The U.S. Geological Survey (USGS) is committed to serve the Nation with accurate and timely scientific information that helps enhance and protect the overall quality of life, and facilitates effective management of water, biological, energy, and mineral resources. (http://www.usgs.gov/). Information on the quality of the Nation's water resources is of critical interest to the USGS because it is so integrally linked to the long-term availability of water that is clean and safe for drinking and recreation and that is suitable for industry, irrigation, and habitat for fish and wildlife. Escalating population growth and increasing demands for the multiple water uses make water availability, now measured in terms of quantity and quality, even more critical to the long-term sustainability of our communities and ecosystems.

The USGS implemented the National Water-Quality Assessment (NAWQA) Program to support national, regional, and local information needs and decisions related to water-quality management and policy. (http://water.usgs.gov/nawqa/). Shaped by and coordinated with ongoing efforts of other Federal, State, and local agencies, the NAWQA Program is designed to answer: What is the condition of our Nation's streams and ground water? How are the conditions changing over time? How do natural features and human activities affect the quality of streams and ground water, and where are those effects most pronounced? By combining information on water chemistry, physical characteristics, stream habitat, and aquatic life, the NAWQA Program aims to provide science-based insights for current and emerging water issues and priorities. NAWQA results can contribute to informed decisions that result in practical and effective water-resource management and strategies that protect and restore water quality.

Since 1991, the NAWQA Program has implemented interdisciplinary assessments in more than 50 of the Nation's most important river basins and aquifers, referred to as Study Units. (http://water.usgs.gov/nawqa/nawqamap.html). Collectively, these Study Units account for more than 60 percent of the overall water use and population served by public water supply, and are representative of the Nation's major hydrologic landscapes, priority ecological resources, and agricultural, urban, and natural sources of contamination.

Each assessment is guided by a nationally consistent study design and methods of sampling and analysis. The assessments thereby build local knowledge about water-quality issues and trends in a particular stream or aquifer while providing an understanding of how and why water quality varies regionally and nationally. The consistent, multi-scale approach helps to determine if certain types of water-quality issues are isolated or pervasive, and allows direct comparisons of how human activities and natural processes affect water quality and ecological health in the Nation's diverse geographic and environmental settings. Comprehensive assessments on pesticides, nutrients, volatile organic compounds, trace metals, and aquatic ecology are developed at the national scale through comparative analysis of the Study-Unit findings. (http://water.usgs.gov/nawqa/natsyn.html).

Robert M. Hirsch (signature)

Robert M. Hirsch

Associate Director for Water

Conversion Factors and Datum

Multiply	By	To obtain
inch (in.)	25.4	millimeter (mm)
mile (mi)	1.609	kilometer (km)
square mile (mi²)	2.590	square kilometer (km²)
cubic foot (ft³)	0.02832	cubic meter (m³)
cubic foot per second (ft³/s)	0.02832	cubic meter per second (m³/s)
pound (lb)	0.4536	kilogram (kg)

Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as follows:
°F=(1.8x°C)+32

Temperature in degrees Fahrenheit (°F) may be converted to degrees Celsius (°C) as follows:
°C=(°F-32)/1.8

Water year is the 12-month period from October 1 through September 30 and is designated by the year in which it ends. For example, the water year ending September 30, 1999 is called water year 1999.

Vertical coordinate information is referenced to the North American Vertical Datum of 1988 (NAVD 88).

Horizontal coordinate information is referenced to the North American Datum of 1983 (NAD 83).

Abbreviated Water-Quality Units

col/100 mL	colonies per 100 milliliters
ft³/sec	cubic feet per second
mg/L	milligrams per liter
µg/g	micrograms per gram
µg/L	micrograms per liter
µm	micrometer
µS/cm	microsiemens per centimeter at 25 degrees Celsius
<	less than
>	greater than
°C	degrees Celsius

Abbreviations

B	Bighorn River at Kane, Wyoming
CF	Clarks Fork Yellowstone River at Edgar, Montana
DOC	Dissolved organic carbon
E. coli	Escherichia coli
GCMS	Gas chromatography and mass spectrometry
IQR	Interquartile range
LP	Little Powder River above Dry Creek near Weston, Wyoming
MCL	Maximum Contaminant Level MRA Mineral resource area
NAWQA	National Water-Quality Assessment Program
NED	National Elevation Database
NLCD	National Land Cover Data
P	Powder River near Locate, Montana
RPD	Relative percent difference
RSD4	Risk-specific dose associated with an excess cancer risk of 1 in 10,000 for a concentration in drinking water equal to the risk-specific dose
RSD5	Risk-specific dose associated with an excess cancer risk of 1 in 100,000 for a concentration in drinking water equal to the risk-specific dose
SB	Soda Butte Creek at Yellowstone National Park boundary near Silver Gate, Montana
SMCL	Secondary Maximum Contaminant Level
T	Tongue River at Dayton, Wyoming
USEPA	U.S. Environmental Protection Agency
USFS	U.S. Forest Service
USGS	U.S. Geological Survey
YELL	Yellowstone River Basin Study Unit
YNP	Yellowstone National Park
YCS	Yellowstone River at Corwin Springs, Montana
YB	Yellowstone River at Billings, Montana
YF	Yellowstone River at Forsyth, Montana
YS	Yellowstone River near Sidney, Montana

Abstract

The National Water-Quality Assessment Program of the U.S. Geological Survey initiated an assessment in 1997 of the quality of water resources in the Yellowstone River Basin. Water-quality samples regularly were collected during 1999-2001 at 10 fixed sites on streams representing the major environmental settings of the basin. Integrator sites, which are heterogeneous in land use and geology, were established on the mainstem of the Yellowstone River (4 sites) and on three major tributaries—Clarks Fork Yellowstone River (1 site), the Bighorn River (1 site), and the Powder River (1 site). Indicator sites, which are more homogeneous in land use and geology than the integrator sites, were located on minor tributaries with important environmental settings—Soda Butte Creek in a mineral resource area (1 site), the Tongue River in a forested area (1 site), and the Little Powder River in a rangeland area (1 site). Water-quality sampling frequency generally was at least monthly and included field measurements and laboratory analyses of fecal-indicator bacteria, major ions, dissolved solids, nutrients, trace elements, pesticides, and suspended sediment.

Median concentrations of fecal coliform and Escherichia coli were largest for basins that were predominantly rangeland and smallest for basins that were predominantly forested. Concentrations of fecal coliform and Escherichia coli significantly varied by season (p-value <0.001); the smallest median concentrations were during January–March and the largest median concentrations were during April–June. Fecal-coliform concentrations exceeded the U.S. Environmental Protection Agency recommended limit for a single sample of 400 colonies per 100 milliliters in 2.6 percent of all samples. Escherichia coli concentrations exceeded the U.S. Environmental Protection Agency recommended limit for a single sample of 298 colonies per 100 milliliters for moderate use, full-body contact recreation in 7.6 percent of all samples.

Variations in water type in the basin are reflective of the diverse geologic terrain in the Yellowstone River Basin. The water type of Soda Butte Creek and the Tongue River was calcium bicarbonate. These two sites are in forested and mountainous areas where igneous rocks and Paleozoic-era and Mesozoic-era sedimentary rocks are the dominant geologic groups. The water type of the Little Powder River was sodium sulfate. The Little Powder River originates in the plains, and geology of the basin is nearly homogenous with Tertiary-period sedimentary rocks. Water type of the Yellowstone River changed from a mixed-cation bicarbonate type upstream to a mixed-cation sulfate type downstream. Dissolved-solids concentrations ranged from fairly dilute in Soda Butte Creek, which had a median concentration of 118 milligrams per liter, to concentrated in the Little Powder River, which had a median concentration of 2,840 milligrams per liter.

Nutrient concentrations generally were small and reflect the relatively undeveloped conditions in the basin; however, some correlations were made with anthropogenic factors. Median dissolved-nitrate concentrations in all samples from the fixed sites ranged from 0.04 milligram per liter to 0.54 milligram per liter. Flow-weighted mean dissolved-nitrate concentrations were positively correlated with increasing agricultural land use and rangeland on alluvial deposits upstream from the sites and negatively correlated with increasing forested land. Ammonia concentrations generally were largest in samples collected from the Yellowstone River at Corwin Springs, Montana, which is downstream from Yellowstone National Park and receives discharge from geothermal waters that are high in ammonia. Median total-phosphorus concentrations ranged from 0.007 to 0.18 milligram per liter. Median total-phosphorus concentrations exceeded the U.S. Environmental Protection Agency's recommended goal of 0.10 milligram per liter for preventing nuisance plant growth for samples collected from the Bighorn River, Powder River, and Yellowstone River. Seasonal variations were observed in nutrient concentrations. Dissolved-nitrate concentrations generally were largest during October to March when plant uptake of nitrate is lowest. In contrast, total-phosphorus concentrations were largest during April–June when sediment concentrations, which contribute to the total-phosphorus concentrations, are largest.

Concentrations of trace elements generally were small in samples for sites in the Yellowstone River Basin. Soda Butte Creek, which is in a mineral resource area with historical mining in the basin upstream from the sampling site, did not have elevated concentrations of trace elements compared to other sites. On the Yellowstone River, median concentrations of dissolved arsenic of 21 micrograms per liter at Corwin Springs, Montana, and 10.5 micrograms per liter at Billings, Montana, exceeded the drinking-water Maximum Contaminant Level of 10 micrograms per liter. Geothermal waters from Yellowstone National Park are a significant source of arsenic in the Yellowstone River. Concentrations of dissolved selenium were largest in the Powder River, ranging from 0.48 microgram per liter to 4.6 micrograms per liter. Concentrations were smaller than the aquatic-chronic criterion of 5 micrograms per liter; however, other studies have shown that concentrations of total selenium larger than 2 micrograms per liter may produce adverse effects on some fish and wildlife species.

Pesticide concentrations generally were small in samples for three sites on the Yellowstone River, one site on the Clarks Fork Yellowstone River, and one site on the Bighorn River. Herbicides were more frequently detected than insecticides. Atrazine was the most commonly detected herbicide and was detected in 74.8 percent of the samples. Concentrations of all compounds generally were smaller than 0.01 microgram per liter and substantially smaller than aquatic-life or human-health criteria. Mixtures of two or more pesticides were detected in 75 percent of the samples.

Suspended-sediment concentrations were seasonally variable and were largest during April–June during snowmelt runoff. Suspended-sediment concentrations were smallest for the fixed sites on Soda Butte Creek and the Tongue River because of the resistant geology in the mountainous settings. Reservoir-adjusted yields were largest for the Clarks Fork Yellowstone River, Bighorn River, and the Powder River, which have large drainage areas with mixed geology that includes Tertiary-period sedimentary rocks. On the Yellowstone River, suspended-sediment loads increased in the downstream direction.

Introduction

The National Water-Quality Assessment (NAWQA) Program of the U.S. Geological Survey (USGS) was designed to: 1) assess the status in quality of the Nation's water resources, 2) describe the trends in quality of these resources over time, and 3) improve the understanding of natural and anthropogenic influences on the quality of these resources (Hirsch and others, 1988). To implement this design, the NAWQA Program initiated more than 50 interdisciplinary assessments of the Nation's most important stream and aquifer systems. These systems—referred to as Study Units—include more than 60 percent of water used by all sectors and populations served by public water supplies.

Study-Unit assessments are implemented using a consistent design to allow for comparison of data and results on a National scale. As such, a Study-Unit assessment consists of four components: retrospective analysis, occurrence and distribution assessment, trends assessment, and detailed studies of selected constituents. Within the occurrence and distribution component, physical and chemical properties of dissolved and total constituents in streams are assessed at fixed sampling sites to address water-column characteristics (Gilliom and others, 1995).

In 1997, the NAWQA Program initiated an assessment of the quality of the streams and aquifers for the Yellowstone River Basin Study Unit (YELL). The YELL was defined as the entire Yellowstone River drainage basin, including the Yellowstone River and its major tributaries: Clarks Fork Yellowstone, Wind/Bighorn, Tongue, and Powder Rivers (Miller and Quinn, 1997). To assess the status of water-quality conditions in the YELL, 10 fixed sites were established on streams representing most of the environmental settings in the Study Unit. Water-quality samples generally were collected at least monthly at the fixed sites from January 1999 through September 2001 for use in describing the occurrence and distribution of water-column characteristics in the YELL.

Purpose and Scope

This report summarizes the environmental settings and water-quality characteristics at 10 fixed sites in the YELL in Montana and Wyoming. Fixed sites were on Soda Butte Creek (1 site), Yellowstone River (4 sites), Clarks Fork Yellowstone River (1 site), Bighorn River (1 site), Tongue River (1 site), Little Powder River (1 site), and Powder River (1 site). The water-quality characteristics described in the report include field measurements, fecal-indicator bacteria, major ions, dissolved solids, nutrients, trace elements, pesticides, and suspended sediment. Analyses of water-quality data for the fixed sites generally are limited to those samples collected during the study period from January 1, 1999 through September 30, 2001. Water-quality characteristics are summarized in the context of the fixed-site environmental settings.

Description of Study Unit

The YELL extends from central Wyoming north to include most of southeastern Montana and a small part of western North Dakota (fig. 1). Drainage area for the YELL is about 70,100 square miles (mi2). The entire Yellowstone River drainage basin defines the YELL boundaries and includes all of the Clarks Fork Yellowstone, Wind/Bighorn, Tongue, and Powder River Basins. Topography of the YELL varies from mountain ranges and high plateaus, including the Wind River Range, Bighorn Mountains, and Absaroka Range, to intermontane basins, such as the Wind River and Bighorn Basins. Elevations in the YELL range from more than 13,800 feet above the North American Vertical Datum of 1988 (NAVD 88) to about 1,900 feet above NAVD 88. Mean annual precipitation ranges from more than 60 inches in the mountains near Yellowstone National Park (YNP) to less than 6 inches in parts of the Bighorn River Basin (Oregon Climate Service, 1998).

Rangeland is the dominant land use and land cover in the YELL, which includes about 47 percent herbaceous grasslands and 27 percent shrublands (U.S. Geological Survey, 1992). The remaining land uses and land covers include about 14 percent forested lands, 9 percent agricultural lands, and 3 percent other uses. Urban areas compose about 0.2 percent of the YELL. Total population of the Yellowstone River Basin was about 323,000 during 1990 (Zelt and others, 1999, p. 57). About 30 active mines remove low-sulfur coal by surface strip-mining methods from the large coal reserves in the Tongue and Powder River Basins. Most of the metals mining was in the mountainous regions of the Study Unit. Oil and gas are produced from reservoirs in the Powder River and Bighorn Basins (Miller and Quinn, 1997).

Streamflow in all the major drainage basins is affected by diversions. Although there are no reservoirs on the Yellowstone River, several reservoirs regulate flow in the Wind/Bighorn and Tongue River drainage basins. About 98 percent of the total water used in the Study Unit in 1990 was surface water (Miller and Quinn, 1997). Most of this water—about 99 percent—was used by the agricultural industry for crops and livestock. Public supply, mining, power generation, and industry made up most of the remaining uses of water in the YELL (Miller and Quinn, 1997).

Figure 1. Location of the Yellowstone River Basin.

Figure 1. Location of the Yellowstone River Basin.

Acknowledgments

USGS hydrologic technicians, in particular Stacy Kinsey, and hydrologists that collected water-quality samples and streamflow data are gratefully acknowledged; without their dedication, this report would not have been possible. All the local, State, and Federal agencies that contribute to the operation of streamflow gages in the basin are acknowledged. Reviews by Kent Becher, Janet Carter, and Norm Spahr improved the report. D. Kirk Nordstrom, USGS, is acknowledged for his insight into the water quality of the geothermal resources of Yellowstone National Park. David Peterson, USGS, is acknowledged for his assistance in evaluating nutrients in the Yellowstone River and his direction as project chief. Thomas Quinn, formerly USGS, is acknowledged for his guidance and support. The compilation and reduction of digital geographic data by Ronald Zelt, USGS, and Robert Kirkwood, formerly USGS, made possible many of the analyses in this report. In addition, Sue Roberts and Emily Sabado are acknowledged for their assistance in the preparation of the report.

Environmental Settings of Fixed Sites

Natural and anthropogenic factors, such as geology and land use, affect the water quality of streams. These factors can be combined to create environmental settings for the drainage basins that can be used for designing sampling networks and making water-quality assessments (Gilliom and others, 1995). This section describes the design and characteristics of the fixed-site network for the YELL and the characteristics for each of the sites in the network.

Design and Characteristics of the Fixed-Site Network

Zelt and others (1999) used three factors in a layered—or stratified—approach to create environmental settings that are believed to be important in assessing water-quality conditions in the YELL (fig. 2). Ecoregions—the first factor—were determined from a modification of mapping by Omernik (1987) (U.S. Environmental Protection Agency, 1996). Geology—the second factor—was determined from a reduction of multiple sources to generalized geologic groups. An additional geologic group—mineral resource area (MRA)—was defined for areas where minerals (for example, metals, coal deposits, or oil and gas fields) occur and (or) extraction activities have occurred or are likely to occur. Alluvium and other Quaternary-period unconsolidated deposits compose a small component of surficial geology and are not included with the consolidated geologic groups in table 1; however, these deposits typically are close to the stream and may be important to water quality, so they are discussed in the text in some places. Land use and land cover classification—the third factor—was determined from Anderson and others (1976).

The fixed-site network has two general types of sites. Integrator sites are used to characterize water quality in larger basins with a heterogeneous environmental setting, potentially consisting of multiple land covers, land uses, and geologic groups in more than one ecoregion. Integrator sites are located near the Study Unit outlet and near important confluences. By comparison, indicator sites are used to characterize water-quality in smaller basins with a relatively homogeneous environmental setting. Indicator sites are selected to represent environmental settings that may have an important influence on water quality in the YELL.

Of the 10 sites that were established as part of the YELL fixed-site network (table 1), 4 integrator sites were established on the Yellowstone River to directly describe downstream water-quality gradients on the mainstem. Sites were established at Corwin Springs (site YCS) downstream from YNP, Billings (site YB) downstream from the Clarks Fork Yellowstone River, Forsyth (site YF) downstream from the Bighorn River, and Sidney (site YS) near the mouth. Three additional integrator sites were established on major tributaries to augment information from the mainstem integrator sites. A site was established near the mouth of the Clarks Fork Yellowstone River (site CF) to characterize the contribution of nutrients and sediment to the mainstem. A site was established on the Bighorn River (site B) to characterize water-quality conditions in the largest and most regulated tributary. A site was established near the mouth of the Powder River (site P), which is a substantial source of sediment to the mainstem. Three indicator sites were established to characterize the water-quality conditions of important environmental settings in more homogeneous basins. A site was established on Soda Butte Creek (site SB) to characterize water-quality conditions of a MRA; on the Tongue River (site T) to characterize water-quality conditions for forested land; and on the Little Powder River (site LP) to characterize water-quality conditions for rangeland.

Environmental settings also were used to help determine sampling frequency and select constituents for analysis. Basic fixed sites were sampled monthly during the study with additional samples collected during extreme flows. Intensive fixed sites were sampled monthly during the study with additional samples collected during extreme flows and during the growing season where agriculture was an important land use. Fixed-site type for each site is listed in table 1. Constituents selected for analysis may be targeted to assess a specific land use. For example, trace elements may be analyzed in samples collected from basins where a MRA is a part of the environmental setting, but not analyzed in samples collected from a forested basin.

The simple one-, two-, or three-letter identifiers that were defined for each site are used in text, figures, and tables throughout the report for the benefit of the reader. Shading used in figures corresponds to the site groups (table 1) of mainstem integrator (gray shading), major tributary integrator (blue shading), and minor tributary indicator (white or no shading).

The land use and land cover distributions in table 1 were determined using the National Land Cover Data (NLCD) (U.S. Geological Survey, 1992). General differences between the NLCD and the Anderson and others (1976) classifications are described by the U.S. Geological Survey (1992). For land use and land cover information over small areas in the YELL, the NLCD probably is more accurate than the previous data, because of enhanced spatial resolution and updated mapping techniques as well as being more contemporaneous with this study. Because of the generalized, composite nature of the stratification process, however, differences between the land use and land cover data and the resulting differences in the environmental settings at the basin scale for the YELL are small. In this report, rangeland is the combined area of the shrubland and herbaceous grassland land covers.

Land-use and land-cover classifications are general descriptors of land characteristics that often are interpreted differently. Such interpretations can exclude detailed characteristics that are important in water-quality assessments. For example, rangelands typically are associated with minimal use as well as livestock grazing of native grasslands. In the YELL, however, land uses on rangelands also include vegetation management to improve forage and dispersed development of oil and gas. Similarly, in addition to the natural conditions in forested lands, forest uses also include logging, livestock grazing, and recreational activities. In addition, for rangelands as well as forested lands, increasing low-density, unsewered domestic development is a changing land use in the YELL. Finally, some land uses that do not compose a large percentage of any given basin can be important in water-quality assessments. For example, much of the agricultural land uses in the YELL are concentrated near water sources because of the semi-arid climate.

Other basin or stream characteristics are useful in describing the environmental setting of the fixed sites. Area-weighted mean basin elevations (table 1) were determined using the National Elevation Database (NED) (U.S. Geological Survey, 1999) and other digital spatial data. Area-weighted mean annual and monthly precipitation values were calculated using digital spatial data by Daly and others (1994) and Oregon Climate Service (1998). Streamflow statistics for the fixed sites were determined using gaging-station data from USGS databases (table 2). Because of the short period of record (1999-2001) for site SB, the streamflow statistics might not be representative of historical or future streamflows. These streamflow data are available from the USGS on the Internet at http://waterdata.usgs.gov/nwis/sw. Stream classifications and corresponding quality or use classifications that are used by the States of Montana and Wyoming for regulatory concerns were simplified for this report and are listed in table 3. The stream classifications and supporting uses are described with the environmental setting for the site.

Table 1. Selected basin characteristics for fixed sites in the Yellowstone River Basin.

[NAVD 88, North American Vertical Datum of 1988; YNP, Yellowstone National Park; p€ , Precambrian crystalline; PZ, Paleozoic and Mesozoic sedimentary; Ks, Cretaceous sedimentary; Ts, Tertiary sedimentary; KTv, Cretaceous and Tertiary volcanic; KTQv, Cretaceous, Tertiary, and Quaternary volcanic]

Site report identifier (fig. 2)	Station number	Station name	Drainage area (square miles)¹	Mean basin elevation (feet above NAVD 88)	Site group	Fixed-site type	Predominant environmental characteristics (percent of drainage area)			Mean annual precipitation (inches)⁴
Site report identifier (fig. 2)	Station number	Station name	Drainage area (square miles)¹	Mean basin elevation (feet above NAVD 88)	Site group	Fixed-site type	Ecoregion²	Simplified geologic groups²	Land use and land cover³	Mean annual precipitation (inches)⁴
SB	06187915	Soda Butte Creek at YNP boundary, near Silver Gate, Mont.	31.2	8,900	Minor tributary indicator	Basic indicator (mineral resource area)	Middle Rocky Mountains (100)	KTv (68), PZ (31), p€(1)	Evergreen forest (77), Transitional (9), Shrubland (9), Herbaceous grassland (3)	38
YCS	06191500	Yellowstone River at Corwin Springs, Mont.	2,619	8,300	Mainstem integrator	Basic integrator	Middle Rocky Mountains (100)	KTQv (77), p€(8), PZ (5), Ks (2)	Evergreen forest (49), Transitional (23), Shrubland (11), Herbaceous grassland (9)	32
CF	06208500	Clarks Fork Yellowstone River at Edgar, Mont.	2,022	6,300	Major tributary integrator	Basic integrator	Middle Rocky Mountains (47), Wyoming Basin (40), Northwestern Great Plains (13)	Ts (33), Ks (19), KTv (18), p€(18), PZ (11)	Shrubland (42), Evergreen forest (27), Herbaceous grassland (19), Transitional (3)	22
YB	06214500	Yellowstone River at Billings, Mont.	11,805	6,500	Mainstem integrator	Basic integrator	Middle Rocky Mountains (55), Northwestern Great Plains (24), Montana Valley and Foothill Prairies (13), Wyoming Basin (7)	KTv (35), Ks (22), p€ (16), Ts (15), PZ (5), KTQv (5)	Herbaceous grassland (33), Evergreen forest (31), Shrubland (15), Transitional (6)	24
B	06279500	Bighorn River at Kane, Wyo.	15,762	6,400	Major tributary integrator	Intensive integrator	Wyoming Basin (74), Middle Rocky Mountains (26)	Ts (43), Ks (23), PZ (18), p€ (9), KTQv (6)	Shrubland (60), Herbaceous grassland (25), Evergreen forest (9), Pasture/hay (2)	14
YF	06295000	Yellowstone River at Forsyth, Mont.	40,146	5,800	Mainstem integrator	Intensive integrator	Wyoming Basin (35), Middle Rocky Mountains (33), Northwestern Great Plains (27), Montana Valley and Foothill Prairies (4)	Ks (34), Ts (27), KTv (16), PZ (13), p€ (8), KTQv (1)	Herbaceous grassland (36), Shrubland (33), Evergreen forest (17), Small grains (3)	18
T	06298000	Tongue River near Dayton, Wyo.	206	8,500	Minor tributary indicator	Basic indicator (forested)	Middle Rocky Mountains (100)	PZ (53), p€(46)	Evergreen forest (60), Herbaceous grassland (23), Shrubland (7), Deciduous forest (5)	27
LP	06324970	Little Powder River above Dry Creek, near Weston, Wyo.	1,237	4,100	Minor tributary indicator	Basic indicator (rangeland)	Northwestern Great Plains (100)	Ts (100)	Herbaceous grassland (81), Shrubland (11), Evergreen forest (3), Small grains (1), Emergent herbaceous wetlands (1)	14
P	06326500	Powder River near Locate, Mont.	13,068	4,600	Major tributary integrator	Basic integrator	Northwestern Great Plains (88), Middle Rocky Mountains (7), Wyoming Basin (5)	Ts (65), Ks (24), PZ (7), p€ (4)	Herbaceous grassland (59), Shrubland (29), Evergreen forest (6), Small grains (1), Fallow (1), Pasture/hay (1)	14
YS	06329500	Yellowstone River near Sidney, Mont.	69,083	5,000	Mainstem integrator	Intensive integrator	Northwestern Great Plains (55), Middle Rocky Mountains (22), Northwestern Great Plains (21), Montana Valley and Foothill Prairies (2)	Ts (48), Ks (26), KTv (9), PZ (9), p€ (6)	Herbaceous grassland (47), Shrubland (27), Evergreen forest (13), Small grains (3), Fallow (3)	17

¹Source U.S. Geological Survey National Water Information System.

²Modified from Zelt and others, 1999; only the four largest (by total basin area) geologic groups are listed.

³Modified from U.S. Geological Survey, 1992; only the four largest (by total basin area) land use/land cover classifications are listed.

⁴Modified from Oregon Climate Service, 1998.

Table 2. Selected streamflow statistics for fixed sites in the Yellowstone River Basin.

Site report identifier (fig. 2)	Period of record (water years)¹	Mean annual streamflow (cubic feet per second)	Streamflow percent exceedance (cubic feet per second)			Annual runoff (acre- feet)
Site report identifier (fig. 2)	Period of record (water years)¹	Mean annual streamflow (cubic feet per second)	10th	50th	90th	Annual runoff (acre- feet)
SB	1999-2001	52.7	203	8.2	1.8	38,000
YCS	1890-2001	3,125	8,490	1,400	760	2,264,000
CF	1922-2001	1,035	2,850	470	275	749,800
YB	1929-2001	6,996	17,700	3,730	2,180	5,069,000
B	1930-2001	2,160	²3,000	²2,100	²1,400	1,565,000
YF	1978-2001	10,800	22,900	7,380	4,600	7,822,000
T	1920-2001	180	479	73	48	130,100
LP	1973-2001	21.8	35	3.0	.03	15,780
P	1939-2001	583	1,370	241	42	422,300
YS	1911-2001	12,550	³27,000	³8,740	³5,000	9,095,000

¹Based on complete water years. May include intervening water years with no record.

²Based on period of record after regulation began by Boysen Reservoir, water years 1952-2001.

³Based on period of record after Bighorn Lake reached operational level, water years 1967-2001.

Table 3. State water-quality classifications of surface waters, Montana and Wyoming.

Class	Quality or use	Comments
Montana¹
A	Very high quality	³Montana surface-water classification includes a nondegredataion policy for "Outstanding Resource Waters"
B	High quality	Majority of surface waters in Montana.
C	Low quality	Same uses as class B waters except unsuitable for drinking.
I	Impaired	Generally unsuitable for any use.
Wyoming²
1	Outstanding waters	Includes all waters in Teton and Yellowstone National Parks and designated wilderness areas.
2	Fisheries and drinking waters	Known to support or have the potential to support fish populations or drinking water supplies.
3	Aquatic life other than fish	Basic default classification for all waters; provides minimum protection for aquatic life and recreation uses.
4	Agriculture, industry, recreation, and wildlife	Aquatic life not protected.

¹Source: Bryan and Kakuk, 1997.

²Source: Wyoming Department of Environmental Quality, 2001a.

³Bryan and Kakuk, p. 11, 1997.

Figure 2. Schematic diagram and map of environmental settings and fixed sites of the Yellowstone River Basin.

Figure 2. Continued.

Figure 2. Schematic diagram and map of environmental settings and fixed sites of the Yellowstone River Basin.

Soda Butte Creek at YNP boundary near Silver Gate, Montana (SB)

Soda Butte Creek is located in the Absaroka Range of southwestern Montana where it flows west-southwest into YNP and Wyoming to its confluence with the Lamar River, a tributary to the Yellowstone River. Site SB (fig. 2) is downstream from Silver Gate, Montana, at the boundary of YNP. The drainage area for the site is located entirely within the Middle Rocky Mountains ecoregion. Mean basin elevation is the highest of the fixed-site basins (8,900 feet above NAVD 88), with elevations ranging from more than 11,000 feet above NAVD 88 in the headwaters to about 7,300 feet above NAVD 88 near site SB. Mean annual precipitation of 38 inches in the Soda Butte Creek Basin is the highest for the fixed-site basins. The geologic groups generally are composed of mostly Cretaceous- and Tertiary-period volcanic rocks and Paleozoic-era and Mesozoic-era sedimentary rocks (Elliott, 1979). About 33 percent of the basin upstream from site SB was included by Zelt and others (1999) in the MRA geologic group, which is predominantly associated with the Paleozoic-era and Mesozoic-era sedimentary rocks, based on descriptions of historical and recent metals mining and other mineral resource assessments.

About 77 percent of the land cover of the basin upstream from site SB is evergreen forest (table 1). An important historical land use in the basin has been the mining of gold, silver, copper, and other metals. Ore from the McLaren Mine was processed at a mill located next to Soda Butte Creek until 1953 (Boughton, 2001). Much of the mining occurred at or near timberline in the upper part of the basin. Disposal of mine tailings included an in-channel impoundment in the headwaters of Soda Butte Creek. Subsequent exploration continued until 1996 when private mining interests were bought out by the U.S. Government (U.S. Forest Service, 2003). The communities of Cooke City and Silver Gate, Montana (fig. 1), which have a combined population of 140 people (U.S. Census Bureau, 2000), are located in the basin along Soda Butte Creek.

The hydrology of site SB is characteristic of small streams with headwaters in the mountainous areas of the YELL. Annual streamflows are dominated by a snowmelt peak of moderate duration during late spring through early summer with low variability in daily mean streamflow throughout the year (fig. 3). Compared to rainstorms in the basins and plains areas, mountain snowpacks are relatively consistent in spatial extent and between years, resulting in low variability in annual streamflows (Miller, 1999).

Soda Butte Creek upstream from site SB is classified by the State of Montana as Class B waters (Montana Department of Environmental Quality, 2002a). Soda Butte Creek upstream from site SB to a tailings impoundment is listed as only partially supporting aquatic life and cold-water fishes because of metals from abandoned mining activities (Montana Department of Environmental Quality, 2002b).

Figure 3. Annual hydrograph (1999-2001).

Figure 3. Annual hydrograph (1999-2001) and photograph for Soda Butte Creek, Montana (site SB).

Photograph for Soda Butte Creek, Montana.

Photograph from U.S. Geological Survey files.

Yellowstone River at Corwin Springs, Montana (YCS)

The headwaters of the Yellowstone River are near the Continental Divide in the Absaroka Range of northwestern Wyoming and from there the river generally flows north-northwest into YNP, through Yellowstone Lake and the Grand Canyon of the Yellowstone, and into Montana. Site YCS (fig. 2) is located at Corwin Springs about 8 miles (mi) north of the Wyoming-Montana state line and about 550 river miles upstream from the confluence with the Missouri River (Shields and others, 2001). The basin upstream from the site is located entirely within the Middle Rocky Mountains ecoregion. Elevations in the basin range from more than 12,100 feet above NAVD 88 in the headwaters to about 5,100 feet above NAVD 88 near site YCS. Mean annual precipitation for the basin is about 32 inches; most of the annual precipitation occurs as snow during the winter months, with November through January accumulations equaling about 30 percent of the total annual precipitation. The geologic groups in the basin upstream from site YCS are composed mostly of Cretaceous-, Tertiary-, and Quaternary-period volcanic rocks (77 percent). YNP and the surrounding region are considered an active volcanic area; Yellowstone Lake is located in a collapsed caldera measuring 45 mi by 30 mi (U.S. Geological Survey, 2003b; Yellowstone National Park, 2001).

About 49 percent of the land cover of the basin upstream from site YCS is evergreen forest and about 23 percent is classified as transitional (table 1). The transitional classification includes areas of sparse vegetative cover that are in the process of changing from one land cover to another. Examples include forest clearcuts and changes due to natural causes (for example, fire). During 1988, forest fires burned large areas of YNP and the surrounding region. About 46 percent of the basin upstream from site YCS was within the perimeters of these fires (Yellowstone National Park, 1995).

The hydrology of site YCS is characteristic of large streams with headwaters in the mountainous areas of the YELL. Annual streamflows are dominated by a single snowmelt peak of moderate duration during late spring through early summer with low variability in daily mean streamflow throughout the year (fig. 4). Compared to rainstorms in the basins and plains areas, mountain snowpacks vary little in spatial extent and between years, resulting in low variability in annual streamflows (Miller, 1999). Annual mean streamflow at site YCS is about 25 percent of the annual mean streamflow at site YS near the mouth; however, drainage area at site YCS accounts for only about 4 percent of the drainage area at site YS.

In YNP, the Yellowstone River and its tributaries are classified by the States of Wyoming and Montana as Class 1 and A waters, respectively, because all waters within national parks and wilderness areas are classified as such (Wyoming Department of Environmental Quality, 2001a; Montana Department of Environmental Quality, 2002a). The Yellowstone River downstream from YNP to site YCS is classified by the State of Montana as Class B waters (Montana Department of Environmental Quality, 2002a).

Figure 4.Annual hydrograph (1890-2001).

Figure 4. Annual hydrograph (1890-2001) and photograph for the Yellowstone River at Corwin Springs, Montana (site YCS).

Photograph for the Yellowstone River at Corwin Springs, Montana.

Photograph by Gregory K. Boughton, U.S. Geological Survey

Clarks Fork Yellowstone River at Edgar, Montana (CF)

The Clarks Fork Yellowstone River flows east-southeast from its headwaters in the Beartooth Mountains and the Absaroka Range of southwestern Montana and northwestern Wyoming before flowing north-northeast to its confluence with the Yellowstone River. Site CF (fig. 2) is located at Edgar, Montana, about 22 river miles upstream from the confluence (Shields and others, 2001). The drainage area for the site is located mostly within the Middle Rocky Mountains (47 percent; upper part of the basin) and Wyoming Basin (40 percent; lower part of the basin) ecoregions. Elevations in the basin range from more than 12,500 feet above NAVD 88 in the headwaters to about 3,500 feet above NAVD 88 near site CF. Annual precipitation in the basin varies substantially, ranging from about 65 inches at the higher elevations in the western part of the basin to about 7 inches in the east-central part of the basin. Precambrian-era crystalline rocks (18 percent of the drainage area) and Cretaceous- and Tertiary-period volcanic rocks (18 percent) are the dominant geologic groups in the upper part of the basin in the Middle Rocky Mountains ecoregion. Cretaceous- (19 percent) and Tertiary-period (33 percent) sedimentary rocks are the dominant geologic groups in the lower part of the basin in the Wyoming Basin ecoregion.

The dominant land cover in the basin upstream from site CF is rangeland, which includes shrubland and herbaceous grassland (table 1). Most of the rangeland is in the lower part of the basin in the Wyoming Basin ecoregion. The agricultural land cover generally is in the lower part of the basin near the streams (fig. 2). About 16 percent of the alluvial deposits in the basin have agricultural uses on them; 53 percent of all agricultural land uses in the basin are on alluvial deposits. The dominant land cover in the upper part of the basin in the Middle Rocky Mountains ecoregion is evergreen forest (table 1).

The hydrology of site CF is characteristic of large streams with headwaters in the mountainous areas of the YELL. Annual streamflows are dominated by a single snowmelt peak of moderate duration during late spring through early summer (fig. 5). Variability in daily mean streamflow during the spring, summer, and fall months is larger at site CF than other snowmelt-dominated streams because of the variability in rainstorms in the lower part of the basin. Irrigation withdrawals in the lower part of the basin also affect variability in daily mean streamflow.

The upper reaches of the Clarks Fork Yellowstone River in Wyoming are the State's only reaches with National Wild and Scenic Rivers designation (Interagency Coordinating Council, 2003) and are classified by the State of Wyoming as Class 1 waters. The Clarks Fork Yellowstone River upstream from site CF is classified by the State of Montana as Class B waters (Montana Department of Environmental Quality, 2002a). From the headwaters to the state line, the river is listed as only partially supporting aquatic life and cold-water fishes, because of metals from abandoned mining activities (Wyoming Department of Environmental Quality, 2002, Montana Department of Environmental Quality, 2002b).

Figure 5. Annual hydrograph (1922-2001).

Figure 5. Annual hydrograph (1922-2001) and photograph for the Clarks Fork Yellowstone River, Montana (site CF).

Photograph for the Clarks Fork Yellowstone River, Montana.

Photograph by Gregory K. Boughton, U.S. Geological Survey

Yellowstone River at Billings, Montana (YB)

From site YCS, the Yellowstone River flows north to Livingston, Montana, then east through southern Montana to Billings, Montana, about 360 river miles upstream from the confluence with the Missouri River (Shields and others, 2001). Billings, Montana, the largest city in the YELL with a population of 89,847 (U.S. Census Bureau, 2000), is located at site YB (fig. 2). The drainage area for the site is located mostly within the Middle Rocky Mountains (55 percent) ecoregion. Mean basin elevation is about 6,500 feet above NAVD 88, with elevations ranging from more than 12,500 feet above NAVD 88 in the headwaters to about 3,100 feet above NAVD 88 near site YB. Mean annual precipitation for the basin is about 24 inches, ranging from about 65 inches at the higher elevations in the basin to about 7 inches in the Clarks Fork Yellowstone River part of the basin. The geologic groups in the basin upstream from site YB are composed mostly of Cretaceous- and Tertiary-period volcanic rocks (35 percent of the basin). Cretaceous-period sedimentary rocks compose 22 percent of the total basin.

Land cover in the basin upstream from site YB is dominated by rangeland and forested land (table 1). About 33 percent of the drainage area land cover is herbaceous grassland and 15 percent is shrubland (table 1). The percentage of forested land decreases and the percentage of rangeland increases downstream along the Yellowstone River between sites YCS and YB as proportionally less of the total drainage area is from mountain areas and more is from basin and plains areas. Land cover adjacent to streams in the basin upstream from site YB is mostly rangeland. About 44 percent of the land cover on alluvial deposits in the basin is herbaceous grasslands and shrublands. Agricultural land uses are more prevalent adjacent to streams than for the distant parts of the basin. About 13 percent of the alluvial deposits in the basin have agricultural uses on them.

The hydrology of site YB is characteristic of large streams with headwaters in the mountainous areas of the YELL. Annual streamflows are dominated by a single snowmelt peak of moderate duration during late spring through early summer with low variability in daily mean streamflow throughout the year (fig. 6). Annual mean streamflow at site YB is about 56 percent of the annual mean streamflow at site YS near the mouth; drainage area at site YB accounts for only about 17 percent of the drainage area at site YS.

The Yellowstone River at site YB is classified by the State of Montana as Class B waters (Montana Department of Environmental Quality, 2002a). The Yellowstone River downstream from Billings is listed as only partially supporting warm-water fishery probably because of habitat alterations (Montana Department of Environmental Quality, 2002b).

Figure 6. Annual hydrograph (1929-2001).

Figure 6. Annual hydrograph (1929-2001) and photograph for the Yellowstone River at Billings, Montana (site YB).

Photograph for the Yellowstone River at Billings, Montana.

Photograph from U.S. Geological Survey files

Bighorn River at Kane, Wyoming (B)

The headwaters for the Bighorn River are the streams in the mountains rimming the southern, eastern, and western margins of the basin. The principal headwater stream—the Wind River—flows east-southeast from the Continental Divide in Wyoming into the Wind River Basin before flowing north into Boysen Reservoir and through Wind River Canyon. Near the mouth of Wind River Canyon at Wedding of the Waters, the Wind River becomes the Bighorn River and flows north through the Bighorn Basin and into Bighorn Lake (a reservoir) and Bighorn Canyon National Recreation Area. Site B (fig. 2) is located at Kane, Wyoming, about 0.5 mi upstream from the normal highwater line for Bighorn Lake (Swanson and others, 2001). The drainage area for site B is mostly within the Wyoming Basin (74 percent) ecoregion (table 1). Elevations in the basin range from more than 13,800 feet above NAVD 88 in the Wind River Range to about 3,700 feet above NAVD 88 near site B. Annual mean precipitation varies with elevation, ranging from about 47 inches in the western mountains of the basin to less than 6 inches in the central part of the Bighorn Basin. The geologic groups in the basin upstream from site B are composed mostly of Tertiary-period sedimentary rocks (43 percent of the drainage area), which occur mostly in the central, lower-elevation parts of the Wind River and Bighorn Basins. Cretaceous-period and Paleozoic- and Mesozoic-era sedimentary rocks are exposed on the flanks of the mountain uplifts at the basin margins.

The dominant land cover in the basin upstream from site B is rangeland. About 60 percent of the drainage area land cover is shrubland and 25 percent is herbaceous grassland (table 1). Evergreen forest land cover is at higher elevations in the basin in contrast to agricultural covers, which generally are located in the lower elevations of the basin. About 70 percent of the land cover on alluvial deposits in the basin is herbaceous grasslands and shrublands. About 16 percent of the alluvial deposits in the basin have agricultural uses on them. About 74 percent of all agricultural land uses in the basin are on alluvial deposits.

Streamflows in the Bighorn River are affected substantially by modifications in the form of releases from Boysen Reservoir as well as irrigation diversions and return flows. Streamflow statistics for the period of record following regulation by Boysen Reservoir (water years 1952-2001) show the effects of streamflow modification in the large variability in streamflows during the summer months (fig. 7). Annual mean streamflow at site B is about 17 percent of the annual mean streamflow at site YS near the mouth of the Yellowstone River; drainage area at site B accounts for about 23 percent of the drainage area at site YS.

The Bighorn River is classified as Class 2 waters (Wyoming Department of Environmental Quality, 2001a). A reach of the Bighorn River downstream from Greybull, Wyoming, to an undetermined distance upstream from Kane, Wyoming, is classified as impaired for contact recreation because of fecal-coliform bacteria (Wyoming Department of Environmental Quality, 2002).

Figure 7. Annual hydrograph (1952-2001).

Figure 7. Annual hydrograph (1952-2001) and photograph for the Bighorn River at Kane, Wyoming (site B).

Photograph for the Bighorn River at Kane, Wyoming.

Photograph from U.S. Geological Survey files

Yellowstone River at Forsyth, Montana (YF)

The Yellowstone River flows east-northeast from site YB through southern Montana. Site YF (fig. 2) is located at Forsyth, Montana, downstream from the confluence with the Bighorn River and about 240 river miles upstream from the confluence with the Missouri River (Shields and others, 2001). The basin upstream from site YF is located mostly within the Wyoming Basin (35 percent), Middle Rocky Mountains (33 percent), and Northwestern Great Plains (27 percent) ecoregions. Mean basin elevation is about 5,800 feet above NAVD 88, with elevations ranging from more than 13,800 feet above NAVD 88 in the Wind River Range to about 2,500 feet above NAVD 88 near site YF. Mean annual precipitation for the basin is about 18 inches. The geologic groups in the basin upstream from site YF are dominated by Cretaceous-period and Tertiary-period sedimentary rocks that generally extend over the basins and plains, including the area along the mainstem of the Yellowstone River and the lower-elevation areas in the Wind River and Bighorn Basins. The volcanic rocks that dominated the geologic groups of the Yellowstone River Basin at site YCS and site YB compose less than 16 percent of the basin at site YF.

The dominant land cover in the basin upstream from site YF is rangeland. About 36 percent of the drainage area land cover is herbaceous grassland and another 33 percent is shrubland (table 1). The shift in land cover from forested land to rangeland along the Yellowstone River between sites YCS and YB continues between sites YB and YF. Land cover adjacent to streams in the basin upstream from site YF also is dominated by rangeland; about 57 percent of the land cover on alluvial deposits in the basin is herbaceous grasslands and shrublands. Agricultural land uses are more prevalent adjacent to streams than for the distant parts of the basin; about 19 percent of the alluvial deposits in the basin have agricultural uses on them.

The hydrology of site YF is similar to the upstream site YB, where annual streamflows are dominated by a single snowmelt peak of moderate duration during late spring through early summer with low variability in daily mean streamflow throughout the year (fig. 8). The larger drainage area at site YF results in lower variability in streamflows at YF compared to YB. Annual mean streamflow at site YF is about 86 percent of the annual mean streamflow at site YS near the mouth; drainage area at site YF accounts for about 58 percent of the drainage area at site YS.

The Yellowstone River at site YF is classified by the State of Montana as Class B waters (Montana Department of Environmental Quality, 2002a). The Yellowstone River in this area only partially supports warm water fishery, possibly because of habitat alteration from stream modification and dam construction (Montana Department of Environmental Quality, 2002b).

Figure 8. Annual hydrograph (1978-2001).

Figure 8. Annual hydrograph (1978-2001) and photograph for the Yellowstone River at Forsyth, Montana (site YF).

Photograph for the Yellowstone River at Forsyth, Montana.

Photograph from U.S. Geological Survey files

Tongue River at Dayton, Wyoming (T)

The Tongue River flows east-northeast from the northeast flank of the Bighorn Mountains in northern Wyoming and through the plains of southeastern Montana to its confluence with the Yellowstone River at Miles City, Montana. Site T (fig. 2) is located near Dayton, Wyoming, at the mouth of a canyon; the drainage area for the site is located entirely within the Middle Rocky Mountains ecoregion (table 1). Mean basin elevation is about 8,500 feet above NAVD 88, with elevations ranging from more than 10,800 feet above NAVD 88 in the headwaters to about 4,100 feet above NAVD 88 near site T. Mean annual precipitation for the basin is about 27 inches, with about 34 percent of the total annual precipitation occurring on average during April through June. The geologic groups are composed of Precambrian-era crystalline rocks (46 percent) in the headwaters of the basin and upturned Paleozoic- and Mesozoic-era sedimentary rocks (53 percent) in the lower reaches of the basin. Included in these sedimentary units are carbonate rocks such as the Madison Limestone. Dissolution of these carbonates along fault zones and fold fractures has developed karstic features in some areas.

The dominant land cover in the basin upstream from site T is forested land (table 1). About 60 percent of the drainage area land cover is evergreen forest and 5 percent is deciduous forest. Other land cover in the basin includes rangeland (about 30 percent). The U.S. Forest Service (USFS) manages about 98 percent of the lands in the basin for multiple uses, including logging, grazing, and recreation (U.S. Bureau of Land Management, 2002).

The hydrology of site T is characteristic of streams with headwaters in the mountainous areas of the YELL. Annual streamflows are dominated by a single mountain snowmelt peak of moderate duration during late spring through early summer with low variability in daily mean streamflow throughout the remainder of the year (fig. 9). Compared to rainstorms in the basins and plains areas, mountain snowpacks are relatively consistent in spatial extent and between years, resulting in low variability in annual streamflows (Miller, 1999). Streamflows resulting from spring rain events upstream from site T generally are attenuated because of the low intensity and moderate duration of those events as well as geologic features. Streamflows at site T also are influenced by karstic features in the basin.

The Tongue River upstream from site T is classified by the State of Wyoming as Class 1 waters. A tributary to the Tongue River was previously listed as only partially supporting aquatic life because of residual chlorine in waste-water discharge from a USFS facility in the basin (Wyoming Department of Environmental Quality, 1998).

Figure 9. Annual hydrograph (1920-2001).

Figure 9. Annual hydrograph (1920-2001) and photograph for the Tongue River, Wyoming (site T).

Photograph for the Tongue River, Wyoming.

Photograph by Gregory K. Boughton, U.S. Geological Survey

Little Powder River above Dry Creek, near Weston, Wyoming (LP)

The Little Powder River flows north through the plains of northeastern Wyoming and southeastern Montana to its confluence with the Powder River. Site LP (fig. 2) is located about 5 mi south of the Wyoming-Montana state line. The drainage area for the site is located entirely within the Northwestern Great Plains ecoregion (table 1). Mean basin elevation is about 4,100 feet above NAVD 88, with elevations ranging from about 4,900 feet above NAVD 88 to about 3,400 feet above NAVD 88 near site LP. Mean annual precipitation for the basin is about 14 inches; about 39 percent of the total annual precipitation occurs on average during May and June. The geologic groups for the basin are composed of flat to gently dipping Tertiary-age sandstones, shales, coal beds, and clinker. Clinker refers to sedimentary rocks that have been thermally altered by fires in underlying coal beds. These distinctive red-colored rocks that cap topographic highs cover about 1,600 mi2 in northeastern Wyoming and southeastern Montana, much of which includes the Tongue and Powder Rivers of the Yellowstone River Basin. Clinker is characterized by high permeability and infiltration rates and large transmissivity and storage values (Heffern and Coates, 1997, 1999; Bartos and Ogle, 2002).

The dominant land cover in the basin upstream from site LP is rangeland (fig 2). About 81 percent of the drainage area land cover is herbaceous grassland and 11 percent is shrubland (table 1). Important land uses of these rangelands include grazing by livestock and wildlife, oil and gas development, and surface coal mining. Land cover adjacent to streams in the basin upstream from site LP also is dominated by rangeland. About 72 percent of the land cover on alluvial deposits in the basin is herbaceous grasslands and shrublands.

The hydrology of site LP generally is characteristic of small streams with headwaters in the plains areas of the YELL and streamflow is highly variable. Isolated precipitation events can substantially influence average daily mean streamflows. Annual streamflows at site LP consist of a plains (lowland) snowmelt peak during late winter through early spring (fig. 10). Several short to moderate duration rainstorm peaks are superimposed on the snowmelt peak and throughout the remainder of the summer. The hydrology of site LP is unique in some respects when compared to similar streams in the YELL. The occurrences and properties of clinker is important to the hydrology of site LP, attenuating streamflows from rainfall and snowmelt runoff, maintaining streamflows during low-flow periods, functioning as local aquifers, recharging underlying regional aquifers, and generally resulting in improved water quality (Lowry and Rankl, 1987; Heffern and Coates, 1997, 1999; Bartos and Ogle, 2002).

The Little Powder River is classified by the State of Wyoming as Class 2 waters (Wyoming Department of Environmental Quality, 2001a). A reach of the Little Powder River upstream from the Wyoming-Montana state line to an undetermined distance upstream is classified as threatened for contact recreation because of fecal-coliform bacteria (Wyoming Department of Environmental Quality, 2002).

Figure 10. Annual hydrograph (1973-2001).

Figure 10. Annual hydrograph (1973-2001) and photograph for the Little Powder River, Wyoming (site LP).

Photograph for the Little Powder River, Wyoming.

Photograph from U.S. Geological Survey files

Powder River near Locate, Montana (P)

Headwaters for the Powder River are in the plains and mountains of central and north-central Wyoming. The Powder River flows north through the plains of northeastern Wyoming and southeastern Montana to its confluence with the Yellowstone River. Site P (fig. 2) is near Locate, Montana, about 25 mi east of Miles City, Montana, and 29 river miles upstream from the confluence with the Yellowstone River (Shields and others, 2001). The drainage area for the site predominantly is located within the Northwestern Great Plains (88 percent) ecoregion (table 1). Mean elevation in the basin is about 4,600 feet above NAVD 88 and ranges from more than 13,100 feet above NAVD 88 in the Bighorn Mountains to about 2,400 feet above NAVD 88 near site P. Mean annual precipitation for the basin is about 14 inches. The geologic groups for the basin include flat to gently dipping Tertiary-period sedimentary rocks that compose 65 percent of the drainage area. Cretaceous-period sedimentary rocks compose 24 percent of the drainage area. Like the Little Powder River Basin, clinker exists in the basin, particularly along the eastern margin of the basin. Paleozoic- and Mesozoic-era sedimentary rocks and Precambrian-era crystalline rocks are exposed in the mountains along the western margin of the basin.

The dominant land cover in the basin upstream from site P is rangeland; about 59 percent of the drainage area land cover is herbaceous grassland and another 29 percent is shrubland (table 1). Important land uses of these rangelands include grazing by livestock and wildlife, oil and gas development, and surface coal mining. Land cover adjacent to streams in the basin upstream from site P also is dominated by rangeland. About 71 percent of the land cover on alluvial deposits in the basin is herbaceous grasslands and shrublands. About 12 percent of the alluvial deposits in the basin have agricultural uses on them.

The hydrology of site P is characteristic of large streams with headwaters in the mountainous areas of the YELL that flow across the basins and plains. Annual streamflow characteristics are a combination of the annual streamflow characteristics of mountain streams and basin and plains streams. Annual streamflows at site P consist of a plains (lowland) snowmelt peak during late winter through early spring followed by a peak from the mountain snowmelt during late spring through early summer (fig. 11). Several short to moderate duration rainstorm peaks are superimposed on the snowmelt peak and throughout the remainder of the summer. Annual mean streamflow at site P is only about 5 percent of the annual mean streamflow at site YS near the mouth of the Yellowstone River; drainage area at site P accounts for about 19 percent of the drainage area at site YS.

The Powder River downstream at site P is classified by the State of Montana as Class B waters (Montana Department of Environmental Quality, 2002a). The Powder River in this reach was not assessed by the State of Montana for impairments (Montana Department of Environmental Quality, 2002b).

Figure 11. Annual hydrograph (1939-2001).

Figure 11. Annual hydrograph (1939-2001) and photograph for the Powder River, Montana (site P).

Photograph for the Powder River, Montana.

Photograph from U.S. Geological Survey files

Yellowstone River near Sidney, Montana (YS)

The Yellowstone River flows east from site YF then northeast through the plains of southeastern Montana and into western North Dakota to its confluence with the Missouri River. The Tongue and Powder Rivers are major tributaries to the Yellowstone River in this part of the YELL. Site YS (fig. 2) is located near Sidney, Montana, about 29 river miles upstream from the confluence with the Missouri River (Shields and others, 2001). About 55 percent of the basin is located within the Northwestern Great Plains ecoregion (table 1). Mean basin elevation is about 5,000 feet above NAVD 88, with elevations ranging from more than 13,800 feet above NAVD 88 in the Wind River Range to about 1,900 feet above NAVD 88 near site YS. Mean annual precipitation for the basin is about 17 inches, which is substantially less than the mean annual precipitation at the mainstem sites (site YCS and site YB) in the upper basin. The geologic groups for the basin at site YS reflect the diverse geology of the sub-basins and ranges from Precambrian-era rocks and Tertiary-period volcanic rocks in the high mountains to predominantly Tertiary-period sedimentary rocks (48 percent) in the eastern part of the basin near Sidney, Montana. Cretaceous-period sedimentary rocks that generally are exposed along the flanks of uplifted areas compose 26 percent of the drainage area for the basin.

The dominant land cover in the basin upstream from site YS is rangeland. About 47 percent of the drainage area land cover is herbaceous grassland and 27 percent is shrubland (table 1). The shift in land cover characteristics downstream along the Yellowstone River from forested land to rangeland continues between sites YF and YS. Land cover adjacent to streams in the basin upstream from site YS also is dominated by rangeland. About 58 percent of the land cover on alluvial deposits in the basin is herbaceous grasslands and shrublands. About 20 percent of the alluvial deposits in the basin have agricultural uses on them.

The hydrology of site YS is characteristic of large streams with headwaters in the mountainous areas of the YELL that flow across the basins and plains. Annual streamflow characteristics are a combination of the annual streamflow characteristics of mountain streams and basin and plains streams. Annual streamflow statistics at site YS (water years 1967-2001) consist of a plains (lowland) snowmelt peak during late winter through early spring followed by a peak from the mountain snowmelt during late spring through early summer (fig. 12). The larger drainage area at site YS results in lower variability in streamflow.

The Yellowstone River at site YS is classified by the State of Montana as Class B waters (Montana Department of Environmental Quality, 2002a). The river upstream from Sidney, Montana is listed as only partially supporting warm-water fishery probably because of habitat alterations from stream modification and dam construction (Montana Department of Environmental Quality, 2002b).

Figure 12. Annual hydrograph (1967-2001).

Figure 12. Annual hydrograph (1967-2001) and photograph for the Yellowstone River near Sidney, Montana (site YS).

Photograph for the Yellowstone River near Sidney, Montana.

Photograph from U.S. Geological Survey files

Data Collection and Analysis

Stream-water quality of the YELL was assessed through the collection of samples generally on a monthly basis from January 1999 to September 2001. A summary of the data set used during analysis of the fixed sites by constituent grouping and calendar year is presented in table 4. In addition to the collection of environmental samples, quality-control samples were collected to estimate the bias and variability that result from sample collection, processing, and analysis. Sampling methods, data-analysis methods, and quality-control data are described in this section of the report.

Sampling Methods

One of the design components of the NAWQA Program is the use of nationally consistent methods for sample collection, field measurements, sample processing, and laboratory analyses. Samples were collected in accordance with established USGS methods and NAWQA protocols (Edwards and Glysson, 1988; Ward and Harr, 1990; Horowitz and others, 1994; Shelton, 1994, 1997; U.S. Geological Survey, 1997-2004). During normal streamflow conditions, samples generally were collected using depth-integrated samplers, using the equal-width-integrated method that is described in Edwards and Glysson (1988) and Ward and Harr (1990). When samples were collected from streams during extreme conditions, such as hazardous ice conditions and very low flows, traditional depth- and width-integrating techniques may not have been used. During these conditions, multiple-vertical or dip-sampling techniques were used. Dissolved organic carbon (DOC) and microbiological samples were collected separately from the primary sample with dip-sampling techniques. When a wading sample could not be collected, a weighted bottle sampler was used to obtain the DOC sample from the centroid of flow.

Field measurements of streams made at the time of sampling included instantaneous streamflow, water temperature, specific conductance, pH, dissolved oxygen, and alkalinity. Air temperature also was measured. All of the fixed sites were located at streamflow-gaging stations, so generally gage heights were measured and the instantaneous streamflow was determined using the most current streamflow rating curve. Stream temperature and dissolved oxygen were measured in-stream using a multi-parameter water-quality probe. A composite sample taken from the cone splitter was analyzed for specific conductance, pH, and alkalinity.

Water samples to be analyzed for inorganic and organic constituents were processed onsite using standard methods and equipment described by Horowitz and others (1994), Shelton (1994, 1997), and the U.S. Geological Survey (1997-2004). Samples that were analyzed for dissolved-inorganic constituents were filtered with a 0.45 micrometer (µm) disposable filter. Samples that were analyzed for dissolved-organic constituents were filtered with a 0.70 µm glass filter. Samples that were analyzed for total-constituent concentrations, which include dissolved and particulate forms, were unfiltered. Samples for fecal coliform and Escherichia coli (E. coli) were processed onsite using sterile techniques and membrane-filtration methods. Subsamples over a range of volumes were filtered in order to obtain countable plates within the ideal ranges of 20 to 60 colonies per 100 milliliters (col/100 mL) for fecal coliform and 20 to 80 col/100 mL for E. coli. Filters were plated, processed, and enumerated according to Myers and Wilde (1997).

Water samples were sent to the USGS National Water Quality Laboratory in Lakewood, Colorado, for analysis using standard USGS analytical methods (Fishman and Friedman, 1989; Faires, 1993; Fishman, 1993; McLain, 1993; Zaugg and others, 1995; Jones and Garbarino, 1999; Patton and Truit, 2000). Sediment samples were sent to the USGS sediment laboratory in Helena, Montana, and were analyzed for concentration in accordance with methods described in Guy (1969) and Lambing and Dodge (1993).

Table 4. Summary of data collected by constituent grouping and calendar year for fixed sites in the Yellowstone River Basin, 1999-2001.

[X; at least one sample collected during calendar year]

Site report identifier (fig. 2)	Field measurements, major ions, nutrients, iron, manganese, and suspended sediment			Bacteria			Trace elements (other than iron and manganese)			Pesticides
	1999	2000	2001	1999	2000	2001	1999	2000	2001	1999	2000	2001
SB	X	X	X		X	X	X	X	X
YCS	X	X	X		X	X	X
CF	X	X	X		X	X	X			X
YB	X	X	X		X	X	X			X
B	X	X	X		X	X	X			X
YF	X	X	X		X	X	X	X	X	X	X	X
T	X	X	X		X	X	X
LP	X	X	X		X	X	X	X	X
P	X	X	X		X	X	X	X	X
YS	X	X	X		X	X	X	X	X	X	X	X

Data-Analysis Methods

Data in this report are summarized using parametric and nonparametric statistics. Descriptive summary statistics were computed using standard methods. Some constituent concentrations were less than laboratory reporting levels (censored data). Statistics of constituent concentrations that included censored data were estimated using robust methods (Helsel and Cohn, 1988; Helsel and Hirsch, 1992). Robust methods use distributions fit to data that are greater than the reporting level(s) to estimate summary statistics. In this report, summary statistics for most data sets with censored values were estimated using log-probability regression. In some cases, data were censored to a consistent reporting level in order to compare data through time or across constituents. Summary statistics are shown using boxplots for some constituents. For boxplots, the lower and upper edges of the box indicate the 25th and 75th percentiles, respectively. The median is a line within the box, and whiskers extend to the 10th and 90th percentiles. For some boxplots, values outside the 10th and 90th percentiles are shown as individual points. Laboratory reporting levels are indicated on the boxplots that were produced from data sets with censored values.

Parametric statistical techniques (Pearson's r) were used to test for correlations between data sets when data were normally distributed. Transformation of a data set may have been used to achieve a normal distribution. Nonparametric statistical techniques were used to test for correlations between data sets when the data distributions were unknown. For data sets where the sample size was less than 20, Kendall's tau was used to measure the strength and direction of the relation between variables. Spearman's correlation coefficient (Spearman's rho) was used to measure the strength and direction of the relation between variables of larger data sets (Helsel and Hirsch, 1992). Kendall's tau and Spearman's Rho use rank-based procedures rather than actual data values and are resistant to the effects of outliers. The nonparametric Kruskal-Wallis test was used to compare whether concentrations for selected constituents varied seasonally or annually. The Kruskal-Wallis test uses data ranks rather than actual data values to reduce the effect of outliers. In the most general form, the Kruskal-Wallis test determines whether three or more groups of ranked data have similar distributions or at least one group differs in its distribution (Helsel and Hirsch, 1992). Data are presented with p-values for correlations and statistical tests. The p-value for correlations indicates the probability of determining the correlation strength by chance alone. Statistical significance for tests was determined using a 95 percent confidence level (p-values <0.05).

Water quality is discussed in terms of constituent concentrations, loads, and yields. Constituent concentrations define the water-quality conditions at a point in time and generally are used for assessing the suitability of the water for aquatic life and other uses. For selected constituents, statistics of flow-weighted mean concentrations and estimated loads and yields also are summarized. Flow-weighted mean concentrations were computed from the sum of estimates of daily concentrations divided by the sum of daily streamflows. Estimates of daily concentrations were computed using the rating-curve method (Cohn and others, 1989) and the computer program LOADEST (Crawford, 1991). The LOADEST program uses robust estimation methods for data that are censored. Because streamflow data are typically skewed, statistics of sample concentrations collected at equal intervals (for example, monthly) can be skewed depending on the relation of the constituent concentration with streamflow. A flow-weighted concentration is influenced less by this skew because it is representative of the concentration of the total mass discharge (load) of the constituent. Similarly, flow-weighted concentrations allow for comparisons between sites with differing sampling frequencies. Estimates of daily concentrations using this method are subject to large errors; as such, only annual and seasonal statistics are used in this report. Estimated constituent loads and yields (load divided by drainage area) also were computed using the rating-curve method (Cohn and others, 1989) and the computer program LOADEST (Crawford, 1991).

Quality Control

The data-quality objectives of the NAWQA Program include the collection of field quality-control samples for inorganic and organic constituents as part of the fixed-site sampling. Quality-control samples, including equipment blanks, replicates, and matrix spikes, are used to estimate the extent to which contamination, measurement variability, and matrix interference affect the interpretation of the environmental data (Mueller and others, 1997). Equipment blank samples were prepared in the field at each of the fixed sites by processing inorganic-grade or pesticide-grade deionized water through the sampling equipment immediately before collecting the environmental sample. Replicate samples were prepared at each of the fixed sites by splitting the environmental sample into duplicate samples. Matrix spikes were prepared by injecting a laboratory-prepared spike solution of pesticide compounds into replicate environmental samples at sites where samples for pesticides were collected. In addition to field quality-control samples for pesticides, surrogate solutions were added to all of the pesticide samples at the laboratory before solid-phase extraction of pesticides and were used to measure the extraction efficiency of each individual sample.

Summary statistics of the concentrations of major ions, nutrients, dissolved organic carbon, and selected trace elements for field equipment blank samples collected at all 10 of the fixed sites are presented in table 5. Concentrations of constituents in equipment blank samples were smaller than the reporting level for the 90th percentile for all the constituents in table 5, except silica. The detectable concentrations of silica in blank samples were an order of magnitude smaller than silica concentrations in environmental samples. The results of the blank samples indicate that sampling collection, processing, and analysis methods and equipment were not routinely introducing contaminants that would cause a substantial positive bias in the environmental samples and affect the interpretation of the environmental data.

Table 5. Statistical summary for concentrations of selected constituents in field equipment blank samples for fixed sites in the Yellowstone River Basin, 1999-2001.

[<, less than]

Constituent	Number of samples	10th percentile	50th percentile (median)	90th percentile
Major ions and dissolved solids, in milligrams per liter
Calcium, dissolved	17	<0.02	<0.02	<0.02
Magnesium, dissolved	17	<.014	<.014	<.014
Potassium, dissolved	16	<.24	<.24	<.24
Sodium, dissolved	17	<.09	<.09	<.09
Chloride, dissolved	16	<.29	<.29	<.29
Flouride, dissolved	16	<.2	<.2	<.2
Silica, dissolved	17	<.09	<.09	.40
Sulfate, dissolved	16	<.3	<.3	<.3
Solids, dissolved, residue on evaporation	16	<10	<10	<10
Nutrients, in milligrams per liter
Ammonia plus organic nitrogen, dissolved as nitrogen	18	<0.1	<0.1	<0.1
Ammonia plus organic nitrogen, total as nitrogen	18	<.1	<.1	<.1
Ammonia, dissolved as nitrogen	19	<.04	<.04	<.04
Nitrite, dissolved as nitrogen	19	<.01	<.01	<.01
Nitrite plus nitrate, dissolved as nitrogen	19	<.05	<.05	<.05
Orthophosphate, dissolved as phosphorus	19	<.02	<.02	<.02
Phosphorus, dissolved as phosphorus	18	<.006	<.006	<.006
Phosphorus, total as phosphorus	18	<.05	<.05	<.05
Organic carbon, dissolved	19	<.3	<.3	<.3
Trace elements, in micrograms per liter
Arsenic, dissolved	12	<2	<2	<2
Copper, dissolved	12	<1	<1	<1
Iron, dissolved	17	<10	<10	<10
Manganese, dissolved	21	<3	<3	<3
Selenium, dissolved	11	<1	<1	<1
Zinc, dissolved	12	<2	<2	<2

Field equipment blanks for pesticide samples were processed at site CF, site YF, and site YS. None of the 47 pesticides that were analyzed were detected at concentrations larger than the reporting level in any of the 9 equipment blank samples. An estimated concentration of 0.001 microgram per liter (µg/L) of one pesticide, tebuthiuron, was detected in one of the equipment blank samples. Tebuthiuron is a broad spectrum herbicide used to control weeds in noncropland areas, rangelands, rights-of-way, and industrial sites (Meister Publishing Company, 2001). Because the environmental sample collected with this blank sample at site YS also contained an estimated concentration of tebuthiuron, the environmental sample analysis of tebuthiuron was not included in the pesticide summary in this report because the sample potentially could have been contaminated with a low-level concentration of tebuthiuron.

Measurement variability was determined for constituents using environmental and replicate samples from the 10 fixed sites by calculating a relative percent difference (RPD) when concentrations of the constituent were detected in both the environmental sample (sample 1) and the replicate sample (sample 2) or the constituent was censored at a common level in both samples. The RPD was calculated using the equation:

Equation

Summary statistics of the RPD for major ions, nutrients, and selected trace elements indicate that variability was minimal for most constituents (table 6). The median RPD in replicate samples was less than 5 percent for 20 of the 24 major-ion, nutrient, and trace-element constituents. A few constituents had an occasional large value for the RPD for some replicate samples, but this generally occurred when concentrations were small and near the reporting level. For example, the RPD for total ammonia plus organic nitrogen was 39.4 percent when the environmental sample had a concentration of 0.218 milligrams per liter (mg/L) and the replicate had a concentration of 0.325 mg/L.

Replicate samples also were collected to measure variability of pesticide compounds. Out of 47 pesticide compounds, 35 compounds were reported as less than the reporting level in both the environmental and replicate sample in all eight of the replicate sample sets. Four compounds were detected in one of the samples but concentrations were less than the reporting levels in the replicate, so the RPDs could not be determined. The RPD was determined for eight compounds that had a measurable concentration in both the environmental sample and the replicate sample (table 6). Atrazine, the most commonly detected pesticide in the replicate samples, had a median RPD of 5.21 percent. The larger median RPD for pesticide compounds compared to the other constituents is probably because concentrations were small and near reporting levels.

Field matrix-spike samples were collected to determine the effects of matrix interference or analyte degradation on concentrations of pesticides in environmental samples. During method development, average mean recovery was about 83 percent in laboratory spike samples for all pesticides combined at low-level concentrations (Zaugg and others, 1995). The average median recovery for the field matrix spikes was 98 percent (fig. 13) and the average mean recovery was 101 percent for all pesticide compounds in the YELL matrix-spike samples. Deethylatrazine demonstrated a small recovery (9 to 19 percent) during method development (Zaugg and others, 1995), but had a mean recovery of 57 percent in field matrix spikes; however, environmental concentrations from this breakdown product may still be biased low. Carbaryl, carbofuran, methyl azinphos, and terbacil were identified during method development as having variable performance (Zaugg and others, 1995). The range in percent recovery for spiked samples from the YELL was variable for these compounds, particularly for carbaryl, carbofuran, and methyl azinphos, which had percent recoveries larger than 200 percent for some samples. These compounds with variable performance, however, were detected infrequently in the environmental samples. The laboratory reported all concentrations for carbaryl, carbofuran, and Deethylatrazine as estimated concentrations in spiked and environmental samples because of their variable performance during method development (Zaugg and others, 1995). The low mean recoveries for p'p-DDE in field matrix spikes were consistent with low mean recoveries during method development. The poor recovery of atrazine in two of the field matrix-spike samples was in a spike set where the concentration in the environmental sample was larger than the associated spike samples.

Table 6. Statistical summary of the relative percent difference determined for selected constituents in replicate samples for fixed sites in the Yellowstone River Basin, 1999-2001.

[--, not applicable]

Constituent	Number of samples	Relative percent difference
Constituent	Number of samples	10th percentile	50th percentile (median)	90th percentile
Major ions and dissolved solids
Calcium, dissolved	20	0.26	0.94	2.36
Magnesium, dissolved	20	.20	.89	2.46
Potassium, dissolved	20	.31	1.14	4.00
Sodium, dissolved	20	.15	1.35	3.03
Chloride, dissolved	20	.05	1.35	4.18
Flouride, dissolved	20	.0	.74	3.19
Silica, dissolved	20	.09	.29	1.69
Sulfate, dissolved	20	.11	.40	1.66
Solids, dissolved, residue on evaporation	20	.0	.51	1.22
Nutrients
Ammonia plus organic nitrogen, dissolved as nitrogen	20	1.15	4.13	23.9
Ammonia plus organic nitrogen, total as nitrogen	21	1.23	8.60	39.4
Ammonia, dissolved as nitrogen	21	.0	.0	2.96
Nitrite, dissolved as nitrogen	21	.0	.0	10.5
Nitrite plus nitrate, dissolved as nitrogen	20	.0	.69	3.36
Orthophosphate, dissolved as phosphorus	20	.0	.0	19.5
Phosphorus, dissolved as phosphorus	21	.0	5.53	28.6
Phosphorus, total as phosphorus	21	.0	3.17	12.9
Organic carbon, dissolved	20	.0	2.14	5.68
Trace elements
Arsenic, dissolved	12	.097	4.46	12.0
Copper, dissolved	9	--	7.18	--
Iron, dissolved	19	.0	.0	16.8
Manganese, dissolved	23	.0	1.22	18.2
Selenium, dissolved	10	.0	.0	25.8
Zinc, dissolved	5	--	32.6	--
Pesticides
Atrazine, dissolved	7	--	5.21	--
Cyanazine, dissolved	2	--	¹6.84	--
Deethylatrazine, dissolved	3	--	.98	--
Malathion, dissolved	1	--	²11.3	--
Metolachlor, dissolved	2	--	¹1.15	--
Prometon, dissolved	2	--	¹14.5	--
Triallate, dissolved	3	--	12.2	--
Trifluralin, dissolved	1	--	²7.87	--

¹Average of two relative percent differences

²Relative percent difference for one sample

In addition to the field matrix-spike samples, surrogate compounds are added to environmental samples at the laboratory and are used for assessing pesticide recoveries. Surrogate compounds are compounds that are not expected to be in the environmental sample, but are expected to behave like pesticide compounds that may be found in environmental samples. The gas chromatography-mass spectrometry (GCMS) method of pesticide analysis for data in this report uses diazinon-d10 and alpha-HCH-d6 as surrogates. The median recoveries for the surrogate compounds in the fixed-site samples were about 105 percent for diazinon-d10 and about 99 percent for alpha-HCH-d6, indicating good extraction efficiency of the environmental samples (fig. 14).

Figure 13. Recovery of pesticide compounds in field matrix spikes for fixed sites in the Yellowstone River Basin, 1999-2001.

Figure 13. Recovery of pesticide compounds in field matrix spikes for fixed sites in the Yellowstone River Basin, 1999-2001.

Figure 14. Recovery of surrogate compounds in pesticide samples for fixed sites in the Yellowstone River Basin, 1999-2001.

Figure 14. Recovery of surrogate compounds in pesticide samples for fixed sites in the Yellowstone River Basin, 1999-2001.

Fixed-Site Water Quality, 1999-2001

The hydrologic conditions during sampling events and results for field measurements and analyses for fecal-indicator bacteria, major ions, dissolved solids, nutrients, trace elements, pesticides, and suspended sediment for the samples collected at the 10 fixed sites in the YELL during 1999-2001 are summarized in this section of the report. Results for 400 samples are included in the data analysis for this report. The number of results for individual constituents may be slightly different because field equipment malfunctioned or only partial laboratory results were received for a few samples. The discrete data values are presented in Swanson and others (2000, 2001, 2002). In addition, data are available online at URL http://nwis.waterdata.usgs.gov/nwis.

Hydrology

Streamflows at the fixed sites during the study period generally ranged from slightly greater than average during water year 1999 to much less than average during the drought conditions of water years 2000 and 2001. Streamflow records indicate that water year 2001 was one of the driest years in the last century for the YELL. For some of the fixed sites, annual mean streamflows for water year 2001 were the lowest on record (site CF, 63 years; site YB, 73 years; site YF, 24 years; site T, 70 years). For the period following regulation by major reservoirs, annual mean streamflows in water year 2001 were the third lowest at site B (50 years) and the lowest at site YS (35 years). For other fixed sites, annual mean streamflows for water year 2001 were the third (site YCS, 95 years; site P, 63 years) or fourth (site LP, 29 years) lowest on record. Streamflow records at site SB were only available for the sampling period (1999-2001). Water-quality samples collected at the fixed sites during the study period generally covered the range of streamflows that have been recorded at the sites (fig. 15) and were representative of historical streamflow conditions in spite of the drought conditions of water years 2000 and 2001.

Figure 15a. Distribution of samples relative to daily mean streamflow duration for fixed sites in the Yellowstone River Basin, 1999-2001.

Figure 15b. Continued.

Figure 15. Distribution of samples relative to daily mean streamflow duration for fixed sites in the Yellowstone River Basin, 1999-2001.

Field Measurements

Field measurements of streams made at the time of sampling included water temperature, specific conductance, pH, dissolved oxygen and alkalinity. All of these measurements were collected onsite and provide a general picture of the water-quality conditions at the fixed sites. Boxplots of water temperature, specific conductance, pH, and dissolved-oxygen concentrations are shown in figure 16. Alkalinity is discussed in the "Major Ions and Dissolved Solids" section of this report.

Water temperature is an important field measurement because it has a direct influence on the physical, chemical, and biological processes that occur in streams. Some of the physical properties that are affected by temperature include density, surface tension, gas solubility, and diffusibility. The gain and loss of dissolved gases, such as dissolved oxygen and dissolved nitrogen, are in part controlled by these physical properties. The rates at which chemical reactions proceed are influenced by temperature, with higher temperatures generally increasing chemical reaction rates (Stevens and others, 1975). Biological processes that are affected by water temperature include the metabolism of aquatic organisms and their ability to survive and reproduce. Water temperatures in streams are affected by solar radiation, precipitation, air and ground temperature, and the temperature of tributary and ground-water inflows.

Median stream temperatures varied among sites in the YELL, ranging from 2.5 degrees Celsius (°C) at site SB to 15°C at site YS (fig. 16). The low median temperatures at site SB, site YCS, and site T result from cooler air temperatures at the higher elevations and less solar radiation reaching streams in forested areas in comparison to conditions at other sites in the YELL. The maximum temperature at site YCS (22°C) was higher compared to the maximum temperatures at site SB (12°C) and site T (17.5°C), which probably reflects inputs from the thermal springs upstream from site YCS. Median stream temperatures in the Yellowstone River increased as it flowed from the forested area onto the plains. The median temperature more than doubled from site YCS (7.0°C) to site YS (15°C). These changes in the stream temperature are likely due to combinations of natural causes, such as increased solar radiation as a result of fewer trees, increased air temperatures at lower elevations, and human causes, including irrigation return flows. Site B had the most variability in stream temperature, ranging from 0°C to 27.5°C, probably mostly due to the wide range of air temperatures that are characteristic of the Bighorn River Basin, which ranged from –10°C to 35°C during sampling events. Irrigation return flows also are common in the Bighorn River. Stream temperatures are not compared to water-quality criteria for temperature in this report because water-quality criteria for stream temperatures in Montana and Wyoming primarily are not a single value, but rather are related to allowable increases to stream temperatures resulting from anthropogenic inputs, such as waste effluent (Wyoming Department of Environmental Quality, 2001b; Montana Department of Environmental Quality, 2002a).

Specific conductance is a measure of a substance's ability to conduct an electrical current at a specific temperature. Pure water has a very low electrical conductance; as concentrations of dissolved solids that carry an ionic charge increase in a stream, the specific conductance increases. Therefore, specific conductance can be used as an indication of the concentration of dissolved solids in a stream. The conversion factor from specific conductance to dissolved-solids concentrations for natural waters is about 0.54 to about 0.96 (Hem, 1985). The median conversion factor between specific conductance and dissolved-solids concentrations varied for sites in the YELL and ranged from about 0.57 for site T to about 0.81 for site LP. For example, a sample with a specific conductance of 100 microsiemens per centimeter at 25°C (µS/cm) for site T would have a corresponding dissolved-solids concentration of about 57 mg/L, and a sample with a specific conductance of 1,000 µS/cm for site LP would have a corresponding dissolved-solids concentration of about 810 mg/L. Conversion factors tend to be larger for streams that have large sulfate concentrations, like site LP. Water-quality criteria for specific conductance have not been established for most streams by the States of Montana and Wyoming (Wyoming Department of Environmental Quality, 2001b, Montana Department of Environmental Quality, 2002a). A numeric standard for a single sample of 2,500 µS/cm has been established for the Powder River and Little Powder River in Montana (Montana Department of Environmental Quality, 2002a).

Median specific conductance values varied substantially at fixed sites, ranging from 202 µS/cm at site SB to 3,400 µS/cm at site LP (fig. 16). Maximum specific-conductance values for sites near mountainous areas (site SB, site YCS, and site T) were lower than median specific-conductance values for streams that flow primarily through the basins and plains (site CF, site YB, site B, site YF, site LP, site P, and site YS), owing to higher precipitation and more resistant rock types in mountainous areas. Specific-conductance values had the largest range at the rangeland indicator site, site LP (358 µS/cm to 4,460 µS/cm), where Tertiary-period sedimentary rocks are the predominant rock type in the basin. Specific conductance was larger than the numeric standard of 2,500 µS/cm in more than 75 percent of the samples from site LP and about 18 percent of the samples at site P.

Hydrogen-ion activity is a measure of the acidic or alkaline character of the stream and is expressed in terms of pH, which is the negative base-10 log of the hydrogen-ion activity. A pH value of 7 is considered neutral, whereas a pH value greater than 7 is considered alkaline and a pH value less than 7 is considered acidic. Water-quality criteria for pH are not a single number. In the State of Montana, criteria for pH for Class B waters are based on levels of induced variation in pH of less than 0.5 pH units, within the range of 6.5 to 8.5 or 9.0 (Montana Department of Environmental Quality, 2002a). In the State of Wyoming, chronic values of pH in the range of 6.5 to 9.0 are allowed (Wyoming Department of Environmental Quality, 2001b). Values for pH that are outside of these ranges are considered harmful to aquatic life.

All stream measurements of pH at fixed sites in the YELL were alkaline, indicating that the soils and rocks naturally buffer the atmospheric waters, which typically have a pH of 5.65 (Hem, 1985). The median pH values at the fixed sites ranged from 7.8 at site T to 8.4 at site YS (fig. 16). The minimum pH value that was measured was 7.2 at site YCS and site T, and the maximum pH value that was measured was 8.9 at site YF. All values for pH were within the water-quality criteria established by the States of Montana and Wyoming.

Dissolved-oxygen analysis is the measure of gaseous oxygen dissolved in an aqueous solution. Oxygen, although soluble in water, is considered poorly soluble because it does not react chemically with water (Stednick, 1991). Concentrations of dissolved oxygen primarily are affected by atmospheric pressure and temperature. Atmospheric pressure and dissolved oxygen are directly proportional to each other, whereas temperature and dissolved oxygen are inversely proportional; so, as pressure increases and temperature decreases, oxygen becomes more soluble in a stream. Concentrations of dissolved oxygen in streams are depleted through respiratory uptake by organisms and by the decaying of organic matter in streams. Oxygen is replenished from the atmosphere and by photosynthesis of aquatic plants. Water-quality criteria for dissolved oxygen that have been determined for the States of Montana and Wyoming are not a single number and depend on aquatic life stages present in a stream (Wyoming Department of Environmental Quality, 2001b, Montana Department of Environmental Quality, 2002a) and, as such, are not used for comparison in this report, except in general terms.

Median dissolved-oxygen concentrations varied slightly among sites and ranged from 9.3 mg/L at site YS to 11.2 mg/L at site SB (fig. 16), which are acceptable concentrations when compared to State criteria for minimum concentrations for all life forms. The median dissolved-oxygen concentrations tended to be larger at sites that are closer to mountainous areas, where stream temperatures also were the lowest; thus, making oxygen more soluble, despite lower atmospheric pressure with elevation. The median dissolved-oxygen concentrations tended to be lower at sites in the basin and plains, like site LP, where stream temperatures generally are warmer, although atmospheric pressure tends to be higher. Dissolved-oxygen conditions for all sites generally were near saturation (100 percent); median values ranged from 94.6 percent of saturation at site LP to 109 percent of saturation at site YCS.

Figure 16. Statistical summary of water temperature, specific conductance, pH, and dissolved-oxygen concentrations for fixed sites in the Yellowstone River Basin, 1999-2001.

Figure 16. Statistical summary of water temperature, specific conductance, pH, and dissolved-oxygen concentrations for fixed sites in the Yellowstone River Basin, 1999-2001.

Fecal-Indicator Bacteria

Analyses for fecal-indicator bacteria of fecal coliform and E. coli were added to the fixed-site sampling during 2000 and 2001 to assess the sanitary quality at the fixed sites in the YELL, because the presence of these bacteria can be an indication that contamination from fecal material has occurred. Fecal coliform was identified as the primary cause of impairment for streams in Wyoming in the 2002 305(b) report, including several reaches in the Bighorn River and Tongue River Basins (Wyoming Department of Environmental Quality, 2002). Antiquated facilities and shifting ground resulted in nine sewage spills in Yellowstone National Park during 1999-2001 (Associated Press, 2001a, 2001b; Wyoming Department of Environmental Quality, 2002). Stream reaches in Montana that were identified with fecal-coliform impairments by the Montana Department of Environmental Quality were outside of the YELL (Montana Department of Environmental Quality, 2002b). Although fecal-indicator bacteria do not necessarily cause illness themselves, large levels of fecal-indicator bacteria can indicate the possible presence of other pathogens that cause such waterborne diseases as gastroenteritis and bacillary dysentery, typhoid fever, and cholera (Myers and Sylvester, 1997).

The occurrence of fecal coliform in recreational waters is not positive confirmation that fecal contamination has occurred because at least one member of the fecal-coliform group, Klebsiella, has non-fecal sources, including pulp and paper mill effluents, textile processing plant effluents, cotton mill wastewaters, and sugar beet wastes (U.S. Environmental Protection Agency, 1986a). The presence of E. coli, another member of the fecal-coliform group, is direct evidence that fecal contamination has occurred because the only sources are humans or other warm-blooded animals (Cabelli, 1977; Dufour, 1977). E. coli was determined to have a stronger relation to swimming-associated gastrointestinal illness than fecal coliform and, as such, was determined to be a better fecal-indicator bacteria for monitoring recreational waters (Dufour, 1984; U.S. Environmental Protection Agency, 1986a). However, because both the States of Montana and Wyoming were using fecal coliform in their water-quality criteria at the time of this study (Wyoming Department of Environmental Quality, 2001b, Montana Department of Environmental Quality, 2002a), it was important to monitor for both of the fecal-indicator bacteria.

The water-quality criteria for fecal coliform for the States of Montana and Wyoming are not a single number and are based on multiple samples in a 30-day period, the class of water, time of year, or location relative to sewage treatment outfalls (Wyoming Department of Environmental Quality, 2001b, Montana Department of Environmental Quality, 2002a). Because multiple samples were not collected during a 30-day period in most cases, recommended limits for single samples that were proposed by the U.S. Environmental Protection Agency (USEPA) are used in this report for assessing the magnitude of fecal-indicator-bacteria concentrations.

Concentrations

The medians and ranges of fecal-indicator-bacteria concentrations varied among the fixed sites (fig. 17). For all sites, the largest median concentrations for fecal coliform of 135 col/100 mL and E. coli of 156 col/100 mL were at site LP, which is the rangeland indicator site. The largest concentration in a single sample from all 156 samples of fecal coliform was 1,400 col/100 mL at site P. For all 145 samples of E. coli, the largest concentration in a single sample was 1,500 col/100 mL at site LP. The Little Powder River and Powder River Basins are both dominated by grasslands that are grazed by cattle; however, small towns, rural domestic homes with septic systems, and agricultural return flows also may contribute to bacteria in these basins. For all sites, the smallest median concentrations were 3 col/100 mL of fecal coliform and 3 col/100 mL for E. coli at site YCS, which drains a predominantly forested basin of Yellowstone National Park. Concentrations of fecal-indicator bacteria for the mineral resource area (site SB) and the forested (site T) indicator sites, which both drain predominantly forested areas, were small as well; median concentrations were 14 col/100 mL or less for fecal coliform and E. coli at both sites. Fecal-indicator- bacteria concentrations at site SB, site YCS, and site T may be good indicators of wildlife contributions in forested drainage basins in the YELL.

The largest median concentrations of fecal coliform (55 col/100 mL) and E. coli (22 col/100 mL) on the mainstem of the Yellowstone River were at site YB, which is downstream from parts of the Billings, Montana urban area. The median fecal-indicator-bacteria concentrations at site YB were smaller than median concentrations of fecal coliform (540 col/100 mL) and E. coli (420 col/100 mL) for sites with urban land use during a NAWQA synoptic study of fecal-indicator bacteria in Wyoming (Clark and Gamper, 2003); however, the streams sampled as part of the synoptic study had substantially smaller streamflow than the Yellowstone River.

Comparison to Recommended Limits

Historically, USEPA studies determined that statistically significant swimming-associated gastrointestinal illness may occur when concentrations of fecal coliform for a single sample are larger than 400 col/100 mL (U.S. Environmental Protection Agency, 1976). The USEPA recommends four different limits for E. coli concentrations for a single sample, depending on the degree of exposure with the source waters. The recommended limit for E. coli for a single sample defined in the USEPA study are: 235 col/100 mL for designated beach areas, 298 col/100 mL for moderate use, full-body contact recreation, 406 col/100 mL for light use, full-body contact recreation, and 576 col/100 mL for infrequent use, full-body contact recreation (U.S. Environmental Protection Agency, 1986a). The recommended limit of 235 col/100 mL was not used for comparison in this report because none of the fixed sites were at designated beaches. Values that exceed recommended limits do not necessarily indicate violations of water-quality criteria for the State of Montana or Wyoming, but are used for comparative purposes.

Concentrations of fecal-indicator bacteria in some samples exceeded recommended criteria. The USEPA recommended single-sample limit for fecal coliform was exceeded in 2.6 percent of the 156 samples for fecal coliform, and the recommended single-sample limit for moderate use, full-body contact recreation for

National Water-Quality Assessment Program

Scientific Investigations Report 2004-5113

Water-Quality Assessment of the Yellowstone River Basin, Montana and Wyoming—Water Quality of Fixed Sites, 1999-2001

By Kirk A. Miller, Melanie L. Clark, and Peter R. Wright

The PDF for the report is 16,798 kb

Table of Contents

Figures

Tables

Foreword

Conversion Factors and Datum

Abbreviated Water-Quality Units

Abbreviations

Abstract

Introduction

Purpose and Scope

Description of Study Unit

Acknowledgments

Environmental Settings of Fixed Sites

Design and Characteristics of the Fixed-Site Network

Soda Butte Creek at YNP boundary near Silver Gate, Montana (SB)

Yellowstone River at Corwin Springs, Montana (YCS)

Clarks Fork Yellowstone River at Edgar, Montana (CF)

Yellowstone River at Billings, Montana (YB)

Bighorn River at Kane, Wyoming (B)

Yellowstone River at Forsyth, Montana (YF)

Tongue River at Dayton, Wyoming (T)

Little Powder River above Dry Creek, near Weston, Wyoming (LP)

Powder River near Locate, Montana (P)

Yellowstone River near Sidney, Montana (YS)

Data Collection and Analysis

Sampling Methods

Data-Analysis Methods

Quality Control

Fixed-Site Water Quality, 1999-2001

Hydrology

Field Measurements

Fecal-Indicator Bacteria

Concentrations

Comparison to Recommended Limits