UNITED STATES ENVIRONMENTAL PROTECTION AGENCY REGION 8
%
DENVER, CO 80202-1129 Phone 800-227-!917
Ref: 8EPR-EP
Mr.Steve Gunderson Diremr Water Quality Control Division Colorado Department of Public Health and Environment 4300 Cherry Creek Drive South Denver, Colorado 80246- 1 53 0
(I L Approvals mison River and Tribdwies, Uncompahgre and Tribasdmies, Selenium
% Dear M/m&rson:
(TMDLs) as subrnitkd by your We have completed our review of the total to this letter. In accordance with office on January 2 1,20 1 1 for the waterbodies of TMDLs as developed for certain the Clban Water Act (33 U.S.C. 1251 et. seq.), 303(d)(l). Based on our review, pollu+ts in water quality limited waterbodies we feel the separate TMDL elements for the pollutants listbd in the enclosed table are adequately a margin of safety. addre& taking into consideration.sasonal variation
1
Thank you for submitting these TMDLs for our rdiew and approval. If you have any questions, the myst knowledgeable person on my staff is Julie Kinseq and she may be reached at (303) 3 12-7065.
Carol
4.Campbell
Assistant Regional Administrator
Ofice of Ecosystems Protection arfd Remediation
@
Printed m Recyded Papw
Total Maximum Daily Load Assessment Gunnison River and Tributaries Uncompahgre River and Tributaries Delta/Mesa/Montrose Counties, Colorado
Colorado Department of Public Health and Environment Water Quality Control Division January, 2011 FINAL
January 2011
1
Final
TMDL Summary
Pollutant Addressed: selenium (dissolved)
Waterbody ID/Waterbody Description
COGUNF03: Mainstem of North Fork of the Gunnison River from the Black Bridge (41.75 Drive) above Paonia to the confluence with the Gunnison River. COGUNF05: Mainstems of Hubbard Creek, Terror Creek, Minnesota Creek, and Leroux Creek from their boundary with national forest land to their confluences with the North Fork of the Gunnison River; mainstem of Jay Creek from its source to its confluence with the North Fork of the Gunnison River; mainstem of Roatcap Creek including all tributaries, wetlands, lakes and reservoirs, from its source to its confluence with the North Fork of the Gunnison. COGUNF06a: All tributaries to the North Fork of the Gunnison River including all lakes, reservoirs, and wetlands which are not on national forest lands, except for the specific listings in Segments 4, 5, 6b and 7. COGUNF06b: Mainstem and all tributaries to Bear Creek, Reynolds Creek, Bell Creek, McDonald Creek, Cottonwood Creek, Love Gulch, Cow Creek, Dever Creek, German Creek, Miller Creek, Stevens Creek, Big Gulch, Stingley Gulch and Alum Gulch including lakes, reservoirs, and wetlands which are not on national forest lands from their source to the North Fork of the Gunnison River. COGUUN04b: Mainstem of the Uncompahgre River from LaSalle Road to Confluence Park. COGUUN04c: Mainstem of the Uncompahgre River from Confluence Park to the confluence with the Gunnison River. COGUUN12: All tributaries to the Uncompahgre River, including all wetlands, lakes and reservoirs, from the South Canal near Uncompahgre to the confluence with the Gunnison River, except for specific listings in Segments 13, 14 and 15a and 15b. COGULG01: Mainstem of the Gunnison River from the outlet of Crystal Reservoir to a point immediately above the confluence with the Uncompahgre River. COGULG02: Mainstem of the Gunnison River from a point immediately above the confluence with the Uncompahgre River to the confluence with the Colorado River. COGULG04a: All tributaries to the Gunnison River, including all wetlands which are not on national forest lands from the outlet of Crystal Reservoir to the confluence with the Colorado River, except for specific listings in the North Fork and Uncompahgre River subbasins and wetlands and in Segments 3, 4b, 5 through 10, 12 and 13. COGULG04b: All lakes and reservoirs that are tributary to the Gunnison River and are not on national forest lands from the outlet of Crystal Reservoir to the confluence with the Colorado River, including all tributaries to Reeder, Hollenbeck and Juniata Reservoirs, except for specific listings in the North Fork and Uncompahgre River subbasins and in Segments 9 and 13. Kannah Creek below the point of diversion for public water supply. COGULG04c: Mainstem of Red Rock Creek from the boundary of Black Canyon of the Gunnison Nation Park to the confluence of the Gunnison River.
Relevant Portion of Segment (as applicable)
COGUNF03: North Fork mainstem below Lazear COGUNF05: Leroux Creek , Jay Creek COGUNF06a: Short Draw COGUNF06b: Big Gulch, Cottonwood Creek COGUUN04b: Mainstem of the Uncompahgre River from LaSalle Road to Confluence Park. COGUUN04c: all COGUUN12: all COGULG01: below the North Fork COGULG02: all COGULG04a: all COGULG04b: lower Kannah Creek COGULG04c: Red Rock Creek
January 2011
2
Final
Antidegradation Designation and Use Classifications
Water Quality Targets (for dissolved selenium fraction) TMDL Goal
January 2011
WBID COGUNF03
Antidegradation Designation “reviewable”
Use Classifications
Aquatic Life Cold 1 Recreation E (Apr-Sept) Recreation N (Oct-Mar) Water Supply Agriculture COGUNF05 “reviewable” Aquatic Life Cold 1 Recreation P Water Supply Agriculture COGUNF06a Use Protected Aquatic Life Warm 2 Recreation P Agriculture COGUNF06b Use Protected Aquatic Life Warm 2 Recreation P Water Supply Agriculture COGUUN04b Use Protected Aquatic Life Warm 2 Recreation N Agriculture COGUUN04c Use Protected Aquatic Life Warm 2 Recreation E Agriculture COGUUN12 Use Protected Aquatic Life Warm 2 Recreation N Agriculture COGULG01 “reviewable” Aquatic Life Cold 1 Recreation E Water Supply Agriculture COGULG02 “reviewable” Aquatic Life Warm 1 Recreation E Water Supply Agriculture COGULG04a Use Protected Aquatic Life Warm 2 Recreation N Water Supply Agriculture COGULG04b Use Protected Aquatic Life Warm 2 Recreation N Water Supply Agriculture COGULG04c Use Protected Aquatic Life Warm 2 Recreation E Water Supply Agriculture Attainment of Table Value Standards for selenium for all assigned uses. Aquatic Life Use-based acute = 18.4 g/L (dissolved), chronic = 4.6 g/L (dissolved) Agriculture Use-based = 20 g/L (total) Water Supply Use-based = 50 g/L (total) “Fully Supporting” all assigned Use Classifications
3
Final
I. EXECUTIVE SUMMARY The Uncompahgre and Gunnison River basins (Figure 1) are extensively underlain by Mancos Shale. These cretaceous shales contain elevated levels of selenium. Various anthropogenic activities like sand and gravel extraction, agricultural and urban landscape irrigation accelerate the mobilization and transport of selenium from shale and shale derived soil to surface water. Consequently selenium concentrations in surface waters throughout much of the Gunnison/Uncompahgre watershed often exceed the assigned Colorado Water Quality Standards. II. INTRODUCTION Section 303(d) of the federal Clean Water Act requires States to periodically submit to the U. S. Environmental Protection Agency (EPA) a list of water bodies that are water-quality impaired. A water-quality impaired segment does not meet the standards associated with its assigned use classification. This list of impaired water bodies is referred to as the “303(d) List”. The List is adopted by the Water Quality Control Commission (WQCC) as Regulation No. 93. For waterbodies and streams on the 303(d) List a Total Maximum Daily Load (TMDL) is used to determine the maximum amount of a pollutant that a water body may receive and still maintain water quality standards. The TMDL is the sum of the Waste Load Allocation (WLA), which is the load from point source discharge, Load Allocation (LA) which is the load attributed to natural background and/or non-point sources, and a Margin of Safety (MOS) (Equation 1). (Equation 1)
TMDL=WLA+LA+MOS
Alternately, a segment or pollutant may be removed from the 303(d) List if water quality improves such that the relevant standard is attained; if implementation of clean-up activities via an alternate means will result in attainment of standards; if the original listing decision is shown to be in error; or if the standards have been changed. The Gunnison and Uncompahgre River basins lie in southwest Colorado (Figure 1). The Uncompahgre basin is tributary to the Gunnison River. The Gunnison basin is, in turn, tributary to the Colorado River at Grand Junction, Colorado. Substantial portions of the waters in both basins exhibit selenium concentrations at levels above their assigned water quality standards. Consequently, surface waters in the basins have been included on various iterations of the 303(d) List. The lower portions of both the Uncompahgre and Gunnison River basins are underlain by bedrock deposits of the cretaceous period, most notably Mancos Shale and Dakota Sandstone. The Mancos Shale is a marine deposit and, as such, contains significant amounts of readily soluble constituent materials, including selenium. Groundwater which leaches to the relatively impermeable shale deposits tends to dissolve selenium and, as it flows atop the bedrock strata towards surface drainages, carries elevated levels of dissolved selenium with it. Various anthropogenic activities like agricultural and urban landscape irrigation, or onsite sewage
January 2011
4
Final
disposal (septic systems) accelerate the mobilization and transport of selenium from shale and shale derived soil to surface water. Sand and gravel excavation, as well as sanitary sewage conveyance systems, may intercept selenium rich groundwater and influence the pathways by which selenium is delivered to the surface water system.
Figure 1. Lower Gunnison River Basin
Selenium readily dissolves in water and moves through the aquatic environment where it can bioaccumulate in organisms to toxic levels (Lemly, 2002). The biological effects of selenium bioaccumulation became a focus of national attention in the early 1980s when selenium was identified as the cause of elevated levels of waterfowl mortality and birth defects at the Kesterton Reservoir Wildlife Refuge in California. The selenium source was found to be water conveyed to the area through constructed irrigation drainages (Ohlendorf et al, 1986, 1988). Within the Gunnison basin the selenium issue is compounded by the presence of four fish species in the lower basin and downstream in the Colorado River that are both state- and federally listed endangered species. These include the Colorado pikeminnow (Ptychocheilus lucius), Humpback chub (Gila cypha), Bonytail chub (Gila Elegans), and Razorback sucker (Xyrauchen texanus). These species may be adversely affected by elevated instream selenium concentrations. Selenium concentrations in the basin exceed those shown to impact fish and wildlife elsewhere and exceed assigned Colorado Water Quality Standards.
January 2011
5
Final
Watershed Description The headwaters of the Gunnison River lie along the western flank of the Continental Divide, in the Sawatch Mountains and Elk Mountains. The Uncompahgre River basin, which is tributary to the Gunnison, lies further to the southwest and originates in the Sneffles mining district. Water quality issues in the upper basins tend to be associated with historic mining activities and are not addressed in this document. The lower portions of the watershed are located in Delta, Mesa and Montrose counties in west-central Colorado. The elevated selenium levels evidenced in much of the lower basins are associated with both the underlying geology and the land uses which dominate as the drainages flatten at these lower elevations. The confluence of the Colorado and Gunnison Rivers is in Grand Junction and is located approximately 50 miles downstream from Delta. The confluence of the Gunnison and Uncompahgre Rivers is located in Delta, Colorado. From Delta, the study area extends approximately 26 miles upstream along the Uncompahgre River to Montrose, Colorado, and approximately 35 miles upstream along the North Fork of the Gunnison River to Paonia, Colorado. The surface elevation at Whitewater is 4659 feet, Delta is 4957 feet, Paonia is 5682 feet, and Montrose is 5807 feet. The study area climate is predominately semi-arid, but some variation in climate occurs at higher elevations (e.g. more precipitation and cooler temperatures at higher elevations). Average annual precipitation ranges from 9 inches at Grand Junction to 9.5 inches at Montrose. Average annual snowfall ranges from 12.3 inches at Grand Junction to 26 inches at Montrose. Average high temperatures are approximately 66 degrees Fahrenheit for Grand Junction and 63 degrees Fahrenheit for Montrose, average lows are 40 and 34 degrees, respectively (Western Regional Climate Center, 2007). Geology The geology of the study area is dominated by two major structural features, the Uncompahgre uplift and the north dipping southwestern flank of the Piceance structural basin (Brooks and Ackerman, 1985). The Uncompahgre uplift forms the Uncompahgre Plateau and is an asymmetrical anticline that plunges northwest to southeast. Mancos shale is the dominant bedrock material outcropping in the study area and outcrops occur to the east of the Uncompahgre uplift (Green, 1992). The Mancos shale is of Cretaceous age and is composed of massive, fossiliferous marine shale with interbedded sandstones, siltstones, and devitrified volcanic ash layers. The Mancos is the lateral equivalent to the Niobrara Formation, Cody Shale, and Pierre Shale in Montana, Nebraska, South Dakota, Wyoming and southeastern Colorado (Wright and Butler, 1993). The other two dominant bedrock materials in the study area are the Dakota sandstone and various Quaternary Alluvium units (Thomas, 2007).
January 2011
6
Final
Figure 2. Geological unit outcroppings in the Gunnison River Basin taken from USGS Scientific Investigations Report 2007-5287: Analysis of Dissolved Selenium Loading for Selected Sites in the Lower Gunnison River Basin, Colorado 1978-2005.
Selenium Mobilization Like the Pierre shale of eastern Colorado, the Mancos shale is classified as a cretaceous marine shale. Such deposits are often referred to as seleniferous shales due to their selenium content and are widely distributed throughout the western United States. Soils derived from underlying seleniferous shales also serve as selenium source material. Selenium is present in several different chemical forms in the soil. In alkaline soils, which are prevalent in much of the Gunnison basin, selenium is predominantly found as selenate (SeO4-2). This species of selenium is not strongly bound to oxides and other minerals in the soil. As such, it is highly soluble. Irrigation of seleniferous soils will dissolve and transport selenium into the surface water system via groundwater and surface return flows (runoff). Municipal sewer systems intended to convey wastewater or stormwater may intercept selenium laden groundwater and provide a direct nexus to surface water.
January 2011
7
Final
Aquatic Life Use Both the Colorado Division of Wildlife and the U.S. Fish and Wildlife Service have designated four fish species in the Gunnison Basin as endangered. These include the Colorado pikeminnow (Ptychocheilus lucius), Humpback chub (Gila cypha), Bonytail chub (Gila Elegans), and Razorback sucker (Xyrauchen texanus). Given the geographic extent of the basin, it is not surprising that a wide variety of other species also occupy portions of the study area. At higher elevations in cold water streams, native Colorado River cutthroat (Colorado State Species of Special Concern), brown and rainbow trout, bluehead sucker (Colorado State Species of Special Concern), mottled sculpin and speckled dace are found. Warm water streams in the watershed host brown and rainbow trout, bluehead sucker (Colorado State Species of Special Concern), flannelmouth sucker (Colorado State Species of Special Concern), white sucker, Colorado roundtail chub (Colorado State Species of Special Concern), speckled dace, common carp, fathead minnow, green sunfish, mirror carp, and common carp. These species may be adversely affected by elevated instream selenium concentrations. Selenium concentrations in the basin exceed those shown to impact fish and wildlife elsewhere and exceed assigned Colorado Water Quality Standards Land Use The primary land use in the area is irrigated agriculture; however residential and urban land use development is increasing as population growth has occurred at a rate of around 25 percent in the study area in the five years between 2000 and 2005. Population estimates for major population centers in the study area (2005) range from 15,479 in Montrose (up from 12,344 in 2000), 8,135 in Delta (up from 6,400 in 2000), 1,584 in Paonia (1,497 in 2000), and 1,402 in Whitewater (U.S. Census Bureau, 2007). Areas near Kannah Creek, Whitewater Creek, and Callow Creek are experiencing an increased demand for residential housing. Changes in land use have resulted in a shift from open range and irrigated agricultural land use to residential and urban land use as well as the use of independent septic drainage systems (Thomas, 2007). These shifts in land use have the potential to introduce new paths of selenium loading to the mainstem Gunnison River (Gunnison Basin Selenium Task Force, 2007). In addition to Aquatic Life uses, all of the segments addressed in this document are classified for Agriculture Use and have been assigned a numeric standard of 20 g/L (total selenium) based upon that use. With the exception of the mainstem Uncompahgre River segments (4a and 4b), and those tributaries to the Uncompahgre River which comprise segment 13, all of the waters within the study area are also designated for Water Supply Use and have been assigned the concomitant 50 g/L selenium standard (as total selenium).
January 2011
8
Final
History The detrimental effects of elevated selenium on waterfowl at the Kesterton Wildlife Refuge in California resulted in several initiatives which focused attention on the Gunnison River basin. The first occurred in 1985 with the formation of a multi-agency program within the Department of the Interior, the National Irrigation Water Quality Program (NIWQP). This group was charged to determine the extent to which various federal irrigation projects in the Western United States was having adverse effects on water quality and on fish and wildlife (USGS RPT 02-4151, 2002). Studies performed under NIWQP reported high levels of selenium in some water, sediment, and biota samples collected in the Uncompahgre River basin and Grand Valley in west-central Colorado. At the same time instream selenium levels in western Colorado were being assessed, the issue was also being examined from a water quality standards perspective. The EPA in 1987 promulgated revised national selenium criteria. Subsequently, in 1995, the WQCC adopted the federal criteria into the state‟s Basic Standards and Methodologies for Surface Water, 5 CCR 1002-31 (WQCC 2007). Two years later, in 1997, the Commission incorporated the revised standards into the Classifications and Numeric Standards for Gunnison and Lower Dolores River Basins, 5 CCR 1002-35 (WQCC 2007a). Prior to 1997, the acute criterion for aquatic life for dissolved selenium was 135 g/L and the chronic criterion was 17 g/L. In 1997, the WQCC adopted an acute standard, intended to protect aquatic species from lethal effects of dissolved selenium, of 20 g/L. The chronic standard of 5 g/L was intended to preclude sub-lethal effects such as birth deformities or depressed reproduction. Existing instream selenium concentrations in much of the lower Gunnison River, lower Uncompahgre River, and numerous tributaries exceeded the newly adopted criteria. The WQCC subsequently promulgated refined Aquatic Life Use-based acute and chronic selenium standards of 18.4 g/L and 4.6 g/L respectively in 2000 (WQCC 2007). Several stream segments were initially identified as impaired with respect to selenium, and were included on the Colorado list of Water Quality Limited Segments Still Requiring TMDLs, Colorado’s 1998 303(d) List and Related Water Quality Management Lists, (WQCD 1998) in 1998. These included the mainstem of the Uncompahgre River from U.S. Highway 34 to the confluence with the Gunnison River (COGUUN04), the mainstem of the Gunnison River from a point immediately above the confluence with the Uncompahgre River to the confluence with the Colorado River (COGULG02), and tributaries to the North Fork of the Gunnison River including Hubbard Creek, Terror Creek, Minnesota Creek, Leroux Creek, Jay Creek, and Roatcap Creek (COGUNF05).
January 2011
9
Final
Figure 3. Waterbodies in the Gunnison River Basin included on the Colorado 2008 List of Impaired Waters due to Non-Attainment of Selenium Standards. Map reflects “all tributary” segments (e.g. COGUNF05, COGUNF06a, COGUNF06b, COGUUN12, and COGULG04b) where only portions of segments are identified on the 303(d) List.
The listings were subsequently changed in 2002 to reflect the re-segmentation of the Uncompahgre River mainstem that had occurred in 2001 and the availability of additional data. Those portions of the Uncompahgre River below La Salle Road to the confluence with the Gunnison River, Uncompahgre River segments 4b and 4c, were retained on the 303(d) List. Hubbard Creek, Terror Creek, Minnesota Creek, Jay Creek, and Roatcap Creek were removed from the List, leaving only Leroux Creek in the North Fork drainage. Additional water quality data resulted in the inclusion of several new listings in 2004. These included all tributaries to the Uncompahgre River from the South Canal to the confluence with the Gunnison River (Uncompahgre River segment 12), all of Lower Gunnison segment 4a, defined as tributaries to the Gunnison River not on national forest lands from the outlet of Crystal Reservoir to the confluence with the Colorado River, and the lower portion of Kannah Creek in Lower Gunnison segment 4b. Cottonwood Creek, located in segment 06 of the North Fork of the Gunnison River, was also added.
January 2011
10
Final
The listings were unchanged in 2006. The 2008 Colorado 303(d) List added Red Rock Creek in Lower Gunnison segment 4c and the mainstem of the North Fork of the Gunnison (COGUNF03). A 2007 USGS study, Analysis of Dissolved Selenium Loading for Selected Sites in the Lower Gunnison River Basin, Colorado, 1978-2005, (Thomas, 2007), which provides the technical underpinnings for this TMDL analysis, focused on selected surface-water features (including tributaries, main stem, canals and laterals) of the North Fork of the Gunnison River, the Upper Gunnison River, the Uncompahgre River, and the Gunnison River from Delta to Grand Junction. The selection of these waters reflects their inclusion on the 2006 Colorado Section 303(d) List of Water-Quality-Limited Segments Requiring TMDLs, 5 CCR 1002-93 (WQCC, 2006). Those waters included on the 2006 303(d) List and addressed in the 2007 USGS study are listed in Table 1. Two changes are reflected in Table 1 as a result of re-segmentation that occurred at a 2006 rulemaking concerning the Classifications and Numeric Standards for Gunnison and Lower Dolores River Basins, 5 CCR 1002-35 (WQCC, 2007). Segment 6 of the North Fork of the Gunnison was divided into two segments such that Short Draw, which had previously been included in segment 5, was now moved into segment 6a, and Cottonwood Creek and Big Gulch were included in segment 6b. References made to these waters in this document reflect the current segmentation as of the 2006 rulemaking. Waterbody Segment Description Identification COGUNF03 Mainstem of North Fork of the Gunnison River from the Black Bridge (41.75 Drive) above Paonia to the confluence with the Gunnison River.
Listed Portion all
COGUNF05
Mainstems of Hubbard Creek, Terror Creek, Minnesota Creek, and Leroux Creek from their Leroux Creek, boundary with national forest land to their confluences with the North Fork of the Gunnison Jay Creek River; mainstem of Jay Creek from its source to its confluence with the North Fork of the Gunnison River; mainstem of Roatcap Creek including all tributaries, wetlands, lakes and reservoirs, from its source to its confluence with the North Fork of the Gunnison. COGUNF06a All tributaries to the North Fork of the Gunnison River including all lakes, reservoirs, and Short Draw wetlands which are not on national forest lands, except for the specific listings in Segments 4, 5, 6b and 7. COGUNF06b Mainstem and all tributaries to Bear Creek, Reynolds Creek, Bell Creek, McDonald Creek, Big Gulch, Cottonwood Creek, Love Gulch, Cow Creek, Dever Creek, German Creek, Miller Creek, Cottonwood Stevens Creek, Big Gulch, Stingley Gulch and Alum Gulch including lakes, reservoirs, and Creek wetlands which are not on national forest lands from their source to the North Fork of the Gunnison River. COGUUN04b Mainstem of the Uncompahgre River from LaSalle Road to Confluence Park.
all
COGUUN04c Mainstem of the Uncompahgre River from Confluence Park to the confluence with the Colorado River.
all
COGUUN12
COGULG01
All tributaries to the Uncompahgre River, including all wetlands, lakes and reservoirs, from all the South Canal near Uncompahgre to the confluence with the Gunnison River, except for specific listings in Segments 13, 14, 15a and 15b. Mainstem of the Gunnison River from the outlet of Crystal Reservoir to the confluence with not listed the Uncompahgre River.
January 2011
11
Final
Waterbody Segment Description Identification COGULG02 Mainstem of the Gunnison River from a point immediately above the confluence with the Uncompahgre River to the confluence with the Colorado River.
Listed Portion all
COGULG04a All tributaries to the Gunnison River, including all wetlands which are not on national forest lands, from the outlet of Crystal Reservoir to the confluence with the Colorado River, except for specific listings in the North Fork and Uncompahgre River subbasins and in Segments 3, 4b, 4c, 5 through 10, 12, and 13. COGULG04b All lakes and reservoirs that are tributary to the Gunnison River and not on national forest lands from the outlet of Crystal Reservoir to the confluence with the Colorado River, including all tributaries to Reeder, Hollenbeck, and Juanita Reservoirs, except for specific listings in the North fork and Uncompahgre River subbasins and in segments 9 and 13. Kannah Creek below the point of diversion for public water supply. COGULG04c Mainstem of Red Rock Creek from the boundary of Black Canyon of the Gunnison Nation Park to the confluence of the Gunnison River.
all
Kannah Creek below USGS Station 09152000 Red Rock Creek
Table 1. Segments within the Gunnison and Uncompahgre River basins addressed in Analysis of Dissolved Selenium Loading for Selected Sites in the Lower Gunnison River Basin, Colorado, 1978-2005, (Thomas, 2007).
Although not included on the Colorado 303(d) List, the portion of the Gunnison River between the confluence with the North Fork of the Gunnison River and the confluence with the Uncompahgre River (the lower portion of Lower Gunnison segment 1) also exceeds selenium standards. This reach of the Gunnison River was included in the USGS study and is therefore also addressed in this TMDL document. An additional waterbody, Sweitzer Lake, is located near Delta Colorado and has been included on the Colorado 303(d) List since 1998. It has been included on successive Lists based upon both elevated water column selenium concentrations and accumulation of selenium in fish tissues to an extent that a Fish Consumption Advisory has been issued for the Lake. It is addressed in a separate TMDL document scheduled for development in mid-2010. Sweitzer Lake is a 137-acre lake located just south of Delta on Highway 50 in Delta County. The park was created in 1972 and is primarily a day-use area that offers swimming, water-skiing, bird watching and waterfowl hunting. The inflow to the lake is from an irrigation canal that carries return water from fields developed on the Mancos Shale. An effort to model the selenium cycle in Sweitzer Lake is currently underway. Since 1985, the National Irrigation Water Quality Program (NIWQP) has done investigations at various federal irrigation projects in the Western United States to determine if irrigation drainage was having adverse effects on water quality and on fish and wildlife. Beginning in 1988, NIWQP studies have been done in the Uncompahgre River Basin, a major tributary of the Gunnison River, and in the Grand Valley in west-central Colorado. High levels of selenium were reported in some water, sediment, and biota samples in both areas (Butler and others, 1996). Selenium concentrations in some fish- and bird-tissue samples were at levels of concern, and in late 1994 NIWQP initiated the planning phase for remediation. One objective of the NIWQP planning phase was to determine what, if any, remediation methods could be used to reduce selenium loading from irrigation sources to the Uncompahgre and Gunnison Rivers.
January 2011
12
Final
The NIWQP was charged with addressing Endangered Species and Migratory Bird Treaty issues, while a local watershed initiative formed independently in February 1998 with the creation of the Gunnison Basin Selenium Task Force (Task Force). The Task Force is a group of private, local, State, and Federal interests, charged with the objective to meet State of Colorado water quality selenium standards. Previous NIWQP studies (Butler and others, 1996) indicated that Cedar Creek (and, notably, the Montrose Arroyo, a tributary to Cedar Creek) and Loutzenhizer Arroyo were the largest contributors to the selenium load in the Uncompahgre River and that the Uncompahgre River was the single largest selenium source to the Gunnison River. The Uncompahgre River accounted for about 38 percent of the load in the Gunnison River (at Whitewater) in water years 1998-2000 (USGS, 2002). However, data for determining selenium loading to the lower Gunnison River from tributary streams other than the Uncompahgre River were limited. Selenium data were available only for a few tributaries in the North Fork Basin, and much of this information was for streams in the upper basin where selenium concentrations are low. Few or no data were available for many of the tributaries to the Gunnison River downstream from the North Fork to the Colorado River.
III. WATER QUALITY STANDARDS Standards Framework Waterbodies in Colorado are divided into discrete units or “segments”. The Colorado Basic Standards and Methodologies for Surface Water, Regulation 31, discusses segmentation of waterbodies in terms of several broad considerations: 31.6(4)(b)…Segments may constitute a specified stretch of a river mainstem, a specific tributary, a specific lake or reservoir, or a generally defined grouping of waters within the basin (e.g., a specific mainstem segment and all tributaries flowing into that mainstem segment. (c) Segments shall generally be delineated according to the points at which the use, physical characteristics or water quality characteristics of a watercourse are determined to change significantly enough to require a change in use classifications and/or water quality standards As noted in paragraph 31.6(4)(c), the use or uses of surface waters are an important consideration with respect to segmentation. In Colorado there are four categories of beneficial use which are recognized. These include Aquatic Life Use, Recreational Use, Agricultural Use and Water Supply Use. A segment may be designated for any or all of these “Use Classifications”: 31.6 Waters shall be classified for the present beneficial uses of the water, or the beneficial uses that may be reasonably expected in the future for which the water is suitable in its present condition or the beneficial uses for which it is to become suitable as a goal.
January 2011
13
Final
Each assigned use is associated with a series of pollutant specific numeric standards. These pollutants may vary and are relevant to a given Classified Use. Numeric pollutant criteria are identified in sections 31.11 and 31.16 of the Basic Standards and Methodologies for Surface Water. Uses and Standards Addressed in this TMDL The segments/waterbodies identified in Table 1 are included on the Section 303(d) List due to non-attainment of numeric standards for selenium. There are numeric standards for selenium associated with Aquatic Life, Water Supply, and Agriculture Uses. With the exception of those segments in the Uncompahgre River Basin, all of the listed segments are designated for all three of these Uses (none of the listed Uncompahgre Basin segments is designated for Water Supply Use). Attainment of the Aquatic Life Use-based chronic selenium standard is assessed by comparison of the 85th percentile value of the ranked discrete sample results against the standard of 4.6 g/L (WQCD, 2007). The numeric standard for the Aquatic Life Use-based chronic selenium standard is not attained in any of the segments listed in Table 1. Attainment of the acute Aquatic Life Use-based selenium standard is assessed by comparison of the discrete sample results to the standard of 18.4 g/L (WQCD, 2007). The acute Aquatic Life Use-based selenium standard is not attained in many of the listed segments (Table 2). Water Supply and Agriculture Uses also have associated numeric selenium standards. Unlike the Aquatic Life Use-based standards, which are expressed as the “dissolved” fraction, the Water Supply and Agriculture Use-based standards are expressed as the “total recoverable” fraction. Most Aquatic Life Use-based standards are expressed as the dissolved fraction because it most closely approximates the ionic form of the metal that is available for biological uptake (i.e. the most toxic fraction). Water Supply and Agriculture Use-based standards are typically expressed in terms of total recoverable metals. The total recoverable fraction generally includes relatively less toxic complexed metals. Attainment of numeric Water Supply and Agriculture Use-based standards is complicated because most water quality data is available for only the dissolved metals fraction. This is addressed in the Section 303(d) Listing Methodology – 2008 Listing Cycle (WQCD, 2007) by comparing the 50th percentile value for the dissolved metals data against the relevant Water Supply and Agriculture Use-based standard. The datasets utilized by the Division in development of the most current (2008) section 303(d) List are summarized in Table 2. Those listings identified in Table 1 remained unchanged on the 2008 List. In addition to non-attainment of chronic, and sometimes, acute Aquatic Life Use-based selenium standards, several tributaries to the Gunnison River downstream of the confluence with the Uncompahgre River also fail to attain the 20 g/L Agriculture Use-based standard, and, in one instance (Alkali Creek) the Water Supply Use-based standard of 50 g/L is not attained. Data utilized for calculation of TMDLs may incorporate additional data not available at the time the 303(d) List was compiled.
January 2011
14
Final
Waterbody Identification
Listed Waterbody
Ambient Sample Period of Se(dis) Size Record concentration
Aquatic Life Use acute
chronic
Water Agriculture Supply Use Use
COGUNF03
North Fork Gunnison River
5.7
114
10/199907/2005
attain
non-attain
attain
attain
COGUNF05
Leroux Creek
8.5
34
10/199903/2005
attain
non-attain
attain
attain
COGUNF05
Jay Creek
16.3
4
5/199903/2000
attain
non-attain
attain
attain
COGUNF06a Short Draw
27.4
2
11/199903/2000
attain
non-attain
attain
attain
COGUNF06b Big Gulch
7.0
2
11/199903/2000
attain
non-attain
attain
attain
COGUNF06b Reynolds Creek
8.0
2
11/199903/2000
attain
non-attain
attain
attain
COGUNF06b Cottonwood Creek
6.9
3
01/200505/2005
attain
non-attain
attain
attain
COGUUN04b Uncompahgre River mainstem COGUUN04c Uncompahgre River mainstem
16.7
67
non-attain non-attain
na
attain
18.5
36
10/199906/2005 10/199906/2005
non-attain non-attain
na
attain
COGUUN12
Dry Creek
6.4
5
12/200004/2005
non-attain
na
attain
COGUUN12
Dry Cedar Creek
27.5
58
11/199904/2005
non-attain non-attain
na
attain
COGUUN12
Loutzenhizer Arroyo
231.4
185
022/200005/2005
non-attain non-attain
na
non-attain
COGULG01
Gunnison River mainstem
5.4
18
11/19999/2002
attain
non-attain
attain
attain
COGULG02
Gunnison River mainstem
8.4
107
10/199907/2005
attain
non-attain
attain
attain
COGULG04a Alkali Creek
121.0
7
11/199904/2000
COGULG04a Callow Creek
15.8
9
08/200005/2002
COGULG04a Currant Creek
36.5
6
11/199903/2001
COGULG04a Deer Creek
9.8
5
01/200003/2000
COGULG04a Peach Valley Arroyo
72.5
3
11/199903/2000
COGULG04a Wells Gulch
9.1
4
06/199903/2000
January 2011
15
attain
non-attain non-attain non-attain attain
non-attain
non-attain
attain
attain
non-attain non-attain
attain
non-attain
non-attain
attain
attain
non-attain non-attain
attain
attain
attain
attain
attain
attain
non-attain
Final
Waterbody Identification
Listed Waterbody
Ambient Sample Period of Se(dis) Size Record concentration
Aquatic Life Use acute
chronic
Water Agriculture Supply Use Use
COGULG04a Whitewater Creek
59.2
22
11/199907/2005
non-attain non-attain
attain
non-attain
COGULG04b lower Kannah Creek
30.8
25
11/199903/2005
non-attain non-attain
attain
attain
COGULG04c Red Rock Canyon
57.1
32
08/200104/2005
non-attain non-attain
attain
non-attain
Table 2. Ambient (85th percentile value) dissolved selenium concentrations for waters included in 303(d) Listed segments (excluding COLGLG01), and attainment status for assigned Uses. Concentrations in g/L.
Several waters are considered to be in attainment of the acute Aquatic Life Use-based standard of 18.4 g/L, although the 85th percentile value defining the ambient condition exceeds the acute standard. The Section 303(d) Listing Methodology – 2008 Listing Cycle (WQCD, 2007) indicates that attainment of the acute standard is assessed by comparison of an individual sample result against the standard. A waterbody is deemed to be in non-attainment of an acute standard if that standard is exceeded more than once in a three year period. Short Draw (COGUNF06a) is considered to be in attainment of the acute standard due to the fact that only two discrete sample results are available (only one of which exceeds the acute standard). Conversely, segment 4b of the Uncompahgre River is considered to be in non-attainment of the acute standard even though the 85th percentile value is 16.7 g/L. Sampling performed at the lower terminus of the segment indicates exceedances of the acute standard occurred at multiple times in 1999 and 2000. Subsequent exceedances were documented in 2002 and 2003. Those waterbodies for which a relatively larger dataset is available are therefore more likely to be in non-attainment of the acute Aquatic Life Use-based standard, even in instances where the 85th percentile value is less than other waterbodies.
IV. PROBLEM IDENTIFICATION Land Use The lower portions of both the Uncompahgre and Gunnison River basins are underlain by extensive deposits of seleniferous Mancos Shale (Figure 2). The Mancos Shale is a marine deposit and, as such, contains significant amounts of readily soluble constituent materials, including selenium. Water which leaches to the relatively impermeable shale deposits tends to dissolve selenium from overlying soils derived from the shale as well as from the bedrock strata. This selenium bearing groundwater then flows atop the bedrock strata towards surface drainages. The groundwater may enter the surface drainage directly or be intercepted by sand and gravel pits, stormwater or sanitary sewer systems, or foundation drains. Some of these features may then discharge to surface water. The “natural” or “background” selenium load is limited by the semi-arid nature of the region. The relative lack of precipitation restricts the amount of groundwater available for selenium transport. Irrigation of agricultural and urban areas
January 2011
16
Final
exacerbates the problem by increasing the amount of water available to leach selenium from soil and bedrock materials. Septic systems may also contribute a significant volume of leachate. The National Irrigation Water Quality Program (NIWQP) was created in 1985. This group was charged to determine the extent to which various federal irrigation projects in the Western United States was having adverse effects on water quality and on fish and wildlife (USGS, 2002). Several extensive irrigation projects have been developed in the Gunnison Basin in areas underlain by Mancos Shale (Figure 4). The NIWQP collected data between 1987 and 2003 in the Uncompahgre Project area and lower Gunnison basin. Samples of water, sediment, food chain items (plants, invertebrates, fish, birds and bird eggs) were collected from ponds, drains, backwaters and streams. Although examined for an extensive list of contaminants (e.g., lead, mercury, and zinc) and pesticides, selenium was found to be the major concern. Sixty percent or more of the selenium loading in the Gunnison basin (as measured at Whitewater) originates from an area encompassing the Uncompahgre River basin and the service area of the Federally-constructed Uncompahgre Project. This figure includes 40 percent from the Uncompahgre River basin and ranges from 3 to 17 percent from portions of the Uncompahgre Project service area in the vicinity (north and east) of Delta. Irrigation water delivery systems and on-farm applications to the Mancos Shale and soils derived from the shale contribute a vast majority (about 90 percent) of the ground-water that mobilizes selenium in this area. (BOR 2006).
Figure 4. Estimated selenium load by sub-basin as taken from Evaluation of Selenium Remediation Concepts for the Lower Gunnison and Lower Uncompahgre Rivers, Colorado, Bureau of Reclamation, 2006.
January 2011
17
Final
Discharge Permits The Colorado Discharge Permit System (CDPS) program controls water pollution by regulating point sources that discharge pollutants into waters of Colorado. Industrial, municipal, and other facilities must obtain permits if their discharges go directly to surface waters. Certain dischargers, because of their size or the nature of the wastewater they generate, require individual permits. These are permits issued to specifically address a given facility‟s discharge. There are six domestic wastewater treatment facilities discharging to listed segments in the Gunnison and Uncompahgre watersheds that have been issued individual CDPS permits (Table 3). These include the Town of Paonia and Town of Hotchkiss wastewater treatment plants which discharge to the North Fork Gunnison River (North Fork segment 3), the Town of Olathe wastewater treatment plant, the City of Montrose and West Montrose Sanitation District, which discharge to the Uncompahgre River segment 4b, and the City of Delta facility which discharges to Lower Gunnison segment 2. Waterbody Identification COGUNF03 COGUNF03
Town of Paonia Town of Hotchkiss
CO0047431 CO0044903
Design Capacity, mgd1 0.50 0.49
COGUUN04b COGUUN04b COGUUN04b
Town of Olathe City of Montrose West Montrose Sanitation District
CO0020907 CO0039624 CO0030449
0.73 4.32 0.70
CO0039641
2.45
Permittee
CDPS No.
COGULG02 City of Delta million gallons per day Table 3. Individual CDPS Permittees 1
Additionally, there are a number of discharging facilities that are authorized via general permit certifications. These are primarily sand and gravel extraction facilities located within the listed portions of the Gunnison Basin. These are covered by a general CDPS permit issued to address such industrial operations. Individual sand and gravel production facilities must receive certification under the terms of the general permit. Each operating sand and gravel operation must comply with a uniform set of discharge requirements as set forth in the general permit. Operations discharging to segments included on the Colorado 303(d) List of Impaired Waters are identified in Table 4. Design capacity refers to the maximum volume of wastewater a facility is authorized to discharge. For municipal wastewater treatment plants, the design capacity reflects an analysis of both the organic and hydraulic treatment limitations of the facility. The design capacities of sand and gravel operations are determined by pump capacities.
January 2011
18
Final
Waterbody Identification COGUNF03 COGUNF03 COGUNF03 COGUNF03 COGUNF03
Tri-County Gravel Diamond Lazy L. Ranch Ent., Janet Pit Oldcastle SW Group, Inc., Campbell Gravel Pit Oldcastle SW Group, Inc., 4D Gravel Pit Oldcastle SW Group, Inc., Tri-County Pit
COG500255 COG500458 COG500397 COG500400 COG500498
Design Capacity, mgd1 0.3 2.9 2.02 0.87 3.0
COGUUN04b
Western Gravel Concrete Facility – North R-34 Pit
COG500486
2.5
COGULG04a COGULG04a
Elam Construction, Dubs Pit Western Gravel, Fredlund Pit
COG500427 COG500451
1.44 3.0
Permittee
CDPS No.
COGULG02 Elam Construction, Mule Farm Gravel Pit COG500210 1.2 COGULG02 Whitewater Building Materials COG500127 2.6 1 million gallons per day Table 4. General CDPS Sand and Gravel Permit Certifications discharging to 303(d) Listed segments
An additional five sand and gravel operations discharge to Segment COGULG01, the mainstem of the Gunnison River in the vicinity of Delta above the confluence with the Uncompahgre River. As was noted earlier, this portion of the Gunnison River, while not currently included on the Colorado 303(d) List, is not in attainment of the Aquatic Life Usebased selenium standard and is therefore included in this analysis. These five facilities are listed in Table 5. Waterbody Design Capacity Permittee CDPS No. Identification mgd1 COGULG01 Elam Construction, Bennett Gravel Pit COG500439 0.65 COGULG01 Grand Junction Pipe & Supply, Delta Paving Gravel Pit COG500444 0.58 COGULG01 Lafarge West, North Delta Pit COG500160 0.58 COGULG01 Oldcastle SW Group, Inc., Anderson Pit COG500464 1.4/4.0 COGULG01 Western Gravel North D-22 Pit COG500358 0.40 1 million gallons per day Table 5. General CDPS Sand and Gravel Permit Certifications discharging to Lower Gunnison segment 1
All CDPS permitted discharges are considered to be “point sources” and are addressed in the “Waste Load Allocation” term of the TMDL.
V.
WATER QUALITY GOALS
The water quality goals for the 303(d) listed segments of the Gunnison River basin are attainment of the Aquatic Life Use classifications and, for several of the listed waters, the Agriculture and Water Supply Use classifications. Attainment of the chronic Aquatic Life Usebased standard of 4.6 g/L dissolved selenium does not preclude the possibility that the Agriculture or Water Supply Use-based total selenium standards of 20 g/L and 50 g/L respectively, are not attained. However, assessments of water quality data compiled for 303(d) List development since 1998 have not documented any instance where this has occurred. The U. S. Geological Survey assessment which is summarized in the following section addresses the
January 2011
19
Final
loading reductions necessary to attain the assigned acute and chronic Aquatic Life Use-based selenium standards as does the subsequent Division analysis. It is presumed that attainment of such will also assure attainment of the assigned Agriculture and Water Supply-based standards.
VI.
U.S. GEOLOGICAL SURVEY LOADING ASSESSMENT
The U.S. Geological Survey in cooperation with the Water Quality Control Division, Colorado Department of Public Health and Environment in 2007 completed an Analysis of Dissolved Selenium Loading for Selected Sites in the Lower Gunnison River Basin, Colorado, 1978-2005 (Thomas, 2007). This study was commissioned by the Division to develop loading assessments for those segments/waterbodies included on the 1998 303(d) list and is available online at http://pubs.usgs.gov/sir/2007/5287/. The USGS study addresses dissolved selenium loads, and not total selenium. Therefore only attainment of the Aquatic Life Use-based acute and chronic standards is considered. However, as noted in the preceding section, it is presumed that attainment of the Aquatic Life Use-based standards will also assure attainment of the assigned Agriculture and Water Supply Use-based standards. Historical data for 54 stream water-quality/water-quantity monitoring sites were assessed. Analytical results from both USGS and Division sampling were included. Three of the 54 sites were located at gaging stations (gaged) and therefore had continuous streamflow data. The remaining 51 sites were not located at streamflow-gaging stations (ungaged); however, instantaneous streamflow was measured at the time selenium concentration data were collected. The amount of available data and the period of data collection varied considerably from site to site. This resulted in a varying ability to estimate selenium loads and selenium load reductions at differing locations (Thomas, 2007). The summaries presented in Tables 6 through 22 reflect the calculated selenium loading, and the concomitant loading reductions necessary to meet the Aquatic Life Use-based standard of 4.6 g/L as calculated by USGS. These summaries do not incorporate all of the TMDL terms required under EPA guidance and regulation, but are included for comparative purposes. TMDL summaries calculated by the Division are illustrated in Tables 69 through 102. North Fork Gunnison River Basin Mean daily selenium loads were calculated for six drainages within the North Fork Gunnison River Basin. These included Leroux Creek and Jay Creek (North Fork segment 5), Short Draw (North Fork segment 6a), and Cottonwood Creek (North Fork segment 6b), all of which are included on Colorado‟s 2008 Section 303(d) List (Thomas, 2007).
January 2011
20
Final
Figure 5. Waterbodies in the Gunnison River Basin, North Fork Sub-basin included on the Colorado 2008 List of Impaired Waters due to Non-Attainment of Selenium Standards. Map reflects “all tributary” segments (e.g. COGUNF05, COGUNF06a, COGUNF06b) where only portions of segments are identified on the 303(d) List.
January 2011
21
Final
WBID
waterbody description Leroux Creek Jay Creek Short Draw Cottonwood Creek Big Gulch Bell Creek
Selenium concentration ( g/L) Range (n1) 85th percentile 1.0-21.0 (21) 15.0 6.4-18.8 (4) 16.3 8.0-29.4 (7) 19.1 4.2-12.8 (7) 9.6 7.0-9.0 (4) 7.0 3.0-7.0 (6) 7.0
mean load (lbs/day) 0.47 0.06 0.37 0.32 0.13 0.21
COGUNF05 COGUNF05 COGUNF06a COGUNF06b COGUNF06b COGUNF06b 1 sample size Table 6. Selenium concentrations and estimated mean daily loads for North Fork Gunnison River tributaries
Ambient selenium concentrations and mean loads for those streams in non-attainment are shown in Table 6. Data are included for Bell Creek although this water has not been included on the state‟s 303(d) List. Mean selenium loads, estimates of the loads necessary to attain the chronic Aquatic Life Use-based selenium standard (4.6 g/L) and load reductions for those streams in non-attainment are shown in Table 7. WBID
waterbody description
mean load (lbs/day)
load to meet load percent standard reduction reduction (lbs/day) (lbs/day) COGUNF05 Leroux Creek 0.47 0.14 0.33 70.2 COGUNF05 Jay Creek 0.06 0.02 0.04 66.7 COGUNF06a Short Draw 0.37 0.09 0.28 75.7 COGUNF06b Cottonwood Creek 0.32 0.15 0.17 53.1 COGUNF06b Big Gulch 0.13 0.09 0.04 30.8 COGUNF06b Bell Creek 0.21 0.14 0.07 33.3 Table 7. Selenium loads and load reductions to attain chronic standard for North Fork Gunnison River tributaries
The cumulative mean selenium load from all of the 12 sites in the North Fork Gunnison River Basin assessed in the USGS report was 2.3 pounds per day (the six 303(d) Listed tributaries contribute 1.56 pounds per day). Assuming that this load is a reasonable representation of the cumulative mean daily load from these sites, the resulting mean annual selenium load would be approximately 840 pounds. The load contributed by Leroux Creek, Jay Creek, Big Gulch, Cottonwood Creek, Short Draw, and Bell Creek amounts to some 562 pounds on a mean annual basis, or, on average, 41.5 percent. Attainment of the 4.6 g/L Aquatic Life Use-based chronic standard in the listed tributaries would result in the removal of some 0.93 pounds per day or 340 pounds per year. The lowermost portion of the North Fork of the Gunnison River immediately above the confluence with the Gunnison River mainstem exhibits elevated selenium levels such that the Aquatic Life Use-based chronic standard is not attained. The 85th percentile dissolved selenium concentration, based upon 54 samples collected by the Division, and the Colorado Division of Wildlife between May 2000 and July 2005 is 6.9 g/L. Upstream, at Lazear, sampling by the USGS and by the Division indicate that at that location the standard is attained (3.7 g/l based upon 46 sample results).
January 2011
22
Final
In a previous report on selenium loading by Butler and Leib (2002), the mean annual selenium load for the North Fork Gunnison River (North Fork Gunnison River segment 3) at the mouth was approximately 1,300 and 1,400 pounds per year for 1999 and 2000, respectively. The cumulative mean annual selenium load calculated for the selected tributary sites represents more than one-half of the mean annual selenium load to the North Fork Gunnison River. The source of the remaining mean annual selenium load is not known, but sources may include unquantified naturally occurring selenium load from ground water and surface water, deep percolation of irrigation water, or septic systems (Thomas, 2007). Red Rock Canyon at the Mouth Site 13, Red Rock Canyon at the mouth near Montrose, represents the cumulative contribution of selenium concentration to 303(d) listed segment COGULG04c. Ambient selenium concentrations and mean loads for Rock Canyon are shown in Table 8. water 85th percentile mean load year (lbs/day) selenium ( g/L) COGULG04c 2004 72.1 0.57 2005 53.1 0.71 average 62.6 0.64 Table 8. Selenium concentrations and estimated mean daily loads for Red Rock Canyon WBID
waterbody description Red Rock Canyon
Mean selenium loads, estimates of the loads necessary to attain the chronic Aquatic Life Use-based selenium standard (4.6 g/L) and load reductions for Red Rock Canyon are shown in Table 9. WBID
waterbody description
water year
mean load (lbs/day)
load to meet load percent standard reduction reduction (lbs/day) (lbs/day) COGULG04c Red Rock Canyon 2004 0.57 0.04 0.53 93.0 2005 0.71 0.06 0.65 92.0 average 0.64 0.04 0.60 93.0 Table 9. Selenium loads and load reductions to attain chronic standard for Red Rock Canyon
The average amount of load that would need to be reduced at this site for water years 2004 and 2005 was 217 pounds or 93 percent of the mean annual load. This high percentage of load reduction represents almost the entire mean annual load at this site (Thomas, 2007). Gunnison River from the North Fork Gunnison River to the Uncompahgre River Leroux Creek, Jay Creek, Short Draw, Cottonwood Creek and Red Rock Canyon represent those tributaries to the Gunnison River located above the confluence with the Uncompahgre River that are known to be in non-attainment of selenium standards. As such they represent discrete drainages for which contributions to the total selenium load at the confluence of the Gunnison and Uncompahgre Rivers can be characterized (Thomas, 2007). The dissolved selenium load associated with each has previously been characterized.
January 2011
23
Final
Figure 6. Waterbodies in the Gunnison River Basin, Lower Gunnison Sub-basin included on the Colorado 2008 List of Impaired Waters due to Non-Attainment of Selenium Standards. Map reflects “all tributary” segment (e.g. COGULG04b) where only portion of tributary segment is identified on the 303(d) List.
January 2011
24
Final
The mainstem of the Gunnison River from Crystal Reservoir to the confluence with the Uncompahgre River (Lower Gunnison River segment 1), when assessed over its entire length, does not exceed the Aquatic Life Use-based selenium standard. It is not included on the 2008 303(d) List. However, the portion of the Gunnison River from the North Fork to the Uncompahgre, when assessed separately, exhibits elevated selenium levels such that the Aquatic Life Use-based chronic standard is not attained. The Division has therefore opted to include a loading assessment for this portion of the basin in this document. Ambient selenium concentrations and mean loads for the Gunnison River at Delta are shown in Table 10. water 85th percentile mean load year (lbs/day) selenium ( g/L) COGULG01 2001 6.9 26.8 2002 6.6 21.2 2003 9.4 18.4 2004 7.2 16.3 2005 5.1 17.9 average 7.0 20.1 Table 10. Selenium concentrations and estimated mean daily loads for the Gunnison River mainstem at Delta WBID
waterbody description Gunnison River at Delta
Mean selenium loads, estimates of the loads necessary to attain the chronic Aquatic Life Use-based selenium standard (4.6 g/L) and load reductions for the Gunnison River at Delta are shown in Table 11. WBID
waterbody description
water year
mean load (lbs/day)
load to meet load percent standard reduction reduction (lbs/day) (lbs/day) COGULG01 Gunnison River at 2001 26.8 18.0 8.8 33.0 Delta 2002 21.2 14.8 6.4 30.0 2003 18.4 9.0 9.4 51.0 2004 16.3 10.4 5.9 36.0 2005 17.9 15.9 2.0 11.0 average 20.1 13.3 6.8 34.0 Table 11. Selenium loads and load reductions to attain chronic standard for the Gunnison River mainstem at Delta
The average amount of load that would need to be reduced for this site for water years 2001 through 2005 was 2,510 pounds or 34 percent of the mean annual load (Thomas, 2007).
January 2011
25
Final
Figure 7. Waterbodies in the Gunnison River Basin, Uncompahgre Sub-basin included on the Colorado 2008 List of Impaired Waters due to Non-Attainment of Selenium Standards. Map reflects “all tributary” segment (e.g. COGUUN12) where only portion of tributary segment is identified on the 303(d) List.
January 2011
26
Final
Uncompahgre River Basin Ambient selenium concentrations and mean loads for the Uncompahgre River at Delta (Confluence Park) are shown in Table 12. water 85th percentile mean load year (lbs/day) selenium ( g/L) COGUUN04b 2001 14.8 16.3 2002 14.7 11.9 2003 15.5 12.5 2004 14.6 15.5 2005 14.1 18.1 average 14.8 14.9 Table 12. Selenium concentrations and estimated mean daily loads for the Uncompahgre River mainstem at Delta WBID
waterbody description Uncompahgre River, LaSalle Road to Confluence Park
Mean selenium loads, estimates of the loads necessary to attain the chronic Aquatic Life Use-based selenium standard (4.6 g/L) and load reductions for the Uncompahgre River at Delta are shown in Table 13. WBID
waterbody description
water year
mean load (lbs/day)
load to load percent meet reduction reduction standard (lbs/day) (lbs/day) COGUUN04b Uncompahgre 2001 16.3 5.1 11.2 69.0 River, LaSalle 2002 11.9 3.7 8.2 69.0 Road to 2003 12.5 3.7 8.8 70.0 Confluence Park 2004 15.5 4.8 10.7 69.0 2005 18.1 6.0 12.1 67.0 average 14.9 4.6 10.3 69.0 Table 13. Selenium loads and load reductions to attain chronic standard for the Uncompahgre River mainstem at Delta
The average amount of load that would need to be reduced for water years 2001 through 2005 was 3,730 pounds or 69 percent of the mean annual load. Data for the Uncompahgre River at the mouth, located at the lower terminus of the Uncompahgre River segment 4c, was limited. Ambient selenium concentrations and mean loads for the Uncompahgre River segment 4c are shown in Table 14. WBID COGUUN04c
waterbody description Uncompahgre River at mouth
Selenium concentration ( g/L) Range (n1) 85th percentile 11.5-17.1(4) 16.6
mean load (lbs/day) 14.9
1
sample size Table 14. Selenium concentrations and estimated mean daily loads for the Uncompahgre River mainstem at Delta
January 2011
27
Final
Mean selenium loads, estimates of the loads necessary to attain the chronic Aquatic Life Use-based selenium standard (4.6 g/L) and load reductions for the Uncompahgre River segment 4c are shown in Table 15. WBID
waterbody description
mean load (lbs/day)
COGUUN04c
load to meet standard (lbs/day) 4.1
load reduction (lbs/day) 10.8
percent reduction
Uncompahgre River 14.9 72.5 at mouth Table 15. Selenium loads and load reductions to attain chronic standard for the Uncompahgre River mainstem at Delta
Despite the relative lack of data for this location, the results are consistent with the analysis performed for the Uncompahgre River at Delta. This is expected given the proximity of the two sites (less than two miles) and the lack of selenium source loads between the sites. Dry Cedar Creek, Cedar Creek, Loutzenhizer Arroyo, and Dry Creek are tributary to the Uncompahgre River and included in Uncompahgre Segment COGUUN12. Ambient selenium concentrations and mean loads for these tributaries are shown in Table 16. water 85th percentile mean load year (lbs/day) selenium ( g/L) COGUUN12 1995 56.5 0.6 1996 67.4 0.8 average 61.9 0.7 Cedar Creek 1992 40.8 6.7 1993 33.0 5.9 1996 40.0 5.1 1997 40.0 5.2 2000 29.5 4.2 average 36.7 5.4 Loutzenhizer Arroyo 1996 180.0 14.4 1997 154.0 12.3 2002 215.0 16.1 average 183.0 14.2 Dry Creek 1992 9.8 3.4 Table 16. Selenium concentrations and estimated mean daily loads for Uncompahgre River tributaries WBID
waterbody description Dry Cedar Creek
Mean selenium loads, estimates of the loads necessary to attain the chronic Aquatic Life Use-based selenium standard (4.6 g/L) and load reductions for Dry Cedar Creek, Cedar Creek, Loutzenhizer Arroyo, and Dry Creek are shown in Table 17. Dry Cedar Creek data spans the 1995 and 1996 water-year periods. The mean annual load would need to be reduced by 90 percent (287 pounds) to bring this site into compliance with the water-quality standard. The average amount of load that would need to be reduced at this site for water years 1995 and 1996 was 250 pounds or 90 percent of the mean annual load (Thomas, 2007).
January 2011
28
Final
WBID
waterbody description
water year
mean load (lbs/day)
load to meet load percent standard reduction reduction (lbs/day) (lbs/day) COGUUN12 Dry Cedar Creek 1995 0.6 0.08 0.52 87.7 1996 0.8 0.07 0.73 91.3 average 0.7 0.07 0.63 90.0 Cedar Creek 1992 6.7 0.11 6.59 98.4 1993 5.9 0.14 5.76 97.6 1996 5.1 0.12 4.98 97.6 1997 5.2 0.12 5.08 97.7 2000 4.2 0.16 4.04 96.2 average 5.4 0.13 5.27 97.6 Loutzenhizer 1996 14.4 0.03 14.37 99.8 Arroyo 1997 12.3 0.03 12.27 99.8 2002 16.1 0.02 16.08 99.9 average 14.2 0.03 14.17 99.8 Dry Creek 1992 3.4 0.47 2.93 86.2 Table 17. Selenium loads and load reductions to attain chronic standard for the Uncompahgre River tributaries
Streamflow quantity and quality affect the results of the loading analysis at Dry Cedar Creek and similar sites like Cedar Creek and Loutzenhizer Arroyo. The percentage of the mean annual selenium load that needs to be reduced at a site to bring it into attainment of the waterquality standard is affected by the amount of streamflow as well as the selenium concentration in the streamflow. Due to geology, land use, and climate in the Lower Gunnison River Basin, the concentration of selenium in streamflow at some sites tends to be high. In the study area, streamflow with low selenium concentrations tends to originate from snowmelt or storm-related streamflow (Butler and others, 1991). Dry Cedar Creek does not receive appreciable amounts of snowmelt or storm-related streamflow which, in combination with geology and land use, would explain the high selenium concentrations observed at this site. Cedar Creek data spans the 1991 through 2001 water-year period. The average amount of load that would need to be reduced for this site for water years 1992, 1993, 1996, 1997, and 2000 was 1,730 pounds or 97 percent of the mean annual load. Montrose Arroyo (Figure 8) is a tributary of Cedar Creek, and is also included in Uncompahgre segment 12. Montrose Arroyo drains mixed land uses of urban, suburban, agricultural, and undeveloped areas. Montrose Arroyo was selected for a demonstration project in the late 1990s that involved replacing some 8.5 miles of open irrigation water laterals with about 7.5 miles of PVC pipe piping. The PVC pipe was placed in trenches and buried, and the original laterals were filled in. About 80 percent of the lateral construction was done in the winter of 1998–99, and the remainder was completed in December 1999.
January 2011
29
Final
Figure 8. City of Montrose Vicinity showing Montrose Arroyo to the Southeast of the City
An assessment of the selenium loading reductions achieved by the Montrose Arroyo demonstration project was performed utilizing post-remediation data collected in 1999 and 2000 (Butler, 2001). Review of available data at that time indicated the project had been successful in reducing the selenium load associated with seepage from the laterals. Based upon the 1999-2000 data, the project resulted in an estimated decrease of about 194 pounds per year, or 28 percent of the pre-project selenium load. A loading analysis for the Montrose Arroyo was not included in the selenium loading analysis prepared by the USGS. The mean daily selenium load for Cedar Creek, however, was calculated at 5.7 pounds for the four years of pre-project sampling (1992-1997). The mean daily load was estimated at 4.2 pounds after the lateral project was completed (however, there was only a single year‟s data available representing the post-remediation dataset). Loutzenhizer Arroyo (Figure 9) data spans the 1991 through 2003 water-year period. Ninety-eight percent of the mean annual load would need to be reduced to bring this site into compliance with the water-quality standard. The average amount of load that would need to be reduced for this site for water years 1996, 1997, and 2002 was 5,070 pounds, or almost 100 percent of the mean annual load. Streamflow at Loutzenhizer Arroyo is almost entirely comprised of irrigation water, which is likely the reason for the high percentage of mean annual load that needs to be reduced to bring the site into compliance with the water-quality standard (Butler and Leib, 2002).
January 2011
30
Final
Currently a lateral piping project similar to that completed along the Montrose Arroyo is underway for Loutzenhizer Arroyo. Construction of the project was initiated in 2009. The project was completed in late 2010.
Figure 9. Loutzenhizer Arroyo and Dry Creek near the Town of Olathe
Dry Creek drains the area west of the Uncompahgre River and joins the mainstem several miles south of the City of Delta. Data for Dry Creek spans the 1991 through 2001 water-year period. Eighty-six percent of the mean annual load would need to be reduced to bring this site into compliance with the water-quality standard. Gunnison River from North Fork Gunnison River to Grand Junction, Colorado Selenium loads were estimated for two tributaries to that reach of the Gunnison River between the confluence with the Uncompahgre River in Delta downstream to the confluence with the Colorado River.
January 2011
31
Final
Figure 10. Lower Gunnison River below the Confluence with the Uncompahgre River to the Confluence with the Colorado River.
Mean selenium loads, estimates of the loads necessary to attain the chronic Aquatic Life Use-based selenium standard (4.6 g/L) and load reductions for tributaries to the Gunnison River downstream of the confluence with the Uncompahgre River at Delta are shown in Table 19. water 85th percentile mean load year (lbs/day) selenium ( g/L) COGULG04 1996 157.0 4.0 2000 36.5 2.6 2001 104.0 3.2 average 99.2 3.3 Whitewater Creek 2001 62.2 0.3 Table 18. Selenium concentrations and estimated mean daily loads for Lower Gunnison River tributaries WBID
January 2011
waterbody description Sunflower Drain
32
Final
WBID
waterbody description
water year
mean load (lbs/day)
load to meet load percent standard reduction reduction (lbs/day) (lbs/day) COGULG04 Sunflower Drain 1996 4.0 0.1 3.9 97 2000 2.6 0.3 2.3 88 2001 3.2 0.1 3.1 96 average 3.3 0.2 3.1 94 Whitewater Creek 2001 0.3 0.02 0.28 93 Table 19. Selenium loads and load reductions to attain chronic standard for the Lower Gunnison River tributaries
Sunflower Drain data spans the 1991 to 2001 water-year period. The average amount of load that would need to be reduced for this site for water years 1996, 2000, and 2001 was 1,130 pounds or 94 percent of the mean annual load. Whitewater Creek data spans the 1999 through 2002 water-year period. The 85th percentile selenium concentration was 62.2 μg/L, and the mean daily load of selenium was 0.332 pound (121 pounds annually). The average amount of load that would need to be reduced for water year 2001 was 112 pounds or 93 percent of the mean annual load. The Gunnison River near Grand Junction represents the lower terminus of the entire Gunnison River basin. Mean selenium loads, estimates of the loads necessary to attain the chronic Aquatic Life Use-based selenium standard (4.6 g/L) and load reductions for the Gunnison River near Grand Junction are shown in Table 21. water 85th percentile mean load year (lbs/day) selenium ( g/L) COGULG02 2001 7.3 45.9 2002 8.2 40.9 2003 11.8 40.1 2004 11.0 44.7 2005 8.9 53.3 average 9.7 45.0 Table 20. Selenium concentrations and estimated mean daily loads for the Gunnison River mainstem at Grand Junction WBID
WBID
waterbody description Gunnison River at Grand Junction
load to meet load percent standard reduction reduction (lbs/day) (lbs/day) COGULG02 Gunnison River at 2001 45.9 28.9 17.0 37 Grand Junction 2002 40.9 22.9 18.0 44 2003 40.1 15.6 24.5 61 2004 44.7 18.8 25.9 58 2005 53.3 27.2 26.1 49 average 45.0 21.2 23.8 53 Table 21. Selenium loads and load reductions to attain chronic standard for the Gunnison River mainstem at Grand Junction
January 2011
waterbody description
water year
mean load (lbs/day)
33
Final
For water years 2001 through 2005 at the Gunnison River near Grand Junction site, the average 85th percentile selenium concentration was 9.70 μg/L, and the mean daily load was 45.0 pounds (16,400 pounds annually). The average amount of load that would need to be reduced for this site for water years 2001 through 2005 was 8,640 pounds or 53 percent of the mean annual load. Mean daily load ranged from 40.1 to 53.3 pounds or 14,700 to 19,500 pounds annually (water years 2003 and 2005, respectively). Load reductions ranged from 6,170 pounds (or 37 percent of the mean annual load) to 9,510 pounds (or 58 percent of the mean annual load) (water years 2001 and 2004, respectively). Load reductions were previously studied from 1997-2001 at the Gunnison River near Grand Junction. Estimated load reduction was 5,000 pounds per year as compared to more than 8,000 pounds per year estimated in this report (2001–2005). Streamflow during the 1997–2001 study period was appreciably higher (33 percent) than that observed during the 2001–2005 study period. Higher streamflow during the 1997–2001 study period was likely the reason for the smaller load reduction (more dilution water). This example illustrates that a different study period can result in a different load reduction at the same site. WBID
COGUNF05 COGUNF05 COGUNF06a COGUNF06b COGUNF06b COGUNF06b COGULG04c COGULG01
waterbody description
mean mean mean daily daily load annual load to meet (lbs) load (lbs) 4.6 g/L (lbs) Gunnison River and tributaries above Delta Leroux Creek 0.47 171.7 0.14 Jay Creek 0.06 21.9 0.02 Short Draw 0.37 135.1 0.09 Cottonwood Creek 0.32 116.9 0.15 Big Gulch 0.13 47.5 0.09 Bell Creek 0.21 76.7 0.14 North Fork tributaries total 1.6 550.0 0.63 Red Rock Canyon .64 233.8 .04 Gunnison River at Delta 21.1 7706.7 13.3 Uncompahgre River and tributaries above Delta
COGUUN12 COGUUN12 COGUUN12 COGUUN12
mean annual load to meet 4.6 g/L (lbs) 51.1 7.3 32.9 54.8 32.9 51.1 230.1 14.6 4857.8
Loutzenhizer Arroyo 14.2 5186.6 Cedar Creek 4.2* 1534.1 Dry Creek 3.4 1241.9 Dry Cedar Creek 0.7 255.7 Uncompahgre tributaries total 23.7 8218.3 COGUUN04b Uncompahgre River at Delta 14.9 5442.3 Gunnison River and tributaries above Grand Junction COGULG01 Gunnison River at Delta 21.1 7706.7 COGUUN04b Uncompahgre River at Delta 14.9 5442.3 COGULG04 Sunflower Drain 3.3 1205.3 COGULG04 Whitewater Creek 0.3 121.0 Lower Gunnison River tributaries total 39.6 14447.7
0.03 0.13 0.47 0.07 0.7 4.6
11.0 47.5 171.7 25.6 255.7 1680.2
13.3 4.6 0.2 0.02 18.1
4857.8 1680.2 73.1 7.3 6618.3
COGULG02
21.2
7743.3
Gunnison River at Grand Junction
45.0
16447.6
* reflects loading after completion of Montrose Arroyo demonstration project
Table 22. Current selenium loads (daily and annual) and allowable loads (daily and annual) for attainment of chronic selenium standards for the Gunnison River and tributaries
January 2011
34
Final
Table 22 summarizes the allowable loads for attainment of the chronic Aquatic Life Usebased selenium standard at various locations within the watershed. An alternate assessment methodology incorporated into the Division assessment, which follows, attempts to characterize the loads, and subsequently the required loading reductions, on a monthly basis. While not possible for all tributaries, monthly reductions were calculated for all mainstem segments, tributaries with sufficient monthly data, and those tributaries to which point source discharges are made. VII.
INSTREAM CONDITIONS
Hydrology The Division utilized historical streamflow data for 32 USGS monitoring stations for this assessment. In general, flow data reflects a period of record between 1990 through 2005. However, data is not available for this entire period of record for most sites. In many instances a more limited dataset is used.
Figure 11. Uncompahgre and Gunnison Rivers in the Vicinity of the City of Delta
January 2011
35
Final
COGULG01, Gunnison River above the Uncompahgre River The hydrograph of the Gunnison River at Delta, USGS Gage #9144250, demonstrates similarities to typical mountain streams, although it is influenced more strongly by irrigation practices on the West Slope of Colorado. Flows generally peak in May, due to snowmelt that tails off through the summer (Table 23 and Figure 12), although there is a much greater variability in stream flows in the months of June and July. The period of record from 1995 through 2007 was used to calculate monthly flows in order to provide a more extensive flow record without bias from the drought years of 2001-2003. Gunnison River at Delta (USGS #9144250) Flow, cfs1 25th% 5th% 95th% 75th% Median 30E3 Jan 767 392 1979 1465 914 407 Feb 771 392 1817 1330 944 407 Mar 807 474 3323 1740 1110 407 Apr 1043 548 4511 2918 1875 451 May 1965 601 7879 5435 3000 523 Jun 1083 573 9575 3603 1985 523 Jul 808 658 9278 1645 980 543 Aug 793 612 2549 1405 1020 532 Sep 812 549 2200 1440 1050 488 Oct 857 512 2318 1540 1140 422 Nov 815 424 2432 1558 1090 407 Dec 780 406 2370 1545 1070 407 1 cubic feet per second Table 23. Summary of Flow Records for the Gunnison River at Delta
1E3 371 368 379 368 444 472 508 501 397 400 378 378
Acute and chronic low flows were calculated using USEPA DFLOW software. Acute (1E3) and chronic (30E3) flows are biologically based low flows. Biologically-based design flows are intended to measure the actual occurrence of low flow events with respect to both the duration and frequency (i.e., the number of days aquatic life is subjected to flows below a certain level within a period of several years). Although the extreme value analytical techniques used to calculate hydrologically-based design flows have been used extensively in the field of hydrology and in state water quality standards, these methods do not capture the cumulative nature of effects of low flow events because they only consider the most extreme low flow in any given year. By considering all low flow events with a year, the biologically-based design flow method accounts for the cumulative nature of the biological effects related to low flow events. Acute low flows (1E3) refer to single low flow events that occur once in a three year period. Chronic low flows (30E3) refer to 30-day low flow periods which occur once in three years. The use of low flows to calculate effluent limits is used to protect a stream‟s “critical condition”.
January 2011
36
Final
Gunnison River at Delta 12000 10000
Flow, cfs
8000 6000 4000 2000 0 Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Month
Figure 12. Box and whisker plots for flows in the Gunnison River at Delta (POR = 1995-2007). Boxes represent upper and lower quartile values while whiskers represent 5 th and 95th percentile values. Red stars indicate median flows.
COGULG02, Gunnison River from the Uncompahgre River to the Colorado River Gunnison River near Grand Junction (USGS #9152500) Flow, cfs1 25th% 5th% 95th% 75th% Median 30E3 1E3 Jan 1040 624 2380 1750 1230 650 547 Feb 1020 611 2445 1595 1210 650 545 Mar 1090 663 3080 1740 1300 650 578 Apr 1470 838 6485 3378 2400 671 545 May 2680 866 10600 5870 3640 825 750 Jun 1425 830 6977 3565 2230 835 740 Jul 1090 918 2930 2000 1450 835 776 Aug 1190 914 3310 1930 1550 916 815 Sep 1490 999 3156 2210 1780 929 921 Oct 1470 1060 3190 2250 1880 865 902 Nov 1130 823 2130 1808 1445 706 721 Dec 973 695 2380 1710 1280 650 633 1 cubic feet per second Table 24. Summary of Flow Records for the Gunnison River near Grand Junction
The hydrograph of the Gunnison River near Grand Junction, USGS Gage #9152500, demonstrates similarities to typical mountain streams, with flows peaking in May, due to snowmelt that tails off through the summer (Table 24 and Figure 13), with a slight increase in flows during fall irrigation. Annual flows were calculated for the entire period of record from 1920 through 2008.
January 2011
37
Final
Figure 13. Box and whisker plots for annual flows in the Gunnison River near Grand Junction (POR = 19202008). Boxes represent upper and lower quartile values while whiskers represent 5 th and 95th percentile values. Red lines indicate median flows.
As illustrated by Figure 13, annual flow has demonstrated a variable trend with an overall decreasing trend in median values in the more recent decades. The period of record from 1998 through 2008 (Figure 14) was therefore used to calculate monthly flows for TMDL calculations in order to provide a more extensive flow record without a significant bias from the drought years of 2001-2003. Gunnison River near Grand Junction, 1998-2008
12000
Flow, cfs
10000 8000 6000
4000 2000
0 Jan
Feb
Mar Apr May Jun
Jul
Aug Sep
Oct Nov Dec
Figure 14. Box and whisker plots for flows in the Gunnison River near Grand Junction (POR = 1998-2008). Boxes represent upper and lower quartile values while whiskers represent 5 th and 95th percentile values. Red lines indicate median flows.
January 2011
38
Final
COGULG04a, Tributaries to the Gunnison River Discharge information is available for 15 locations on streams included in segment COGULG04a. Station descriptions and, where available, geolocational data, where available, is summarized in Table 25. Station Name Station ID Latitude Currant Creek near Read, Co. 9137050 38.78471 Callow Creek at Whitewater, Co. 9152520 38.98915 Alkali Creek below Hwy 50, near Delta, Co. 384510108111801 38.752762 Dry Gulch above confluence with Heartland Ditch near Delta, Co. 384557108024300 38.77034 Peach Valley Arroyo near mouth 384604107570701 38.76776 Alfalfa Run at Austin 384649107570501 38.78026 Sulphur Gulch 384752107502201 -Lawhead Gulch 384802107522201 -Wells Gulch 384813108184301 -Negro Creek 384830108001401 -Deer Creek 385130108202301 -North Fork Kannah Creek 385612108202801 -Whitewater Creek .4 mi above mouth, at Whitewater, Co. 385839108264401 38.97748 Cummings Gulch at mouth 384448108070301 38.74665 Kannah Creek about .1 mi below Indian Creek 385600108250301 38.93333 Table 25. USGS Discharge Monitoring Stations Located in Segment COGULG04a.
Longitude -107.939 -108.449 -108.189 -108.04835 -107.953 -107.952 -------108.446 -108.118 -108.418
Flows in Currant Creek near Read, CO, USGS Station #9137050, were taken during nine sampling events from the period 1991 through 2001. The highest flows were measured in May and June while the lowest flows occurred in July. Median flows were calculated to be 4.3 cubic feet per second. Individual results are shown in Table 26. Date Discharge, cfs1 Date Discharge, cfs1 Date Discharge, cfs1 7/16/1991 0.9 7/20/1999 4.9 3/14/2000 4.3 5/17/1999 14.0 8/31/1999 2.9 7/7/2000 1.2 6/24/1999 11.0 11/9/1999 8.3 3/28/2001 3.0 1 cubic feet per second Table 26. USGS Discharge Monitoring Data for Currant Creek near Read, CO.
Flows in Callow Creek at Whitewater, CO, USGS Station #9152520, were taken during each of the sampling events. The highest flows were recorded in May and June while the lowest flows occurred in fall and early winter. Ten samples were collected from the period 1999 through 2002. Median flows were calculated to be 0.04 cubic feet per second (Table 27). Date Discharge, cfs1 Date Discharge, cfs1 Date Discharge, cfs1 4/29/1999 0.28 2/14/2001 0.01 9/25/2001 0.04 8/21/2000 0.02 5/16/2001 0.03 5/23/2002 0.14 9/27/2000 0.09 8/29/2001 0.04 --1 cubic feet per second Table 27. USGS Discharge Monitoring Data for Callow Creek at Whitewater, CO
January 2011
39
Final
Flows in Alkali Creek below Hwy 50, near Delta, Co., USGS Station #384510108111801, were taken during each of the sampling events. The highest flows were recorded in winter which corresponded to the months of lowest flows. Ten samples were collected from the period 1995 through 2000. Median flows were calculated to be 0.11 cubic feet per second. Date Discharge, cfs1 Date Discharge, cfs1 Date Discharge, cfs1 11/24/1995 0.14 1/17/2000 0.12 3/20/2000 0.14 2/14/1996 0.13 1/19/2000 0.17 4/4/2000 0.03 4/7/1999 0.05 1/26/2000 0.09 --11/17/1999 0.03 2/7/2000 0.04 --1 cubic feet per second Table 28. USGS Discharge Monitoring Data for Alkali Creek below Hwy 50, near Delta, CO
Flows in Dry Gulch above the confluence with Heartland Ditch near Delta, CO, USGS Station #384557108024300, were taken during two sampling events. The higher flow was recorded in July (0.45 cfs) while the month of lower flow occurred in March (0.08 cfs). Both samples were collected in 2002. Median flows were calculated to be 0.27 cubic feet per second. Flows in the Peach Valley Arroyo near the mouth, USGS Station #384604107570701, were taken during each of the sampling events. The highest flows were recorded in July and August while the month of lowest flow occurred in March. Ten samples were collected from the period 1991 through 2000. Median flows were calculated to be 4.9 cubic feet per second. Individual results are shown in Table 29. Date Discharge, cfs1 Date Discharge, cfs1 Date 7/22/1991 14.0 9/9/1993 6.1 3/15/2000 3/12/1992 0.6 5/18/1999 7.1 11/9/1999 7/14/1993 13.0 8/31/1999 3.7 -8/11/1993 24.0 12/10/1999 0.7 -1 cubic feet per second Table 29. USGS Discharge Monitoring Data for Peach Valley Arroyo
Discharge, cfs1 0.2 3.2 ---
Flows in Alfalfa Run at Austin, USGS Station #384649107570501, were taken during each of the seven sampling events (Table 13). The highest flows were recorded in July while the month of lowest flow occurred in March. Seven samples were collected from the period 1991 through 2000. Median flows were calculated to be 4.8 cubic feet per second. Date Discharge, cfs1 Date Discharge, cfs1 Date 7/22/1991 13.0 7/20/1999 6.0 3/15/2000 5/18/1999 5.5 8/31/1999 4.4 -6/24/1999 4.8 11/9/1999 2.4 -1 cubic feet per second Table 30. USGS Discharge Monitoring Data for Alfalfa Run at Austin
January 2011
40
Discharge, cfs1 0.8 ---
Final
Flows in Sulphur Gulch, USGS Station #384752107502201, were taken during five of the seven sampling events. The highest flow was recorded in March while the lowest flow occurred in August. Seven samples were collected from the period 1999 through 2003. Median flows were calculated to be 0.1 cubic feet per second. Date Discharge, cfs1 Date Discharge, cfs1 5/17/1999 0.33 3/14/2000 0.09 8/31/1999 0.06 3/11/2002 0.48 11/9/1999 0.10 3/21/2002 -1 cubic feet per second Table 31. USGS Discharge Monitoring Data for Sulphur Gulch
Date 7/22/2003 ---
Discharge, cfs1 ----
Flows in Lawhead Gulch, USGS Station #384802107522201, were taken during all of the four sampling events. The highest flow was recorded in May while the lowest flow occurred in March. Four samples were collected from the period 1999 through 2000. Median flows were calculated to be 0.17 cubic feet per second. Date Discharge, cfs1 Date Discharge, cfs1 5/17/1999 0.80 11/10/1999 0.19 8/31/1999 0.14 3/14/2000 0.04 1 cubic feet per second Table 32. USGS Discharge Monitoring Data for Lawhead Gulch
Flows in Wells Gulch, USGS Station #384813108184301, were taken during all of the eight sampling events. The highest flow was recorded in March while the lowest flows occurred in winter months. Eight samples were collected from the period 1995 through 2000. Median flows were calculated to be 0.02 cubic feet per second. Date Discharge, cfs1 Date Discharge, cfs1 11/24/1995 0.01 7/15/1999 0.04 2/14/1996 0.01 1/17/2000 0.03 6/17/1999 0.60 1/19/2000 0.01 1 cubic feet per second Table 33. USGS Discharge Monitoring Data for Wells Gulch
Date 1/26/2000 3/21/2000 --
Discharge, cfs1 0.01 0.74 --
There was only one flow sample in Negro Creek, USGS Station #384830108001401, taken during the one recorded sampling event in January of 2000. This flow was calculated to be 0.07 cubic feet per second and was used in TMDL calculations. Flows in Deer Creek, USGS Station #385104108213501, were taken during all of the five sampling events. The highest flow was recorded in March while the lowest flow occurred in February. Five samples were collected in the year 2000. Median flows were calculated to be 0.05 cubic feet per second. Individual results are shown in Table 34.
January 2011
41
Final
Date Discharge, cfs1 Date Discharge, cfs1 1/17/2000 0.05 1/26/2000 0.04 1/19/2000 0.08 2/7/2000 0.01 1 cubic feet per second Table 34. USGS Discharge Monitoring Data for Deer Creek
Date 3/21/2000 --
Discharge, cfs1 0.35 --
There was only one flow sample in the North Fork of Kannah Creek, USGS Station #385612108202801, taken during the one recorded sampling event in October of 2003. This flow was calculated to be 0.15 cubic feet per second and was used in TMDL calculations. Kannah Creek is segmented such that the upper portion is included in this segment, COGULG04a. The lower portion is included in Segment COGULG04b. No discharge measurements were taken in Kannah Creek within Segment COGULG04a. Median flows were calculated to be 2.8 cubic feet per second within Segment COGULG04b. This value was assumed to be representative of flows in upper Kannah Creek for this TMDL evaluation. Flows in Whitewater Creek 0.4 miles above the mouth at Whitewater, USGS Station #385839108264401, were taken during all of the twenty sampling events (Table 35). The highest flows were recorded in April through July while the lowest flows occurred during late summer and winter months. Twenty samples were collected from the period 1999 through 2002. Median flows were calculated to be 1.7 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 36 below. Date Discharge, cfs1 Date Discharge, cfs1 Date Discharge, cfs1 4/29/1999 7.3 8/22/2000 1.5 9/25/2001 0.4 6/29/1999 6.9 9/27/2000 1.3 11/19/2001 0.7 7/21/1999 8.6 2/15/2001 2.7 1/15/2002 1.6 9/8/1999 6.2 4/17/2001 0.8 3/12/2002 1.7 11/17/1999 3.6 5/15/2001 7.6 5/23/2002 2.4 3/21/2000 4.7 7/30/2001 0.6 1/15/2002 1.6 7/25/2000 1.2 8/29/2001 0.3 --1 cubic feet per second Table 35. USGS Discharge Monitoring Data for Whitewater Creek near mouth Jan Feb Mar Apr May Jun Jul Aug 1.61 2.7 3.2 4.0 5.0 6.9 1.2 0.9 1 cubic feet per second Table 36. Monthly Median Flows for Whitewater Creek near mouth
Sep 1.3
Oct 2.5
Nov 3.6
Dec 2.6
Flows in Cummings Gulch at the mouth, USGS Station #384448108070301, were taken during all of thirteen sampling events (Table 37). The highest flows were recorded in April and August while the lowest flows occurred during January through March. Twenty samples were collected from the period 1991 through 2000. Median flows were calculated to be 7.8 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 38 below.
January 2011
42
Final
Date Discharge, cfs1 Date Discharge, cfs1 Date 7/15/1991 35.0 12/13/1994 6.5 9/7/1999 1/28/1992 7.8 1/17/1995 3.8 11/17/1999 7/15/1993 26.0 2/15/1995 3.5 3/20/2000 8/12/1993 84.0 3/8/1995 3.5 -9/10/1993 40.0 4/27/1999 61.0 -1 cubic feet per second Table 37. USGS Discharge Monitoring Data for Cummings Gulch at mouth Jan Feb Mar Apr May Jun Jul 5.81 3.5 2.9 61.0 50.8 40.7 30.5 1 cubic feet per second Table 38. Monthly Median Flows for Cummings Gulch at mouth
Aug 84.0
Sep 54.5
Discharge, cfs1 69.0 6.0 2.3 ---
Oct 30.3
Nov 6.0
Dec 6.5
Flows in the Sunflower Drain at Highway 92 near Read, USGS Station # 384551107591901, were taken during all of one hundred six sampling events. The highest median flows were measured in September and October while the lowest median flows occurred during January through March. All of the samples were collected from the period 1992 through 2002. Median monthly flows were calculated to be 23 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 39 below. Jan Feb Mar Apr May Jun Jul Aug Sep 2.51 2.6 2.2 38.0 42.0 46.0 33.0 45.0 57.5 1 cubic feet per second Table 39. Monthly Median Flows for Sunflower Drain at Highway 92 near Read
Oct 59.5
Nov 11.0
Dec 3.9
COGULG04b, Kannah Creek below the point of diversion for public water supply Flows in Kannah Creek below USGS Station 09152000, at USGS Station #385600108250301, were taken during seventeen of the nineteen sampling events (Table 40). The highest flows were recorded in May while the lowest flows occurred during September and October. Seventeen flow samples were collected from the period 1999 through 2002. Median flows were calculated to be 2.8 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 41 below. Date Discharge, cfs1 Date Discharge, cfs1 Date Discharge, cfs1 4/29/1999 1.4 3/21/2000 5.7 4/17/2001 1.0 5/27/1999 51.0 8/21/2000 0.8 5/15/2001 64.0 6/29/1999 0.4 9/27/2000 0.5 9/25/2001 0.0 7/21/1999 1.8 11/14/2000 0.9 11/19/2001 1.1 9/8/1999 1.1 2/15/2001 2.0 1/15/2002 3.6 11/17/1999 4.7 3/12/2002 1.4 --1 cubic feet per second Table 40. USGS Discharge Monitoring Data for Kannah Creek below Indian Creek Jan Feb Mar Apr May Jun Jul Aug 3.61 2.0 3.6 1.2 57.5 0.4 1.8 0.8 1 cubic feet per second Table 41. Monthly Median Flows for Kannah Creek below Indian Creek
January 2011
43
Sep 0.5
Oct 0.8
Nov 1.1
Final
Dec 2.4
COGULG04c, Red Rock Canyon Creek Flows in Red Rock Canyon Creek at the mouth near Montrose, CO (Figure 15), at USGS Station #383537107471500, were taken during twenty-two of twenty-three sampling events (Table 42). The highest median flows were recorded in August and September while the lowest median flows occurred during October and April. Twenty-two flow samples were collected between 2001 through 2005. Median flows were calculated to be 2.3 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 43. Date Discharge, cfs1 Date Discharge, cfs1 Date Discharge, cfs1 8/23/2001 5.0 7/16/2003 2.5 9/16/2004 4.0 10/18/2001 1.6 8/7/2003 3.0 11/2/2004 2.4 11/6/2001 -8/27/2003 6.5 4/28/2005 1.7 5/21/2002 2.1 3/31/2004 1.8 5/31/2005 1.6 6/13/2002 1.3 5/13/2004 2.3 6/22/2005 2.2 8/15/2002 0.9 6/15/2004 4.2 7/18/2005 5.0 5/22/2003 1.3 7/8/2004 1.0 9/15/2005 12.0 6/17/2003 2.3 8/12/2004 4.1 --1 cubic feet per second Table 42. USGS Discharge Monitoring Data for Red Rock Canyon Creek at Mouth Jan Feb Mar Apr May Jun Jul Aug Sep Oct 2.11 2.0 1.8 1.7 1.9 2.3 2.5 4.1 8.0 1.6 1 cubic feet per second Table 43. Summary of Median Monthly Flows for Red Rock Canyon Creek at Mouth
Nov 2.4
Figure 15. Red Rock Canyon Creek
January 2011
44
Final
Dec 2.3
COGUNF03, Mainstem of North Fork of the Gunnison River below Lazear to the Gunnison River The hydrograph of the North Fork Gunnison River near Hotchkiss (see Figure 16), USGS Gage #9135950, is more typical of hydrographs of mountain streams, with peak flows occurring in May due to snowmelt which tails off through the summer. The period of record from 2000 through 2007 was used in order to most accurately represent flows on the North Fork Gunnison River. Lapses in continuous flow data were estimated from the USGS Gage #9134100, North Fork Gunnison River below Paonia, CO (R2 = 0.83) using Equation 2, whose period of record begins in March of 2000. Eq. 2
Flow = (0.9242* USGS Gage #9134100) + 72.941
North Fork Gunnison River near Hotchkiss (USGS #9135950) Flow, cfs 1 25th% 5th% 95th% 75th% Median 30E3 1E3 Jan 122 114 211 175 131 113 109 Feb 123 112 201 163 136 113 106 Mar 151 113 1071 277 210 115 111 Apr 453 231 2755 1358 688 152 135 May 853 288 4178 2435 1560 111 185 Jun 195 82 2291 878 440 31 61 Jul 45 28 443 89 60 31 24 Aug 53 28 149 89 69 31 22 Sep 70 52 270 127 86 31 23 Oct 102 81 405 250 138 48 70 Nov 142 121 275 198 157 97 92 Dec 127 112 209 167 139 115 99 1 cubic feet per second Table 44. Summary of Flow Records for the North Fork Gunnison River near Hotchkiss
January 2011
45
Final
North Fork Gunnison 4500 4000 3500
Flow, cfs
3000 2500 2000 1500 1000
500 0 Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Figure 16. Box and whisker plots for annual flows in the North Fork Gunnison River near Hotchkiss (POR = 2001-2008). Boxes represent upper and lower quartile values while whiskers represent 5th and 95th percentile values. Red stars indicate median flows.
COGUNF05, Leroux Creek and Jay Creek Flows in Leroux Creek at Hotchkiss, CO, and at the mouth, at USGS Stations # 9135900 and # 384732107434801, were taken during twenty-one of the fifty sampling events (Table 45). The highest median flows were recorded in October and November while the lowest median flows occurred during May and March. Twenty-one flow samples were collected between 1990 through 2000. Median flows were calculated to be 7.2 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 46 below. Date Discharge, cfs1 Date Discharge, cfs1 Date 12/18/1990 6.1 10/14/1998 12.0 3/14/2000 3/19/1991 3.9 10/26/1992 12 6/20/2000 6/5/1991 6.1 4/8/1993 10 6/20/2000 8/27/1991 6.2 6/16/1993 272 12/5/20002 11/5/1991 12.0 8/11/1993 9.6 12/5/20002 3/17/1992 4.7 5/12/1999 4 -6/16/1992 4.9 8/30/1999 11 -8/11/1992 8.6 11/9/1999 11 -1 cubic feet per second 2 flows measured at both stations Table 45. USGS Discharge Monitoring Data for Leroux Creek at Hotchkiss
Jan Feb Mar Apr May Jun Jul Aug Sep 6.41 5.5 4.7 10.0 4.0 4.9 7.0 9.1 10.6 1 cubic feet per second Table 46. Summary of Median Monthly Flows for Leroux Creek at Hotchkiss
January 2011
46
Discharge, cfs1 5.0 1.9 3.7 7.3 7.2 ----
Oct 12.0
Nov 11.5
Final
Dec 7.2
Figure 17. Leroux Creek and Jay Creek near Hotchkiss, CO
Flows in Jay Creek at Jay Creek at Highway 133 near the mouth, at USGS Station # 384915107412101, were taken during four sampling events. The highest flow was recorded in May while the lowest flow occurred during November. Four flow samples were collected between 1999 through 2000. Median flows were calculated to be 0.58 cubic feet per second. Date Discharge, cfs1 Date Discharge, cfs1 5/11/1999 4.80 11/8/1999 0.07 8/30/1999 0.84 3/13/2000 0.32 1 cubic feet per second Table 47. USGS Discharge Monitoring Data for Jay Creek
COGUNF06a, Short Draw Flows in Short Draw west of the city fairgrounds at Hotchkiss, CO, at USGS Station # 384747107430501, were taken during seven sampling events (Table 48). The highest flow was recorded in July while the lowest flow occurred during March. Seven flow samples were collected between 1998 through 2000. Median flows were calculated to be 5.6 cubic feet per second.
January 2011
47
Final
Date Discharge, cfs1 Date Discharge, cfs1 Date 10/15/1998 7.4 7/19/1999 12.0 3/14/2000 5/12/1999 4.6 8/30/1999 5.6 -6/23/1999 8.8 11/9/1999 2.9 -1 cubic feet per second Table 48. USGS Discharge Monitoring Data for Short Draw at Hotchkiss
Discharge, cfs1 0.7 ---
COGUNF06b, Big Gulch, Cottonwood Creek Flows in Big Gulch at Highway 92, at USGS Station # 384756107490801, were taken during four sampling events (Table 49). The highest flow was recorded in March while the lowest flow occurred during August. Four flow samples were collected between 1999 through 2000. Median flows were calculated to be 3.3 cubic feet per second. Date Discharge, cfs1 Date Discharge, cfs1 11/9/1999 5.5 8/31/1999 0.4 5/12/1999 1.1 3/14/2000 6.3 1 cubic feet per second Table 49. USGS Discharge Monitoring Data for Big Gulch
Flows in Cottonwood Creek near Hotchkiss, CO, at USGS Station # 9134200, were taken during seven of the eight total sampling events. The highest flow was recorded in June while the lowest flow occurred during November. Seven flow samples were collected between 1999 through 2000. Median flows were calculated to be 7.0 cubic feet per second. Individual results are tabulated below. Date Discharge, cfs1 Date Discharge, cfs1 Date Discharge, cfs1 5/11/1999 9.5 8/30/1999 13.0 7/25/2000 4.4 6/23/1999 15.0 11/8/1999 4.9 --7/19/1999 7.0 3/13/2000 5.6 --1 cubic feet per second Table 50. USGS Discharge Monitoring Data for Cottonwood Creek near Hotchkiss
Flows in Bell Creek at County Road and railroad tracks near the mouth, at USGS Station # 384922107402001, were taken during all six of the sampling events (Table 51). The highest flow was recorded in June while the lowest flow occurred during November. Six flow samples were collected between 1999 through 2000. Median flows were calculated to be 10.3 cubic feet per second. Date Discharge, cfs1 Date Discharge, cfs1 5/11/1999 11.0 7/19/1999 13.0 6/23/1999 24.0 8/30/1999 9.5 1 cubic feet per second Table 51. USGS Discharge Monitoring Data for Bell Creek
January 2011
48
Date 11/8/1999 3/13/2000
Discharge, cfs1 1.6 1.8
Final
COGUUN04b, Uncompahgre River from LaSalle Road to Confluence Park The hydrograph of the Uncompahgre River near Olathe, UNCOLACO Gage, is unlike hydrographs of typical mountain streams (Table 52, Figure 18). Peak flows occur in the fall, likely due to irrigation return flows. Highest median flows occur in November. Variability in streamflow is evident for most of the year. Acute low flows fall below 1 cubic foot per second for the months April through October. The period of record from 1998 through 2008 was used to calculate monthly flows for TMDL calculations in order to provide a more extensive flow record without a significant bias from the drought years of 2001-2003. Uncompahgre River near Olathe (UNCOLACO) Flow, cfs1 25th% 5th% 95th% 75th% Median 30E3 Jan 120 90 181 150 130 89 Feb 112 88 198 146 122 51 Mar 95 51 351 173 117 0.4 Apr 13 1 361 146 65 0.4 May 17 1 508 188 84 0.4 Jun 5 1 545 229 95 0.5 Jul 2 0 322 141 23 0.5 Aug 7 1 494 161 53 0.5 Sep 44 2 473 199 110 1.3 Oct 26 1 581 337 110 1.3 Nov 169 150 394 230 199 1.4 Dec 137 113 215 179 152 94 1 cubic feet per second Table 52. USGS Discharge Statistics for the Uncompahgre River near Olathe
1E3 86 85 4.1 0.1 0.6 0.5 0.3 0.3 0.4 0.4 134 101
Uncompahgre near Olathe 700 600
Flow, cfs
500
400 300 200 100 0 Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Figure 18. Box and whisker plots for annual flows in the Uncompahgre River near Olathe (POR = 1998-2008). Boxes represent upper and lower quartile values while whiskers represent 5 th and 95th percentile values. Red lines indicate median flows.
January 2011
49
Final
COGUUN04c, Uncompahgre River below Delta to the Gunnison River Confluence The hydrograph of the Uncompahgre River at Delta, USGS Gage #9149500, also deviates from hydrographs of typical mountain streams (Table 53, Figure 15). Peak flows occur in the fall, likely due to irrigation return flows. The period of record from 1995 through 2007 was used to calculate monthly flows for TMDL calculations in order to provide a more extensive flow record without a significant bias from the drought years of 2001-2003. Uncompahgre River at Delta (USGS #9149500) Flow, cfs 1 25th% 5th% 95th% 75th% Median 30E3 Jan 168 151 234 208 186 129 Feb 153 133 225 201 166 122 Mar 148 117 475 281 179 78 Apr 142 55 623 352 238 78 May 199 83 890 487 332 78 Jun 160 83 1141 487 292 79 Jul 150 83 791 324 218 79 Aug 179 110 850 372 255 82 Sep 332 223 943 640 451 110 Oct 339 269 937 632 452 215 Nov 271 235 488 342 309 184 Dec 207 183 299 257 228 151 1 cubic feet per second Table 53. USGS Discharge Statistics for the Uncompahgre River at Delta
1E3 138 119 75 24 66 72 69 76 107 211 210 158
Uncompahgre at Delta 1200 1000 800 600 Flow, cfs
400 200 0 Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Month
Figure 19. Box and whisker plots for annual flows in the Uncompahgre River at Delta (POR = 19952007). Boxes represent upper and lower quartile values while whiskers represent 5 th and 95th percentile values. Red lines indicate median flows.
January 2011
50
Final
COGUUN12, tributaries to the Uncompahgre River Flows in the Montrose Arroyo at Niagara, at USGS Station # 382802107513301, were taken during one hundred thirty sampling events. The highest median flows were recorded in September and November while the lowest median flows occurred during March and April. This is likely due to seepage of irrigation return flows into the arroyo. One hundred thirty flow samples were collected between 1993 through 2008. Median flows were calculated to be 2.3 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 54 below. Jan Feb Mar Apr May Jun Jul Aug Sep 2.21 2.0 1.6 1.8 5.9 8.2 15.0 15.0 22.5 1 cubic feet per second Table 54. Summary of Median Monthly Flows for Montrose Arroyo at Niagara
Oct 16.0
Nov 22.0
Figure 20. Dry Creek, Dry Cedar Creek, and Loutzenhizer Arroyo near Montrose, CO
January 2011
51
Final
Dec 15.0
Similar to the Uncompahgre mainstem, the hydrograph of Dry Creek near Delta, USGS Gage #9149480, is unlike hydrographs of typical mountain streams. Peak flows occur in the fall, likely due to irrigation return flows. (Tables 55 and 56, Figure 17). Highest median flows occur in the months of September and October due to irrigation practices. The months of June and July reflect the lowest median monthly flows. The period of record from 1995 through 1998 was used to calculate monthly flows for TMDL allocations. Monthly medians were also calculated from instantaneous discharge measurements from the period of record 1991 through 2001 and are illustrated in Table 56. Dry Creek near Delta (USGS #9149480) Flow, cfs1 25th% 5th% 95th% 75th% Median Jan 24 22 36 29 26 Feb 21 18 30 26 23 Mar 21 19 155 85 25 Apr 43 16 223 131 86 May 67 17 243 209 147 Jun 17 14 184 39 20 Jul 16 14 64 23 18 Aug 21 18 125 51 32 Sep 101 30 233 184 165 Oct 109 72 243 206 165 Nov 46 37 122 55 48 Dec 30 25 45 39 34 1 cubic feet per second Table 55. USGS Discharge Statistics for Dry Creek near Delta
Jan Feb Mar Apr May Jun Jul Aug 551 55 57 129 200 96 80 110 1 cubic feet per second Table 56. Summary of Median Monthly Flows for Dry Creek near Delta
January 2011
52
30E3 20 19 19 29 17 17 17 18 22 45 32 26
Sep 139
1E3 23 19 18 9 14 12 12 17 22 67 33 25
Oct 169
Nov 128
Final
Dec 60
Dry Creek near Delta 300 250 200 150 100 50
0 Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
b whisker plots for annual flows in DrygCreek near Delta (POR = 1995-1998). Figure 21. Box and Boxes represent upper and lower quartile values while whiskers represent 5 th and 95th percentile values. Red lines indicate median flows.
Flows in Dry Cedar Creek at the mouth, at USGS Station # 382711107520101, were taken during thirty-one sampling events (Table 57). The highest median flows were recorded in May and September while the lowest median flows occurred during January and February. Thirty-one flow samples were collected between 1991 through 2000. Median flows were calculated to be 9.7 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 58 below. Date Discharge, cfs1 Date Discharge, cfs1 7/17/1991 11.0 4/25/1995 8.3 2/20/1992 2.0 5/23/1995 9.7 4/28/1992 6.4 6/22/1995 8.9 6/9/1992 22.0 7/25/1995 9.3 7/13/1993 23.0 8/23/1995 17.0 8/12/1993 12.0 9/26/1995 27.0 9/8/1993 24.0 10/25/1995 16.0 12/20/1994 2.1 11/22/1995 2.7 1/19/1995 2.0 12/11/1995 2.4 2/16/1995 1.4 1/18/1996 1.9 3/15/1995 2.7 2/15/1996 3.1 1 cubic feet per second Table 57. USGS Discharge Monitoring Data for Dry Cedar Creek
Jan Feb Mar Apr May Jun Jul Aug 2.01 2.0 2.3 13.7 8.8 20.0 13.0 15.0 1 cubic feet per second Table 58. Summary of Median Monthly Flows for Dry Cedar Creek
January 2011
53
Date 3/13/1996 4/10/1996 5/21/1996 6/26/1996 7/23/1996 8/21/1996 9/17/1996 4/11/2000 7/21/2000 ---
Sep 24.0
Discharge, cfs1 1.9 19.0 7.8 20.0 26.0 15.0 20.0 29.0 13.0 ---
Oct 16.0
Nov 2.7
Final
Dec 2.3
Flows in Cedar Creek near the mouth, at USGS Station # 383041107544201, were taken during forty-five sampling events (Table 59). The highest median flows were recorded in June and August while the lowest median flows occurred during December and February. Forty-five flow samples were collected between 1991 through 2000. Median flows were calculated to be 37 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 60 below. Date Discharge, cfs1 Date Discharge, cfs1 7/24/1991 150.0 11/21/1995 18.0 10/29/1991 45.0 12/11/1995 18.0 2/20/1992 15.0 2/15/1996 16.0 3/3/1992 18.0 3/13/1996 15.0 3/27/1992 22.0 4/16/1996 151.0 4/28/1992 68.0 5/21/1996 149.0 6/9/1992 200.0 7/23/1996 179.0 6/23/1992 213.0 8/21/1996 191.0 7/14/1992 208.0 9/17/1996 197.0 8/11/1992 178.0 10/16/1996 41.0 3/3/1993 19.0 11/6/1996 27.0 3/23/1993 37.0 12/5/1996 21.0 7/13/1993 162.0 1/8/1997 20.0 8/13/1993 208.0 3/11/1997 19.0 9/8/1993 128.0 5/15/1997 92.0 1 cubic feet per second Table 59. USGS Discharge Monitoring Data for Cedar Creek
Jan Feb Mar Apr May Jun Jul 201 16 19 68 149 207 179 1 cubic feet per second Table 60. Summary of Median Monthly Flows for Cedar Creek
Aug 189
Date 7/10/1997 8/12/1997 11/10/1998 12/10/1998 1/27/1999 3/9/1999 11/16/1999 1/21/2000 3/15/2000 4/10/2000 5/22/2000 8/1/2000 12/19/2000 12/20/2000 2/13/2001
Sep 163
Discharge, cfs1 199.0 180.0 24.0 17.0 20.0 17.0 24.0 18.0 16.0 55.0 198.0 189.0 8.4 21.0 20.0
Oct 43
Nov 24
Dec 18
Flows in the Loutzenhizer Arroyo at North River Road, at USGS Station # 383946107595301, were taken during eighty-six sampling events. The highest median flows were recorded in June and August while the lowest median flows occurred during January through March. Eighty-six flow samples were collected between 1991 through 2003. Median flows were calculated to be 60 cubic feet per second. Monthly median flows are illustrated as cubic feet per second in Table 61 below. Jan Feb Mar Apr May Jun Jul Aug 131 10 12 67 90 96 82 103 1 cubic feet per second Table 61. Summary of Median Monthly Flows for Loutzenhizer Arroyo
January 2011
54
Sep 94
Oct 70
Nov 24
Final
Dec 18
VII. TECHNICAL ANALYSIS SOURCE ANALYSIS The TMDL is the sum of the Waste Load Allocation (WLA), which is the load from point source discharge, Load Allocation (LA) which is the load attributed to natural background and/or non-point sources, and a Margin of Safety (MOS). Section IV. PROBLEM IDENTIFICATION includes a discussion of the various permitted point sources that were identified through the EPA Envirofacts Data Warehouse site (http://oaspub.epa.gov/enviro/ef_home2.water) and Colorado Discharge Permit System files. Point Source Contributions There are six domestic wastewater treatment facilities discharging to listed segments in the Gunnison and Uncompahgre watersheds that have been issued individual CDPS permits (Table 3). These include the Town of Paonia and Town of Hotchkiss wastewater treatment plants which discharge to the North Fork Gunnison River (North Fork segment 3), the Town of Olathe wastewater treatment plant, the City of Montrose and West Montrose Sanitation District which discharge to the Uncompahgre River segment 4b, and the City of Delta facility which discharges to Lower Gunnison segment 2. The Paonia and Hotchkiss facilities discharge to upstream reaches of the North Fork of the Gunnison River (Segment 3) that are in attainment of assigned selenium standards, however they will be addressed further as wasteload allocations in the TMDL analysis of North Fork Segment 3. The Town of Olathe‟s domestic wastewater treatment plant began operation in 2005. The design capacity identified in Table 3 is the facility‟s maximum design flow. The CDPS permit actually includes “tiered” effluent limits. These effluent limits are applied at discharge volumes of up to 0.35 mgd, 0.351 to 0.49 mgd, and greater than 0.49 up to the facility‟s rated capacity of 0.73 mgd. To date the facility has reported a maximum monthly average flow of 0.482 mgd. Selenium is not a constituent typically found in domestic or industrial wastewater. However, selenium bearing groundwater may enter a sanitary sewer system by seeping through cracks in the system (infiltration), via foundation drain/sump pump systems, or through the Town‟s water supply. Portions of the Town of Olathe‟s collection system date back to the 1900s. In this instance, excessive infiltration into the Town‟s sanitary sewer system is likely the most significant pathway for selenium to enter the system. The Town has began a program intended to identify areas within the collection system that were most susceptible to excess infiltration, and to remedy those situations under the terms of their previous CDPS permit. This program has successfully addressed the infiltration problem in much of the collection system.
January 2011
55
Final
Selenium loads for the Olathe facility were calculated based upon nine quarters of selenium monitoring required under the current CDPS permit. Average flows were calculated based upon the monthly discharges for the quarter during which selenium data was reported (Table 62). quarter selenium flow, average load, lb/day 4/05 0.035 0.32 mgd1 13.0 g/L 1/06 0.29 mgd 0.058 24.0 g/L 2/06 0.27 mgd 0.016 7.0 g/L 3/06 0.39 mgd 0.026 8.0 g/L 4/06 0.42 mgd 0.070 20.0 g/L 1/07 0.32 mgd 0.043 16.0 g/L 2/07 0.35 mgd 0.023 8.0 g/L 3/07 0.42 mgd 0.028 0.8 g/L 4/07 0.32 mgd 0.019 7.0 g/L 1 million gallons per day Table 62. Town of Olathe selenium discharge loads
The CDPS permit for the Delta domestic wastewater treatment plant requires monthly selenium monitoring. Data is available for the most recent two years of monitoring. Results exceed the chronic limit of 4.6 g/L in all but one sampling event. For purposes of calculating the discharge load, flows were converted to cubic feet per second (Table 63). Selenium Flow, (dis), average Date g/L (cfs) load, lb/day 31-Mar-09 13.60 1.48 0.109 30-Apr-09 7.30 1.51 0.059 31-May-09 4.04 1.67 0.036 30-Jun-09 5.85 1.74 0.055 31-Jul-09 9.60 1.72 0.089 31-Aug-09 9.18 1.75 0.087 30-Sep-09 9.81 1.80 0.095 31-Oct-09 8.67 1.80 0.084 30-Nov-09 8.49 1.57 0.072 31-Dec-09 7.38 1.45 0.058 31-Jan-10 7.00 1.36 0.051 28-Feb-10 7.44 1.33 0.053 31-Mar-10 8.08 1.33 0.058 30-Apr-10 6.98 1.28 0.048 31-May-10 7.90 1.44 0.061 30-Jun-10 8.28 1.50 0.067 31-Jul-10 8.06 1.58 0.069 31-Aug-10 12.10 1.84 0.120 30-Sep-10 9.32 1.70 0.086 Table 63. Town of Delta selenium discharge loads
January 2011
56
Final
Within a domestic treatment plant selenium tends to be partitioned into the residual treatment solids, or “biosolids” produced at the plant. EPA (2003) estimates some 50 percent of the selenium entering a mechanical treatment plant, such as the City of Delta‟s facility, is removed into the biosolids. The Olathe facility is a lagoon system, estimated by EPA to be more efficient (77 percent) in terms of selenium removal, however it should be noted that the information that EPA had available was limited to a single lagoon system. That the City of Delta discharge contains a slightly lower concentration of selenium than that of the Town of Olathe‟s lagoon system suggests that infiltration into the collection system continues to introduce a substantial amount of selenium into the Town‟s treatment system. The Town of Olathe consistently exceeds the standard, as does the City of Delta. The City of Delta is also under a compliance schedule (issued in 2009 with permit renewal) to meet the 4.6 g/L dissolved selenium water quality based effluent limit by the year 2013. The city is currently in the process of installing a diffuser instream in order to provide more assimilative capacity below the City‟s discharge in order to protect critical T & E habitat. Additionally, there are a number of discharging facilities that are authorized to discharge to listed segments via general permit certifications. These are primarily sand and gravel extraction facilities located within the listed portions of the Gunnison Basin. These are covered by a general CDPS permit issued to address such industrial operations. Individual sand and gravel production facilities must receive certification under the terms of the general permit. Each operating sand and gravel operation must comply with a uniform set of discharge requirements as set forth in the general permit. These operations are identified in Table 64. Waterbody Identification COGUNF03 COGUNF03 COGUNF03 COGUNF03 COGUNF03
Tri-County Gravel Diamond Lazy L. Ranch Enterprises, Janet Pit Oldcastle SW Group, Inc., Campbell Gravel Pit Oldcastle SW Group, Inc., 4D Gravel Pit Oldcastle SW Group, Inc., Tri-County Pit
CDPS No. COG500255 COG500458 COG500397 COG500400 COG500498
Design Capacity, mgd1 0.30 2.90 2.02 0.87 3.00
COGUUN04b
Western Gravel Concrete Facility – North R-34 Pit
COG500486
2.50
COGULG04a COGULG04a
Elam Construction, Dubs Pit Western Gravel, Fredlund Pit
COG500427 COG500451
1.44 3.00
Permittee
COGULG02 Elam Construction, Mule Farm Gravel Pit COG500210 1.20 COGULG02 Whitewater Building Materials COG500127 2.60 1 million gallons per day Table 64 General CDPS Sand and Gravel Permit Certifications discharging to 303(d) Listed segments
Five additional sand and gravel operations have been identified that discharge to segment 1 of the lower Gunnison River (Table 65). This segment is not currently listed; however, the Aquatic Life Use–based chronic selenium standard is not attained in the lower portion (North Fork to Uncompahgre River) on this segment (Table 2). Discharges from these facilities potentially contribute to the downstream selenium load.
January 2011
57
Final
Waterbody Identification
Design Capacity mgd1 0.65 0.58 0.58 1.4/4.0 0.4
Permittee CDPS No. COGULG01 Elam Construction, Bennett Gravel Pit COG500439 COGULG01 Grand Junction Pipe & Supply, Delta Paving Gravel Pit COG500444 COGULG01 Lafarge West, North Delta Pit COG500160 COGULG01 Oldcastle SW Group, Inc., Anderson Pit COG500464 COGULG01 Western Gravel North D-22 Pit/Concrete Plant #2 COG500358 1 million gallons per day Table 65. General CDPS Sand and Gravel Permit Certifications discharging to Lower Gunnison segment 1
As is the case with the municipal treatment facilities within the watershed, sand and gravel extraction facilities do not add selenium as a result of their operations. Wastewater discharged from sand and gravel operations typically results from the pumping of groundwater that accumulates in the pits and from the washing of rock/gravel. The selenium concentration in the groundwater is influenced by local geology and upgradient land use. The selenium load contributed by sand and gravel facilities is a function of selenium concentration and the volume of groundwater collected in the pits. Wastewater treatment typically consists of retention of the wastewater in settling ponds prior to discharge. Little, if any, selenium removal is accomplished. Data available with which to characterize selenium loads associated with discharging sand and gravel operations in the Gunnison River Basin is extremely limited. The Division conducted discharge sampling at nine sand and gravel operations in the Gunnison and Lower Colorado River basins in June of 2003. Two of the operations sampled discharge to segment 3 of the North Fork of the Gunnison. The 4D Pit, CDPS permit number COG500400 is the further upstream of the two, located north of the Town of Paonia. The Campbell Pit, CDPS permit number COG500397, is located some three miles downstream. Two other permittees, the Oldcastle Delta Sand and Gravel Pit and the Grand Junction Pipe and Supply Delta Paving Gravel Pit are located east of the Town of Delta, on Lower Gunnison River segment 1 above the confluence with the Uncompahgre River. Selenium discharge concentrations for these four facilities are shown in Table 66. Waterbody Selenium, Identification Permittee CDPS No. g/L COGUNF03 Oldcastle SW Group, Inc., Campbell Gravel Pit COG500397 4.3 COGUNF03 Oldcastle SW Group, Inc., 4D Gravel Pit COG500400 8.3 COGULG01 Oldcastle SW Group, DS&G Pit COG500136 15.0-16.0 COGULG01 Grand Junction Pipe & Supply, Delta Paving Gravel Pit COG500444 12.0-42.0 Table 66. Analytical Results for Selenium Sampling of Sand and Gravel Production Discharges - 2003
The remaining six facilities sampled in 2003 discharged either to the Colorado River mainstem below the confluence with the Gunnison River, or to tributaries to the Colorado River located downstream of the confluence. Selenium concentrations in these discharges ranged from a high value of 54.0 g/L for a discharging facility in the vicinity of Grand Junction to 2.6 g/L for a facility located downstream in the Town of Fruita.
January 2011
58
Final
Subsequent to the 2003 sampling exercise, Grand Junction Pipe and Supply provided the Division with analytical results for sampling performed at three sand and gravel extraction facilities which they operate. One of these, the Delta Paving Pit (CDPS permit number COG500444) discharges to segment COGULG01 near the City of Delta upstream of the Gunnison-Uncompahgre confluence. Analytical results for 11 samples collected between August 2004 and May 2006 were reported. Effluent selenium concentrations ranged from 12 g/L on August 30, 2004 to 42 g/L on May 5, 2006. Waterbody Selenium, Identification Permittee CDPS No. g/L COGULG01 Oldcastle SW Group, Inc., Anderson Pit COG500464 19.0-48.0 COGUNF03 Oldcastle SW Group, Inc., Tri-County Pit COG500498 9.7-44.0 Table 67. Analytical Results for Selenium Self Monitoring of Sand and Gravel Production Discharges – 2008/9
In 2008, quarterly monitoring for selenium was inserted into re-issued sand and gravel certifications as a monitoring requirement. Quarterly results for several of the affected permittees from September 2008 through June 2009 are included in Table 67. Daily maximum and 30-day average values were reported. Effluent selenium concentrations range from 9.7 g/L to 48 g/L as eighty-fifth percentile concentrations. That data which is available suggests selenium discharge levels associated with sand and gravel extraction operations likely vary within the Gunnison River Basin dependent upon localized geology (i.e. depth to the underlying Mancos shale) and groundwater hydrology. The variability reflected in the discharge data for the Delta Paving Gravel Pit also suggests that seasonal changes influence selenium concentrations in groundwater. Permitted discharges, including sand and gravel operations as well as municipal wastewater treatment plants represent a significant, previously unquantified group of sources. Selenium loads were calculated for domestic wastewater treatment facilities including the Delta and Paonia facilities (Tables 62 and 63). Estimates of selenium loads associated with permitted sand and gravel discharges (Table 68) are based upon the maximum discharge rates identified in the General Permit Certification and eighty-fifth percentile selenium discharge concentrations where available. In cases where selenium discharge data were absent, eighty-fifth percentile discharge concentrations were calculated for the given segment (which assumes similar discharge concentrations for sand and gravel facilities on a given segment). The eighty-fifth percentile selenium concentration for the Anderson Pit is 43.3 g/L. The eighty-fifth percentile selenium concentration for the Grand Junction Pipe & Supply, Delta Paving Gravel Pit is 34.0 g/L. This latter value is fairly consistent with the limited data available for discharges located further downstream, below the Gunnison River Basin but underlain by similar geologic formations. Estimates of selenium loads associated with sand and gravel discharges for the remainder of the affected portion of COGULG01 were calculated based upon a 38.7 g/L effluent concentration (i.e. average of eighty-fifth percentile concentrations from Anderson and Delta Paving Pits).
January 2011
59
Final
Waterbody Identification COGUNF03 COGUNF03 COGUNF03 COGUNF03 COGUNF03
Permittee Tri-County Gravel Diamond Lazy L. Ranch Ent., Janet Pit Oldcastle SW Group, Inc., Campbell Gravel Pit Oldcastle SW Group, Inc., 4D Gravel Pit Oldcastle SW Group, Inc. , Tri-County Pit
design capacity, cfs1 0.5 4.5 3.1 1.3 1.5
selenium concentration g/L 14.2 12.7 4.3 8.3 24.0
daily load lbs. 0.04 0.48 0.07 0.06 0.20
COGUNF03 total COGUUN04b
Western Gravel Concrete Facility – North R-34 Pit
0.85 3.9
12.7
0.27
COGUUN04b total COGULG04a COGULG04a
Elam Construction, Dubs Pit Western Gravel, Fredlund Pit
0.27 2.23 4.64
67.0 63.0
0.81 1.58
COGULG04a total COGULG01 COGULG01 COGULG01 COGULG01 COGULG01
Elam Construction, Bennett Gravel Pit Grand Junction Pipe & Supply, Delta Paving Gravel Pit Lafarge West, North Delta Pit Oldcastle SW Group, Inc., Anderson Pit Western Gravel North D-22 Pit/Concrete Plant #2
2.39 1.01 0.90 0.89 6.2 1.4
38.7 34.0 38.7 43.3 38.7
0.21 0.16 0.19 1.45 0.13
COGULG01 total COGULG02 COGULG02
Elam Construction, Mule Farm Gravel Pit Whitewater Building Materials
2.14 1.9 4.0
24.4 24.4
0.24 0.53
COGULG02 total
0.87
1.
Cubic feet per second Table 68. Estimated Selenium Loads for Sand and Gravel Discharges
The calculated selenium loads in Table 68 potentially overestimate the actual selenium load as the facilities discharge at their permitted maximum volume infrequently, if at all. Within the North Fork of the Gunnison, segment COGUNF03, these discharges potentially contribute a significant percentage of the total load. The Butler and Leib (2002) report estimated a mean annual selenium load for the North Fork Gunnison River of 1,300 and 1,400 pounds per year for 1999 and 2000, respectively. The cumulative annual selenium load for the permitted discharges may potentially contribute some 460 pounds or about 32 percent of the total load. All CDPS permitted discharges are considered to be “point sources” and are addressed in the “Waste Load Allocation” term of the TMDL.
January 2011
60
Final
Non-Point Source Contributions An important pathway by which selenium is introduced into surface water within the Gunnison Basin is associated with irrigation projects constructed within the basin. Seepage from irrigation canals, particularly when unlined, and deep percolation of irrigation waters dissolve selenium from Mancos shale derived soils as well as from the underlying parent material. As illustrated in Figure 13, extensive portions of the basin, particularly in the lowlands, are serviced by federal irrigation projects. Other pathways include surface runoff of irrigation water (tailwater), whether associated with agriculture or urban land uses. Irrigation of residential, commercial and recreational facilities all result in some degree of surface runoff and/or deep percolation. Sixty percent or more of the selenium loading in the Gunnison basin (as measured at Whitewater) originates from an area encompassing the Uncompahgre River basin and the service area of the Federally-constructed Uncompahgre Project (see Figure 4, Page 17). This figure includes 40 percent from the Uncompahgre River basin and 17 percent or more from portions of the Uncompahgre Project service area in the vicinity of Delta. Irrigation water delivery systems and on-farm applications to the Mancos Shale and soils derived from the shale contribute a vast majority (about 90 percent) of the ground-water that mobilizes selenium in this area. (BOR 2006). These are considered “non-point sources” and are included in the Load Allocation term of the TMDL.
VIII. TOTAL MAXIMUM DAILY LOADS TMDLs are required in instances where waterbodies fail to support classified uses and/or attain assigned numeric water quality standards. The TMDL calculates the requisite pollutant loading reduction to be achieved in order for the waterbody to attain those uses and standards. The loading reductions are apportioned among Margin of Safety (MOS), Wasteload Allocation (WLA), and Load Allocation (LA) terms. The WLA term represents pollutant contributions arising from permitted and non-permitted point source discharges. The LA term is comprised of non-point source and/or background contributions. The TMDL may be expressed as the sum of the LA, WLA and MOS. TMDL = WLA + LA + MOS TMDL = Sum of Waste Load Allocations + Sum of Load Allocations + Margin of Safety Waste Load Allocations “(WLA)” The selenium loads to the Gunnison, North Fork Gunnison, and Uncompahgre Rivers are derived from natural geology associated with the Mancos shale. The selenium loads to the different watersheds vary based upon the underlying geology and proximity to the seleniferous shales, as do the ambient concentrations of selenium. Selenium enters sanitary sewage collection systems and sand and gravel excavations via groundwater and is present in wastewater discharges in varying amounts.
January 2011
61
Final
There are six permitted domestic treatment facilities that discharge to surface water within the watershed and sixteen permitted sand and gravel discharges (Tables 3, 4 and 5). Modeled discharges for domestic wastewater facilities and sand and gravel discharges were given waste load allocations (WLAs). Waste load allocations for permitted dischargers were calculated as Water Quality Based Effluent Limits (WQBELs) according to Regulation 61, the Colorado Discharge Permit Regulations. The WQBEL is a calculation of what an effluent limitation may be in a permit. The WQBELs for any given parameter is compared to other potential limitations (Federal Effluent Limitations Guidelines, State Effluent Limitations, or other applicable limitation). Typically, the more stringent limit is incorporated into a permit. If the WQBEL is the more stringent limitation, incorporation into a permit is dependent upon a reasonable potential analysis (i.e. the potential for a constituent in the discharge to be present at a level such that it might cause or contribute to an instream exceedance of a water quality standard). In-stream background data and chronic low flows were used to determine the assimilative capacities of the Uncompahgre River, North Fork Gunnison River, and Gunnison River for pollutants of concern, and to calculate the WQBELs. It is the Division‟s approach to calculate the WQBELs using the lowest of the monthly low flows (referred to as the annual low flow) as determined in the low flow analysis. The Division‟s standard analysis consists of steady-state, mass-balance calculations for the pollutants considered in this TMDL. The mass-balance equation is used by the Division to calculate the WQBELs, and accounts for the upstream concentration of a pollutant at the existing quality, critical low flow (minimal dilution), effluent flow and the water quality standard. The mass-balance equation is expressed as:
M2
M 3Q3 M 1Q1 Q2
Where, Q1 = Upstream low flow (1E3 or 30E3) Q2 = Average daily effluent flow (design capacity) Q3 = Downstream flow (Q1 + Q2) M1 = In-stream background pollutant concentrations at the existing quality M2 = Calculated WQBEL M3 = Water Quality Standard, or other maximum allowable pollutant concentration The upstream background pollutant concentrations used in the mass-balance equation will vary based on the regulatory definition of existing ambient water quality. For the dissolved pollutants in this TMDL, existing quality is determined to be the 85th percentile. Where upstream concentrations exceed stream standards, no assimilative capacity is available and WQBELs are set to stream standards.
January 2011
62
Final
The calculations of waste load allocations are based on the protocol for setting effluent limits outlined in The Colorado Discharge Permit System Regulation, 5 CCR 1002-61. Waste load allocations were calculated for the current underlying standard of 4.6 g/L selenium. Load Allocations “(LA)” Remaining background sources that were examined are considered non-point sources and are therefore accountable to load allocations. An important pathway by which selenium is introduced into surface water within the Gunnison Basin is associated with irrigation projects constructed within the basin. Seepage from irrigation canals, particularly when unlined, and deep percolation of irrigation waters dissolve selenium from Mancos shale derived soils as well as from the underlying parent material. Extensive portions of the basin, particularly in the lowlands, are serviced by federal irrigation projects. Other pathways include surface runoff of irrigation water (tailwater), whether associated with agriculture or urban land uses. Irrigation of residential, commercial and recreational facilities all result in some degree of surface runoff and/or deep percolation. Margin of Safety “(MOS)” According to the Federal Clean Water Act, TMDLs require a margin of safety (MOS) component that accounts for the uncertainty about the relationship between the pollutant loads and the receiving waterbody. The MOS can be implicit or explicit. A 10% explicit margin of safety was included in this TMDL. This MOS is included to account for the uncertainty in the analysis of the relationship between the TMDL loading allocations and the desired water quality target. For those segments where there are point source discharges, and therefore the TMDL calculation includes a WLA term, the MOS is reflected in the WLA Reserve Capacity column. TMDLs for those segments that do not receive a point source discharge do not include a WLA term. The allowable load value (the LA) identified in Tables 69 through 102 already incorporates an explicit 10% margin of safety. Load and Waste Load Allocation Options The TMDL analyses incorporated into this document involve both point and non-point source contributions of selenium. The sum of these contributions is such that numeric selenium standards are exceeded. For some listed segments addressed in this document, only non-point sources are identified. For those segments, the allocation need only reflect the MOS and LA terms. Other listed segments must also include an allocation for point source contributions, the WLA term. Allocations to point sources may possibly vary due to the volume of a given discharge as well as the pollutant concentration. Were point source contributions an unspecified but minimal portion of the overall load, it might be possible to accommodate the current contributions from those point sources by achieving substantial reduction of the non-point source load. In this way an offset might be created, either on a site-specific (i.e. at the discharge point) or even on a watershed-wide basis. This would require that the selenium concentrations above a discharge point be reduced to a level below the standard. The difference between the instream level and the standard would represent assimilative capacity and the effluent limit for the facility
January 2011
63
Final
could be set at a level greater than the standard. This approach is consistent with federal regulation as stated at 40 CFR 122.44(d): 122.44(d)(1)(i) Limitations must control all pollutants or pollutant parameters (either conventional, nonconventional, or toxic pollutants) which the Director determines are or may be discharged at a level which will cause, have the reasonable potential to cause, or contribute to an excursion above any State water quality standard, including State narrative criteria for water quality. (ii) When determining whether a discharge causes, has the reasonable potential to cause, or contributes to an in-stream excursion above a narrative or numeric criteria within a State water quality standard, the permitting authority shall use procedures which account for existing controls on point and nonpoint sources of pollution, the variability of the pollutant or pollutant parameter in the effluent, the sensitivity of the species to toxicity testing (when evaluating whole effluent toxicity), and where appropriate, the dilution of the effluent in the receiving water. (iii) When the permitting authority determines, using the procedures in paragraph (d)(1)(ii) of this section, that a discharge causes, has the reasonable potential to cause, or contributes to an in-stream excursion above the allowable ambient concentration of a State numeric criteria within a State water quality standard for an individual pollutant, the permit must contain effluent limits for that pollutant. Corresponding prohibitions are included in the Colorado Discharge Permit System Regulation, 5 CCR 1002-61, at section 61.8 (1). Per both federal and state regulations, existing point source discharges might conceivably continue to discharge at currently authorized (via CDPS permit) volumes and concentrations if existing controls on point and nonpoint sources of pollution, and the dilution of the effluent in the receiving water can accommodate the discharge without causing or contributing to an exceedance of numeric water quality standards. Characterization of the domestic wastewater treatment facilities within the watershed indicates that several currently discharge at a concentration that is below the chronic selenium standard of 4.6 g/L, or discharge to upstream portions of the watershed at locations where some degree of assimilative capacity has been documented. The Town of Olathe facility‟s selenium discharge concentration currently exceeds the chronic standard. This is apparently due to excessive seepage of selenium-bearing groundwater into the Town‟s sanitary sewer system. The Town‟s current CDPS permit incorporates a compliance schedule intended to remedy this problem. It is anticipated that upon completion of the compliance schedule, the quality of Olathe‟s discharge will be similar to that of the other municipal facilities in the watershed. Discharges from sand and gravel facilities present a somewhat similar problem. Sand and gravel pits intercept groundwater migrating from upland areas towards the various watercourses. The selenium concentration in the groundwater is influenced by local geology and upgradient land use. The selenium load contributed by sand and gravel facilities is a function of selenium concentration and the volume of groundwater collected in the pits. Unlike domestic wastewater treatment plants, however, wastewater treatment typically consists of retention of the pumped water in settling ponds prior to discharge. The discharge concentration of dissolved selenium is relatively unaffected by this form of treatment.
January 2011
64
Final
As noted above, sand and gravel discharges are, potentially, a significant source of selenium. Upstream selenium concentrations are typically at levels such that there is no assimilative capacity available to allow discharge concentrations in excess of the 4.6 g/L standard. The Colorado Discharge Permit System Regulation, 5 CCR 1002-61, at section 61.8 (1)(e) precludes issuance of a discharge permit in this situation subject to certain requirements: 61.8(1)(e) Subject to the provisions of subsection 31.14(15)(b), no permit shall be issued which allows a discharge that by itself or in combination with other pollution will result in pollution of the receiving waters in excess of the pollution permitted by an applicable water quality standard or applicable antidegradation requirement unless the permit contains effluent limitations and a schedule of compliance specifying treatment requirements or the Division has granted a variance from the water quality standard. The current general permit for sand and gravel operations includes a compliance schedule requirement that effluent selenium monitoring be implemented in order to allow assessment of localized geologic and land use influences on discharge quality. Discharges to that portion of the Gunnison River mainstem (segment COGULG02) and any tributaries that are within the 100 year floodplain are subject to additional permitting restrictions. This is because that portion of the watershed is identified by the U. S. Fish and Wildlife Service (USFWS) as habitat for a number of threatened and endangered (T&E) aquatic species. The Division has entered into a Memorandum of Agreement (MOA) with the USFWS and EPA which identifies specific restrictions that the Division has agreed to incorporate into CDPS permits issued for discharges to these reaches. The MOA identifies three possible options for discharge to T&E habitat: Permit limits may be based on mixing that will result from the installation of a diffuser covering part of the low flow channel while allowing for an appropriate zone of passage. Permit limits may be based on relocation of the discharge to a waterbody or waterbody segment where there is neither occurrence of listed aquatic species nor designated critical habitat. Permit limits may be based on denial of a regulatory mixing zone with the requirement that water quality standards will be met at the point of discharge. There are two sand and gravel operations identified in the TMDL assessment that discharge to segment COGULG02. These are subject to the provisions of the MOA. For purposes of the wasteload allocation, however, all of the sand and gravel operations discharging to COGULG02 and the lower portion of COGULG01 will receive an allocation based upon application of the 4.6 g/L standard as the effluent limit. This approach is consistent with that required by the MOA. The TMDL and existing stream loads were calculated using a monthly median flow multiplied by the existing stream standard or ambient eighty-fifth percentile concentration and a conversion factor (0.0054) to approximate a load in pounds/day.
January 2011
65
Final
Acute and chronic low flows were calculated using USEPA DFLOW software. Acute (1E3) and chronic (30E3) flows are biologically based low flows. Biologically-based design flows are intended to measure the actual occurrence of low flow events with respect to both the duration and frequency (i.e., the number of days aquatic life is subjected to flows below a certain level within a period of several years). Although the extreme value analytical techniques used to calculate hydrologically-based design flows have been used extensively in the field of hydrology and in state water quality standards, these methods do not capture the cumulative nature of effects of low flow events because they only consider the most extreme low flow in any given year. By considering all low flow events within a year, the biologically-based design flow method accounts for the cumulative nature of the biological effects related to low flow events. Acute low flows (1E3) refer to single low flow events that occur once in a three year period. Chronic low flows (30E3) refer to 30-day low flow periods which occur once in three years. Chronic low flows are used in the calculation of the WQBELs and permitted waste load allocations. The TMDL was calculated using median monthly stream flow from the nearest gage where available. For those segments without gaged streamflow, instantaneous streamflow measurements were used to generate annual medians. Where streamflow represented only historic data, regression analyses were used to predict current stream flow. Selenium loads from dischargers were calculated using the design capacity for flow, and the effluent limits listed in their current permit. Where no limits were given, the limit was calculated based on the protocol for setting effluent limits outlined in Regulation 61. Modeled discharges were then given waste load allocations (WLA). Waste load allocations for sand and gravel discharges were calculated using the design capacity for flow where available, and the existing stream standard of 4.6 micrograms per liter, or calculated WQBEL, where applicable. Load allocations (LA) were then calculated by subtracting the total of the WLAs from the TMDL. Where the ambient stream load was higher than the TMDL a load reduction was calculated. Attainment of the Acute Aquatic Life Use-based Selenium Standard As discussed, existing or ambient water quality is determined using the 85th percentile value of the ranked data. The necessary pollutant loading reduction necessary to attain the chronic standard is calculated by subtracting the allowable load, based upon the 4.6 g/L standard, from the ambient load based upon the 85th percentile value. The loading reduction this calculation yields is that necessary to assure the chronic standard is attained. In order to determine whether the chronic standard-based loading reduction is adequate to protect the acute standard of 18.4 g/L. the maximum selenium value in any segment-specific dataset is multiplied by the chronic standard-based loading reduction. If the resultant value is less than the 18.4 g/L value, it is assumed the loading reduction is adequate to assure attainment of the acute standard. If not, the load reduction must be increased as necessary to attain the acute standard. Because attainment of the acute standard is evaluated on a sample specific-basis while attainment of the chronic standard is assessed against the 85th percentile value, the maximum calculated selenium value may exceed the 4.6 g/L chronic standard and remain protective of the chronic standard.
January 2011
66
Final
TMDL for Segment COGULG01, Gunnison River above the Uncompahgre River Background levels of selenium, above the confluence of the Gunnison with the North Fork Gunnison River, were in attainment of stream standards. Selenium reductions for Lower Gunnison Segment 1, below the confluence with the North Fork, are greatest from July through November, with required reductions ranging from 25 percent in August and October to 45 percent in July (Table 69). In addition, a 33 percent reduction is required in April. On average, 17 percent of the total selenium load needs to be reduced in order to attain chronic selenium standards. However, the months of January-March, May, and December are currently attaining their allowable load. However, since four of the five sand and gravel facilities are located more than one mile upstream of the beginning of COGULG02, which has been characterized as threatened and endangered species habitat and is thus subject to the Memorandum of Agreement (MOA) with the U.S. Fish and Wildlife Service (USFWS) and USEPA, ambient upstream concentrations (with available assimilative capacity) were used to establish waste load allocations. The fifth permitted sand and gravel facility has not discharged since 2000. Because of the continued growth in and around Delta and Montrose Counties, a reserve capacity was included in the TMDL in order to reserve allocations for future permitted discharges to this segment. Estimated population growth for Delta County is projected to increase by approximately ten percent by the year 2015; therefore ten percent of the available TMDL was set aside for future growth. The percent reduction reflects reductions required from nonpoint sources. TMDL for COGULG01 at Median Flows Median flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
WLA, lbs/day
WLA Reserve Capacity
LA, lbs/day
85th% Conc.
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
914
4.6
22.7
2.3
7.3
2.0
11.1
3.0
14.8
-5.6
0%
Feb*
944
4.6
21.1
2.1
7.3
1.9
9.8
3.5
17.8
-1.2
0%
Mar
1110
4.6
24.8
2.5
7.3
2.2
12.8
4.0
23.8
1.4
0%
Apr
1875
4.6
41.9
4.2
8.0
3.8
25.9
5.6
56.2
18.5
33%
May
3000
4.6
67.1
6.7
9.3
6.0
45.0
1.9
31.2
-29.2
0%
Jun
1985
4.6
44.4
4.4
9.3
4.0
26.7
4.6
49.3
9.4
0%
Jul
980
4.6
21.9
2.2
9.6
2.0
8.1
6.8
36.0
16.3
45%
Aug
1020
4.6
22.8
2.3
9.4
2.1
9.0
5.0
27.5
7.0
25%
Sep
1050
4.6
23.5
2.3
8.7
2.1
10.3
6.2
35.4
14.3
40%
Oct
1140
4.6
25.5
2.5
7.5
2.3
13.1
5.0
30.8
7.8
25%
Nov
1090
4.6
24.4
2.4
7.3
2.2
12.5
6.3
36.8
14.9
40%
Dec 1070 4.6 23.9 2.4 7.3 2.2 12.1 4.0 23.1 1.6 *85th concentrations were estimated from adjacent months Table 69. Total Maximum Daily Load for Segment COGULG01, Gunnison River above the Uncompahgre River. Period of record 1987-2002.
January 2011
67
Final
0%
Annual loading reductions calculated by USGS for the Gunnison River at Delta ranged from 11 to 51 percent dependent upon the water year (see Table 11). These results are not inconsistent with the monthly reductions identified in Table 69 given the differences in assessment methodologies utilized. Existing loads in the mainstem Gunnison River at Delta total approximately 11,630 pounds annually. An annual load reduction of 2,436 pounds is required in order for Lower Gunnison Segment 1 to be in attainment of its annual TMDL. No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for this location. Therefore no loading reduction is required to attain the acute standard. TMDL for Segment COGULG02, Gunnison River from the Uncompahgre River to the Colorado River Selenium reductions for Lower Gunnison Segment 2 are greatest in December and February, with required reductions of 69 percent and 71 percent, respectively (Table 70). Reductions in the remaining months range from 28 percent in April to 57 percent in October. No reduction in selenium load is required in May. TMDL for COGULG02 at Median Flows
Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
WLA (Delta), lbs/day
WLA (Sand & Gravel), lbs/day
Jan
1230
4.6
30.6
3.1
0.094
0.146
2.7
24.5
6.3
42.0
14.5
35%
Feb*
1210
4.6
30.1
3.0
0.094
0.146
2.7
24.1
14.4
94.2
67.2
71%
Mar
1300
4.6
32.3
3.2
0.094
0.146
2.9
25.9
8.3
58.5
29.5
50%
Apr
2400
4.6
59.6
6.0
0.094
0.146
5.4
48.0
5.7
74.2
20.5
28%
May
3640
4.6
90.4
9.0
0.094
0.146
8.1
73.0
3.5
68.0
-13.4
0%
Jun
2230
4.6
55.4
5.5
0.094
0.146
5.0
44.6
6.2
75.0
25.1
33%
Jul
1450
4.6
36.0
3.6
0.094
0.146
3.2
28.9
6.3
49.0
16.6
34%
Aug
1550
4.6
38.5
3.9
0.094
0.146
3.5
30.9
7.1
59.8
25.1
42%
Sep
1780
4.6
44.2
4.4
0.094
0.146
4.0
35.6
7.7
73.6
33.8
46%
Oct
1880
4.6
46.7
4.7
0.094
0.146
4.2
37.6
9.7
98.3
56.2
57%
Nov
1445
4.6
35.9
3.6
0.094
0.146
3.2
28.8
7.9
61.6
29.3
48%
Dec
WLA Reserve Capacity
LA, lbs/day
85th% Conc.
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
1280 4.6 31.8 3.2 0.094 0.146 2.9 25.5 13.4 92.3 63.7 Table 70. Total Maximum Daily Load for Segment COGULG02, Gunnison River from the Uncompahgre River to the Colorado River. Period of record 2000-2006.
Currently, there is one individually permitted discharger to Lower Gunnison Segment 2, the City of Delta. Due to exceedances of ambient water quality at the beginning of Segment 2 and the presence of threatened and endangered species, the City of Delta was given the chronic selenium standard of 4.6 micrograms per liter as their chronic effluent limit. Similarly, the allowable waste load allocation for current sand and gravel discharges was also set to 4.6 micrograms per liter. Reductions in current discharge concentrations of 88 percent from sand and gravel operations are required in order to attain stream standards.
January 2011
68
Final
69%
Because of the continued growth in and around Delta and Montrose Counties, a reserve capacity was included in the TMDL in order to reserve allocations for future permitted discharges in this segment. Estimated population growth for Delta County is projected to increase by approximately ten percent by the year 2015; therefore ten percent of the available TMDL was set aside for future growth. Load reductions in Table 70 reflect reductions already taken from sand and gravel facilities. The percent reduction remaining reflects reductions required from nonpoint sources. Annual loading reductions calculated by USGS for the Gunnison River at Grand Junction ranged from 37 to 61 percent dependent upon the water year used (see Table 21). These results are not inconsistent with the monthly reductions identified in Table 70 given the differences in assessment methodologies utilized. However, due to the inherent differences in methodologies (i.e. 85th% -based annual loads versus mean annual loads), estimated ambient loads are almost 1.5 times that of reported USGS loads. TMDL estimates are inherently conservative in magnitude in order to support existing uses. Existing loads in the mainstem Gunnison River at Grand Junction total 25,701 pounds annually. An annual load reduction of 11,538 pounds is required from non-point sources and 390 pounds from sand and gravel facilities in order for Lower Gunnison Segment 2 to be in attainment of its annual TMDL. No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for this location (most recent ten years). Therefore no loading reduction is required to attain the acute standard. TMDLs for Segment COGULG04a, Tributaries to the Gunnison River Currant Creek Selenium TMDLs for sixteen tributaries included in Lower Gunnison Segment 4a are provided below (Tables 71-86). Instantaneous discharge measurements were used to generate median flows. Where gaged flows were present, monthly TMDLs were calculated. An annual reduction of 88 percent is required for Currant Creek to be in attainment of its selenium TMDL. A reduction of 0.67 pounds per day (246 pounds annually) would result in attainment of the chronic selenium standard in Currant Creek. Currently, there are no point source discharges to Currant Creek; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
Annual 4.3 4.6 0.11 0.011 0.10 33.2 0.77 0.67 Table 71. Total Maximum Daily Load for Segment COGULG04a, Currant Creek. Period of record 19912000.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Currant Creek is 44 g/L (March 14, 2000). An 88 percent loading reduction
January 2011
69
Final
% Reduction 88%
would result in an instream concentration of 5.3 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. The 85th percentile value of 33.2 g/L for Currant Creek also exceeds the Agriculture Use-based standard of 20 g/L. The dataset for which the 33.2 g/L value is calculated is reported as dissolved selenium whereas the Agriculture Use-based selenium standard is expressed as the total recoverable species. Therefore, direct comparison of any available analytical results against the Agriculture Use-based standard is not possible. Were an 88 percent loading reduction achieved, however, it is reasonable to assume that the 20 g/L Agriculture Use-based standard would not be exceeded. Callow Creek An annual reduction of 73 percent is required for Callow Creek to be in attainment of its selenium TMDL. A reduction of 0.0024 pounds per day (0.9 pounds annually) would result in attainment of the chronic selenium standard in Callow Creek. Currently, there are no point source discharges to Callow Creek; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 0.04 4.6 0.001 0.0001 0.001 15.3 0.003 0.0024 Table 72. Total Maximum Daily Load for Segment COGULG04a, Callow Creek. Period of record 19992002.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Callow Creek is 48.5 g/L (February 14, 2001). A 73 percent loading reduction would result in an instream concentration of 13.1 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. Alkali Creek An annual reduction of 96 percent is required for Alkali Creek to be in attainment of its selenium TMDL. A reduction of 0.064 pounds per day (23.4 pounds annually) would result in attainment of the chronic selenium standard in Alkali Creek. Currently, there are no point source discharges to Alkali Creek; therefore load reductions would come from reductions in non-point source loads.
January 2011
70
Final
73%
Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 0.11 4.6 0.0026 0.0003 0.0023 117.0 0.0663 0.064 96% Table 73. Total Maximum Daily Load for Segment COGULG04a, Alkali Creek. Period of record 19952000.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Alkali Creek is 150.0 g/L (January 19, 2000). A 96 percent loading reduction would result in an instream concentration of 4.2 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. The 85th percentile value of 117.0 g/L for Alkali Creek also exceeds both the Agriculture Use and the Water Supply Use-based standards of 20 g/L and 50 g/L respectively. The dataset for which the 117.0 g/L value is calculated is reported as dissolved selenium whereas the Agriculture Use and Water Supply Use-based selenium standards are expressed as the total recoverable species. Therefore, direct comparisons of any available analytical results against the Agriculture and Water Supply Use-based standards are not possible. Were a 96 percent loading reduction achieved, however, it is reasonable to assume that the Agriculture and Water Supply Use-based selenium standards would not be exceeded. Dry Gulch An annual reduction of 100 percent is required for Dry Gulch to be in attainment of its selenium TMDL. A reduction of 5.6 pounds per day (2057.8 pounds annually) would result in attainment of the chronic selenium standard in Dry Gulch. Currently, there are no point source discharges to Dry Gulch; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
Annual 0.27 4.6 0.007 0.0007 0.006 3941.2 5.640 5.634 Table 74. Total Maximum Daily Load for Segment COGULG04a, Dry Gulch. Period of record 2002.
The annual loading reduction calculated by USGS for Dry Gulch is 86 percent or 1,070 pounds of selenium per year (see Table 17). These results are not inconsistent with the annual reduction (2,578 pounds) identified in Table 74 given the differences in assessment methodologies utilized.
January 2011
71
Final
% Reduction 100%
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Dry Gulch is 4580 g/L (March 2, 2002). A 99.9 percent loading reduction would result in an instream concentration of 4.58 g/L. Therefore the loading reduction necessary to attain the acute standard in all instances (based upon the existing dataset) would be calculated in excess of 99.9 percent. Thus a loading reduction of virtually 100 percent is required in order to assure that the acute standard of 18.4 g/L is not exceeded. The 85th percentile value of 3941.2 g/L for Dry Gulch also exceeds both the Agriculture Use and the Water Supply Use-based standards of 20 g/L and 50 g/L respectively. The dataset for which the 3941.2 g/L value is calculated is reported as dissolved selenium whereas the Agriculture Use and Water Supply Use-based selenium standards are expressed as the total recoverable species. Therefore, direct comparisons of any available analytical results against the Agriculture and Water Supple Use-based standards are not possible. Were a 99.9 percent loading reduction achieved, however, it is reasonable to assume that the Agriculture and Water Supply Use-based selenium standards would not be exceeded. Peach Valley Arroyo An annual reduction of 85 percent is required for the Peach Valley Arroyo to be in attainment of its selenium TMDL. A reduction of 0.6 pounds per day (229 pounds annually) would result in attainment of the chronic selenium standard in the Peach Valley Arroyo. Currently, there are no point source discharges to the Peach Valley Arroyo; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 4.90 4.6 0.122 0.012 0.1095 27.8 0.736 0.626 Table 75. Total Maximum Daily Load for Segment COGULG04a, Peach Valley Arroyo. Period of record 1991-2000.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Peach Valley Arroyo is 95.0 g/L (November 9, 1995). An 85 percent loading reduction would result in an instream concentration of 14.3 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded.
January 2011
72
Final
85%
Alfalfa Run An annual reduction of 76 percent is required for Alfalfa Run to be in attainment of its selenium TMDL. A reduction of 0.3 pounds per day (123 pounds annually) would result in attainment of the chronic selenium standard in Alfalfa Run. Currently, there are no point source discharges to Alfalfa Run; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 4.80 4.6 0.119 0.012 0.107 17.1 0.443 0.336 76% Table 76. Total Maximum Daily Load for Segment COGULG04a, Alfalfa Run. Period of record 1991-2000.
No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Alfalfa Run. Therefore no loading reduction is required to attain the acute standard. Sulphur Gulch An annual reduction of 74 percent is required for Sulphur Gulch to be in attainment of its selenium TMDL. A reduction of 0.006 pounds per day (2.3 pounds annually) would result in attainment of the chronic selenium standard in Sulphur Gulch. Currently, there are no point source discharges to Sulphur Gulch; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
Annual 0.10 4.6 0.002 0.0002 0.002 15.8 0.009 0.006 Table 77. Total Maximum Daily Load for Segment COGULG04a, Sulphur Gulch. Period of record 19992003.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Sulphur Gulch is 21.0 g/L (March 14, 2000). A 74 percent loading reduction would result in an instream concentration of 5.5 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. Lawhead Gulch An annual reduction of 45 percent is required for Lawhead Gulch to be in attainment of its selenium TMDL. A reduction of 0.003 pounds per day (1.1 pounds annually) would result in attainment of the chronic selenium standard in Lawhead Gulch. Currently, there are no point source discharges to Lawhead Gulch; therefore load reductions would come from reductions in non-point source loads.
January 2011
73
Final
% Reduction 74%
Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 0.17 4.6 0.004 0.0004 0.004 7.6 0.007 0.003 Table 78. Total Maximum Daily Load for Segment COGULG04a, Lawhead Gulch. Period of record 19992000.
45%
No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Lawhead Gulch. Therefore no loading reduction is required to attain the acute standard. Wells Gulch An annual reduction of 51 percent is required for Wells Gulch to be in attainment of its selenium TMDL. A reduction of 0.0005 pounds per day (0.2 pounds annually) would result in attainment of the chronic selenium standard in Wells Gulch. Currently, there are no point source discharges to Wells Gulch; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 0.02 4.6 0.0005 0.00005 0.0004 8.5 0.001 0.0005 51% Table 79. Total Maximum Daily Load for Segment COGULG04a, Wells Gulch Period of record 1995-2000.
No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Wells Gulch. Therefore no loading reduction is required to attain the acute standard. Negro Creek An annual reduction of 54 percent is required for Negro Creek to be in attainment of its selenium TMDL. A reduction of 0.002 pounds per day (0.7 pounds annually) would result in attainment of the chronic selenium standard in Negro Creek. Currently, there are no point source discharges to Negro Creek; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
Annual 0.07 4.6 0.002 0.0002 0.002 9.0 0.003 0.001 Table 80. Total Maximum Daily Load for Segment COGULG04a, Negro Creek. Period of record 2000.
No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Negro Creek. Therefore no loading reduction is required to attain the acute standard.
January 2011
74
Final
% Reduction 54%
Deer Creek An annual reduction of 58 percent is required for Deer Creek to be in attainment of its selenium TMDL. A reduction of 0.002 pounds per day (0.6 pounds annually) would result in attainment of the chronic selenium standard in Deer Creek. Currently, there are no point source discharges to Deer Creek; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 0.05 4.6 0.001 0.0001 0.0011 9.8 0.003 0.002 Table 81. Total Maximum Daily Load for Segment COGULG04a, Deer Creek. Period of record 2000.
58%
No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Deer Creek. Therefore no loading reduction is required to attain the acute standard. North Fork Kannah Creek An annual reduction of 95 percent is required for the North Fork of Kannah Creek to be in attainment of its selenium TMDL. A reduction of 0.06 pounds per day (21.2 pounds annually) would result in attainment of the chronic selenium standard in the North Fork Kannah Creek. Currently, there are no point source discharges to the North Fork Kannah Creek; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 0.15 4.6 0.004 0.0004 0.003 75.9 0.061 0.058 Table 82. Total Maximum Daily Load for Segment COGULG04a, North Fork Kannah Creek. Period of record 2003.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for the North Fork of Kannah Creek is 75.9 g/L (October 17, 2003). A 95 percent loading reduction would result in an instream concentration of 3.8 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. Kannah Creek An annual reduction of 68 percent is required for Kannah Creek to be in attainment of its selenium TMDL. A reduction of 0.13 pounds per day (48.3 pounds annually) would result in attainment of the chronic selenium standard in Kannah Creek. Currently, there are no point source discharges to Kannah Creek; therefore load reductions would come from reductions in non-point source loads.
January 2011
75
Final
95%
Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Annual 2.80 4.6 0.070 0.007 0.063 12.9 0.195 0.132 68% Table 83. Total Maximum Daily Load for Segment COGULG04a, Kannah Creek. Period of record 20042005.
No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for the upper reaches of Kannah Creek. Therefore no loading reduction is required to attain the acute standard. Whitewater Creek Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
1.60
4.6
0.040
0.004
0.036
30.8
0.266
0.230
87%
Feb
2.70
4.6
0.067
0.007
0.060
47.3
0.690
0.629
91%
Mar
3.20
4.6
0.079
0.008
0.072
45.2
0.782
0.710
91%
Apr
4.03
4.6
0.100
0.010
0.090
55.7
1.213
1.123
93%
May
5.00
4.6
0.124
0.012
0.112
24.1
0.650
0.539
83%
Jun
6.90
4.6
0.171
0.017
0.154
16.0
0.596
0.442
74%
Jul
1.20
4.6
0.030
0.003
0.027
40.4
0.262
0.235
90%
Aug
0.88
4.6
0.022
0.002
0.020
45.8
0.216
0.197
91%
Sep
1.30
4.6
0.032
0.003
0.029
62.4
0.438
0.409
93%
Oct
2.45
4.6
0.061
0.006
0.055
65.7
0.869
0.814
94%
Nov
3.60
4.6
0.089
0.009
0.080
69.0
1.341
1.260
94%
Dec 2.60 4.6 0.065 0.006 0.058 49.9 0.700 0.642 92% Table 84. Total Maximum Daily Load for Segment COGULG04a, Whitewater Creek. Period of record 1999-2002.
Selenium reductions for Whitewater Creek are greater than or equal to 90 percent for nine months of the year, July-December and February-April (Table 84). Reductions in the remaining months range from 74 percent in June to 87 percent in January. If monthly load reductions ranging from 0.2 to 1.3 pounds per day (219 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Whitewater Creek. Currently, there are no point source discharges to Whitewater Creek; therefore load reductions would come from reductions in non-point source loads. The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for the Whitewater Creek is 69.9 g/L (October 17, 2003). A 95 percent loading reduction would result in an instream concentration of 3.8 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded.
January 2011
76
Final
The maximum monthly 85th percentile value of 69.0 g/L for Whitewater Creek also exceeds both the Agriculture Use and the Water Supply Use-based standards of 20 g/L and 50 g/L respectively. The dataset for Whitewater Creek is reported as dissolved selenium whereas the Agriculture Use and the Water Supply Use-based selenium standards are expressed as the total recoverable species. Therefore, direct comparison of any available analytical results against the Agriculture Use and the Water Supply Use-based standards is not possible. A 94 percent loading reduction is necessary to attain the Aquatic Life Use-based standard during November, when the maximum monthly 85th percentile value was recorded. Were a 94 percent loading reduction achieved, however, it is reasonable to assume that the Agriculture and Water Supply Use-based selenium standards would not be exceeded. Cummings Gulch Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
5.8
4.6
0.144
0.014
0.130
15.7
0.492
0.362
74%
Feb
3.5
4.6
0.087
0.009
0.078
14.0
0.265
0.186
70%
Mar
2.9
4.6
0.072
0.007
0.065
12.6
0.197
0.132
67%
Apr
61.0
4.6
1.515
0.152
1.364
3.0
0.988
-0.376
0%
May
50.8
4.6
1.263
0.126
1.136
4.9
1.345
0.209
16%
Jun
40.7
4.6
1.010
0.101
0.909
6.8
1.493
0.584
39%
Jul
30.5
4.6
0.758
0.076
0.682
8.7
1.433
0.751
52%
Aug
84.0
4.6
2.087
0.209
1.878
7.0
3.175
1.297
41%
Sep
54.5
4.6
1.354
0.135
1.218
6.6
1.928
0.709
37%
Oct
30.3
4.6
0.751
0.075
0.676
7.3
1.188
0.512
43%
Nov
6.0
4.6
0.149
0.015
0.134
8.0
0.259
0.125
48%
Dec 6.5 4.6 0.161 0.016 0.145 7.0 0.246 0.100 41% Table 85. Total Maximum Daily Load for Segment COGULG04a, Cummings Gulch. Period of record 1991-2000.
Selenium reductions for Cummings Gulch are required for eleven months of the year (Table 85). Reductions range from 16 percent in May to 74 percent in January. If monthly load reductions ranging from 0.1 to 1.3 pounds per day (152 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Cummings Gulch. Currently, there are no point source discharges to Cummings Gulch; therefore load reductions would come from reductions in non-point source loads. No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Cummings Gulch. Therefore no loading reduction is required to attain the acute standard.
January 2011
77
Final
Sunflower Drain Selenium reductions for Sunflower Drain are required during all months of the year (Table 86). Reductions range from 76 percent in May to 98 percent in January-March. If monthly load reductions ranging from 2.2 to 6.4 pounds per day (1421 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Sunflower Drain. Currently, there are no point source discharges to Sunflower Drain; therefore load reductions would come from reductions in non-point source loads. The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for the Sunflower Drain is 209.0 g/L (January 17, 2002). A 98 percent loading reduction would result in an instream concentration of 4.2 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
2.5
4.6
0.061
0.006
0.055
184.1
2.435
2.380
98%
Feb
2.6
4.6
0.065
0.006
0.058
192.9
2.708
2.649
98%
Mar
2.2
4.6
0.053
0.005
0.048
196.5
2.281
2.233
98%
Apr
38.0
4.6
0.944
0.094
0.850
17.9
3.663
2.813
77%
May
42.0
4.6
1.043
0.104
0.939
17.0
3.850
2.911
76%
Jun
46.0
4.6
1.143
0.114
1.028
19.0
4.713
3.685
78%
Jul
33.0
4.6
0.820
0.082
0.738
29.0
5.171
4.434
86%
Aug
45.0
4.6
1.118
0.112
1.006
22.7
5.511
4.505
82%
Sep
57.5
4.6
1.428
0.143
1.285
24.9
7.731
6.446
83%
Oct
59.5
4.6
1.478
0.148
1.330
23.0
7.382
6.052
82%
Nov
11.0
4.6
0.273
0.027
0.246
92.9
5.521
5.275
96%
Dec 3.9 4.6 0.097 0.010 0.087 160.5 3.379 3.292 97% Table 86. Total Maximum Daily Load for Segment COGULG04a, Sunflower Drain. Period of record 19952003.
The maximum monthly 85th percentile value of 196.5 g/L for the Sunflower Drain also exceeds both the Agriculture Use and the Water Supply Use-based standards of 20 g/L and 50 g/L respectively. The dataset for the Sunflower Drain is reported as dissolved selenium whereas the Agriculture Use and the Water Supply Use-based selenium standards are expressed as the total recoverable species. Therefore, direct comparison of any available analytical results against the Agriculture Use and the Water Supply Use-based standards is not possible. A 98 percent loading reduction is necessary to attain the Aquatic Life Use-based standard during March, when the maximum monthly 85th percentile value was recorded. Were a 98 percent loading reduction achieved, however, it is reasonable to assume that the Agriculture and Water Supply Use-based selenium standards would not be exceeded.
January 2011
78
Final
Annual loads from lower Gunnison tributaries total 5,167 pounds annually. Measured tributaries contribute approximately 44 percent of the annual selenium load to the lower Gunnison River. Tributary load reductions total 4,547 pounds of selenium annually, which accounts for 39 percent of the annual load reduction in the Gunnison River at Grand Junction. The two tributaries, Dry Gulch and Sunflower Drain, contribute 3,686 pounds (81 percent) of the total measured tributary load. If standards were attained in these two tributaries alone, loads in the lower Gunnison could be reduced by approximately 32 percent annually. TMDL for Segment COGULG04b, Kannah Creek below the point of diversion for public water supply Selenium reductions for Kannah Creek below USGS Station #09152000 are required during all months of the year except May (Table 87). Reductions range from 31 percent in June to 94 percent in February. If monthly load reductions ranging from 0.02 to 0.9 pounds per day (105 pounds annually) are attained, it would result in attainment of the chronic selenium standard in lower Kannah Creek. Currently, there are no point source discharges to lower Kannah Creek; therefore load reductions would come from reductions in non-point source loads. Existing loads in Kannah Creek total 155 pounds annually. An annual load reduction of 105 pounds is required in order for Kannah Creek, Lower Gunnison Segment 4b, to be in attainment of its annual TMDL.
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Median Existing Load Flows, Standard, TMDL, LA, 85th% Load, Reduction, % cfs g/L lbs/day 10% MOS lbs/day Conc lbs/day lbs/day Reduction 3.6 4.6 0.089 0.009 0.080 49.3 0.958 0.878 92% 2.0 4.6 0.050 0.005 0.045 65.0 0.702 0.657 94% 3.6 4.6 0.088 0.009 0.079 52.2 1.000 0.921 92% 1.2 4.6 0.030 0.003 0.027 15.2 0.098 0.072 73% 57.5 4.6 1.428 0.143 1.285 4.0 1.233 -0.053 0% 0.4 4.6 0.010 0.001 0.009 6.0 0.013 0.004 31% 1.8 4.6 0.045 0.004 0.040 32.2 0.313 0.273 87% 0.8 4.6 0.021 0.002 0.019 8.0 0.036 0.017 48% 0.5 4.6 0.011 0.001 0.010 13.1 0.032 0.022 68% 0.8 4.6 0.019 0.002 0.017 10.2 0.043 0.025 59% 1.1 4.6 0.027 0.003 0.025 28.1 0.167 0.142 85% 2.4 4.6 0.058 0.006 0.053 38.7 0.491 0.439 89% Table 87. Total Maximum Daily Load for Segment COGULG04b, Kannah Creek. Period of record 19992005.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Kannah Creek in the most recent ten year period of record is 19.0 g/L (multiple dates in January, February and March). A 92 percent loading reduction would result in an instream concentration of 1.5 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded.
January 2011
79
Final
TMDL for Segment COGULG04c, Red Rock Canyon Creek Selenium reductions for Red Rock Canyon Creek exceed 90 percent for all months of the year except September (87 percent reduction) (Table 88). Reductions range from 91 percent in November to 95 percent in April. If monthly load reductions ranging from 0.5 to 1.4 pounds per day (240 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Red Rock Canyon Creek. Currently, there are no point source discharges to Red Rock Canyon Creek; therefore load reductions would come from reductions in non-point source loads. Existing loads in Red Rock Canyon Creek total 277 pounds annually. An annual load reduction of 240 pounds is required in order for Red Rock Canyon Creek, Lower Gunnison Segment 4c, to be in attainment of its annual TMDL. Median Existing Load Flows, Standard, TMDL, 10% LA, 85th% Load, Reduction, % cfs g/L lbs/day MOS lbs/day Conc lbs/day lbs/day Reduction Jan 2.1 4.6 0.052 0.005 0.047 57.5 0.652 0.605 93% Feb 2.0 4.6 0.048 0.005 0.044 64.2 0.676 0.633 94% Mar 1.8 4.6 0.045 0.004 0.040 71.0 0.690 0.650 94% Apr 1.7 4.6 0.042 0.004 0.038 78.8 0.723 0.685 95% May 1.9 4.6 0.046 0.005 0.041 53.2 0.531 0.490 92% Jun 2.3 4.6 0.056 0.006 0.050 54.6 0.664 0.613 92% Jul 2.5 4.6 0.062 0.006 0.056 67.4 0.910 0.854 94% Aug 4.1 4.6 0.102 0.010 0.092 54.9 1.215 1.124 92% Sep 8.0 4.6 0.199 0.020 0.179 32.6 1.409 1.230 87% Oct 1.6 4.6 0.040 0.004 0.036 52.4 0.453 0.417 92% Nov 2.4 4.6 0.060 0.006 0.054 44.0 0.570 0.516 91% Dec 2.3 4.6 0.056 0.006 0.050 50.7 0.616 0.566 92% Table 88. Total Maximum Daily Load for Segment COGULG04c, Red Rock Canyon Creek. Period of record 2001-2005.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Red Rock Canyon Creek is 78.8 g/L (April 28, 2003). A 95 percent loading reduction would result in an instream concentration of 3.9 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. The maximum monthly 85th percentile value of 78.8 g/L for Red Rock Canyon Creek also exceeds both the Agriculture Use and the Water Supply Use-based standards of 20 g/L and 50 g/L respectively. The dataset for Red Rock Canyon Creek is reported as dissolved selenium whereas the Agriculture Use and the Water Supply Use-based selenium standards are expressed as the total recoverable species. Therefore, direct comparison of any available analytical results against the Agriculture Use and the Water Supply Use-based standards is not possible. A 95 percent loading reduction is necessary to attain the Aquatic Life Use-based standard during April when the maximum monthly 85th percentile value was recorded. Were a 95 percent loading
January 2011
80
Final
reduction achieved, however, it is reasonable to assume that the Agriculture and Water Supply Use-based selenium standards would not be exceeded. TMDL for Segment COGUNF03, North Fork of the Gunnison River Selenium reductions for North Fork of the Gunnison River Segment 3 are required for ten months of the year (May – February), with required reductions ranging from 2 percent in May to 46 percent in February (Table 89). No reductions are required in the months of March and April. Background levels of selenium, below Paonia, were in attainment of stream standards (85th percentile = 2.5 g/L). Waste load allocations for the towns of Hotchkiss and Paonia were set according to Colorado Discharge Permit Regulations, Regulation 61. If selenium discharges from sand and gravel operations were to continue at the current rate, the contribution from sand and gravel discharges could account for almost 100 percent of the allowable stream load at chronic low flows. Similar to calculation of the effluent limits for the towns of Hotchkiss and Paonia, a mass balance was performed for the sum of the five sand and gravel facilities on the segment. Ambient eighty-fifth percentile concentrations of 2.5 g/L from the North Fork below Paonia were used in the calculation. Due to the influence of flow in providing available assimilative capacity for selenium dilution, seasonal limits of 6 g/L were calculated for the months of May through November, and 12 g/L were calculated for the months of December through April. Because of the continued growth in and around Delta and Montrose Counties, a reserve capacity was included in the TMDL in order to reserve allocations for future permitted discharges to this segment. Estimated population growth for Delta County is projected to increase by approximately ten percent by the year 2015; therefore ten percent of the available TMDL was set aside for future growth. Reductions in current discharge of 48 percent from sand and gravel operations are required in order to attain standards instream. Instream, the months of March and April are currently attaining their allowable load. Load reductions in Table 89 reflect reductions already taken from sand and gravel facilities. The percent reduction remaining reflects reductions required from nonpoint sources. Annual loads from the North Fork Gunnison River total 3,124 pounds annually. Currently, the North Fork contributes approximately 12 percent of the annual selenium load to the lower Gunnison River. Load reductions for the North Fork total 568 pounds of selenium annually, which account for about 5 percent of the annual load reduction required in the Gunnison River. No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for the North Fork of the Gunnison River. Therefore no loading reduction is required to attain the acute standard.
January 2011
81
Final
TMDL for COGUNF03 at Chronic Low Flows WLA WLA WLA (Sand & (individuals), Reserve Gravel), LA, lbs/day Capacity lbs/day lbs/day
Median Flows, cfs
Std., g/L
TMDL, lbs/day
10% MOS
Jan
131
4.6
3.3
0.33
0.32
0.29
0.87
1.4
5.3
3.7
0.8
21%
Feb*
136
4.6
3.4
0.34
0.32
0.30
0.87
1.5
7.7
5.6
2.6
46%
Mar
210
4.6
5.2
0.52
0.32
0.47
0.87
3.0
2.2
2.5
-2.2
0%
Apr
688
4.6
17.1
1.71
0.32
1.54
0.87
12.6
4.1
15.3
-0.1
0%
May
1560
4.6
38.8
3.88
0.32
3.49
0.44
30.6
4.2
35.6
0.7
2%
Jun
440
4.6
10.9
1.09
0.32
0.98
0.44
8.1
6.7
15.8
6.0
38%
Jul
60
4.6
1.5
0.15
0.32
0.13
0.44
0.4
7.0
2.3
0.9
41%
Aug
69
4.6
1.7
0.17
0.32
0.15
0.44
0.6
7.5
2.8
1.3
45%
Sep
86
4.6
2.1
0.21
0.32
0.19
0.44
1.0
6.7
3.1
1.2
38%
Oct
138
4.6
3.4
0.34
0.32
0.31
0.44
2.0
7.6
5.7
2.6
45%
Nov
157
4.6
3.9
0.39
0.32
0.35
0.44
2.4
6.4
5.4
1.9
35%
139 4.6 3.5 0.35 0.32 0.31 0.87 1.6 6.3 4.7 1.6 Table 89. Total Maximum Daily Load for Segment COGUNF03, North Fork Gunnison River. Period of record 20002005.
34%
Dec
85th% Conc.
Existing Load Load, Reduction, % lbs/day lbs/day Reduction
TMDLs for Segment COGUNF05, Leroux Creek and Jay Creek Leroux Creek Selenium reductions for Leroux Creek range from 39 percent in January to 72 percent in June (Table 90). If monthly load reductions ranging from 0.1 to 0.5 pounds per day (96 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Leroux Creek. Currently, there are no point source discharges to Leroux Creek; therefore load reductions would come from reductions in non-point source loads. Existing monthly loads in Leroux Creek total 159 pounds annually. An annual load reduction of 96 pounds is required in order for Leroux Creek, North Fork Segment 5, to be in attainment of its annual TMDL.
January 2011
82
Final
Median Existing Load Flows, Standard, TMDL, 10% LA, 85th% Load, Reduction, % cfs g/L lbs/day MOS lbs/day Conc lbs/day lbs/day Reduction Jan 6.4 4.6 0.16 0.02 0.14 6.8 0.23 0.09 39% Feb* 5.5 4.6 0.14 0.01 0.12 8.0 0.24 0.12 48% Mar 4.7 4.6 0.12 0.01 0.11 13.6 0.35 0.24 70% Apr 10.0 4.6 0.25 0.02 0.22 13.8 0.75 0.52 70% May 4.0 4.6 0.10 0.01 0.09 7.6 0.16 0.07 45% Jun 4.9 4.6 0.12 0.01 0.11 15.0 0.40 0.29 72% Jul 7.0 4.6 0.17 0.02 0.16 7.7 0.29 0.13 46% Aug 9.1 4.6 0.23 0.02 0.20 13.8 0.68 0.47 70% Sep 10.6 4.6 0.26 0.03 0.24 8.7 0.50 0.26 52% Oct 12.0 4.6 0.30 0.03 0.27 11.9 0.77 0.50 65% Nov 11.5 4.6 0.29 0.03 0.26 8.7 0.54 0.28 52% Dec 7.2 4.6 0.18 0.02 0.16 8.2 0.32 0.16 50% Table 90. Total Maximum Daily Load for Segment COGUNF05, Leroux Creek. Period of record 19912005.
No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Leroux Creek. Therefore no loading reduction is required to attain the acute standard. Jay Creek Annual selenium reductions for Jay Creek are approximately 75 percent (Table 91). If load reductions of 0.04 pounds per day (13.9 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Jay Creek. Currently, there are no point source discharges to Jay Creek; therefore load reductions would come from reductions in non-point source loads. Existing loads in Jay Creek total 18.6 pounds annually. An annual load reduction of 13.9 pounds is required in order for Jay Creek, North Fork Segment 5, to be in attainment of its annual TMDL. Existing Load Standard, TMDL, Median 10% LA, 85th% Load, Reduction, % Flows, cfs g/L lbs/day MOS lbs/day Conc lbs/day lbs/day Reduction Annual 0.58 4.6 0.0144 0.0014 0.0130 16.3 0.0511 0.0381 75% Table 91. Total Maximum Daily Load for Segment COGUNF05, Jay Creek. Period of record 1999-2000.
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Jay Creek is 19.0 g/L (March 13, 2000). A 75 percent loading reduction would result in an instream concentration of 4.8 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded.
January 2011
83
Final
TMDL for Segment COGUNF06a, Short Draw Annual selenium reductions for Short Draw are approximately 78 percent (Table 92). If load reductions of 0.45 pounds per day (165 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Short Draw. Currently there are no point source discharges to Short Draw; therefore load reductions would come from reductions in non-point source loads. Median Existing Load Flows, Standard, TMDL, 10% LA, 85th% Load, Reduction, % cfs g/L lbs/day MOS lbs/day Conc lbs/day lbs/day Reduction Annual 5.6 4.6 0.139 0.014 0.13 19.1 0.58 0.45 78% Table 92. Total Maximum Daily Load for Segment COGUNF06a, Short Draw. Period of record 19992000.
Existing loads in Short Draw total 211 pounds annually. An annual load reduction of 165 pounds is required in order for Short Draw, North Fork Segment 6a, to be in attainment of its annual TMDL. The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Short Draw is 29.0 g/L (March 14, 2000). A 78 percent loading reduction would result in an instream concentration of 6.4 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. TMDLs for Segment COGUNF06b TMDL, Big Gulch, Cottonwood Creek, Bell Creek Big Gulch Annual selenium reductions for Big Gulch are approximately 52 percent (Table 93). If load reductions of 0.07 pounds per day (29 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Big Gulch. Currently, there are no point source discharges to Big Gulch; therefore load reductions would come from reductions in non-point source loads. Existing Load Standard, TMDL, Median 10% LA, 85th% Load, Reduction, % Flows, cfs g/L lbs/day MOS lbs/day Conc lbs/day lbs/day Reduction Annual 3.30 4.6 0.082 0.008 0.074 8.6 0.152 0.079 52% Table 93. Total Maximum Daily Load for Segment COGUNF06b, Big Gulch. Period of record 1999-2000.
Existing loads in Big Gulch total 56 pounds annually. An annual load reduction of 29 pounds is required in order for Big Gulch, North Fork Segment 6b, to be in attainment of its annual TMDL.
January 2011
84
Final
No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Big Gulch. Therefore no loading reduction is required to attain the acute standard. Cottonwood Creek Annual selenium reductions for Cottonwood Creek are approximately 54 percent (Table 94). If load reductions of 0.18 pounds per day (66 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Cottonwood Creek. Currently, there are no point source discharges to Cottonwood Creek; therefore load reductions would come from reductions in non-point source loads. Median Existing Load Standard, TMDL, Flows, 10% LA, 85th% Load, Reduction, % cfs g/L lbs/day MOS lbs/day Conc lbs/day lbs/day Reduction Annual 7.0 4.6 0.17 0.02 0.16 9.0 0.34 0.18 54% Table 94 Total Maximum Daily Load for Segment COGUNF06b, Cottonwood Creek. Period of record 1999-2000.
Existing loads in Cottonwood Creek total 124 pounds annually. An annual load reduction of 66 pounds is required in order for Cottonwood Creek, North Fork Segment 6b, to be in attainment of its annual TMDL. No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Cottonwood Creek. Therefore no loading reduction is required to attain the acute standard. Bell Creek Annual selenium reductions for Bell Creek are approximately 41 percent (Table 95). If load reductions of 0.16 pounds per day (58 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Bell Creek. Currently, there are no point source discharges to Bell Creek; therefore load reductions would come from reductions in non-point source loads. Median Existing Load Standard, TMDL, Flows, 10% LA, 85th% Load, Reduction, % cfs g/L lbs/day MOS lbs/day Conc lbs/day lbs/day Reduction Annual 10.3 4.6 0.255 0.025 0.229 7.0 0.387 0.158 41% Table 95. Total Maximum Daily Load for Segment COGUNF06b, Bell Creek. Period of record 1999-2000.
Existing loads in Bell Creek total 142 pounds annually. An annual load reduction of 58 pounds is required in order for Bell Creek, North Fork Segment 6b, to be in attainment of its annual TMDL. No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Bell Creek. Therefore no loading reduction is required to attain the acute standard.
January 2011
85
Final
TMDL for Segment COGUUN04b, Uncompahgre River from LaSalle Road to Confluence Park Annual selenium reductions for the mainstem Uncompahgre River, Segment 4b, are greater than 60 percent year round (Table 96). Required load reductions range from 61 percent in May to 81 percent in January. If load reductions of 0.9 to 15.3 pounds per day (2,279 pounds annually) are attained, it would result in attainment of the chronic selenium standard in the mainstem Uncompahgre, Segment 4b. Currently, there are three domestic point source discharges to the Uncompahgre River. Permits for the City of Montrose and West Montrose were recently renewed in 2009. Waste load allocations for these two facilities were calculated as a joint allocation and were set according to Colorado Discharge Permit Regulations, Regulation 61. Currently, the third permitted discharge, the town of Olathe water treatment facility has a compliance schedule to address inflow and infiltration problems; therefore load reductions in Table 96 reflect reductions from non-point source loads. Annual loading reductions calculated by USGS for the Uncompahgre River at Confluence Park ranged from 67 to 70 percent dependent upon the water year (see Table 13). These results are not inconsistent with the monthly reductions identified in Table 69 given the differences in assessment methodologies utilized. Existing loads in this segment of the Uncompahgre River total 3,133 pounds annually. An annual load reduction of 2,279 pounds is required in order for the Uncompahgre River, Uncompahgre Segment 4b, to be in attainment of its annual TMDL.
Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
WLA (Olathe, Montrose, West Montrose), lbs/day
Jan
130
4.6
3.2
0.3
0.361
0.098
2.45
21.8
15.3
12.4
81%
Feb
122
4.6
3.0
0.3
0.361
0.098
2.27
19.0
12.5
9.8
78%
Mar
117
4.6
2.9
0.3
0.361
0.098
2.15
17.2
10.8
8.2
76%
Apr
65
4.6
1.6
0.2
0.361
0.098
0.98
13.8
4.8
3.4
70%
May
84
4.6
2.1
0.2
0.361
0.098
1.42
10.6
4.8
2.9
61%
Jun
95
4.6
2.4
0.2
0.361
0.098
1.66
12.1
6.2
4.1
66%
Jul
23
4.6
0.6
0.1
0.361
0.098
0.04
11.7
1.4
0.9
65%
Aug
53
4.6
1.3
0.1
0.361
0.098
0.73
11.4
3.3
2.1
64%
Sep
110
4.6
2.7
0.3
0.361
0.098
2.00
11.8
7.0
4.5
65%
Oct
110
4.6
2.7
0.3
0.361
0.098
2.00
13.5
8.0
5.6
69%
Nov
199
4.6
4.9
0.5
0.361
0.098
3.99
13.7
14.7
10.3
70%
152 4.6 3.8 0.4 0.361 0.098 2.93 17.6 14.4 11.0 Table 96. Total Maximum Daily Load for Segment COGUUN04b, Uncompahgre River from LaSalle Road to Confluence Park. Period of record 2000-2003.
76%
Dec
January 2011
86
WLA (Sand & Gravel), lbs/day
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Final
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for this reach of the Uncompahgre River is 23.0 g/L (January 7, 2000). An 81 percent loading reduction would result in an instream concentration of 4.4 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. TMDL for Segment COGUUN04c, Uncompahgre River from Confluence Park to Colorado River Annual selenium reductions for the mainstem Uncompahgre River, Segment 4c, are required in all months of the year except September (Table 97). Required load reductions range from 56 percent in March to 82 percent in February. If load reductions of 1.4 to 12.5 pounds per day (2,129 pounds annually) are attained, it would result in attainment of the chronic selenium standard in the mainstem Uncompahgre, Segment 4c. The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for segment COGUUN04c of the Uncompahgre River is 23.0 g/L (February 8, 2001 and July 19, 2005). A 73 percent loading reduction (July) would result in an instream concentration of 6.2 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded.
Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
130
4.6
3.2
0.32
2.9
19.9
14.0
11.1
79%
Feb
122
4.6
3.0
0.30
2.7
22.4
14.8
12.0
82%
Mar
117
4.6
2.9
0.29
2.6
9.5
6.0
3.4
56%
Apr
65
4.6
1.6
0.16
1.4
10.8
3.8
2.3
62%
May
84
4.6
2.1
0.21
1.9
10.2
4.6
2.7
59%
Jun
95
4.6
2.4
0.24
2.1
12.3
6.3
4.2
66%
Jul
23
4.6
0.6
0.06
0.5
15.5
1.9
1.4
73%
Aug
53
4.6
1.3
0.13
1.2
16.3
4.7
3.5
75%
Sep
110
4.6
2.7
0.27
2.5
0.0
0.0
-2.5
0%
Oct
110
4.6
2.7
0.27
2.5
15.3
9.1
6.6
73%
Nov
199
4.6
4.9
0.49
4.4
14.0
15.0
10.6
70%
Dec 152 4.6 3.8 0.38 3.4 19.4 15.9 12.5 79% Table 97. Total Maximum Daily Load for Segment COGUUN04c, Uncompahgre River from Confluence Park to Colorado River. Period of record 1999-2005.
January 2011
87
Final
Existing loads in the Uncompahgre River, Segment 4c, total 2,909 pounds annually. An annual load reduction of 2,129 pounds is required in order for the Uncompahgre River, Uncompahgre Segment 4c, to be in attainment of its annual TMDL. TMDLs for Segment COGUUN12, tributaries to the Uncompahgre River Cedar Creek Annual selenium reductions for Cedar Creek (tributary to the Uncompahgre), Segment 12, are required in all months of the year (Table 98). Required load reductions range from 31 percent in June to 90 percent in December-January. If load reductions of 2.1 to 6.6 pounds per day (1,472 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Cedar Creek, Segment 12. Currently, there are no point source discharges to Cedar Creek; therefore load reductions would come from reductions in non-point source loads. Existing loads in Cedar Creek, Segment 12, total 2,219 pounds annually. An annual load reduction of 1,472 pounds is required in order for Cedar Creek to be in attainment of its annual TMDL. The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Cedar Creek is 45.0 g/L (March 19, 1992). Additional high values in the dataset tend to occur in March and December. A 90 percent loading reduction would result in an instream concentration of 4.5 g/L for the March 19, 1992 value. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
20.0
4.6
0.50
0.05
0.45
41.4
4.47
4.02
90%
Feb
16.0
4.6
0.40
0.04
0.36
32.8
2.83
2.48
87%
Mar
18.5
4.6
0.46
0.05
0.41
43.0
4.29
3.88
90%
Apr
68.0
4.6
1.69
0.17
1.52
14.1
5.18
3.66
71%
May
149.0
4.6
3.70
0.37
3.33
9.0
7.24
3.91
54%
Jun
206.5
4.6
5.13
0.51
4.62
6.0
6.69
2.07
31%
Jul
179.0
4.6
4.45
0.44
4.00
7.8
7.54
3.54
47%
Aug
189.0
4.6
4.69
0.47
4.23
9.0
9.19
4.96
54%
Sep
162.5
4.6
4.04
0.40
3.63
10.3
8.99
5.36
60%
Oct
43.0
4.6
1.07
0.11
0.96
32.7
7.59
6.63
87%
Nov
24.0
4.6
0.60
0.06
0.54
36.4
4.72
4.18
89%
Dec 18.0 4.6 0.45 0.04 0.40 40.2 3.91 3.51 90% Table 98. Total Maximum Daily Load for Segment COGUUN12, Cedar Creek. Period of record 19912001.
January 2011
88
Final
The maximum monthly 85th percentile value of 41.4 g/L for Cedar Creek also exceeds the Agriculture Use-based standard of 20 g/L. The dataset for Cedar Creek is reported as dissolved selenium whereas the Agriculture Use-based selenium standard is expressed as the total recoverable species. Therefore, direct comparison of any available analytical results against the Agriculture Use-based standard is not possible. A 90 percent loading reduction is necessary to attain the Aquatic Life Use-based standard during January when the maximum monthly 85th percentile value was recorded. Were a 90 percent loading reduction achieved, however, it is reasonable to assume that the Agriculture Use-based selenium standard would not be exceeded. Dry Cedar Creek Annual selenium reductions for Dry Cedar Creek (tributary to the Uncompahgre), Segment 12, are required in all months of the year (Table 99). Required load reductions range from 52 percent in June to 95 percent in December. If load reductions of 2.1 to 6.6 pounds per day (1,472 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Cedar Creek, Segment 12. Currently, there are no point source discharges to Cedar Creek; therefore load reductions would come from reductions in non-point source loads. Existing loads in Dry Cedar Creek, Uncompahgre sub-basin Segment 12, total 342 pounds annually. An annual load reduction of 260 pounds is required in order for Dry Cedar Creek to be in attainment of its annual TMDL. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
2.0
4.6
0.048
0.005
0.044
65.3
0.687
0.643
94%
Feb
2.0
4.6
0.050
0.005
0.045
66.5
0.718
0.673
94%
Mar
2.3
4.6
0.057
0.006
0.051
68.7
0.853
0.801
94%
Apr
13.7
4.6
0.339
0.034
0.305
17.3
1.275
0.970
76%
May
8.8
4.6
0.217
0.022
0.196
12.6
0.593
0.397
67%
Jun
20.0
4.6
0.497
0.050
0.447
8.7
0.940
0.492
52%
Jul
13.0
4.6
0.323
0.032
0.291
13.0
0.913
0.622
68%
Aug
15.0
4.6
0.373
0.037
0.335
13.8
1.118
0.782
70%
Sep
24.0
4.6
0.596
0.060
0.537
11.0
1.426
0.889
62%
Oct
16.0
4.6
0.397
0.040
0.358
11.0
0.950
0.593
62%
Nov
2.7
4.6
0.067
0.007
0.060
56.0
0.816
0.756
93%
Dec
2.3 4.6 0.056 0.006 0.050 79.6 0.967 0.916 95% Table 99. Total Maximum Daily Load for Segment COGUUN12, Dry Cedar Creek. Period of record 19912000.
January 2011
89
Final
The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Dry Cedar Creek is 45.0 g/L (March 3, 1992). A 94 percent loading reduction would result in an instream concentration of 2.7 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. The maximum monthly 85th percentile value of 79.6 g/L for Dry Cedar Creek also exceeds both the Agriculture Use and the Water Supply Use-based standards of 20 g/L and 50 g/L respectively. The dataset for Dry Cedar Creek is reported as dissolved selenium whereas the Agriculture Use and the Water Supply Use-based selenium standards are expressed as the total recoverable species. Therefore, direct comparison of any available analytical results against the Agriculture Use and the Water Supply Use-based standards is not possible. A 95 percent loading reduction is necessary to attain the Aquatic Life Use-based standard during December when the maximum monthly 85th percentile value was recorded. Were a 95 percent loading reduction achieved, however, it is reasonable to assume that the Agriculture and Water Supply Use-based selenium standards would not be exceeded. Loutzenhizer Arroyo Annual selenium reductions for Loutzenhizer Arroyo (tributary to the Uncompahgre), Segment 12, are required in all months of the year (Table 100). Required load reductions range from 89 percent in May-August to 98 percent in November-March. If load reductions of 12.1 to 32.1 pounds per day (6,625 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Loutzenhizer Arroyo, Segment 12. Currently, there are no point source discharges to Loutzenhizer Arroyo; therefore load reductions would come from reductions in non-point source loads. Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
13.0
4.6
0.323
0.032
0.291
201.9
14.173
13.883
98%
Feb
10.0
4.6
0.248
0.025
0.224
230.0
12.420
12.196
98%
Mar
12.0
4.6
0.298
0.030
0.268
194.6
12.607
12.339
98%
Apr
67.0
4.6
1.664
0.166
1.498
57.5
20.794
19.297
93%
May
90.0
4.6
2.236
0.224
2.012
37.3
18.104
16.091
89%
Jun
96.0
4.6
2.385
0.238
2.146
38.8
20.124
17.978
89%
Jul
81.5
4.6
2.024
0.202
1.822
37.3
16.394
14.572
89%
Aug
102.5
4.6
2.546
0.255
2.291
39.0
21.567
19.276
89%
Sep
93.5
4.6
2.323
0.232
2.090
45.9
23.188
21.097
91%
Oct 70.0 4.6 1.739 0.174 1.565 89.0 33.642 32.077 95% Nov 24.0 4.6 0.596 0.060 0.537 172.0 22.291 21.755 98% Dec 17.5 4.6 0.435 0.043 0.391 182.0 17.199 16.808 98% Table 100. Total Maximum Daily Load for Segment COGUUN12, Loutzenhizer Arroyo. Period of record 1991-2003.
January 2011
90
Final
Existing loads in Loutzenhizer Arroyo, Segment 12, total 7,087 pounds annually. An annual load reduction of 6,625 pounds is required in order for Loutzenhizer Arroyo to be in attainment of its annual TMDL. The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Loutzenhizer Arroyo is 250.0 g/L (January 17, 2002). A 98 percent loading reduction would result in an instream concentration of 5.0 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded. The maximum monthly 85th percentile value of 230.0 g/L for Loutzenhizer Arroyo also exceeds both the Agriculture Use and the Water Supply Use-based standards of 20 g/L and 50 g/L respectively. The dataset for Loutzenhizer Arroyo is reported as dissolved selenium whereas the Agriculture Use and the Water Supply Use-based selenium standards are expressed as the total recoverable species. Therefore, direct comparison of any available analytical results against the Agriculture Use and the Water Supply Use-based standards is not possible. A 98 percent loading reduction is necessary to attain the Aquatic Life Use-based standard during February when the maximum monthly 85th percentile value was recorded. Were a 98 percent loading reduction achieved, however, it is reasonable to assume that the Agriculture and Water Supply Use-based selenium standards would not be exceeded. Montrose Arroyo Annual selenium reductions for the Montrose Arroyo (tributary to Cedar Creek), Segment 12, post 2000, are required in all months of the year (Table 101). Required load reductions range from 82 percent in July to 97 percent in December and January. If load reductions of 0.4 to 11 pounds per day (1,133 pounds annually) are attained, it would result in attainment of the chronic selenium standard in the Montrose Arroyo, Segment 12. Currently, there are no point source discharges to the Montrose Arroyo; therefore load reductions would come from reductions in non-point source loads. Existing loads in the Montrose Arroyo, Segment 12, totals 1,230 pounds annually. An annual load reduction of 1,142 pounds is required in order for the Montrose Arroyo to be in attainment of its annual TMDL. The adequacy of the loading reduction necessary to attain the chronic standard to also result in attainment of the acute standard was evaluated by application of the calculated percent load reduction to the maximum observed value. The maximum observed instream selenium concentration for Montrose Arroyo (after completion of the piping project) is 166.0 g/L (January 6, 2009). A 97 percent loading reduction would result in an instream concentration of 5.0 g/L. The required loading reduction, if achieved, would also ensure the acute standard of 18.4 g/L is not exceeded.
January 2011
91
Final
Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
2.2
4.6
0.053
0.005
0.053
139
1.61
1.56
97%
Feb
2.0
4.6
0.050
0.005
0.050
114
1.23
1.19
96%
Mar
1.6
4.6
0.040
0.004
0.040
86
0.75
0.71
95%
Apr
1.8
4.6
0.043
0.004
0.043
48
0.46
0.42
91%
May
5.9
4.6
0.145
0.015
0.145
37
1.18
1.05
89%
Jun
8.2
4.6
0.204
0.020
0.204
24
1.08
0.90
83%
Jul
15.0
4.6
0.373
0.037
0.373
23
1.84
1.50
82%
Aug
15.0
4.6
0.373
0.037
0.373
27
2.19
1.85
85%
Sep
22.5
4.6
0.559
0.056
0.559
34
4.17
3.67
88%
Oct
16.0
4.6
0.397
0.040
0.397
51
4.37
4.01
92%
Nov
22.0
4.6
0.546
0.055
0.546
97
11.51
11.02
96%
Dec 15.0 4.6 0.373 0.037 0.373 123 9.94 9.60 97% Table 101. Total Maximum Daily Load for Segment COGUUN12, Montrose Arroyo. Period of record 2000-2008.
The maximum monthly 85th percentile value of 139.0 g/L (post 2000) for the Montrose Arroyo also exceeds both the Agriculture Use and the Water Supply Use-based standards of 20 g/L and 50 g/L respectively. The dataset for the Montrose Arroyo is reported as dissolved selenium whereas the Agriculture Use and the Water Supply Use-based selenium standards are expressed as the total recoverable species. Therefore, direct comparison of any available analytical results against the Agriculture Use and the Water Supply Use-based standards is not possible. A 97 percent loading reduction is necessary to attain the Aquatic Life Use-based standard during January when the maximum monthly 85th percentile value was recorded. Were a 97 percent loading reduction achieved, however, it is reasonable to assume that the Agriculture and Water Supply Use-based selenium standards would not be exceeded. Dry Creek Annual selenium reductions for Dry Creek (tributary to the Uncompahgre), Segment 12, are required in all months of the year except May (Table 102). Required load reductions range from 26 percent in December to 59 percent in February. If load reductions of 0.5 to 1.8 pounds per day (349 pounds annually) are attained, it would result in attainment of the chronic selenium standard in Dry Creek, Segment 12. Currently, there are no point source discharges to Dry Creek; therefore load reductions would come from reductions in non-point source loads. Existing loads in Dry Creek, Segment 12, totals 1,297 pounds annually. An annual load reduction of 430 pounds is required in order for Dry Creek to be in attainment of its annual TMDL. No exceedances of the acute Aquatic Life Use-based standard of 18.4 g/L appear in the dataset for Dry Creek. Therefore no loading reduction is required to attain the acute standard.
January 2011
92
Final
Median Flows, cfs
Standard, g/L
TMDL, lbs/day
10% MOS
LA, lbs/day
85th% Conc
Existing Load, lbs/day
Load Reduction, lbs/day
% Reduction
Jan
55
4.6
1.37
0.14
1.37
8.3
2.45
1.22
50%
Feb
55
4.6
1.37
0.14
1.37
10.0
2.97
1.74
59%
Mar
57
4.6
1.42
0.14
1.42
7.6
2.32
1.05
45%
Apr
129
4.6
3.19
0.32
3.19
5.8
4.01
1.13
28%
May
200
4.6
4.97
0.50
4.97
4.0
4.32
-0.15
0%
Jun
96
4.6
2.38
0.24
2.38
5.0
2.59
0.45
17%
Jul
80
4.6
1.99
0.20
1.99
7.7
3.33
1.54
46%
Aug
110
4.6
2.72
0.27
2.72
7.1
4.22
1.77
42%
Sep
139
4.6
3.46
0.35
3.46
6.6
4.94
1.83
37%
Oct
169
4.6
4.20
0.42
4.20
6.0
5.48
1.70
31%
Nov
128
4.6
3.18
0.32
3.18
6.0
4.15
1.29
31%
Dec 60 4.6 1.49 0.15 1.49 5.6 1.82 0.48 26% Table 102. Total Maximum Daily Load for Segment COGUUN12, Dry Creek. Period of record 19912001.
IX. IMPLEMENTATION Since 1985, a multi-agency program within the Department of the Interior, the National Irrigation Water Quality Program (NIWQP), has done investigations of various federal irrigation projects in the Western United States to determine if irrigation drainage was having adverse effects on water quality and on fish and wildlife (USGS RPT 02-4151, 2002). High levels of selenium were reported in some water, sediment, and biota samples in the Uncompahgre River basin and Grand Valley in west-central Colorado.
The NIWQP collected data between 1987 and 2003 in the Uncompahgre Project area and lower Gunnison basin. Samples of water, sediment, food chain items (plants, invertebrates), fish, birds and bird eggs were collected from ponds, drains, backwaters and streams. Although examined for an extensive list of contaminants (e.g., lead, mercury, and zinc) and pesticides, selenium was found to be the major concern. Sixty percent or more of the selenium loading in the Gunnison basin (as measured at Whitewater) originates from an area encompassing the Uncompahgre River basin and the service area of the Federally-constructed Uncompahgre Project. This figure includes 40 percent from the Uncompahgre River basin and ranges from 3 to 17 percent from portions of the Uncompahgre Project service area in the vicinity of Delta (see Figure 4, page 17). Irrigation water delivery systems and on-farm applications to the Mancos Shale and soils derived from the shale contribute a vast majority (about 90 percent) of the ground-water that mobilizes selenium in this area (BOR 2006).
January 2011
93
Final
The Gunnison Basin Selenium Task Force was formed in 1998, and their goal has been to examine projects or methods that could be feasible for reducing selenium levels in the Uncompahgre and Gunnison Rivers. The task force and NIWQP (now inactive) worked together to identify and resolve selenium contamination issues while supporting the economic and lifestyle needs of the Gunnison basin citizens (USEPA, 2007). The USGS has conducted several focused studies to aid in the understanding of selenium-related water-quality issues. From these studies, it has been determined that the magnitude of selenium concentrations and loads in surface-water features are directly related to the application of irrigation water. Links to these studies are posted on the Task Force website at: http://www.seleniumtaskforce.org/projects.html. Several remediation projects have been completed or are underway. Perhaps the most significant project undertaken to date involved piping about 8.5 miles of irrigation laterals (ditches) that are part of the Federal Uncompahgre Irrigation Project within the Montrose Arroyo drainage basin. The Montrose Arroyo is a tributary of Cedar Creek, which is in turn tributary to the Uncompahgre River in the Montrose area (see figure 16). One reason Montrose Arroyo was selected for a demonstration project was because the area included several laterals of moderate size that could be lined or placed in pipe at a cost that was acceptable to the Salinity Control Program and the NIWQP. The project replaced 8.5 miles of open laterals with about 7.5 miles of PVC pipe. The PVC pipe was placed in trenches and buried, and the original laterals were filled in. About 80 percent of the lateral construction was done in the winter of 1998–99, and the remainder was completed in December 1999. (Butler, 2001). An assessment of the selenium loading reductions achieved by the Montrose Arroyo demonstration project was performed utilizing post-remediation data collected in 1999 and 2000 (Butler, 2001). Review of available data at that time indicated the project had been successful in reducing the selenium load associated with seepage from the laterals. Based upon the 1999-2000 data, the project resulted in an estimated decrease of about 194 pounds per year, or 28 percent of the pre-project selenium load. Currently a similar project has been initiated involving Loutzenhizer Arroyo. The document entitled “Evaluation of Selenium Remediation Concepts for the Lower Gunnison and Lower Uncompahgre Rivers” (http://www.seleniumtaskforce.org/images/Final_Gunn-Unc._Rivers_remediation_plan.pdf) discusses a variety of potential remedial actions under consideration by the Task Force.
X. RESPONSES TO COMMENTS RECEIVED CONCERNING DECEMBER 2009 PUBLIC NOTICE DRAFT Public comment was solicited relative to these TMDLs over a 45 day period from December 1, 2009 through January 15, 2010. Comments were received from five parties and are summarized, along with the Division responses, as follows:
January 2011
94
Final
Issue: Sand and gravel operations and municipal wastewater treatment facilities as sources of selenium Several commenters indicated that sand and gravel operations and domestic wastewater treatment plants are not a source of selenium to the river, and that the amount of selenium contributed to the river remains the same regardless of sand and gravel operations. The Colorado River District noted ―…sand and gravel operators, as well as waste water treatment plant operators do not add selenium loads to the river systems; they simply intercept local groundwater that may be laden with elevated selenium levels.‖ Oldcastle Group SW, Inc. commented that ―Experts in water quality issues involving selenium associated with the Uncompahgre and Gunnison River basins agree that sand and gravel operations located along the river are not a source of selenium…Gravel pits simply pass through groundwater containing selenium that would normally discharge to the river whether the gravel pit existed or not.‖ Colorado Stone, Sand & Gravel Association reiterated that ―It is absolutely clear that sand and gravel operators are not loading selenium during normal day to day operations or during daily, weekly, monthly or annual discharge processes.‖ Grand Junction Concrete Pipe Company believes that ―Even if these gravel pits discontinued discharging for a period of time, they do not believe that any measurable difference in the selenium loading in the river would be detected.‖ Gunnison Basin and Grand Valley Selenium Task Forces (STF) states that ―It is the opinion of the STF that sand and gravel operations do accelerate the mobilization, add, or load selenium to the river system.‖ Division Response: These comments touch upon several issues that dictate how point source discharges are addressed in not only this document, but in any TMDL analysis. This and the following portions of this chapter discuss a series of interrelated comments and the Division‟s responses. A common theme among the comments is the argument that sand and gravel facilities, and domestic waste water treatment facilities, do not add selenium to their discharges as a result of their operations. Nonetheless, there is ample data which indicates these discharges contain selenium, often at levels well above the standard. The TMDL includes a specific statement to that effect: As is the case with the municipal treatment facilities within the watershed, sand and gravel extraction facilities do not add selenium as a result of their operations. Wastewater discharged from sand and gravel operations typically results from the pumping of groundwater that accumulates in the pits and from the washing of
January 2011
95
Final
rock/gravel. The selenium concentration in the groundwater is influenced by local geology and upgradient land use. The selenium load contributed by sand and gravel facilities is a function of selenium concentration and the volume of groundwater collected in the pits. Wastewater treatment typically consists of retention of the wastewater in settling ponds prior to discharge. Little, if any, selenium removal is accomplished. (Page 57). The Division notes that this statement reflects the assumption that sand and gravel operations conform with Best Management Practices (BMPs) identified by the Colorado Division of Reclamation and Mining Safety (DRMS) and do not excavate to a depth such that any underlying shale is disturbed. Disturbance of the shale layers would exacerbate the mobilization of selenium. Sand and gravel operations would, in that case, clearly be adding selenium to the river system. Assuming that sand and gravel operations conform to the DRMS requirements, the statement that ―…sand and gravel operators, as well as waste water treatment plant operators … simply intercept local groundwater that may be laden with elevated selenium levels.‖ is generally valid. Nonetheless, several comment letters frame this argument in an overly broad manner. The CSSGA letter states: “Gravel producers simply intake and use selenium laden water for their crushing and wash plant operations, pump the water to holding ponds, and ultimately discharge water, with the same selenium levels, back to the river. Groundwater diverted to gravel pits contains the same amount of selenium that is pumped to the river by a permitted discharge. The only difference being that gravel operators take a non-point source (natural groundwater discharge to the river) and turn it into a point-source discharge from a dewatering pump through a pipe. The amount of selenium remains the same regardless of the gravel mining operation.‖ The first sentence states that Sand and Gravel producers divert and use groundwater in their crushing and washing operations, pump the water to holding ponds, and then discharge the groundwater to surface water. This conflicts with the statement that “The only difference being that gravel operators take a non-point source (natural groundwater discharge to the river) and turn it into a point-source discharge from a dewatering pump through a pipe”, which indicates that only intercepted groundwater is discharged. Once the diverted groundwater is used in the producer's crushing and washing operations, the groundwater now becomes or is considered "process water". Process water is defined in Regulation 61 Colorado Discharge Permit System Regulations. This discharge of process water from a sand and gravel operation to surface water contradicts the argument that sand and gravel operations simply intercept groundwater and convey it to the river through a discrete conveyance.
January 2011
96
Final
The letter also references the use of diverted waters for “crushing and wash plant operations‖. It is likely these processes add to dissolved selenium levels in groundwater entering the pit but qualitative or quantitative data documenting selenium additions due to these activities is lacking. Further, there has been no documentation provided which characterizes selenium concentrations in groundwater upgradient of these sand and gravel operations. Such data is necessary to document this assertion. The River District, in framing a similar argument, states “…sand and gravel operators, as well as waste water treatment plant operators do not add selenium loads to the river systems”. Similarly, the Colorado Stone, Sand & Gravel Association (CSSGA) states “It is absolutely clear that sand and gravel operators are not loading selenium during normal day to day operations or during daily, weekly, monthly or annual discharge processes.” The term “load” or „loading” has a specific meaning when utilized within the context of TMDL development. The pollutant load is a unit of mass-based measurement. In calculating pollutant load contributions, the pollutant mass is determined by multiplying the pollutant concentration times the discharge volume. Unless either of these terms is set to zero, this calculation will yield a quantified pollutant mass load. In other words, unless the pollutant is absent or present only in very low, undetectable amounts, or there is no measurable discharge, a quantifiable mass or load will always be calculated. Section VII – Point Source Contributions discusses the calculation of selenium loads based upon effluent monitoring data provided by municipal wastewater treatment facilities. This section also discusses pollutant loads calculated for sand and gravel discharges. During the public notice period several comment letters incorporated additional sand and gravel discharge data that had not been available previously. Currently available discharge quality as compared to ambient instream selenium is summarized below.
WBID COGULG01
Instream selenium 5.0
COGULG02
8.4
COGULG04a
8.4
Delta Paving Pit 34 Mule Farm Pit 38
Discharger Anderson Pit 31
Dubs Pit
Fredlund Pit
67
63
TriCounty Pit COGUNF03 6.9 2.05 8.3 24 th Table 103. Ambient and 85 Percentile Selenium Values (in g/L) Campbell Pit
4-D Pit
Janet Pit 24
With one exception, the Campbell Pit which discharges to the North Fork of the Gunnison River, the 85th percentile selenium concentration value exceeds the instream concentration. Discharge concentration is usually an order of magnitude greater than the instream level. Each of the discharges contributes a selenium load to the receiving waters because they contain a quantifiable amount of selenium. This additional information is useful in gauging the impact of TMDL implementation through the CDPS permitting process and is discussed in detail in the next section.
January 2011
97
Final
Finally, the CSSGA ―…strongly challenges the CDPHE’s finding that sand and gravel operators have discharges with selenium concentrations that may be greater than the standards.‖ This statement may arise from confusion as to the applicable water quality standards addressed in the TMDLs. Comments provided by Old Castle Group SW contend that the 50.0 g/L Water Supply Use-based standard should be used as the water quality target for these TMDLs. The numeric standards assigned to water bodies in Colorado are dictated by the beneficial uses for which such waters are utilized. The potential Use categories that the Water Quality Control Commission has assigned to this water body are Water Supply, Agriculture and Aquatic Life Uses. Each use classification carries with it an associated set of numeric water quality standards. Each has a Use-specific selenium standard or standards. Old Castle is correct in that the Water Supply Use-based standard is set at 50.0 g/L (as total recoverable selenium). Most, but not all, of the water bodies addressed in the TMDL document are assigned a Water Supply Use. All of the waters addressed are also classified for Agriculture Use as well as Aquatic Life Use. The numeric selenium standard associated with Agriculture Use is 20.0 g/L. As this standard is more stringent than the Water Supply Use-based standard of 50.0 g/L, the TMDLs would have been developed so as to address the 20 g/L standard had those been the only designated Uses in place. However, all of the water bodies addressed in the TMDLs are also classified for Aquatic Life Use. The Aquatic Life Use-based standards for selenium are 4.6 g/L as a chronic standard and 18.4 g/L as an acute standard. The classified Uses for all of the waters addressed in the TMDL document are summarized on page 3, Table 1. Section III – STANDARDS FRAMEWORK (pages 12 through 15) explains this issue in greater detail. Effluent monitoring data for several sand and gravel discharges within the Gunnison River basis are summarized in Section VII, Tables 66 and 67. All of the samples collected both by the Division and reported self-monitoring results of samples collected by the operators themselves exceed the 4.6 g/L Aquatic Life Use-based chronic selenium standard. Additional discharge data was included in the summary provided above in Table 103. The Division believes that the assertions that have been cited above lack supporting data
Issue: Influent and Effluent selenium concentrations The Colorado River District indicates that ―…available data and conceptual analyses indicate that the inflow concentrations and related mass of selenium is the same as that which flows out of these plants and operations.‖ The Colorado Stone, Sand & Gravel Association states that ―Our members have sampled for selenium for a number of years and the influent and effluent concentrations are relatively constant within our areas of operation….Groundwater diverted to gravel pits contains the same amount of selenium that is pumped to the river by a permitted discharge.‖
January 2011
98
Final
Division response: The Colorado River District and the Colorado Stone, Sand & Gravel Association have not presented information characterizing the quality of groundwater that is influent to any aggregate extraction pit. In the absence of any data, the Division continues to support the assumptions in the TMDL. There have been similar instances where other point sources (primarily domestic treatment plants) are located on waterbodies that exceed chronic and acute metals standards. In these cases, significant inflow and infiltration (I & I) problems may exist, which is not an end result of plant operations. It is the current practice of the Division that these point sources receive waste load allocations and compliance schedules in order to attain the underlying standards that the TMDLs are based upon, regardless of the “pass through” nature of the plant. Consequently, the Division feels the same is true for sand and gravel operations that “intercept” selenium laden groundwater and in turn, discharge it directly into the river. The Colorado River District and the Colorado Stone, Sand & Gravel Association have not supported their arguments with data to the contrary.
Issue: Attainment of TMDL Water Quality Targets The Colorado River District speculated that ― ... the burden of compliance falls on the shoulders of two industries that do not add or load selenium to the river system.‖ Oldcastle SW Group SW, Inc. stated that: ―The proposed TMDL would use a quirk of the State of Colorado’s regulatory authority to enforce point source discharge limits that would provide no significant benefit to the water quality of the river while imposing an unreasonable financial burden on the sand and gravel industry.‖ The Selenium Task Force (STF) commented that ―Much of the burden of compliance within the TMDL lies with sand and gravel operators and municipal wastewater service providers. The STF does not believe this is a reasonable approach to dealing with needed selenium load reductions and that it could have significant economic impacts to local communities.‖ The Colorado Stone, Sand & Gravel Association indicated that ―Once the TMDL is promulgated, then the burden of compliance, or demonstration that attainment of selenium allocations is neither practiced, nor technically feasible, shifts to dischargers. In this case the burden will shift to the sand and gravel industry who will bear significant costs to correct poor assumptions upon which the TMDLs were premised.‖
January 2011
99
Final
Division Response: The Federal Clean Water Act (CWA) conveys regulatory authority to the State with respect to control of point source discharges. An overview of the Colorado Discharge Permit System (CDPS) program with respect to TMDL implementation issues is included in the TMDL document (see SECTION IV. – PROBLEM IDENTIFICATION ( page 16) and Section VII. TECHNICAL ANALYSIS (page 55)). It is correct to assume that CDPS permits issued to point source dischargers within the lower Gunnison basin will consider the potential for those discharges to cause or contribute to exceedences of the water quality standards for selenium as appropriate. However, it is not correct to assume that existing dischargers will bear a disproportionate burden with respect to financing of selenium control efforts. There are six domestic wastewater treatment facilities discharging to listed segments in the Gunnison and Uncompahgre watersheds that have been issued individual CDPS permits. Five of these facilities have performed sufficient monitoring to demonstrate effluent concentrations of selenium that fall consistently below the water quality standard of 4.6 g/L. These discharges cannot cause or contribute to exceedences of the water quality standards for selenium. Effluent selenium Municipal Permittee CDPS No. ( g/L) 1 Town of Paonia CO0047431 1 Town of Hotchkiss CO0044903 Town of Olathe CO0020907 <1.0 – 24.0 1 City of Montrose CO0039624 1 West Montrose Sanitation COGUUN04b CO0030449 District COGULG02 City of Delta CO0039641 4.0-13.6 1 effluent selenium monitoring not required Table 104. Domestic Wastewater Treatment Facilities within Lower Gunnison Basin Waterbody Identification COGUNF03 COGUNF03 COGUUN04b COGUUN04b
Two domestic facilities (Town of Olathe wastewater treatment plant and City of Delta) are operating under a compliance schedule incorporated into their current CDPDS permit. This compliance schedule addresses the fact that selenium enters the Town‟s sanitary sewer system via infiltration into the Town‟s collection system. In addition to introducing elevated levels of selenium into the Town‟s treatment plant influent, infiltration also dilutes the influent, reducing the wastewater treatment plant‟s efficiency and increasing treatment costs. Implementation of the TMDLs will have no effect on the cost of operating the Town‟s wastewater treatment facility. The City of Delta‟s compliance schedule was issued in 2009 with its permit renewal to meet the 4.6 g/L dissolved selenium water quality based effluent limit by the year 2013. The city is currently in the process of installing a diffuser instream in order to provide more assimilative capacity below the city‟s discharge in order to protect critical T & E habitat. There are ten permitted sand and gravel discharges to 303(d) listed segments within the basin. These are identified in Table 4 of the TMDL document. An additional six operations are permitted for discharge to that portion of the Gunnison River between the confluences with the
January 2011
100
Final
North Fork of the Gunnison and the Uncompahgre River (Table 6). Although not included on the 303(d) List this reach is also in non-attainment of water quality standards for selenium (see Table 2, pages 15-16). Six of the 16 permitted sand and gravel operations report that they have not discharged since 2006. Several have not discharged in ten or more years. Because there is no wastewater discharged by these facilities, they will not be impacted by these TMDLs. One of these non-discharging operations, Tri-County Gravel (COG500255) is permitted for discharge to the North Fork of the Gunnison River. The remaining North Fork facilities report discharge volumes that are typically an order of magnitude less than their permitted design capacity and/or discharge concentrations of selenium that are less than the standard. These include the Oldcastle SW Group‟s Campbell (COG500397), 4D (COG500400), and Tri-County (COG500498) Pits, and the Diamond Lazy L. Ranch Enterprises Janet Pit (COG500458). The TMDL includes Waste Load Allocations (WLAs) for these facilities that have been calculated based upon their design capacity and standards applied as end-of-pipe limits. Although these facilities report discharge volumes that are an order of magnitude less than their permitted maximum design flow, the potential selenium load is calculated using the permitted maximum flow. Therefore, the calculated selenium loads from these facilities is overestimated, but, their selenium load may be appreciable during periods of low flow. These facilities‟ operations should not be affected as a result of TMDL implementation, although it may be advantageous for the facilities to apply for an individual CDPS permit. Because the ambient instream selenium levels do not typically exceed the standard in that reach above Lazear (where these facilities are located), and because the facilities‟ historic discharge volume is a fraction of their currently permitted design flow, it is likely that adequate assimilative capacity is available to allow effluent limits to be set at a concentration higher than the standard. However, in order to take advantage of this possibility, the facilities would need to receive individual permits in lieu of general permit certifications. A single aggregate extraction facility holds a permit authorizing discharge to segment 4b of the Uncompahgre River in the vicinity of Montrose, the Western Gravel Concrete Facility North R-34 Pit (COG500486). Currently this operation is not required to perform selenium monitoring as a permit requirement. Consequently, there is no information available with which to predict any potential change in discharge requirements. Ambient instream selenium levels recorded at the Division‟s monitoring station located immediately upstream of the Montrose wastewater treatment plant discharge and in the vicinity of the North R-34 Pit W are less than the standards (Se = 2.47 g/L, n=12, POR 2001-2005). If sampling of the R-34 Pit discharge indicate selenium concentrations greater than the 4.6 g/L standard, there is an opportunity as with the North Fork facilities to establish effluent limitations a concentration higher than the standard through the individual permit process. Two sand and gravel operations, the Elam Construction Dubs Pit (COG500427) and the Western Gravel Fredlund Pit (COG500451) discharge to Lower Gunnison River segment 4a. This segment includes tributaries to the Gunnison River from below the outlet of Crystal
January 2011
101
Final
Reservoir to the confluence with the Colorado River. Neither operation has reported a discharge since 2004 (although the Dubs Pit reports effluent concentrations ranged from 39 to 104 g/L in 2003 and 2004 prior to cessation of the discharge). If these facilities continue to operate without a discharge, the TMDL will have no effect upon their continuing operation. If these facilities recommence discharge, it would likely be appropriate to pursue individual permits in order to ascertain whether there is available assimilative capacity at their discharge locations. Three facilities, the Elam Construction Mule Farm Gravel Pit (COG500210), the Western Gravel North D-22 Pit (COG500358) and the Whitewater Building Materials Pit (COG500127) discharge to segment 2 of the Lower Gunnison Basin. This segment includes the mainstem of the Gunnison River from below the confluence with the Uncompahgre River to its terminus at the Colorado River confluence. This portion of the watershed is identified by the U. S. Fish and Wildlife Service (USFWS) as habitat for a number of threatened and endangered aquatic species. The Division has entered into a Memorandum of Agreement with the USFWS and EPA that identifies specific restrictions that the Division has agreed to incorporate into CDPS permits issued for discharges to any such designated habitat. This is discussed in greater detail on page 65 of the TMDL assessment. The MOA is posted online and may be found at: http://www.cdphe.state.co.us/wq/PermitsUnit/POLICYGUIDANCEFACTSHEETS/policyandgui dance/MOA_TandE.pdf. Finally, there are five sand and gravel facilities located along segment 1 of the Lower Gunnison River. This includes that reach of the Gunnison River below the confluence with the North Fork of the Gunnison River to the confluence with the Uncompahgre River. These facilities are located to the north and northeast of the City of Delta. Two of these facilities (Elam Construction Bennett Gravel Pit (COG500439) and the Lafarge West North Delta Pit (COG500160)) have reported “no discharge” since 1994 and 2000 respectively. The remaining three facilities include the Grand Junction Pipe & Supply Delta Paving Gravel Pit (COG500444), the Oldcastle SW Group, Inc., Anderson Pit (COG500464), and the Oldcastle SW Group, DS&G Pit (COG500136). The annual ambient selenium concentration for the Gunnison River at Delta is approximately 5.4 g/L (Table 2). The TMDL for Lower Gunnison River segment 1 (Table 69) indicates that loading reductions are necessary in order to attain the 4.6 g/L standard for six months of the year. There is no assimilative capacity available for those six months. Any discharge that occurs within those six months would need to achieve the standard as an end of pipe limit. Monitoring data for the Grand Junction Pipe & Supply Delta Paving Gravel Pit indicates that the lowest recorded selenium discharge concentration was 8.0 g/L in July of 2007. The lowest reported value for the Oldcastle SW Group, Inc., Anderson Pit is 19 g/L. There is assimilative capacity during the remaining six months. Ongoing operation of these facilities will require issuance of individual permits which may restrict the facilities‟ ability to discharge untreated wastewater during a portion of the year, or incorporate a compliance schedule intended to result in attainment of appropriate effluent limits at some future point.
January 2011
102
Final
The Division believes that implementation of the TMDL will be a multifaceted effort between the agricultural community, point source dischargers, and city, county, state and federal representatives to adapt management strategies for a workable solution to address elevated river selenium levels. Such management strategies may identify non-point source controls intended to reduce selenium loading to a level less than the assigned standard of 4.6 g/L. Successful implementation of non-point source control strategies could conceivably result in the development of assimilative capacity for point source discharges at locations or periods where none currently exists. Ongoing assessment of changing water quality trends relative to land uses and point source and non-point sources; an “adaptive implementation” plan coordinated among the various stakeholders represents the best approach for successful TMDL implementation. Issue: Intermittent discharge of sand and gravel operations Grand Junction Pipe & Supply Company clarified that ―The four gravel pits that discharge to the Lower Gunnison Segment 1 do not all discharge continuously on an annual basis.‖ The Colorado Stone, Sand & Gravel Association provided discharge values for the Delta Paving Pit for the years 2004 through 2009. Division response: The Division would like to thank Grand Junction Pipe & Supply and the Colorado Stone, Sand & Gravel Association for providing discharge information for the Delta Paving Pit. As both of these entities pointed out, discharge from sand and gravel pits does not occur on a daily basis year-round. This is noted in the Division‟s response to comments regarding the “attainment of TMDL water quality targets”. This is relevant for the calculation of waste load allocations for sand and gravel dischargers, as discussed previously, and an implicit margin of safety. TMDL waste load allocations were based on a continuous discharge at design capacity at the chronic selenium standard of 4.6 g/L, and are expressed in units of pounds per day in the TMDL document as a joint waste load allocation. However, after further discussion with the Division‟s Permits Unit, the waste load allocation for sand and gravel operations on the North Fork Gunnison River may be implemented into permits as a seasonal allowable load. Wasteload allocations for the months when there is little assimilative capacity available (May-November) were calculated to be approximately 6 micrograms per liter, while allocations during higher flow months (December-April) were calculated to be approximately 12 micrograms per liter. Therefore, if the sand and gravel operations are neither discharging continuously, nor at design capacity, there is a built-in flexibility to the allowable waste load on a seasonal basis. This results in an additional element of conservatism, or an implicit Margin of Safety in addition to the ten percent explicit Margin of Safety included in the TMDL calculations. Individual estimates of waste loads can be calculated for specific facilities if individual permits are required. However, if the Colorado Stone, Sand & Gravel Association would like the Division to clarify their “error” in the waste load calculation, the Division is able to recalculate allowable loads based on a discontinuous discharge.
January 2011
103
Final
Issue: Attainability/Feasibility of waste load allocations The Colorado Stone, Sand & Gravel Association commented that ―The Division does not consider attainability or feasibility of controls for selenium during the TMDL development, which is incredibly shortsighted.‖ Division response: The Federal Clean Water Act 40 CFR §130.7(c)(1) requires that “Each State shall establish TMDLs for the water quality limited segments identified in paragraph (b)(1) of this section…For pollutants other than heat, TMDLs shall be established at levels necessary to attain and maintain the applicable narrative and numerical WQS with seasonal variations and a margin of safety…Determination of TMDLs shall take into account critical conditions for stream flow, loading, and water quality parameters.‖ The Clean Water Act requires TMDLs to address attainment of applicable narrative and numerical WQS. The applicable Water Quality Standards are those assigned to a given waterbody by the Colorado Water Quality Control Commission. The Commission may consider the attainability of a particular standard as part of the surface water standards rulemaking process. Once a given standard is promulgated, however, the Division is obligated to develop TMDLs as necessary to achieve the assigned standard. TMDL implementation and the associated planning that may parallel TMDL development should consider both the economic and technical feasibility of load and waste-load reduction alternatives as management practices are evaluated and selected. If attainment of an assigned standard proves infeasible, the Commission‟s standards rulemaking hearing process is the public forum within which alternate standards may be proposed. Issue: Promulgation of new selenium criteria The Colorado Stone, Sand & Gravel Association suggested that “It is a waste of state resources and private funds for the Division to promulgate selenium TMDLs based upon out-dated and soon to be replaced selenium standards.‖ Division response: On August 24, 1999, a Settlement Agreement was signed by the parties in Colorado Environmental Coalition, et. al. v. United State Environmental Protection Agency, et. al., Civ. Act. No. 97-S-1841 (D. Colo.). The terms of this Agreement require the State to complete TMDLs for all water quality limited segments on the State‟s 1998 Section 303(d) List (dated March 24, 1998) according to the schedule set forth in Exhibit B of the Agreement. According to the schedule in Exhibit B, the State is to have completed (as that term is defined in the Agreement) a cumulative number during the June 1998 - June 2008 timeframe of no less than 198 TMDLs. This term defines “completing a TMDL” for purposes of measuring State compliance with the schedule given in Exhibit B as 1) the State submitting a TMDL to EPA for approval under CWA Section 303(d)(1); 2) final completion of a TMDL by EPA; or 3) a written determination by EPA or the State that a TMDL is not needed. This term in the Agreement also
January 2011
104
Final
provides certain provisions each TMDL must address. The table below identifies segments remaining in order to complete the terms of the 1999 Settlement Agreement. Waterbody Identification Number
Waterbody
Pollutant
Action
COGULG02
Gunnison River, Uncompahgre River to Colorado River
Se
TMDL not completed
COGUNF05
N. Fork Gunnison River tributaries
Se
TMDL not completed
COGUUN04
Uncompahgre River, Highway 550 to Gunnison River
Se
TMDL not completed
Table 105. 1999 Settlement Agreement – Outstanding TMDLs
The Division concurs that promulgation of the new selenium criteria is likely imminent. However, EPA circulated its first update to the 304(a) selenium criteria in 2004, and due to problems associated with one of its toxicity studies it has yet to be finalized. Therefore, it is not appropriate to delay base TMDL development pending the promulgation of new criteria when there is no indication as to when such criteria will be finalized. In addition, if EPA were to finalize new 304(a) criteria for selenium in 2010, it would not be proposed for incorporation into Colorado‟s Basic Standards until the year 2016, and would not be incorporated into the San Juan and Gunnison basins until the year 2017. Following the incorporation of revised selenium criteria into the Gunnison basin standards, the new selenium criteria would not be integrated into sand and gravel permits until their general permit renewal date in 2018, which would allow eight years of additional monitoring and adaptive implementation strategies to occur. Issue: Phased TMDL and uncertainty in TMDL analysis The Division received comments from various sources indicating concerns about the uncertainty in the TMDL calculations. The Colorado River District commented “While we believe the report is generally sound, the analysis and data that this report is based upon, is subject to significant uncertainty. As such, the report would benefit from additional data collection, analysis, and stakeholder collaboration.” The Colorado Stone, Sand & Gravel Association suggested that ―Simply stated, the proposed TMDL does not align with the current realities of the Gunnison and Uncompahgre River and tributaries. Therefore, we propose that the TMDLs be adopted, if at all, as phased TMDLs which include a plan for collecting additional scientific evidence and monitoring data.‖ The Selenium Task Force believes that ―The Gunnison TMDL has far reaching economic and environmental implications for local and regional communities…..the STF believes the TMDL report could benefit from additional data collection, analyses, and stakeholder collaboration, especially with point source dischargers.‖
January 2011
105
Final
Division response: The Division does not believe the “phased TMDL” approach for the Gunnison and Uncompahgre River basins is appropriate. The phased TMDL approach is generally used when additional data based on better analytical techniques would increase the accuracy of the TMDL calculations and merit development of a second phase. Phased TMDLs are often used when a surrogate is used to identify TMDL loading reductions (e.g. using phosphorus to regulate chlorophyll concentrations). Regardless of the terminology used, waste load allocations are required to be written to the stream standard (i.e. 4.6 g/L). Allocation of additional Division resources to development of a second phase TMDL, when it would generate the same outcome, is not appropriate. The Division believes that an “adaptive implementation” approach would be more useful in this particular case. Adaptive implementation is a tool used to improve the implementation strategies in a TMDL document. Using this approach, new information generated from monitoring following initial TMDL implementation efforts can be utilized to appropriately target the next set of implementation activities. An iterative plan will be established for this TMDL in a coordinated effort with the Permits unit and stakeholders. Issue: Margin of Safety The Division received comments from various sources indicating concerns about the margin of safety in the TMDL calculations. The Colorado River District indicated that ―We support the use of a Margin of Safety (MOS), in concept, but it is unclear how this MOS changes in different reaches or under differing data conditions, or how it affects the specific waste load allocation (WLA) calculations.‖ The Selenium Task Force states that ―The STF is unclear about what the MOS is. Is it a percentage? How does the MOS change in different reaches and how does it affect the specific waste load allocation calculations?‖ Division response: The Division would like to clarify its use of a margin of safety in the TMDL. Due to the inherent complexity of a natural system, a margin of safety is required in order to ensure attainment of water quality standards. Some level of uncertainty is intrinsic when a mathematical relationship is used to quantify stressor/response relationships between pollutant loading rates and the resultant water quality. The Division utilized an explicit 10 percent margin of safety in its TMDL calculations. The TMDL was calculated using median monthly flow rates (when available) multiplied by the existing chronic selenium stream standard (4.6 g/L) and a conversion factor (0.0054) in order to determine a load in pounds per day. Ten percent of this TMDL calculation was then subtracted in order to quantify some of the uncertainty inherent in the calculation. This number was then incorporated as the allowable load portion for each of the listed stream segments. The explicit margin of safety does not change in each of the affected reaches, but remains consistent at ten percent. In the case of waste load allocations, they are
January 2011
106
Final
affected only in that there is ten percent less of an allowable load available to point source dischargers as assimilative capacity. Additionally, the conservative assumptions utilized in the TMDL make up an implicit margin of safety. Issue: Implementation Plan The Colorado Stone, Sand & Gravel Association believe that ―development of the TMDL implementation plans has already commenced...and this violates due process requirements.‖ Division response: Development of a TMDL implementation plan has been identified for coordination with the Selenium Management Plan that is currently under development for the Gunnison River Basin. Current participants in the development of the Selenium Management Plan include: State of Colorado, Selenium Task Force (STF), Colorado River District, National Resource Conservation Service (NRCS), US Geological Survey (USGS), UVWUA, City of Grand Junction, Colorado State University Extension (CSU), Shavano Conservation District, Bureau of Land Management (BLM), US Fish & Wildlife Service (USFWS), Colorado Stone, Sand & Gravel Association (CSSGA), and the US Bureau of Reclamation (USBR). Components of the Programmatic Biological Opinion (PBO) include: accelerated implementation of salinity/selenium, reduction/prevention of other non-point source selenium loading, technology development, water quality monitoring, monitoring of endangered fish populations, coordination with lower Gunnison River Basin watershed management plan, regulatory support, public information and education, adaptive management and monitoring, institutional arrangements, and control projects for irrigated agriculture. A Memorandum of Understanding (MOU) between parties is a required outcome of the PBO. The implementation portion of the TMDL would be coordinated with this process in addition to looking at possible remediation alternatives for point source operations where needed. The US Bureau of Reclamation is currently serving in a leadership role for the Selenium Management Plan, but there has been no finalization of a process for TMDL implementation. There is still an important opportunity to work with all of the interested parties to develop and coordinate a TMDL implementation plan with the affected parties. General Comments The USEPA commented relative to the description of the USGS loading assessment that was incorporated into the TMDL document. EPA felt some confusion could arise between the USGS summaries and the actual TMDLs calculated by the Division. Consequently the description of the USGS study has been somewhat pared down. The Division does believe there is value in presenting the USGS loading estimates in the TMDL document as they illustrate a fairly consistent result with the Division‟s TMDL calculations despite the differences in the overall statistical approaches.
January 2011
107
Final
XI. REFERENCES Bureau of Reclamation, and Gunnison Basin Selenium Task Force, 2006, Evaluation of Selenium Remediation Concepts for the Lower Gunnison and Lower Uncompahgre Rivers, Colorado, 43 p. Butler, D.L., Wright, W.G., Stewart, K.C., Osmundson, B.C., Krueger, R.P., and Crabtree, D.W., 1996, Detailed Study of Selenium and Other Constituents in Water, Bottom Sediment, Soil, Alfalfa, and Biota Associated with Irrigation Drainage in the Uncompahgre Project Area and in the Grand Valley, West-Central Colorado, 1991–93: U.S. Geological Survey Water-Resources Investigations Report 96–4138, 136 p. Butler, D.L., and Osmundson, B.C., 2000, Physical, Chemical, and Biological Data for the Uncompahgre Project area and the Grand Valley, West-Central Colorado, 1993–98: U.S. Geological Survey Open-File Report 99–453, 216 p. Butler, D.L., 2001, Effects of Piping Irrigation Laterals on Selenium and Salt Loads in the Montrose Arroyo Basin, Western Colorado: U.S. Geological Survey Water-Resources Investigations Report 01–4204, 14 p. Butler, D.L., and Lieb, K. J., 2002, Characterization of Selenium in the Lower Gunnison River Basin, Colorado: U.S. Geological Survey Water-Resources Investigations Report 02–4151, 26 p. Environmental Protection Agency, 2008, EnviroFacts, http://oaspub.epa.gov/enviro/ef_home2.water. Ohlendorf, H. M., Hoffman, D. J., Saiki, M. K., and Aldrich, T. W., 1986, Embryonic Mortality and Abnormalities of Aquatic Birds – Apparent Impact of Selenium from Irrigation Drainwater: Science of the Total Environment, v. 52, p. 49-63. Ohlendorf, H. M., Kilness, A. W., Simmons, J. L., Stroud, R. K., Hoffman, D. J., and Moore, J. F., 1988, Selenium Toxicosis in Wild Aquatic Birds; Journal of Toxicology and Environmental Health, v. 24, p. 67-92. Thomas, J. C., Leib, K. J., and Mayo, J. W., 2007, Analysis of Dissolved Selenium Loading for Selected Sites in the Lower Gunnison River Basin, Colorado, 1978-2995: U.S. Geological Survey Scientific Investigations Report 2007-5287, 25 p. United States Environmental Protection Agency, 2003, Technically-based Local Limits Development Strategy, USEPA Region VIII, 132 p. Water Quality Control Division, 1998, Water Quality Limited Segments Still Requiring TMDLs, Colorado’s 1998 303(d) List and Related Water Quality Management Lists, 46 p.
January 2011
108
Final
Water Quality Control Division, U. S. Fish and Wildlife Service, and U. S. Environmental Protection Agency, 2005, Memorandum of Agreement among the Colorado Department of Public Health and Environment‟s Water Quality Control Division, U. S. Fish and Wildlife Service‟s Colorado Field Office and the U. S. Environmental Protection Agency Region 8‟s Ecosystem Protection and Water Programs regarding Enhanced Coordination in Implementing Colorado‟s Mixing Zone Rule/Implementation Guidance and the Service‟s August 11, 2003 Biological Opinion on this Matter, http://www.cdphe.state.co.us/wq/PermitsUnit/PolicyandGuidance/MOA_TandE.pdf, 11p. Water Quality Control Commission, 2006, Section 303(d) List Water-Quality-Limited Segments Requiring TMDLs, 5 CCR 1002-93, 19 p. Water Quality Control Division, 2007, Section 303(d) Listing Methodology – 2008 Listing Cycle, 33 p. Water Quality Control Commission, 2007, The Basic Standards and Methodologies for Surface Water, 5 CCR 1002-31, 174 p. Water Quality Control Commission, 2007b, Classifications and Numeric Standards for Gunnison and Lower Dolores River Basins, 5 CCR 1002-35, 82 p. Water Quality Control Commission, 2009, Colorado Discharge Permit System Regulations, 5 CCR 1002-61, 286 p. Wright, W.G., and Butler, D.L., 1993, Distribution and Mobilization of Dissolved Selenium in Ground Water of the Irrigated Grand and Uncompahgre Valleys, Western Colorado in Allen, R.G., and Neale, C.M.U., eds., Management of Irrigation and Drainage Systems—Integrated Perspectives: Proceedings of the 1993 National Conference on Irrigation and Drainage Engineering, Park City, Utah, July 21–23, 1993, American Society of Civil Engineers, New York, p. 770–777.
January 2011
109
Final
