, and Analysis
Quality Control
Trend Analysis

Sediment Deposition in Tuttle Creek Lake
Concentrations and Trends in Selected Chemical Constituents

Nutrients and Total Organic Carbon
Metals and Trace Elements
Organochlorine Compounds

Comparison of Depositional Trends to Streamflow and Human Activity
Summary and Conclusions
References
Supplemental Information

For additional information about Reservoir Sediment in Kansas, please visit:

FIGURES

1-4. Maps showing:

Figure 1. Location of Big Blue River Basin, physiography, streamflow-gaging stations, and Tuttle Creek Lake, northeast Kansas
Figure 2. Location of bathymetric range lines and sediment-coring sites in Tuttle Creek Lake
Figure 3. Depth-weighted, mean soil permeability in Big Blue River Basin
Figure 4. Land use in Big Blue River Basin, circa 1992

Figure 5. Graph showing variation in cesium-137 concentrations with depth of bottom-sediment core samples collected from site TIC-2 in Tuttle Creek Lake, September 1999
Figure 6. Graph showing estimated bottom-sediment thickness in submerged Big Blue River channel of Tuttle Creek Lake, 1999
Figure 7. Graph showing estimated bottom-sediment thickness on submerged Big Blue River flood plain in Tuttle Creek Lake, 1999
Figure 8. Map showing mean concentrations of total ammonia plus organic nitrogen in bottom-sediment cores collected from Tuttle Creek Lake, 1999
Figure 9. Graph showing variation in total phosphorus concentration with depth of bottom-sediment core samples collected from site TIC-2 in Tuttle Creek Lake, September 1999
Figure 10. Map showing mean concentrations of total phosphorus in bottom-sediment cores collected from Tuttle Creek Lake, 1999
Figure 11. Graph showing variation in total organic carbon concentration with depth of bottom-sediment core samples collected from site TIC-2 in Tuttle Creek Lake, September 1999
Figure 12. Map showing mean concentrations of total organic carbon in bottom-sediment cores collected from Tuttle Creek Lake, 1999
13-16. Graph showing variation in concentrations with depth of bottom-sediment core samples collected from site TIC-2 in Tuttle Creek Lake, September 1999, for:

Figure 13. Chromium, cobalt, gallium, and iron
Figure 14. Lithium, magnesium, mercury, and molybdenum
Figure 15. Nickel, scandium, selenium, and tantalum
Figure 16. Tin and zinc

17-22. Map showing mean concentrations in bottom-sediment cores collected from Tuttle Creek Lake, August and September 1999, for:

Figure 17. Arsenic
Figure 18. Chromium
Figure 19. Copper
Figure 20. Nickel
Figure 21. Silver
Figure 22. Zinc

Figure 23. Graph showing variation in concentrations with depth of bottom-sediment core samples collected from site TIC-2 in Tuttle Creek Lake, September 1999, for DDE
Figure 24. Graph showing estimated population in Big Blue River Basin, 1960-2000
25-29. Graphs showing approximate number of acres harvested and irrigated in Big Blue River Basin:

Figure 25. 1960-99, for alfalfa
Figure 26. 1960-2000, for grain corn
Figure 27. 1960-2000, for grain sorghum
Figure 28. 1960-2000, for soybeans
Figure 29. 1960-2000, for wheat

Figure 30. Graph showing approximate number of cattle in Big Blue River Basin, 1960-2000
Figure 31. Graph showing approximate number of hogs in Big Blue River Basin, 1960-98
32-41. Graphs showing comparison of 1962 U.S. Army Corps of Engineers bathymetric data and 1999 U.S. Geological Survey bathymetric data for Tuttle Creek Lake:

Figure 32. Range line 1
Figure 33. Range line 2
Figure 34. Range line 3
Figure 35. Range line 4
Figure 36. Range line 5
Figure 37. Range line 6
Figure 38. Range line 7
Figure 39. Range line 8
Figure 40. Range line 9
Figure 41. Range line 13

42-50. Graphs showing comparison of 1962 and 2000 U.S. Army Corps of Engineers bathymetric data for Tuttle Creek Lake:

Figure 42. Range line 10
Figure 43. Range line 14
Figure 44. Range line 15
Figure 45. Range line 16
Figure 46. Range line 17
Figure 47. Range line 18
Figure 48. Range line 19
Figure 49. Range line 20
Figure 50. Range line 21

TABLES

Table 1. Analyses performed on bottom-sediment samples from Tuttle Creek Lake, northeast Kansas
Table 2. Mean relative percentage differences for constituent concentrations in sequential replicate samples of bottom sediment from coring sites TOC-1A, TOC-3A, and TOC-7A in Tuttle Creek Lake, northeast Kansas, August and September 1999
Table 3. Mean relative percentage differences for constituent concentrations in split replicate samples of bottom sediment from coring sites TOC-2A, TOC-4A, and TOC-5A in Tuttle Creek Lake, northeast Kansas, August and September 1999
Table 4. Estimated channel length, mean channel width, mean bottom-sediment thickness, and computed bottom-sediment volume in submerged in-channel components of Tuttle Creek Lake, northeast Kansas, 1999
Table 5. Estimated surface area, mean bottom-sediment thickness, and computed bottom-sediment volume in out-of-channel components of Tuttle Creek Lake, northeast Kansas, 1999
Table 6. Estimated mean bulk density of bottom sediment at in- and out-of-channel coring sites in Tuttle Creek Lake, northeast Kansas, 1999
Table 7. Estimated bottom-sediment volume, representative bulk density, and computed bottom-sediment mass in submerged in-channel components of Tuttle Creek Lake, northeast Kansas, 1999
Table 8. Estimated bottom-sediment volume, representative bulk density, and computed bottom-sediment mass in out-of-channel components of Tuttle Creek Lake, northeast Kansas, 1999
Table 9. Statistical summary of concentrations, and comparison to sediment-quality guidelines, for selected constituents in bottom-sediment samples from Tuttle Creek Lake, northeast Kansas, August and September 1999
Table 10. Estimated mean annual net loads and yields of constituents deposited in bottom sediment of Tuttle Creek Lake, northeast Kansas
Table 11. Results of trend tests on concentrations of selected constituents in bottom-sediment core samples collected from site TIC-2 in Tuttle Creek Lake, northeast Kansas, September 1999
Table 12. Latitude and longitude coordinates for end points of U.S. Army Corps of Engineers range lines used in bathymetric surveys of Tuttle Creek Lake, northeast Kansas
Table 13. Latitude and longitude coordinates for sediment-coring sites in Tuttle Creek Lake, northeast Kansas

CONVERSION FACTORS, ABBREVIATIONS, AND DEFINITION

Temperature can be converted to degrees Celsius (÷C) or degrees Fahrenheit (÷F) by the equations:

Sea level: In this report, "sea level" refers to the National Geodetic Vertical Datum of 1929-a geodetic datum derived from a general adjustment of the first-order level nets of the United States and Canada, formerly called Sea Level Datum of 1929.

Sediment Deposition and Occurrence of Selected Nutrients and Other Chemical Constituents in Bottom Sediment, Tuttle Creek Lake, Northeast Kansas, 1962-99

Abstract

INTRODUCTION

Tuttle Creek Lake is a Federal impoundment on the Big Blue River in Riley, Pottawatomie, and Marshall (only at flood pool) Counties, northeast Kansas (fig. 1). Although the reservoir was officially completed in 1962 by the U.S. Army Corps of Engineers (USCOE), water storage actually began in 1959. The reservoir is used for flood control, recreation, water supply, fish and wildlife habitat, water-quality control, and navigation supplementation.

Some early water-quality information for Tuttle Creek Lake is available from USCOE (U.S. Army Corps of Engineers, circa 1966). Additional historical information on water quality and (or) streambed sediment quality for the Big Blue River Basin is available from several sources including Tanner and others (1990), Fallon and McChesney (1993), Frankforter (1994), Stamer and Zelt (1994), Bevans and others (1995), and Jordan and Stamer (1995). From 1996 through 2000, USCOE sampled the water in Tuttle Creek Lake and analyzed it for nutrients (total nitrogen and total phosphorus) and herbicides (alachlor, atrazine, cyanazine, and metolachlor) (U.S. Army Corps of Engineers, 2000).

During a 5-year period from 1993-97, the water in Tuttle Creek Lake was sampled by the Kansas Department of Health and Environment (KDHE) and analyzed for selected organic and inorganic constituents. Detections of the herbicides alachlor and atrazine exceeded water-quality standards (Tom Stiles, Kansas Department of Health and Environment, oral commun., 2000), which resulted in the listing of Tuttle Creek Lake under Section 303(d) of the Federal Clean Water Act of 1972. The 303(d) list is a priority list that identifies water bodies that do not meet water-quality standards developed on the basis of the use of the water bodies. For each impaired water body on the 303(d) list, a State is required by the Clean Water Act to develop a total maximum daily load (TMDL), which is an estimate of the maximum pollutant load (material transported during a specified time period) from point and nonpoint sources that a receiving water can accept without exceeding water-quality standards (U.S. Environmental Protection Agency, 1991). Other concerns affecting the reservoir, in addition to alachlor and atrazine, were sedimentation, eutrophication, and bacteria (Tom Stiles, Kansas Department of Health and Environment, oral commun., 2000).

A 2-year study by the U.S. Geological Survey (USGS), in cooperation with KDHE, was begun in 1999 to estimate sedimentation in Tuttle Creek Lake as well as to determine the deposition of, and trends in, various chemical constituents. The specific study objectives were to:

1. Estimate the volume and mass of bottom sediment deposited in the reservoir as well as the mean annual net deposition and yield since the original USCOE bathymetric survey of Tuttle Creek Lake in 1962;
2. Determine the occurrence and mass of, as well as trends in, selected chemical constituents in the bottom sediment;
3. Determine the mean annual net load and yield for selected chemical constituents in the bottom sediment;
4. Relate, to the extent possible, any observed constituent trends in the bottom sediment to documented historical changes in human activity in the basin; and
5. Provide a baseline of information on reservoir conditions with which to compare future conditions that may represent a response to changes in human activity in the basin.

The purpose of this report is to present the results of the USGS study to estimate sediment deposition and occurrence of selected chemical constituents in the bottom sediment of Tuttle Creek Lake. Results of this study will assist KDHE in evaluating the implementation of existing TMDLs and in developing new TMDLs for constituents found to contribute to water-quality impairment in Tuttle Creek Lake. From a national perspective, the methods and results presented in this report provide guidance and perspective for future reservoir studies concerned with the issues of sedimentation and water quality.

Description of Tuttle Creek Lake Basin

The Tuttle Creek Lake Basin, which is essentially synonymous with the Big Blue River Basin (except for the small area located downstream from the dam), is an area of 9,628 miî that drains parts of southeast Nebraska and northeast Kansas (fig. 1). About 75 percent of the basin is located in Nebraska. In 2000, the lake had a surface area of about 12,620 acres and a water-storage capacity of about 280,000 acre-ft at the conservation-pool elevation of 1,075 ft above sea level (Phil Snell, U.S. Army Corps of Engineers, written commun., 2000). Originally, the lake had a surface area of about 15,830 acres and a water-storage capacity of about 425,000 acre-ft at the conservation-pool elevation (U.S. Army Corps of Engineers, 1966). The decreases in surface area and storage capacity are due to the effects of ongoing sedimentation. At the flood-control pool elevation of 1,136 ft above sea level, the lake in 2000 had a surface area of about 53,050 acres and a water-storage capacity of about 2,150,000 acre-ft (Phil Snell, U.S. Army Corps of Engineers, written commun., 2000). In addition to the Big Blue River, a number of small tributaries contribute flow directly to Tuttle Creek Lake including Baldwin, Booth, Carnahan, Dry, Fancy, McIntire, Mill, and Tuttle Creeks (fig. 2).

Physiographically, the Big Blue River Basin can be characterized with reference to physical divisions as defined by Fenneman (1946) (fig. 1). The upstream half of the basin is located mostly in the High Plains and Plains Border Sections of the Great Plains Province. The High Plains Section is typified by flat plains with limited stream dissection and little local relief. This section is underlain by fluvial (stream) and eolian (windblown) deposits that consist of sand, gravel, silt, and clay. The Plains Border Section is more dissected than the High Plains Section and thus has greater local relief. This section is underlain by shale, sandstone, and limestone, with minor fluvial and eolian deposits. The downstream half of the basin is located mostly in the Dissected Till Plains Section of the Central Lowland Province. This section is characterized by dissected deposits of glacial till that consist of silt, clay, sand, gravel, and boulders that overlie bedrock of primarily shale and limestone, with some sandstone (Jordan and Stamer, 1995).

Maximum local relief is from 300 to 500 ft in the downstream part of the Big Blue River Basin and generally less than 300 ft elsewhere. Slopes in the basin range from typically less than 3 percent in the High Plains Section to as much as 10 percent or more elsewhere in the basin (Jordan and Stamer, 1995). Slope, along with soil permeability and land use (discussed in the following paragraphs), are important determinants of storm runoff in a basin. In general, soil erodibility is somewhat higher in the upstream half of the Big Blue River Basin (U.S. Department of Agriculture, 1994).

Depth-weighted, mean soil permeability in the Big Blue River Basin ranges from about 0.3 to 13. 0 in/hr with a mean of about 0.9 in/hr. In general, soil permeability is less in the uplands (typically less than 1.0 in/hr) and somewhat greater in the flood plains of the principal rivers and streams (typically between 1.0 and 1.5 in/hr). Isolated areas with much greater soil permeability (4.0 to 13.0 in/hr) mostly are located along the Little Blue River and its tributaries (fig. 3) (U.S. Department of Agriculture, 1994).

Long-term mean annual precipitation in the Big Blue River Basin ranges from about 28 in. at Hastings, Nebraska (period of record 1948-99), in the northwest part of the basin (fig. 1), to about 33 in. at Manhattan, Kansas (period of record 1900-99), in the southeast (High Plains Regional Climate Center, 2000). Most of the annual precipitation is received during the growing season (generally April-September).

Land use (circa 1992) in the Big Blue River Basin is mostly agricultural with cropland accounting for about 66 percent of the basin. Grassland and pasture account for about 29 percent of the basin. Woodland and urban land use cover about 3 and 1 percent of the basin, respectively (fig. 4) (U.S. Geological Survey, 2000).

Acknowledgments

METHODS

Bathymetric Survey

To provide the additional information necessary for estimating the volume of bottom sediment in Tuttle Creek Lake, a bathymetric (lake-bed elevation) survey was performed by the USGS during June and July 1999. The bathymetric survey involved the use of global-positioning-system (GPS) technology to record the geographic location of the boat on the lake and a fathometer system to determine the depth to the sediment/water interface (top of lake bed). The GPS and fathometer data were recorded digitally using a data-logging unit. The bathymetric survey was conducted along 10 range lines (1-9, 13) that were established by USCOE in 1962 (fig. 2; see also figs. 32-41 in the "Supplemental Information" section at the back of this report). The latitude and longitude coordinates for the end points of USCOE range lines used in this study are provided in table 12 in the "Supplemental Information" section. The reliability of the fathometer was verified at the start and end of each day (weather conditions permitting) by suspending a metal plate at known depths directly below the transducer. Additional bathymetric data from a USCOE survey of the lake in September 2000 also were used to provide information for nine range lines (10, 14-21) that were not surveyed in 1999 due to insufficient lake levels (fig. 2; see also figs. 42-50 in the "Supplemental Information" section). Given that 2000 was a very dry year (with minimal sediment inflow to the lake), the 1999 and 2000 bathymetrically derived lake-bed elevations may be considered comparable for the purposes of estimating total sediment deposition within the lake. The upstream limits of the original conservation-pool surface area of the lake were defined approximately by range lines 10 and 21 (fig. 2).

Estimation of Bottom-Sediment Volume, Mass, and Mean Annual Net Deposition and Yield

Total bottom-sediment volume (sediment plus water and gases) in Tuttle Creek Lake was estimated using a partitioning approach in which the original conservation-pool surface area of the lake (15,830 acres) was divided into segments as determined by the locations where the bathymetric range lines crossed the lake (fig. 2). These segments were divided further into in-channel and out-of-channel components to improve the precision of the bottom-sediment volume estimates. Bottom-sediment volume was computed separately for all components as the total surface area multiplied by the mean thickness of the bottom sediment.

The total conservation-pool surface area for each segment was determined by digitizing the lake boundary and range lines using USGS 1:24,000-scale topographic quadrangles as the source maps. Mean channel widths of the Big Blue River and Fancy Creek were estimated using USCOE 1962 bathymetric information. With two exceptions, the mean channel width for each segment was computed as the average of the channel widths determined using the two range lines that defined the segment. The exceptions were the segments from the dam to range line 1 and from range line 9 to the confluence with the submerged Big Blue River channel (fig. 2), for which the respective channel widths computed for range lines 1 and 9 were used. The main channel length for each segment was estimated by measuring the channel centerline on USGS 1:24,000-scale topographic quadrangles. The surface area for each in-channel component was computed (mean channel width multiplied by channel length) and subtracted from the total surface area for each segment to determine the surface area for each out-of-channel component.

For range lines 1-9 and 13, the mean thickness of the bottom sediment was computed as the difference between the 1999 and 1962 lake-bed elevations (figs. 32-41 at the back of this report). For range lines 10 and 14-21, the mean thickness of the bottom sediment was computed as the difference between the 2000 and 1962 lake-bed elevations (figs. 42-50 at the back of this report). With three exceptions, the mean sediment thickness was computed for each in- and out-of-channel component as the average of the sediment thicknesses determined using the range lines that defined the segment. The in-channel exceptions were the components from the dam to range line 1 and from range line 9 to the confluence with the submerged Big Blue River channel, for which the respective in-channel sediment thicknesses computed for range lines 1 and 9 were used. The out-of-channel exception was the component from the dam to range line 1, for which the out-of-channel sediment thickness computed for range line 1 was used. The component results then were combined to provide an estimate of the total volume of bottom sediment in the lake.

The total mass (dry weight) of bottom sediment in the lake was estimated using the same in- and out-of-channel components as described previously. For each component, a representative bulk density was computed using the bulk densities that were determined from sediment cores (see discussion in following section). Bottom-sediment mass then was computed for each component as bottom-sediment volume multiplied by the representative bulk density. The component results then were combined to provide an estimate of the total mass of bottom sediment in the lake.

With several exceptions, the representative bulk density for each component was computed as the average of the bulk densities determined for the coring sites located along or near the range lines that defined the reservoir segments. In-channel exceptions included the component from the dam to range line 1 and the components upstream from range lines 9 and 16, for which limited information dictated that the representative bulk density was assigned the value for the nearest coring site for which a bulk density value was available. Likewise, for the out-of-channel components upstream from range lines 9 and 13, limited information dictated that the representative bulk density was assigned the value for the nearest coring site for which a bulk density value was available. Also, for some components, the representative bulk density was determined on the basis of coring site(s) for a single range line. Overall, the approach taken was to assign a bulk density value to each component, using available values outright or averaged, to provide what was believed to be the most representative value in the absence of more complete information. Typically, the bulk densities of reservoir sediment tend to be lowest downstream near the dam where the fine sediment is deposited and highest in the upstream part of the lake where the coarse delta deposits are located (Morris and Fan, 1997).

Because it was not feasible to accurately distinguish annual layers of sediment deposition in the lake, mean annual sediment deposition was estimated by dividing the total mass of bottom sediment in the lake by the number of years of deposition. Given that the original bathymetric survey was completed in 1962, about 37 years of sediment deposition had occurred in the lake at the time the new bathymetric survey was completed in July 1999. Therefore, mean annual sediment deposition was estimated as the total mass of bottom sediment divided by 37. Mean annual sediment yield from the Tuttle Creek Lake Basin was estimated as the mean annual sediment deposition divided by the area of the basin (9,628 miî). Because losses are not accounted for (for example, due to reservoir outflow), the computed sediment deposition and yield represent net, rather than total, values.

Sediment-Core Collection, Processing, and Analysis

To determine the occurrence and mass of, as well as trends in, selected chemical constituents, bottom-sediment cores were collected in August and September of 1999 at 22 sites (fig. 2) within Tuttle Creek Lake using a gravity corer. The liner used in the corer was cellulose acetate butyrate transparent tubing with a 2.625-in. inside diameter. The coring sites were located to provide a spatially representative sample of bottom sediment both in and outside of the submerged river and stream channels. A total of 8 in-channel and 14 out-of-channel sites were cored. In most cases, the cores were collected on or near the range lines used in the bathymetric surveys (fig. 2). One to three cores were collected at each site to provide sufficient sediment material for laboratory analyses. The latitude and longitude for each coring site, obtained using GPS technology, are provided in table 13 in the section at the back of this report.

When using a gravity corer, a phenomenon referred to as "core shortening" occurs that results in a recovered sediment core that may be only about one-half of the actual thickness of sediment penetrated (Emery and Hulsemann, 1964). Core shortening is caused by the friction of the sediment against the inner wall of the liner as the gravity corer penetrates the sediment (Emery and Hulsemann, 1964; Hongve and Erlandsen, 1979; Blomqvist, 1985; Blomqvist and Bostrom, 1987). In "normal" lake-bottom sediment at Tuttle Creek Lake, which is characterized by uniform texture with decreasing water content at depth, core shortening results in a core sample that provides a thinned but complete representation of all of the sediment layers that were penetrated (Emery and Hulsemann, 1964; Hongve and Erlandsen, 1979). However, there is some evidence to suggest that the use of a gravity corer may result in the complete loss of some of the uppermost soft surficial sediment on the lake bed (Crusius and Anderson, 1991).

The sediment cores were refrigerated and processed about 1 week after collection at the USGS laboratory in Lawrence, Kansas. The core liners were cut lengthwise in two places 180 degrees apart. The cuts were completed with a 4-in. hand-held circular saw with its blade set at a depth to minimize penetration of the sediment cores. The cores were split in half by pulling a tightly held nylon string through the length of the cores and allowing the halves to separate. Once split, the relatively undisturbed inner parts of the cores were exposed for examination and sampling. On the basis of differences in moisture content, texture, and organic matter content (for example, root hairs, sticks, seed pods, leaves), the boundary between the lake bottom sediment and the underlying original (pre-reservoir) land-surface (or channel-bed) material was determined. Typically, the bottom sediment was characterized by higher moisture content, finer texture, and little if any visible organic matter as compared to the original material. Due to the substantial thickness of the bottom sediment in Tuttle Creek Lake, penetration of the entire thickness of sediment was not achieved for all cores.

The number of samples extracted from each core depended on the length of the core, the intended use of the core, and the amount of material required for chemical analyses. Typically, a core was divided into three or five intervals of equal length. An exception was the in-channel core collected at site TIC-2 (fig. 2), which was divided into 15 intervals of equal length for the purpose of trend analyses. An equal volume of sediment (defined as the space occupied by the sediment particles, water, and gases as measured in cubic units) was extracted lengthwise from both halves of the core for each interval and combined. The combined sediment volume was homogenized and sampled for subsequent chemical analyses. For the in-channel sites, the intervals were sampled and analyzed separately. For the out-of-channel sites, the sediment from all intervals was combined and homogenized to produce a composite for the core that then was sampled and analyzed.

The sediment samples were analyzed for nutrients (total ammonia plus organic nitrogen and total phosphorus), 44 metals and trace elements, 15 organochlorine compounds, and 1 radionuclide. A complete list of the constituents for which analyses were performed is provided in table 1. All constituent analyses of bottom-sediment samples were performed at the USGS National Water-Quality Laboratory in Denver, Colorado. Analyses of sediment samples for total ammonia plus organic nitrogen and total phosphorus concentrations were performed using the methods described by Fishman and Friedman (1989). Analyses for metals and trace elements were performed using the methods described by Arbogast (1996) and Briggs and Meier (1999). Analyses for organochlorine compounds were performed using the methods described by Wershaw and others (1987).

Age dating of the bottom sediment at site TIC-2 was accomplished by determining the concentration of cesium-137 (¹³⁷Cs) by gamma-ray spectrometry (American Society for Testing and Materials, 2000). ¹³⁷Cs is a radioactive isotope that is a by-product of nuclear weapons testing. Measurable concentrations of this isotope first appeared in the atmosphere in about 1952, peaked during 1963-64, and have since declined. ¹³⁷Cs is an effective marker for age dating bottom sediment in reservoirs constructed before 1963-64 (Van Metre and others, 1997). It also can be used to demonstrate that the sediment is relatively undisturbed if a generally uniform, exponential decrease in ¹³⁷Cs concentrations follows the 1963-64 peak. As shown in figure 5, the ¹³⁷Cs profile for site TIC-2 indicated that the sediment is relatively undisturbed. Thus, trend analyses performed for this site may be considered meaningful.

The sediment cores also were analyzed to determine bulk density. For this purpose, each core was divided into 3 to 10 intervals of equal length depending on the length of the core. From each interval a 1-in. thick sediment sample was extracted, weighed to the nearest 0.10 g, oven dried at about 45 ÷C for 96 hours, and reweighed. Oven drying of the sample continued as it was reweighed on a daily basis until no additional moisture loss was observed.

Bulk density was computed as follows:

where Db is the bulk density (in grams per cubic centimeter), m is the mass (dry weight) of the sample (in grams), and v is the volume of the sample (in cubic centimeters). The volume for a cylindrical core sample was computed as:

where v is the volume of the core sample (in cubic centimeters), h is the height (length) of the core sample (in centimeters), and d is the diameter of the core sample (in centimeters) (Gordon and others, 1992). The bulk densities then were converted to pounds per cubic foot for use in subsequent computations. Analyses of sediment samples for bulk density were performed at the USGS laboratory in Lawrence, Kansas.

For the in-channel cores, bulk densities were computed separately for each interval. The results then were averaged to estimate a representative mean bulk density for the site. For the out-of-channel cores, the sediment samples for all intervals were combined and homogenized. The homogenized material then was weighed, and a representative volume of material was extracted that was equal to the total weight of the material divided by the number of intervals sampled. The bulk density determined for this sample provided the representative mean value for the site.

A particle-size analysis was performed to determine the percentage of sand (that is, particles larger than 0.062 mm in diameter) and silt and (or) clay (that is, particles smaller than 0.062 mm in diameter) in the sediment cores. The particle-size analyses were completed at the USGS sediment laboratory in Iowa City, Iowa, according to the methods presented in Guy (1969).

Quality Control

Quality control was provided by an evaluation of within-site and analytical variability. At selected sampling sites during this study, multiple sediment cores were collected to provide the required amount of material for planned physical and chemical determinations. As a result, it was necessary to evaluate "within-site" variability (sediment-quality variability among cores). Within-site variability was evaluated through the collection and analysis of sequential replicate sediment cores at three sites (TOC-1A, TOC-3A, and TOC-7A) (fig. 2). For each core a composite sample was prepared, as described previously, and analyzed for the various constituents. For each pair of cores, the relative percentage differences between the replicate sample concentrations were computed as the quotient of the absolute value of the difference in the replicate concentrations divided by the summation of the replicate concentrations, multiplied by 100. Results indicated that, with the exception of total ammonia plus organic nitrogen, total phosphorus, and inorganic carbon, within-site variability was minimal with mean relative percentage differences less than 8 percent. Total ammonia plus organic nitrogen, total phosphorus, and inorganic carbon had mean relative percentage differences of 29.0, 34.2, and 12.0 percent, respectively. The mean relative percentage differences computed for all constituents detected in the sequential replicate cores are provided in table 2.

Analytical variability was evaluated through the collection and analysis of split replicate samples from an individual core at three sites (TOC-2A, TOC-4A, and TOC-5A) (fig. 2). Two composite samples were prepared for each core and analyzed for the various constituents. The relative percentage differences between the split replicate sample concentrations were computed as previously described. Results indicated that, with the exception of total ammonia plus organic nitrogen, total phosphorus, and tantalum, analytical variability was minimal with mean relative percentage differences generally less than 5 percent. Total ammonia plus organic nitrogen, total phosphorus, and tantalum had mean relative percentage differences of 33.8, 10.0, and 16.7 percent, respectively. The mean relative percentage differences computed for all constituents detected in the split replicate samples are provided in table 3. On the basis of these results, most of the within-site variability determined for total ammonia plus organic nitrogen and some of the within-site variability determined for total phosphorus may be due to analytical variability. Analytical variability can be attributed to sample preparation and, to a lesser degree, instrument variability.

Trend Analysis

The in-channel core collected at site TIC-2 (fig. 2), which was divided into 15 sample intervals, was used for the purpose of trend analyses. Site TIC-2 was selected because it is in relatively deep water near the dam where the sediment was least likely to be disturbed. Trends in constituent concentrations were examined by computing a nonparametric Spearman's rho correlation coefficient. An advantage of Spearman's rho is that, because it is based on ranks, it is more resistant to outlier effects than the more commonly used Pearson's r correlation coefficient (Helsel and Hirsch, 1992). Trends were considered to be significantly positive (constituent concentration increased with decreasing depth in the sediment core) or negative (constituent concentration decreased with decreasing depth in the sediment core) if the probability (two-sided p-value) of rejecting a correct hypothesis (in this case, no trend) was less than or equal to 0.05. Rho was not computed for any constituent having more than five samples with concentrations less than the detection limit.

SEDIMENT DEPOSITION IN TUTTLE CREEK LAKE

The total volume of bottom sediment in Tuttle Creek Lake was estimated by partitioning the original conservation-pool surface area of the lake into in- and out-of-channel components (as segmented by the bathymetric range lines), computing bottom-sediment volume separately for each component, and then summing all component results. The total in- and out-of-channel bottom-sediment volumes were 1,600 and 4,570 million ft, respectively. Therefore, the total estimated volume of bottom sediment in the original conservation-pool area of the lake was 6,170 million ft or about 142,000 acre-ft. In comparison, the USCOE bathymetry-based estimate of the total volume of bottom sediment in 2000 was 145,000 acre-ft (Phil Snell, U.S. Army Corps of Engineers, written commun., 2000). The 142,000 acre-ft of sediment occupies about 33 percent of the lake's original water-storage capacity of 425,000 acre-ft at conservation pool. Within the present (2000) conservation pool area of the lake, the upstream limits of which are approximately indicated by range lines 9 and 13 (fig. 2), the total estimated volume of bottom sediment was 3,640 million ft or about 83,600 acre-ft. Table 4 provides the estimated channel length, mean channel width, mean bottom-sediment thickness, and computed bottom-sediment volume for each in-channel component. Table 5 provides the estimated surface area, mean bottom-sediment thickness, and computed bottom-sediment volume for each out-of-channel component. Longitudinal views of in-channel and out-of-channel sediment deposition in Tuttle Creek Lake are provided in figures 6 and 7, respectively.

Bottom-sediment mass was estimated as the bottom-sediment volume multiplied by the representative bulk density of the sediment. Bulk densities were estimated at all 8 in-channel and all 14 out-of-channel sites in the lake (table 6, fig. 2). Estimated bulk densities ranged from a mean of 18.7 lb/ft for core samples from site TOC-2A (out-of-channel site) to 64.3 lb/ft for core samples from site TOC-13 (out-of-channel site) with an overall mean of 38.2 lb/ft. The particle-size composition of the bottom sediment in Tuttle Creek Lake was very uniform. At virtually every site and sampling depth, the silt and (or) clay content of the sediment was 98 percent or greater.

Annual net sediment deposition was estimated by dividing the total mass of bottom sediment in the lake by the number of years of deposition (that is, 37). The mean annual net sediment deposition was estimated to be 7,900 million lb or 3,600 million kg. Mean annual net sediment yield from the Tuttle Creek Lake Basin, computed as the mean annual net sediment deposition divided by the area of the basin (9,628 miî), was estimated to be 821,000 lb/miî or 1,440 kg/ha.

CONCENTRATIONS AND TRENDS IN SELECTED CHEMICAL CONSTITUENTS

Nutrients and Total Organic Carbon

Total ammonia plus organic nitrogen concentrations in the bottom-sediment samples collected from Tuttle Creek Lake ranged from 600 to 5,200 mg/kg with a median concentration of 800 mg/kg (table 9). Of the 26 samples analyzed, concentrations of total ammonia plus organic nitrogen in 24 were in the range of 600 to 1,100 mg/kg. Because total ammonia plus organic nitrogen was not analyzed in the sediment core from site TIC-2 (fig. 2), a trend analysis was not performed. Total ammonia plus organic nitrogen concentrations were determined for several other sites; however, none of the cores analyzed for total ammonia plus organic nitrogen were divided into a sufficient number of intervals to enable trend analysis. Mean concentrations of total ammonia plus organic nitrogen in bottom-sediment cores collected from Tuttle Creek Lake are shown in figure 8. The estimated mean annual net load of total ammonia plus organic nitrogen deposited in the bottom sediment was 6,350,000 lb/yr (2,880,000 kg/yr). The estimated mean annual net yield of total ammonia plus organic nitrogen from the Tuttle Creek Lake Basin was 657 (lb/miî)/yr [1.15 (kg/ha)/yr] (table 10).

Total phosphorus concentrations ranged from 198 to 952 mg/kg with a median concentration of 419 mg/kg (table 9). No statistically significant trend in total phosphorus deposition was indicated (table 11, fig. 9). Mean concentrations of total phosphorus in bottom-sediment cores collected from Tuttle Creek Lake are shown in figure 10. The estimated mean annual net load of total phosphorus deposited in the bottom sediment was 3,330,000 lb/yr (1,510,000 kg/yr). The estimated mean annual net yield of total phosphorus from the Tuttle Creek Lake Basin was 348 (lb/miî)/yr [0.61 (kg/ha)/yr] (table 10).

TOC concentrations ranged from 0.84 to 2.0 percent with a median concentration of 1.7 percent (table 9). No statistically significant trend in TOC deposition was indicated (table 11, fig. 11). Mean concentrations of TOC in bottom-sediment cores collected from Tuttle Creek Lake are shown in figure 12. The estimated mean annual net load of TOC deposited in the bottom sediment was 138,000,000 lb/yr (62,600,000 kg/yr). The estimated mean annual net yield of TOC from the Tuttle Creek Lake Basin was 14,300 (lb/miî)/yr [25.1 (kg/ha)/yr] (table 10).

Metals and Trace Elements

Toxicity is a function of several factors including the type of organism, availability of a metal or trace element in the environment, and its potential to bioaccumulate once in the food chain. The daily intake of metals and trace elements by animals and humans may be classified as deficient, optimal, or toxic. Most, if not all, metals and trace elements may be toxic in animals and humans if the concentrations are sufficiently large (Pais and Jones, 1997). Information on the bioaccumulation index for most of the metals and trace elements that were analyzed in this study is provided in table 10.

Of the 44 elements analyzed, 40 were detected in all or virtually all of the sediment samples analyzed and at all or virtually all sites for which the analyses were performed (sites TOC-1, TOC-1A, TOC-2, TOC-2A, TOC-3, TOC-3A, TOC-4, TOC-4A, TOC-5, TOC-5A, TOC-7, TOC-7A, TOC-9, TOC-13, TIC-2, TIC-9, and TIC-13; (fig. 2). The exceptions included bismuth and gold, which were not detected in any of the 41 sediment samples analyzed. Sulfur and thallium were detected in 16 of 41 and 15 of 41 samples analyzed, respectively (table 9).

Trend analyses, with a significance level of 0.05, indicated a statistically significant positive depositional trend (constituent concentration increased with decreasing depth in the sediment core) for chromium, cobalt, gallium, iron, lithium, magnesium, mercury, molybdenum, nickel, scandium, selenium, tantalum, tin, and zinc (table 11, figs. 13, 14, 15, and 16). However, despite the statistical significance, the indicated trends may not be representative of actual conditions for two reasons. First, an assessment of the trends with respect to analytical variance (defined here as the mean element concentration for the core plus or minus 10 percent) indicated that many of the trends may be due to analytical variance. This conclusion was based on the fact that, for most of the elements for which a positive trend was indicated, the majority of the concentrations were within 10 percent of the mean value. Second, visual inspection of the vertical profiles (figs. 13, 14, 15, and 16) indicated that most of the constituents for which a positive trend was indicated actually had a bimodal distribution rather than a trend. An exception was mercury for which the indication of a positive trend was due to the effect of a single, anomalously large value (fig. 14).

In general, the bimodal distribution was characterized by relatively small concentrations in the bottom one-fifth of the core and relatively large and uniform concentrations in the upper four-fifths of the core. Examples of the bimodal distribution include the vertical profiles for chromium (fig. 13), lithium (fig. 14), nickel (fig. 15), scandium (fig. 15), and zinc (fig. 16). A possible explanation for the smaller concentrations at the bottom of the core may be disturbance of the Big Blue River flood plain during construction of the dam coupled with a mobilization and redeposition of flood-plain and channel-bank materials during the initial filling of the reservoir. Such locally derived sediment may be chemically different from sediment originating elsewhere in the basin.

The vertical profiles of element concentrations for the sediment core from site TIC-2 (fig. 2) indicated the possibility of unusual conditions during the deposition of sediment at core intervals 3 and 11. For interval 3, 25 elements had notably smaller concentrations (for example, see figs. 13, 14, 15, and 16). For interval 11, anomalously large concentrations were detected for antimony, barium, chromium (fig. 13), mercury ( fig. 14), molybdenum (fig. 14), nickel (fig. 15), selenium (fig. 15), sodium, and strontium. Conversely, anomalously small concentrations were detected for cerium, holmium, lanthanum, neodymium, niobium, silver, tantalum (fig. 15), tin (fig. 16), ytterbium, and yttrium. Several other elements also exhibited larger or smaller concentrations for interval 11 but to a lesser degree. One possible explanation for the divergent concentrations observed for intervals 3 and 11 may be floods. During floods, different processes and sediment sources may have been activated that would account for the anomalous concentrations. Noteworthy was the fact that the organic carbon concentrations for intervals 3 and 11 were much less than the other intervals in the core (fig. 11). Moody and others (2000), in a study of the effects of the 1993 flood on bed sediment in the upper Mississippi River, found that decreased organic carbon concentrations were attributable to an increase in the deposition of coarser sediment during the flood.

Because of uncertainty as to whether or not the statistically significant positive depositional trends determined may be due, in part, to the anomalous values of intervals 3 and 11, the trend analyses for the sediment core from site TIC-2 were repeated with the data for intervals 3 and 11 excluded. Results indicated that, with the exception of mercury (for which no trend was indicated), all elements for which a statistically significant positive depositional trend was originally indicated also had a statistically significant positive depositional trend when data from intervals 3 and 11 were excluded. Also, the repeated analyses indicated a statistically significant negative depositional trend for sodium and statistically significant positive depositional trends for thorium and vanadium table 11. USEPA has not established sediment-quality guidelines for thorium or vanadium. However, for the reasons stated earlier, the indicated trends may not be representative of actual changes.

Sediment-quality guidelines have been established by USEPA for arsenic, cadmium, chromium, copper, lead, mercury, nickel, silver, and zinc (table 9). Arsenic concentrations in the bottom sediment of Tuttle Creek Lake ranged from 6.9 to 18 Îg/g with a median concentration of 14 Îg/g (table 9). Of the 41 detections, 2 were less than the TEL (7.24 Îg/g), and 39 exceeded the TEL but were less than the PEL (41.6 Îg/g). Mean concentrations of arsenic in bottom-sediment cores collected from Tuttle Creek Lake are shown in figure 17.

Cadmium concentrations ranged from 0.26 to 0.61 Îg/g with a median concentration of 0.44 Îg/g (table 9). All detections of cadmium were less than the TEL (0.676 Îg/g).

Chromium concentrations ranged from 48 to 120 Îg/g with a median concentration of 81 Îg/g (table 9). In general, the concentrations exceeded the TEL (52.3 Îg/g) but were less than the PEL (160 Îg/g). Figure 13 shows the variation in chromium concentrations with sediment depth at coring site TIC-2 (fig. 2). Mean concentrations of chromium in bottom-sediment cores collected from Tuttle Creek Lake are shown in figure 18.

Copper concentrations ranged from 20 to 44 Îg/g with a median concentration of 34 Îg/g (table 9). All detections exceeded the TEL (18.7 Îg/g) but were less than the PEL (108 Îg/g). Mean concentrations of copper in bottom-sediment cores collected from Tuttle Creek Lake are shown in figure 19.

Lead concentrations ranged from 16 to 160 Îg/g with a median concentration of 25 Îg/g (table 9). Aside from the 160-Îg/g value, which was measured for interval 5 of the core collected from site TIC-2 (fig. 2), all other detections ranged between 16 and 34 Îg/g. Of the 41 detections, 33 were less than the TEL (30.2 Îg/g), 7 exceeded the TEL but were less than the PEL (112 Îg/g), and 1 exceeded the PEL.

Mercury is a very toxic element for plants, animals, and humans. Once in the food chain it can readily bioaccumulate (Forstner and Wittman, 1981; Adriano, 1986; Pais and Jones, 1997). With the exception of the anomalously large concentration (1.4 Îg/g) determined for interval 11 of the core collected from site TIC-2 (fig. 2), mercury concentrations at the site appeared stable over time and were consistently in the range of 0.02 to 0.05 Îg/g (fig. 14). Overall, mercury concentrations in bottom sediment from Tuttle Creek Lake ranged from less than 0.02 to 1.4 Îg/g with a median concentration of 0.04 Îg/g (table 9). Aside from the 1.4-Îg/g value, all other detections measured 0.07 Îg/g or less. With the exception of the 1.4-Îg/g value, which exceeded the PEL (0.696 Îg/g), all other detections were less than the TEL (0.13 Îg/g).

Nickel concentrations ranged from 19 to 77 Îg/g with a median concentration of 38 Îg/g (table 9). Aside from the 77-Îg/g value, which was measured for interval 11 of the core collected from site TIC-2 (fig. 2), all other detections ranged between 19 and 43 Îg/g. The 77-Îg/g value exceeded the PEL (42.8 Îg/g). All other detections exceeded the TEL (15.9 Îg/g) with several values at or near the PEL. Figure 15 shows the variation in nickel concentrations with sediment depth at coring site TIC-2 (fig. 2). Mean concentrations of nickel in bottom-sediment cores collected from Tuttle Creek Lake are shown in figure 20.

Silver concentrations ranged from 0.4 to 1.2 Îg/g with a median concentration of 0.73 Îg/g (table 9). Of the 41 detections, 21 were less than the TEL (0.733 Îg/g), and 20 exceeded the TEL but were less than the PEL (1.77 Îg/g). Mean concentrations of silver in bottom-sediment cores collected from Tuttle Creek Lake are shown in Figure 21.

Zinc concentrations ranged from 65 to 150 Îg/g with a median concentration of 120 Îg/g (table 9). Of the 41 detections, 21 were less than the TEL (124 Îg/g), and 20 exceeded the TEL but were less than the PEL (271 Îg/g). Figure 16 shows the variation in zinc concentrations with sediment depth at coring site TIC-2 (fig. 2). Mean concentrations of zinc in bottom-sediment cores collected from Tuttle Creek Lake are shown in Figure 22.

USEPA has not established sediment-quality guidelines for selenium. However, concentrations equal to or greater than 4.0 Îg/g in sediment are a concern for fish and wildlife because of the potential for bioaccumulation (Lemly and Smith, 1987). Concentrations in bottom-sediment cores collected from Tuttle Creek Lake ranged from 0.34 to 1.5 Îg/g with a median concentration of 0.81 Îg/g (table 9). Figure 15 shows the variation in selenium concentrations with sediment depth at coring site TIC-2 (fig. 2).

The estimated mean annual net loads and yields for most of the metals and trace elements are provided in (table 10).

Organochlorine Compounds

Of the 15 organochlorine compounds analyzed in this study (14 insecticides and PCBs), only 4 were detected in the bottom sediment of Tuttle Creek Lake. Aldrin was detected in 2 of 34 sediment samples analyzed (table 9). Both detections, with concentrations of 0.6 and 0.3 Îg/kg, were from coring site TIC-2 (fig. 2). No sediment-quality guidelines have been established for aldrin.

DDT was not detected in any of the 34 sediment samples analyzed. However, its degradation products DDD and DDE were detected at several sites. DDD was detected in 10 of 34 sediment samples analyzed (table 9) and at 3 of 10 sites for which the analysis was performed (sites TOC-1A, TIC-2, and TIC-9; fig. 2). Detectable concentrations of DDD ranged from 0.6 to 1.4 Îg/kg. The median concentration of DDD was less than the detection limit of 0.5 Îg/kg. Because DDD was not detected in 8 of 15 intervals of the sediment core from site TIC-2, a trend analysis was not performed. With one exception, the detectable concentrations of DDD were less than the TEL of 1.22 Îg/kg.

The pattern of DDE deposition (fig. 23) was consistent with the history of DDT usage. DDT was used extensively in agriculture during the 1950s and 1960s. Then, with the ban of DDT in 1972, its use declined (Manahan, 2000). DDE was detected in 30 of 34 sediment samples analyzed (table 9) and at all 10 sites for which the analysis was performed (sites TOC-1A, TOC-2A, TOC-3A, TOC-4A, TOC-5A, TOC-7A, TIC-2, TIC-5, TIC-9, and TIC-13; fig. 2). Concentrations of DDE ranged from less than the detection limit of 0.2 to 5.2 Îg/kg with a median concentration of 0.4 Îg/kg. Overall, no significant trend was indicated for DDE (table 11, fig. 23). However, when the intervals for the sediment core from site TIC-2 were grouped and analyzed for trend separately, a different result was indicated. For intervals 1-5 (that is, the oldest sediment in the core), a Spearman's rho of 0.70 was computed. Thus, a positive trend was indicated in which DDE concentrations increased with decreasing depth in the sediment core. However, this trend was not significant at the 0.05 level. For intervals 5-15 (that is, the youngest sediment in the core), a Spearman's rho of -0.90 was computed. Thus, a negative trend, which was significant at the 0.001 level, was indicated in which DDE concentrations decreased with decreasing depth (fig. 23). These results indicated that, over the life of the lake, a reversal in the trend of DDE deposition occurred. With one exception, the detectable concentrations of DDE were less than the TEL of 2.07 Îg/kg. The estimated mean annual net load of DDE deposited in the bottom sediment was 3.18 lb/yr (1.44 kg/yr). The estimated mean annual net yield of DDE from the Tuttle Creek Lake Basin was 0.0003 (lb/miî)/yr [0.0000006 (kg/ha)/yr] (table 10).

Dieldrin was detected in 10 of 34 sediment samples analyzed (table 9) and in samples from 6 of 10 sites for which the analysis was performed (sites TOC-1A, TOC-4A, TOC-5A, TIC-2, TIC-9, and TIC-13; fig. 2). Detectable concentrations of dieldrin ranged from 0.2 to 0.4 Îg/kg. The median concentration of dieldrin was less than the detection limit of 0.2 Îg/kg. Because dieldrin was not detected in 10 of 15 intervals for the sediment core from site TIC-2, a trend analysis was not performed. In all cases, concentrations of dieldrin were less than the TEL of 0.715 Îg/kg.

PCBs were not detected in any of the 34 sediment samples analyzed. Because PCBs are associated with industrial or commercial activities such as those located in intensively urbanized areas, PCBs generally would not be expected in reservoir bottom sediment from a predominantly agricultural basin.

COMPARISON OF DEPOSITIONAL TRENDS TO STREAMFLOW AND HUMAN ACTIVITY

In this study, trend analyses indicated a statistically significant positive depositional trend over time for chromium, cobalt, gallium, iron, lithium, magnesium, mercury, molybdenum, nickel, scandium, selenium, tantalum, tin, and zinc (table 10, figs. 13, 14, 15, and 16). However, for reasons previously discussed, the indicated trends may or may not be representative of actual conditions. In general, differences in the deposition of nutrients, metals, trace elements, and organochlorine compounds over time may be due, in part, to several factors including changes in the particle size and composition of the sediment, redox conditions (oxidation/reduction potential of the environment), precipitation, streamflow, land use, irrigation (source of water and method of application), and land-management practices.

To describe conditions for the mostly agricultural Big Blue River Basin that may affect the deposition of constituents in Tuttle Creek Lake, the factors streamflow, population, crop production, irrigation, and livestock production were investigated. The time period of interest was 1960 through 2000. County-level data were used to assess changes in population, crop production, irrigation, and livestock production. For this study, only counties that were located completely or mostly within the Big Blue River Basin were included. For Kansas, data were compiled for Marshall and Washington Counties (fig. 1). For Nebraska, data were compiled for Adams, Clay, Fillmore, Gage, Hamilton, Jefferson, Nuckolls, Saline, Seward, Thayer, and York Counties (fig. 1). Because complete information on crop production and irrigation was not available for the Kansas counties for the time period of interest (1960-2000), only the Nebraska counties were used in the compilation of data for these activities. However, because about 75 percent of the Big Blue River Basin is located in Nebraska, the results may be considered as representative of crop production and irrigation activities throughout the basin.

Streamflow, specifically mean annual discharge for the period of record, was examined for five USGS streamflow-gaging stations located in the Big Blue River Basin-the Big Blue River at Marysville, Kansas (station 06882510, period of record 1985-2000), the Little Blue River near Barnes, Kansas (station 06884400, period of record 1959-2000), Mill Creek at Washington, Kansas (station 06884200, period of record 1960-99), the Black Vermillion River near Frankfort, Kansas (station 06885500, period of record 1954-2000), and the Big Blue River near Crete, Nebraska (station 06881000, period of record 1954-2000) (fig. 1). In each case, the year-to-year variability in mean annual discharge was evident. No statistically significant trends in streamflow were determined.

The mean annual discharge for 1993 was the largest for the period of record at the four Kansas gaging stations and the second largest for the period of record at the Nebraska gaging station. The Big Blue River Basin was affected by major flooding in 1993, which may account, in part, for the anomalous constituent concentrations that were measured for interval 11 of the sediment core collected at site TIC-2 (fig. 2). Likewise, major flooding in 1973 may account, in part, for the anomalous constituent concentrations that were measured for interval 3 of the same core.

Population change over time can provide a general indication of human activity and the associated potential for environmental pollution. For the Big Blue River Basin, population estimates for the years 1960, 1970, 1980, 1990, and 2000 were compiled for the 13 counties listed previously (U.S. Census Bureau, 2001). Population in the basin has declined from 1960 to 2000 by about 10 percent (fig. 24).

Crop production can have a substantial effect on water and sediment quality in a basin due to repeated disturbance of the land surface, application of pesticides and fertilizers, and irrigation. Cropland is typified by higher runoff volumes and sediment yields than grassland and woodland (Novotny and Chesters, 1981; Novotny, 1995; Morris and Fan, 1997). Increased runoff from cropland is attributable to several factors, including the removal of native vegetation and soil compaction, which decrease surface permeability.

Pesticides and fertilizers are potential sources of metal and trace element pollution. Pesticides may contain arsenic, copper, mercury, lead, manganese, and zinc. Impurities in fertilizers may include cadmium, chromium, molybdenum, uranium, vanadium, and zinc (Alloway and Ayres, 1997).

Irrigation can degrade water and sediment quality through the mobilization, transport, and deposition of various constituents. For example, metals and trace elements often are redistributed by irrigated agriculture. Also, the use of ground water for irrigation may provide an additional source of metals and trace elements. In the Big Blue River Basin, ground water from the High Plains aquifer (fig. 1) is used for most of the irrigation (Jordan and Stamer, 1995). Among the metals and trace elements that have been detected in ground water from the High Plains aquifer are arsenic, barium, cadmium, chromium, copper, iron, lead, manganese, mercury, selenium, silver, and zinc (Engberg, 1984).

Crop production and irrigation data were compiled at 5-year intervals for five principal crops in the Big Blue River Basin--alfalfa, grain corn, grain sorghum, soybeans, and wheat. Alfalfa production declined from almost 200,000 acres harvested in 1965 to about 85,000 acres harvested in 1999 (fig. 25). Typically, alfalfa was not irrigated, with total irrigated acres less than 25,000 acres (U.S. Department of Agriculture, 2001).

Following a decrease in 1965, grain corn production steadily increased from about 400,000 acres harvested in 1965 to almost 1.5 million acres harvested in 2000 (fig. 26). Most of the grain corn produced was irrigated. Following a decrease in 1965, total irrigated acres for grain corn increased from about 300,000 acres in 1965 to about 1.2 million acres in 1995. In 2000, about 1 million acres were irrigated (U.S. Department of Agriculture, 2001).

Grain sorghum production in the Big Blue River Basin peaked at about 930,000 acres harvested in 1980 and then steadily declined to about 310,000 acres harvested in 2000 (fig. 27). Most of the grain sorghum was not irrigated. Total irrigated acres declined from about 210,000 acres in 1965 to only about 12,000 acres in 2000 (U.S. Department of Agriculture, 2001).

Soybean production increased substantially in the basin from less than 2,000 acres harvested in 1960 to about 1 million acres harvested in 2000 (fig. 28). Likewise, the total acres of soybeans irrigated increased from only about 300 to almost 540,000 over the same time period (U.S. Department of Agriculture, 2001).

Wheat production decreased substantially from about 665,000 acres harvested in 1965 to about 140,000 acres harvested in 2000 (fig. 29). Typically, wheat was not irrigated, with total irrigated acres generally less than 10,000 acres (U.S. Department of Agriculture, 2001).

Livestock production data were compiled at 5-year intervals for cattle and hogs. Cattle production increased from about 560,000 animals in 1960 to a peak of about 810,000 animals in 1975, then decreased to about 600,000 animals in 1995 and 2000 (fig. 30). Hog production increased from about 350,000 animals in 1960 to a peak of about 970,000 animals in 1990, then decreased to about 725,000 animals in 1998 (fig. 31) (U.S. Department of Agriculture, 2001). Wastes from hog production may include arsenic, cadmium, copper, lead, nickel, and zinc (Alloway and Ayres, 1997).

SUMMARY AND CONCLUSIONS

REFERENCES

Kyle Juracek
U.S. Geological Survey
4821 Quail Crest Place
Lawrence, KS 66049-3839
Telephone: (785) 832-3527
Fax: (785) 832-3500
Email: kjuracek@usgs.gov