Final, November 2012

TOTAL MAXIMUM DAILY LOAD ASSESSMENT FRUITGROWERS RESERVOIR COGULG09 Dissolved Oxygen DELTA COUNTY, COLORADO November 2012

Waterbody Description / WBID Pollutants Addressed Relevant Portion of Segment (as applicable) Use Classifications / Designation Water Quality Target

TMDL Goal

TMDL Summary Fruitgrowers Reservoir, COGULG09 Dissolved oxygen Fruitgrowers Reservoir

Aquatic Life Warm 2, Agriculture, April 1 to Oct. 31 Recreation E, Nov. 1 to March 31 Recreation N Segment 9 Dissolved 5.0 mg/L (minimum) Oxygen Total 0.043 mg/L Phosphorus Attainment of Aquatic Life use classification standards for DO.

Final TMDL, October 2012

Fruitgrowers Reservoir. Summer bloom EXECUTIVE SUMMARY Fruitgrowers Reservoir (FGR) is Lower Gunnison Segment 09 (COGULG09) and is located in Hart’s Basin below the Town of Cedaredge WWTP (Cedaredge) near Orchard City in Delta County, Colorado. FGR has been on the State’s 303(d) list of water-quality impaired waterbodies for nonattainment of water quality standards for Dissolved Oxygen since 2002, when it was given a high priority (Table 1). Low Dissolved Oxygen impairs the Aquatic Life Warm 2 classification for Segment 9. Low Dissolved Oxygen (hypoxia) is primarily the result of nutrient enrichment, also known as eutrophication.

Segment # Segment 9

Segment Description Fruitgrowers Reservoir, COGULG09

303(d) Listed Portion Contaminants Fruitgrowers DO Reservoir

Table 1. Segment within the Lower Gunnison watershed that appears on the 2002, 2004, 2006, 2008, 2010 303(d) list of impaired water bodies.

2

Final TMDL, October 2012

BACKGROUND: I. INTRODUCTION Section 303(d) of the federal Clean Water Act (CWA) requires states to periodically submit to the U.S. Environmental Protection Agency (EPA) a list of water bodies that are water-quality impaired. Water-quality limited segments are those water bodies for which one or more assigned use classifications or standards, the classification or standard is not fully achieved. This list of water bodies is referred to as the “303(d) List”. In Colorado, the agency responsible for developing the 303(d) list is the Water Quality Control Division (WQCD). The List is adopted by the Water Quality Control Commission (WQCC) as Regulation No. 93. The WQCC adopted the current 303(d) list in March of 2010. For waterbodies on the 303(d) list a Total Maximum Daily Load (TMDL) is used to determine the maximum amount of a pollutant that a waterbody may receive and still maintain the water quality standard. The TMDL is the sum of the Waste Load Allocation (WLA), which is the load from point source discharges, 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

Alternatively, a segment or pollutant may be removed from the list if the applicable standard is attained, if implementation of clean-up activities via 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 as the result of a Use Attainability Analysis (UAA), or other EPA approved recalculation method. Fruitgrowers Reservoir, Segment 9 of the Lower Gunnison sub-basin, is identified on the 2002, 2004, 2006, 2008 and 2010 303(d) list for exceeding the water quality standards for dissolved oxygen (Table 1) (WQCC, 2006b). The impairment status for designated uses in Fruitgrowers Reservoir is presented in Table 2. Date (Cycle Year) of Current Approved 303(d) list: 2010 WBID Segment Description Designated Uses & Impairment Status

COGULG09

Aquatic Life Warm 2: Impaired Recreation E: Impaired Agriculture: Impaired

Fruitgrowers Reservoir

Table 2. Designated uses and impairment status for Segment 09, Fruitgrowers Reservoir

3

Final TMDL, October 2012 II. SEGMENT DESCRIPTIONS Geographical Extent Fruitgrowers Reservoir is Segment 9 of the Lower Gunnison sub-basin (COGULG09), Hydrologic Unit Code (HUC) 14020005. FGR was built in Hart’s Basin, Delta County, Colorado, which makes up the reservoir’s surrounding watershed. The reservoir’s watershed is approximately 12 square miles (7680 acres). The reservoir surface area is 437 acres. The spillway is at an elevation of 5485 feet (USBOR Fruitgrowers Reservoir 1998 survey). FGR is filled with water imported from Surface Creek via the Alfalfa Ditch diversion. It also receives water imported from other creeks via irrigation diversions. Hart’s Basin is sparsely populated but is affected by agriculture, runoff, and domestic wastewater effluent from Cedaredge via the Cedaredge WWTP. The drainage area of the FGR watershed is 12 square miles (7680 acres). The reservoir surface area is 437 acres and an elevation of 5485 feet. The maximum storage is 4540 Ac-Ft (1998 survey). The long-term mean annual precipitation is approximately 11 inches (NRCS data from original 1998 project). The predominant source of water to the reservoir is imported from other basins. Nutrient pollution results from a combination of both natural and anthropogenic sources, heavily dominated by agricultural practices and domestic wastewater effluent from Cedaredge. Landuse within FGR’s watershed is primarily agricultural. Approximately 80% of the area is irrigated. Approximately 95% of the irrigated acreage is grass and hay, and approximately 2% is corn (Randy Kramer, Delta Conservation Service, Pers. Comm., August 2010). The remaining 3% is undefined for this report. The National Pollutant Discharge Elimination System (NPDES) module of the Compliance Information System (ICIS) tracks surface water permits issued under the Clean Water Act. Under NPDES, all facilities that discharge pollutants from any point source into waters of the United States are required to obtain a permit. The permit will likely contain limits on what can be discharged, impose monitoring and reporting requirements, and include other provisions to ensure that the discharge does not adversely affect water quality The Town of Cedaredge (Permit # CO0031984) is the only NPDES permitted facility in the Fruitgrowers watershed. This facility uses aerated lagoons for wastewater treatment. The current design capacity of Cedaredge is 0.218 MGD (0.34 cfs). Proposed seasonal flows are 0.275 MGD (0.43 cfs) from April1 through October 31, and 0.26 MGD (0.40 cfs) from November 1 through March 31 (WQCD Cedaredge WWTF Water Quality Assessment, 2005). For the past 5 years (2005-2010), the permit requirements have included quarterly monitoring of total phosphorus in the facility’s effluent.

4

Final TMDL, October 2012 Background/History FGR is an irrigation reservoir located in Delta County, Colorado. The Reservoir serves as an impoundment for water from Surface Creek and Alfalfa Ditch for the purpose of providing irrigation water for the Orchard City Irrigation District. Surface Creek via Alfalfa Ditch serves as the major source of water to FGR. The initial Fruitgrowers Dam was constructed in 1904. Water was impounded from Alfalfa Run. In 1935, Alfalfa Ditch was completed for diversion of Surface Creek water to Alfalfa Ditch upstream of the reservoir. The dam failed in 1937 causing extensive property damage in Austin, approximately 2 miles downstream from FGR. In 1938, the Bureau of Reclamation (BOR) began to reconstruct the Fruitgrowers Dam. The dam was completed in 1939, and impounded 437 surface acres when completely filled. In 1940, the Orchard City Irrigation District assumed the operation and maintenance of the project. In the 1950s, FGR was a popular recreational area. Local residents enjoyed swimming, boating, and fishing. The City of Cedaredge constructed a wastewater treatment plant in 1978 that discharged into Alfalfa Ditch. Users of the reservoir began noticing a decline in the water quality soon thereafter. In 1984, the Delta County Health Department began receiving complaints from reservoir users regarding disagreeable odors, potential health problems following water contact activities, and fish kills. In 1989 the Delta County Health Department closed FGR to all water contact activities after testing indicated coliform bacterial counts exceeded state water quality standards for natural swimming areas. A year later, FGR was drawn down to very low levels, and the game fish died. In 1991, the BOR agreed to develop a wildlife area, and currently the reservoir is a popular area for bird watching and is a migration flyway for many bird species, including Sandhill Cranes. In 1996, a break in one of the treatment lagoons spilled raw sewage into FGR. This lead to inclusion of FGR on the 1998 303(d) List for fecal coliform and NH3. Subsequently, in 2002, the fecal coliform and NH3 listings were dropped when monitoring results indicated both standards were attained. However, this monitoring also documented non-attainment of the Dissolved Oxygen criteria. This was the catalyst that brought the community together to talk about water quality issues affecting FGR. Aerial images of the Fruitgrowers study area are shown in different resolutions in Figure 1 and 2. A map of associated sampling sites is shown in Figure 3. Table 3 lists the site codes and descriptions.

5

Final TMDL, October 2012

Figure 1. Location of Fruitgrowers Reservoir in the Gunnison Basin, Colorado.

Figure 2. Aerial imagery of the sub-watershed for Fruitgrowers Reservoir.

6

Final TMDL, October 2012 Figure 3. Fruitgrowers Reservoir watershed sampling sites.

Table 2. Fruitgrowers Reservoir Watershed Sample Locations. AD-1 AD-2 AD-2A AD-2B AD-3 AD-4 AD-5 AD-6 BD-1A BD-2 TD-1 PG-1 PG-2 CMD-0 CMD-2

Alfalfa Ditch at reservoir outlet. Alfalfa Ditch at reservoir inlet at culvert under Road N. Alfalfa Ditch ¾ to 1 mile upstream of confluence with Fruitgrower’s Reservoir at Western States Ranch property. Alfalfa Ditch at Western States Ranch upstream of CMD-0 confluence. Alfalfa Ditch downstream of confluence with WWTP effluent. WWTP effluent upstream of confluence with Alfalfa Ditch. Alfalfa Ditch upstream of confluence with WWTP effluent. Alfalfa Ditch at headgate downstream of confluence with Surface Creek. Burgess Ditch about 150 meters upstream of confluence with Fruitgrower’s Reservor. Burgess Ditch at County Road N crossing. Transfer Ditch at gaging station. Posin Gulch upstream of confluence with Cedar Mesa Ditch. Western (left) fork of Posin Gulch upstream of confluence with Eastern fork. Cedar Mesa Ditch about 25 meters upstream of confluence with Alfalfa Ditch. Cedar Mesa Ditch upstream Posin Gulch.

III. WATER QUALITY STANDARDS 7

Final TMDL, October 2012 Standards Framework Waterbodies in Colorado are divided into discrete units or segments. The Colorado Basic Standards and Methodologies for Surface Water, Regulation 31(WQCC 2006b), 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. 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 relevant standard for the segment addressed in this document is a Table Value Standard (TVS) for dissolved oxygen. FGR is 303(d) listed for non-attainment of the Aquatic Life Warm 2 Use-based dissolved oxygen standard, which is 5.0 mg/L (minimum). Therefore, the pollutant of concern for this TMDL is dissolved oxygen. For lakes, exceedance of the dissolved oxygen standard is the result of eutrophication (nutrient enrichment), and therefore, all TMDLs for dissolved oxygen are nutrient (total phosphorus) TMDLs. Total phosphorus targets will be identified that, when implemented, will result in attainment of dissolved oxygen standards. Setting TMDL targets is discussed in a following section. The use classifications applied to FGR are Aquatic Life Warm 2, Recreation E (seasonally), and Agriculture. Water quality standards assigned to protect these uses are 8

Final TMDL, October 2012 identified in Colorado Basic Standards and Methodologies for Surface Water, Regulation 31 (WQCC 2010) and in Regulation 35 - Classifications and Numeric Standards for Gunnison and Lower Dolores River Basins (amended 2/8/10, effective 6/30/10). FGR exceeds the Aquatic Life Use-based dissolved oxygen standards intended to protect against short-term, acute conditions (acute) and longer-term, sub-lethal (chronic) effects. Although dissolved oxygen standards apply to all uses assigned to this segment, the Aquatic Life Use is the most stringent standard and is the focus of this TMDL. FGR does not attain the dissolved oxygen standards for any assigned uses (Tables 4 and 5) and attainment of the most stringent standard will protect all uses. Table 4. Ambient water quality criteria and status for Segment 09, Fruitgrowers Reservoir Water Quality Criteria for Impaired Designated Uses WBID Impaired Designated Use Applicable Water Quality Criteria and Status COGULG09 Aquatic Life Warm 2 Dissolved Oxygen (1) / Not Attained COGULG09 Recreation Dissolved Oxygen (1) / Not Attained COGULG09 Agriculture Dissolved Oxygen (1) / Not Attained Applicable State or Federal Regulations: (1) Classifications and Numeric Standards for the Gunnison and Lower Dolores River Basin (Regulation 35)

Table 5. Use-Based table value standards for Dissolved Oxygen for 303(d) listed segment FGR.

Dissolved Oxygen

mg/L

Aquatic Life Warm 2

Recreation

Agriculture

5

3

3

Many water quality standards are applied as chronic and acute standards. Typically, this refers to the application of these standards for protection of the Aquatic Life Use: acute standards protect against short-term acute (lethal) exposure to a pollutant while chronic standards protect against longer term sub-lethal exposure to low levels of a pollutant. The terms chronic and acute do not apply to the dissolved oxygen standard, as there is not really toxicity implied from low dissolved oxygen. However the Aquatic Life Use dissolved oxygen standard does protect against acute lethal events, as well as low levels of dissolved oxygen which, while not lethal, might affect spawning as well as optimal growth of aquatic life. The dissolved oxygen standards assigned to the other Uses (Agriculture, Recreation and Water Supply) are to protect against anoxic conditions. Note that the Water Supply Use is not assigned to FGR. Several statements on dissolved oxygen standards found in Regulation 31 Table 1 9

Final TMDL, October 2012 Footnotes are important for understanding the standard impairment for this TMDL. These include Footnotes 1 and 9. Table I – Footnotes 1.) Standards for dissolved oxygen are minima, unless specified otherwise. For the purposes of permitting, dissolved oxygen may be modeled for average conditions of temperature and flow for the worst case time period. Where dissolved oxygen levels less than these levels occur naturally, a discharge shall not cause a further reduction in dissolved oxygen in receiving water. (For lakes, also see footnote 9.) 9.) The dissolved oxygen standard applies to lakes and reservoirs as follows. a. Recreation: In the upper portion of a lake or reservoir, dissolved oxygen shall not be less than the criteria in Table 1 or the applicable site-specific standard. In the lower portion of a lake or reservoir, dissolved oxygen may be less than the applicable standard except where a site-specific standard has been adopted. A site-specific dissolved oxygen standard will be established for the lower portion of a lake or reservoir where there is evidence that primary contact occurs within the lower portion. b. Agriculture: In the upper portion of a lake or reservoir, dissolved oxygen shall not be less than the criteria in Table 1 or the applicable site-specific standard. In the lower portion of a lake or reservoir, dissolved oxygen may be less than the applicable standard except where a site-specific standard has been adopted. A site-specific dissolved oxygen standard will be established for the lower portion of a lake or reservoir where there is evidence that livestock watering or irrigation water is pumped from the lower portion. c. Aquatic Life: In the upper portion of a lake or reservoir, dissolved oxygen shall not be less than the criteria in Table 1 or the applicable site-specific standard. In the lower portion of a lake or reservoir, dissolved oxygen may be less than the applicable standard except where footnote 5(c)(iii) applies or a sitespecific standard has been adopted. A site-specific dissolved oxygen standard will be established for the lower portion of a lake or reservoir where the expected aquatic community has habitat requirements within the lower portion. i. Fall turnover exclusion: Dissolved oxygen may drop 1 mg/l below the criteria in Table 1 in the upper portion of a lake or reservoir for up to seven consecutive days during fall turnover provided that profile measurements are taken at a consistent location within the lake or reservoir 7-days before, and 7-days after the profile with low dissolved oxygen. The profile measurements taken before and after the profile with low dissolved oxygen must attain the criteria in Table 1 in the upper portion of the lake or reservoir. The fall turnover exclusion does not apply to lakes or reservoirs with fish species that spawn in the fall unless there are data to show that adequate dissolved oxygen is maintained in all spawning areas, for the entire duration of fall turnover. d. Water Supply: The dissolved oxygen criteria is intended to apply to the epilmnion and metalimnion strata of lakes and reservoirs. Dissolved oxygen in the 10

Final TMDL, October 2012 hypolimnion may, due to the natural conditions, be less than the table criteria. No reductions in dissolved oxygen levels due to controllable sources is allowed. Additional language clarifying the application of the DO standard is found in Regulation 31 in the Statement of Basis and Purpose language: 31.48 STATEMENT OF BASIS, SPECIFIC STATUTORY AUTHORITY AND PURPOSE; JUNE 7-8, 2010 RULEMAKING; FINAL ACTION AUGUST 9, 2010; EFFECTIVE DATE JANUARY 1, 2011. In summary, the dissolved oxygen standards apply as minima to individual profiles. If more than one profile is measured on a lake on a particular sampling day, each profile will be assessed independently and not averaged together. Furthermore, for a lake or reservoir greater than 5 meters deep, the dissolved oxygen in the upper portion of each profile, of the lake or reservoir is generally assessed as the average of all measurements from 0.5 meters to 2.0 meters. For a lake or reservoir less than 5 meters deep, the dissolved oxygen in the upper portion of each profile is assessed as the average of all measurements from 0.5 meters to a depth equal to 40% of the total depth. The specific numeric standards assigned to the listed reservoir segment are contained in REGULATION NO. 35 CLASSIFICATIONS AND NUMERIC STANDARDS FOR GUNNISON AND LOWER DOLORES RIVER BASINS (5 CCR 1002-35) (WQCC, 2010, Section 35.6(4)). All remaining assigned numeric standards associated with Aquatic Life, Recreational and Agricultural Use Classifications are attained.

IV. PROBLEM IDENTIFICATION FGR does not attain the Aquatic Life Use-Based dissolved oxygen standard. Fruitgrowers exhibits hypoxia (low dissolved oxygen) and anoxia (dissolved oxygen depletion). This impairment is attributed to the extremely eutrophic conditions in the reservoir. Eutrophic conditions result from nutrient enrichment of the water. Nutrients (specifically phosphorus) fertilize the lake and cause algae to grow, which increases the biomass in the lake. The algae die and the resulting organic material decays, exerting an oxygen demand that results in depressed levels of dissolved oxygen. Although the algal photosynthesis produces oxygen during the day, at night the algal respiration consumes oxygen, further exacerbating the oxygen demand. The key to reducing the oxygen demand is to reduce algal production through control (reduction) of total phosphorus. Contributions of nutrients to the Fruitgrowers Reservoir watershed include background nutrients in imported water, wastewater effluent, and nonpoint source runoff. There is one permitted discharger in the basin upstream from FGR. Anecdotal information indicates that water quality in FGR declined rapidly after the Cedaredge Wastewater Treatment Plant began discharging into Alfalfa Ditch a short distance upstream from the reservoir. The point and nonpoint sources of phosphorus must be 11

Final TMDL, October 2012 reduced to attain the dissolved oxygen standard in the reservoir. A two year study by the EPA Region 8 and WQCD on nutrients, bacteria, metals and pesticides in the watershed and the reservoir was conducted during 1997-98. Data from this study are used for the development of this total phosphorus TMDL. A mass balance approach was used to develop the existing nutrient budget and to identify and quantify point source and nonpoint sources of phosphorus. Modeling of the watershed and reservoir is used to develop the TMDL nutrient budget and to identify point source and non-point source reductions of phosphorus.

12

Final TMDL, October 2012

Water Quality Target/Expected Condition: The water quality goal for Fruitgrowers Reservoir is attainment of the Aquatic Life Warm 2 use classification standard for dissolved oxygen. Attainment of this standard will result in attainment of the Recreation and Agriculture use standards. In order to attain the dissolved oxygen standard, a target must be set to address the underlying cause of the low dissolved oxygen. Eutrophication TMDLs Low dissolved oxygen is a symptom of eutrophic conditions. Eutrophication results from excess nutrient enrichment. Low dissolved oxygen results from oxygen demands such as organismal respiration and the microbial decomposition of organic matter. As a lake receives larger nutrient loads and becomes enriched, more algal biomass, as measured by chlorophyll a, is produced. The increased algal biomass is also organic matter that is decomposed when the algae die and sink to the lake bottom. The decomposition process exerts an oxygen demand in the water. Other oxygen demands are due to algal respiration at night or when algae are in a light-limited environment. Although algae produce oxygen through photosynthesis during the day, at night or in light-limited conditions the algae merely respire (use oxygen). Reduction in nutrients, specifically phosphorus will result in reduced algal biomass and algal productivity. It follows that less biomass will result in less organic matter subjected to decomposition. All of these result in reduced respiration and therefore diminished oxygen demand. Because low dissolved oxygen is a symptom of eutrophication and is clearly related to excess biomass which is linked to nutrients, this TMDL targets reduction of phosphorus to reduce algal biomass in order to attain the dissolved oxygen standard. Ultimately, the intent is to identify an appropriate level of chlorophyll a and link that to a concentration of total phosphorus. With very few exceptions, Colorado does not have statewide nutrient standards, and until such are adopted, 303(d) lake listings for exceedances of dissolved oxygen standards (eutrophication indicators) will be written as eutrophication TMDLs and will have nutrient (TP) and chlorophyll targets. The term “nutrients” refers to nitrogen as well as phosphorus, both of which can limit algal production. However, it is not important from a lake-management perspective to determine the limiting factor/nutrient for the listed lake. Although lakes may be nitrogen limited, the nitrogen limitation usually is due to an excess of phosphorus. Also, phosphorus is the focus because nitrogen mitigation generally is not effective. It is generally accepted in the field of limnology that controlling (reducing) nitrogen encourages growth of Cyanobacteria (blue-green algae). Because many species of Cyanobacteria can fix nitrogen, these algae can relieve their nitrogen-limited condition and can outcompete other algae. Algal communities dominated by Cyanobacteria are undesirable for a number of reasons, but primarily because these algae may produce toxins. Controlling nitrogen could encourage increases in Cyanobacteria and result in 13

Final TMDL, October 2012 additional problems. Finally, phosphorus typically is easier to control. Therefore, in Colorado, eutrophication TMDLs will focus on reduction of phosphorus. Setting a TMDL target for Fruitgrowers Reservoir As mentioned previously, the State of Colorado currently does not have statewide nutrient standards, which makes setting chlorophyll and total phosphorus targets for TMDLs somewhat challenging. Ideally attainment of dissolved oxygen would be linked to a chlorophyll target which would be linked to a phosphorus target. We also can link the DO to total phosphorus and by-pass the chlorophyll link. In this section, we examine several approaches. Several lines of evidence were used to determine the chlorophyll a and total phosphorus TMDL targets, and the required total phosphorus reductions. Approaches used for identifying a chlorophyll target and related total phosphorus target are discussed below. 1. WQCD proposed interim values for total phosphorus and chlorophyll. 2. Response ratios (chlorophyll:total phosphorus) 3. Compare to other Colorado lakes 4. EPA 304(a) Criteria Recommendations 1. One approach would be to use the Water Quality Control Division’s proposed interim nutrient values for the TMDL targets. It must be emphasized that these have not been adopted by the Water Quality Control Commission as criteria, let alone as water quality standards. As of the date of this TMDL (November 2011) these proposed interim numeric values, were identified by the Water Quality Control Division as thresholds above which excessive algal abundance has an adverse impact on uses while still optimizing use support. Because these proposed interim values would be protective of Aquatic life uses, we assume in this TMDL that these levels will dovetail with other Aquatic Life Usebased standards related to eutrophication (dissolved oxygen and pH standards), and thus represent a reasonable target for Fruitgrowers Reservoir. Colorado’s proposed interim nutrient values for lakes, as of November 2011 are shown in Table 6. Classification Cold Warm

Recreation2 Chlorophyll (ug/L) 20 30

Chlorophyll (ug/L) 8 20

Aquatic Life1 Total P (ug/L) 20 80

Total N (ug/L) 410 850

1 – summer (July1-September 30) average Total Phosphorus (ug/L) in the mixed layer of lakes (median of multiple depths), allowable exceedance frequency 1-in-5 years. 2 – March-November average chlorophyll (ug/L) in the mixed layer of lakes (median of multiple depths), allowable exceedance frequency 1-in-5 years.

Table 6. Colorado proposed interim nutrient values for lakes. 2. Response ratios (chlorophyll:total phosphorus) for FGR also can be examined. Chlorophyll represents the algal biomass or biological manifestation of the nutrients in a lake. Although there is clearly a relationship between total phosphorus and algal production in lakes, when nutrients are extremely high, the relationship of these within a 14

Final TMDL, October 2012 particular lake is not always readily apparent using regression techniques. Often a regression relationship is masked by exceedingly high nutrient (total phosphorus) concentrations as well as other factors including light limitation, and limitation by other nutrients, etc. As a result, the chlorophyll concentrations are not as high as would be expected from the concentration of total phosphorus (Hern, et al., 1981). Within a particular lake, the ratio of chlorophyll to total phosphorus is expected to be relatively consistent. This ratio, termed response ratio (RA) quantifies the algal biomass per unit of total phosphorus, and can be used to predict the lake’s algal response to management efforts to reduce total phosphorus (Hern, et al., 1981). In the case of TMDLs, the RA is used to back-calculate the total phosphorus concentration related to the desired chlorophyll concentration. For FGR, the RA was calculated as the average annual growing season mean chlorophyll to average annual growing season mean total phosphorus. The RA was 0.60. Using the WQCD proposed interim value with this response ratio of 0.60, for the chlorophyll criterion of 20 ug/L, total phosphorus is estimated/expected to be 33 ug/L.

3. Another approach to setting a TMDL target is to compare FGR to other Colorado lakes with similar characteristics. The data record for FGR is limited to only a few years of monitoring and includes only one complete year of watershed monitoring and growing season data. Dissolved oxygen values are consistently low in FGR and all of the chlorophyll and total phosphorus data from FGR are extremely high. The FGR data are out of the range of what are likely levels for attaining the dissolved oxygen standards and therefore are well out of the desired range for modeling. However, data from other Colorado lakes similar to FGR show attainment of the dissolved oxygen standard and the concomitant chlorophyll and total phosphorus data are generally in a lower range than those for FGR. Therefore, it is appropriate to compare the reservoir to other similar Colorado lakes and pool data from the comparable lakes to determine chlorophyll and total phosphorus concentrations observed when the dissolved oxygen standard is attained. FGR is a warm water reservoir with a median depth of 7 meters. FGR stratifies intermittently. Colorado warm water lakes were grouped as those that stratify persistently, stratify intermittently, or do not stratify. Of the group that stratifies intermittently, the comparison was narrowed to those that were in the range of 7-9 meters median depth. The lakes used were: Boulder Reservoir, Cherry Creek Reservoir, Fruitgrowers Reservoir, Lonetree Reservoir, Loveland Reservoir, Totten Reservoir, Puett Reservoir, Quincy Reservoir, Sweitzer Reservoir and Union Reservoir. These 10 lakes yielded 365 growing season sampling events. Scatter plots of dissolved oxygen vs. chlorophyll, and chlorophyll vs. total phosphorus were examined for relationships between the variables. Regressions were not strong; however thresholds were observed that represent potential targets for this TMDL. In determining thresholds for attaining the dissolved oxygen standards, data for 15

Final TMDL, October 2012 individual sampling events for all of the selected lakes were used because dissolved oxygen attainment is assessed based on individual profiles. The analysis was limited to sampling events during the growing season of July through September, as this typically is the critical period for Colorado lakes. As described in an earlier section, the Division’s assessment approach for dissolved oxygen in lakes is to evaluate the upper layer defined as the average of profile measurements from 0.5 meters to 2.0 meters for lakes 5 meters or greater in depth. Using this assessment methodology, dissolved oxygen was plotted against chlorophyll in order to identify a maximum chlorophyll concentration for which the dissolved oxygen standard of 5.0 mg/L was attained. The regression was not strong. However by close examination of the lower left side of the plot (by fixing the maximum values for the X and Y axes), it is observed that below 25 ug/L chlorophyll, dissolved oxygen rarely falls below the standard of 5.0. (Figure 4) A scatter plot was also examined of dissolved oxygen vs. chlorophyll where dissolved oxygen was calculated for a deeper interval. The profile average was based on measurements from 0.5 meters to 3.0 meters (40% of the median depth). Again, dissolved oxygen was not exceeded when chlorophyll was below 25 ug/L (Figure 5).

Figure 4. Dissolved oxygen (0.5- 2.0 m interval) vs. chlorophyll.

16

Final TMDL, October 2012

Figure 5. Dissolved oxygen (0.5- 3.0 m interval) vs. chlorophyll.

Although this chlorophyll threshold (25 ug/L) is somewhat (25%) higher than the WQCD’s proposed interim value for chlorophyll a (20 ug/L for warm lakes), it should be noted that the WQCD proposed chlorophyll value is the 80th percentile of summer averages, while the 25 ug/L identified above represents an instantaneous maximum threshold. The proposed interim value is based on summer averages rather than individual samples within one season. Therefore more than one year of data would be required to evaluate attainment of the interim values. Using the 80th percentile allows an exceedance frequency of 1 in 5 years for the interim nutrient values. Note that there is no exceedance frequency associated with the dissolved oxygen standard. If a profile exceeds the standard, the lake is out of attainment. To be consistent with the WQCD proposed approach, the WQCD proposed interim value of 20 ug/L chlorophyll as 80th percentile of summer average is selected for this TMDL’s chlorophyll target. The median seasonal average chlorophyll associated with the 80th percentile for the seasonal chlorophyll averages in the data set used for this TMDL is approximately15 ug/L. The median represents what the typical summer average would be. In conclusion, the TMDL chlorophyll target for Fruitgrowers will be 20 ug/l chlorophyll as an 80th percentile of summer averages. This target may be refined in the future, if necessary, when state nutrient standards are adopted. Several approaches were used to relate a total phosphorus concentration to the 20 ug/L chlorophyll target. One approach was to use an inverse regression in which summer average total phosphorus was regressed on summer average chlorophyll. The power 17

Final TMDL, October 2012 equation for the regression predicts 61 ug/L total phosphorus for the target 20 ug/L chlorophyll (as 80th percentiles). Using the same relationship but assuming the median chlorophyll value of 15 results in a TP value of 50 ug/L. The regression was relatively strong with an r2=0.8667 (Fig 6).

Figure 6. Inverse regression of Total Phosphorus vs Chlorophyll

Another approach to setting the TP target, the Dillon-Rigler model (1974), was also considered. Chlorophyll a = 0.0731* TP1.449 (ug/L) This model is widely used and was developed by correlating summer chlorophyll with spring total phosphorus from many North American lakes. Although data from individual samples from an individual lake would not be expected to fit the model line exactly, it is useful for supporting the development of a total phosphorus target. In this case, the FGR total phosphorus data used are summer (July-September) values rather than spring values. As expected, Fruitgrowers Reservoir individual sample data do not fit the DillonRigler line (calculated from observed total phosphorus data from FGR) (Figure 7). However, the growing season averages for 1997 and 1998 for Fruitgrowers Reservoir are close to the Dillon-Rigler line. The Dillon-Rigler model was used to back-calculate a TP value for 20 ug/L chlorophyll a and for 15 ug/L chlorophyll a. The model result for 20 ug/L chlorophyll a is 48 ug/L and 41 ug/L TP, respectively. These estimates seem reasonable for an uncalibrated model, althought there is uncertainty (unquantified) because there are only two growing season averages for FGR. 18

Final TMDL, October 2012

19

Final TMDL, October 2012 Figure 7. Fruitgrowers Reservoir Chlorophyll a vs TP: Individual sample scattergram plotted with Dillon-Rigler model line.

Finally, the data set for intermittently stratified lakes was used to determine the median (typical) and 80th percentile for summer averages for each lake for chlorophyll and total phosphorus. These values were used to relate phosphorus to chlorophyll in order to calculate total phosphorus concentrations related to the chlorophyll targets (15 and 20 ug/L ). The data set was made up of 53 lake summer averages for 10 lakes. Using this approach, the median (or typical) summer average total phosphorus related to the 80th percentile chlorophyll of 20 ug/L was approximately 43 ug/L.

Another approach for setting targets for this TMDL could be to use EPA 304(a) nutrient criteria recommendations. EPA published nutrient criteria development guidance and included 304(a) criteria recommendations. Recommendations for nutrient criteria levels were based on Level III Ecoregions (EPA, 2001). FGR is in the Colorado Plateau Subecoretion (20). Tables 7 and 8 illustrate the EPA nutrient levels based on Aggregate Nutrient Ecoregions and Level 3 Ecoregions in Colorado, respectively. Table 7. Aggregate Nutrient Ecoregion II. Western Forested Mountains III. Xeric West IV. Great Plains Grass and Shrublands V. South Central Cultivated Great Plains

TP, mg/L

Chl a, ug/L

0.009

1.9

0.017 0.020

3.4 2.0

0.033

2.3

20

Final TMDL, October 2012 Table 8. Level 3 Ecoregion 18: Wyoming Basin 21: Southern Rockies 20: Colorado Plateaus 22: Arizona/New Mexico Plateau 25: Western High Plains 26: Southwestern Tablelands

TP, mg/L

Chl a, ug/L

0.010 0.015 0.003 0.015

1.4 1.7 1.4 2.0

0.024 0.020

2.4 1.2

The nutrient levels for this subecoregion are: Chlorophyll a = 1.4 ug/L Total Phosphorus = 3.0 ug/L Although Colorado opted to develop its own nutrient criteria rather than using EPA’s recommended criteria, it is worth including these values in this discussion to fully illustrate the range of possible nutrient targets for FGR. Discussion of Colorado’s rationale for not using the EPA values can be found in Colorado’s Nutrient Criteria Workgroup documents. The results from the approaches described above are reasonable given the limited data and the uncertainty inherent in these approaches. The models used to calculate loading to the reservoirs are based on typical conditions. Therefore, the lake and watershed modeling will be based on the median total phosphorus related to the chlorophyll target.

The TP estimates from the various approaches described above range from 33 ug/L to 82 ug/L. Note that these values are as 80th percentile, medians and averages. The preferred approach for selecting the target for this TMDL is based on using the data set of intermittently stratifying Colorado lakes, as this best represents a site-specific target for FGR. Therefore, the TMDL target is the 43 ug/L TP as the median of summer averages. In summary, the chlorophyll target selected for this TMDL is 20 ug/L as an 80th percentile of summer averages. The total phosphorus target is 43 ug/L as a median of summer averages. Implicit in these targets is a seasonal component. Also implicit is the need for a data period of record of at least 5 years.

21

Final TMDL, October 2012 V. WATER-QUALITY GOAL AND TARGET The water quality goal for the 303(d)-listed FGR segment is attainment of the Aquatic Life Warm 2 use classification standard for dissolved oxygen, through attainment of the defined targets for total phosphorus (43 ug/L as median of summer averages) and chlorophyll (20 ug/L as 80th percentile of summer averages). Implicit in these targets is a seasonal component. Also implicit is the need for a data period of record of at least 5 years.

VI. INSTREAM CONDITIONS Water Quality Studies on Fruitgrowers Reservoir Beginning in 1997 through 1998, EPA and WQCD conducted an intensive study on Fruitgrowers Reservoir and its watershed. The most frequent sampling (nearly monthly) occurred during the 1998 water year between November 1997 and October 1998, although a few samples were collected before and after that period. The WQCD conducted additional reservoir sampling, once in 1999 and once in 2000. Because the project is no longer active, no watershed and reservoir samples have been collected since approximately 2001. The objectives of the study, relevant to this TMDL report, were to characterize the water quality of FGR, and to identify and quantify the watershed sources of nutrients which contribute to the eutrophication of FGR. FGR receives water from diversions from Surface Creek and Cedar Mesa Ditch via Alfalfa Ditch, and Dry Creek, via Transfer Ditch. The sampling design included several stations along Alfalfa Ditch selected to bracket potential nutrient sources to Alfalfa Ditch upstream of FGR. The potential sources include a golf course, Town of Cedaredge WWTP, and Cedar Mesa Ditch. The water budget along Alfalfa Ditch did not balance, so it was not possible to determine contributions from specific nonpoint sources along Alfalfa Ditch such as the golf course. However, enough information exists to develop a water loading budget and nutrient loading budget for FGR. Additional sampling stations were set on Burgess Ditch, and Posin Gulch. The WQCD initially collected reservoir samples from 3 sites along each of 4 transects with the intent of studying the heterogeneity of the reservoir. However, this effort was not continued throughout the study. Currently the WQCD typically collects and assesses reservoir data from one site near the dam of a reservoir. Therefore, for development of this TMDL, only the center dam site is used. The sample sites are shown on the map in Figure 3. The Burgess Ditch flows through a property that is a biosolids application site. Stations on Burgess Ditch bracketed (were above and below) the biosolids application site to evaluate nutrient contributions from the biosolids. Not enough data were collected to determine the load contributions from the biosolids. However, Burgess Ditch nutrient 22

Final TMDL, October 2012 loading to FGR are represented by the data for the sampling station closest to FGR. Posin Gulch is tributary to Cedar Mesa Ditch. Nutrient and water contributions from Posin Gulch are accounted for in the Cedar Mesa Ditch data. Posin Gulch stations were included to balance phosphorus loads along Cedar Mesa Ditch. For the reservoir, 3 sampling sites were set along 4 transects. The intent of sampling at many sites was to characterize spatial variability in the reservoir, and to support modeling of internal phosphorus loading. Because of water releases, the size of the reservoir decreased during the summer, and sampling of transects was not continued. Currently the WQCD typically collects and assesses reservoir data from one site near the dam of a reservoir. To maintain consistency with WQCD’s approach, data for the center dam site are used for development of this TMDL. The modeling approach used for internal loading does not require multiple sampling sites in-lake. The following sections explain the hydrology (water load) and water quality of the FGR system. Hydrology The hydrology of FGR is highly managed. Most of the water in the reservoir is imported from outside of Hart’s Basin, primarily through Surface Creek diversions to Alfalfa Ditch. As the predominant source of water to the reservoir, the hydrograph of Alfalfa Ditch at the diversion from Surface Creek adequately illustrates the pattern of water loading to FGR (Figure 7). The reservoir also receives water from other tributaries, ditches, and Cedaredge, as well as snowmelt and other precipitation events. Average annual precipitation from 1994-1998 was approximately 14 inches (NRCS). A volume-balance approach was used to develop the water load to FGR. This approach produced calculated inflows to the reservoir based on outflow estimates provided by Orchard City Irrigation District (OCID) and U.S. Bureau of Reclamation (USBOR), change in storage, precipitation and evaporation. Flow measurements were available for the water quality sampling sites during the 1997-1998 study. Gage records for Surface Creek Diversion to Alfalfa Ditch were obtained from the Colorado Decision Support System (CDSS). Flows for other sources, Transfer Ditch, Cedar Mesa Ditch, Cedaredge and Burgess Ditch, were measured for each sampling event during the 19971998 study period. However, long-term records were not available. Additional data are available for Cedaredge from their Discharge Monthly Reports (DMR). Figure 8 compares the hydrograph resulting from Surface Creek Diversion to Alfalfa Ditch and the calculated inflow. Although the calculated inflow data differ from the Surface Creek Diversion data because the calculated inflow includes all sources of water to the Reservoir (ungaged runoff, Transfer Ditch, Cedar Mesa Ditch, Burgess Ditch, Cedaredge, and precipitation), the pattern is representative of annual water loading to FGR.

23

Final TMDL, October 2012 Figure 8. Hydrographs for Fruitgrowers Reservoir water sources to Fruitgrowers Reservoir, WY1998.

Not all of the water diverted from Surface Creek shows up in the reservoir right away. There are diversions from Alfalfa Ditch between the Surface Creek Diversion and the reservoir. While these are not accounted for in this water budget, adequate data from identified sources were available to develop the water budget for the reservoir. Due to the highly managed water rights system, the pattern of diversions and filling of the reservoir are fairly regular (Figure 9). A longer-term hydrograph illustrates the repetitive pattern. Very little water is diverted during the winter. The amount of water that is diverted increases in the spring as the reservoir is filled and more water is available from snow-melt. Once the reservoir is filled, the diversions decrease and remain relatively constant in the summer. In the fall, diversions decrease to minimal amounts again.

24

Final TMDL, October 2012 Figure 9. Long-term (5-year) hydrograph for Alfalfa Ditch.

Figures 10 illustrates 5-year patterns for reservoir storage from 1994-1998 and from 2005-2009. Data for years between these were not readily available. However, this figure clearly illustrates that patterns of storage are mostly consistent over time. Figure 11 illustrates the 5-year pattern for surface area. The reservoir is filled to approximately the same level each year, but there is some variability in the amount of drawdown.

25

Final TMDL, October 2012 Figure 10. Fruitgrowers Reservoir storage (AF)

Figure 11. Fruitgrowers Reservoir surface area.

Another source of water to FGR is the Town of Cedaredge. Effluent flows presented in Figure 12 illustrate little change in this source over time. Although Cedaredge has proposed an increase in the design capacity of the plant since the 1998 study, the quantity of discharged water has not increased to any appreciable extent. The current design capacity is 0.218 MGD. The proposed design capacity is seasonal at 0.275 MGD (April 1-Octtober 31) and 0.26 MGD (November 1-March 31). .

26

Final TMDL, October 2012 Figure 12. Effluent flows from Town of Cedaredge WWTP

The regular patterns in the hydrologic loading to FGR are emphasized because the hydrologic budget is an essential underlying component for modeling the reservoir. The data for this TMDL is more than 10 years old. There has been little change (development and increase in population, etc.) in the basin to suggest that water quality has changed. The inflow quantity to the reservoir shows no increasing trends. Therefore, it is assumed that if the water load to the reservoir and agricultural practices in the basin haven’t changed, that water quality would be relatively consistent and probably hasn’t changed. Determining the regularity of the hydrologic budget supports the assumption that although this TMDL is based on little more than one year of sampling, the data are valid and are representative of the long-term conditions. The 1998 water year annual water load from each source was determined for purposes of modeling the lake and for determining total phosphorus load from each source. Quantity and quality of many of the sources was available for only the period of the study from 1997 through 1998. To determine watershed phosphorus loading to the reservoir and to model reservoir chemistry, the water budget for the reservoir was evaluated. The water budget is based on flows entering the reservoir and flows leaving the reservoir. Flows entering the reservoir include gaged and ungaged flows. Gaged inflows include Surface Creek Diversion to Alfalfa Ditch, Transfer Ditch, Burgess Ditch and Cedar Mesa Ditch. Posin Gulch is tributary to Cedar Mesa Ditch, and is accounted for in flows from Cedar Mesa. The WQCD water budget was determined using data provided by OCID and instantaneous flow measurements provided from the EPA’s watershed sampling program, as well as the U.S. Bureau of Reclamation (USBR) report “Draft Working Paper: FGR 1998 Water Budget Evaluation” (Ozga, 1999). The USBR report provided additional information on outflows from the reservoir, as well as some gaged inflow values to the reservoir. 27

Final TMDL, October 2012

For this TMDL, the WQCD used USBR outflows, precipitation and evaporation data for 1998 (obtained from the NRCS), and changes in reservoir volume (based on 1998 Area Capacity tables) to calculate monthly and annual total inflows to the reservoir. (The USBR water budget did not account for precipitation and evaporation.) Gaged inflows were subtracted from the calculated total inflows to determine ungaged inflows to the reservoir. Figure 13 illustrates the relative contributions of sources of inflow to FGR. Figure 13.

Hydrology along Alfalfa Ditch from Surface Creek to FGR did not balance. Gains and losses along Alfalfa Ditch, from return flows and diversions have not been identified for this study. More information is needed for better understanding of the system for future work. However, phosphorus loading to the reservoir could be determined with the available data. Alfalfa Ditch carried the largest water load (58%) in the system. Alfalfa Ditch carries water from Surface Creek, Cedar Mesa Ditch, Cedaredge, as well as some nonpoint source runoff. For Figure 13, nonpoint sources were separated from Alfalfa Ditch.

VII. ANALYSIS OF POLLUTANT SOURCES Ambient Water Quality Data Reservoir Data: Fruitgrowers Reservoir is on the 303(d) list of impaired waters for exceedance of the dissolved oxygen standard. The link between dissolved oxygen, chlorophyll a and 28

Final TMDL, October 2012 nutrients was discussed in the section on eutrophication TMDLs. Ambient data for dissolved oxygen, chlorophyll a and total phosphorus are presented here. The dissolved oxygen assessment for the 2010 303(d) List assessed 9 dissolved oxygen profiles for the growing season (July-September). The table below was taken from that report (Table 9). Table 9. 303(d) assessment of Fruitgrowers Reservoir dissolved oxygen data. Table 1. Physical Standards Applicable Number of Profiles Not In Attainment to Lake Profiles Applicable Applicable Total # of DO** – DO** – Temperature* Standard Standard Profiles Epilimnion Metalimnion DO (mg/l) Temp. (°C) 5.0+ 26.5 9 0 2*** 3 *Lakes are not listed for temperature if an adequate refuge exists in lower layers of the lake. Adequate refuge requires attainment of both DO and temperature standards. If an adequate refuge does not exist because of low DO in the lower layers, the lakes is listed for DO, not temperature. **Attainment of DO standard in lakes is assessed for the epilimnion and metalimnion only, unless temperature is not attained in the epilimnion, and the lower layers are being assessed for the presence of an adequate refuge with respect to the temperature standard.

The 2010 report included the following attainment conclusions: Attainment Conclusions: No recent data is available for Fruitgrowers Reservoir. Previous data assessments show that Fruitgrowers Reservoir is not in attainment of the Aquatic Life Use DO standard. DO was below the standard in the epilimnion on 2 of 8 dates: 8/26/98, 9/16/98. Fruitgrowers Reservoir is a shallow waterbody that does not show clear signs of stratification. Wind and wave action break layer boundaries on a regular basis throughout the growing season. Surficial heating is evident along with hypersaturation of dissolved oxygen at the surface due to high levels of algal productivity. Dissolved oxygen regularly falls less than 5 mg/l below one meter depth, and remains depleted to the bottom of the reservoir. The isopleth below displays dissolved oxygen concentrations (mg/l) in Fruitgrowers Reservoir between July 1997 and July of 2000 (Figure 14). The red and orange portions of the graph represent portions of the water column where the DO was below the standard (5mg/l) for that date. The white arrows point to two events in which dissolved oxygen was below the standard in the top layer of the water column (epilimnion), indicating non-attainment.

29

Final TMDL, October 2012

Figure 14. Isopleths of FGR dissolved oxgyen profiles. Individual profiles are shown in the figure below (Figure 15). Depth is plotted on the X-axis and dissolved oxygen concentration (mg/L) is plotted on the Y-axis. The dissolved oxygen standard of 5.0 mg/L is shown in orange. Several profiles illustrate conditions of super-saturation in the upper 1 meter with rapid decline to below the standard below the first meter in depth.

30

Final TMDL, October 2012 Figure 15. Dissolved oxgen profiles for Fruitgrowers Reservoir.

Chlorophyll and Total Phosphorus Samples for chlorophyll a and total phosphorus were collected on the same dates as the dissolved oxygen profiles. A total of 9 summer growing season (July-September) samples were available from 1997-2000 (Table 10). Table 10. Seasonal chlorophyll a and total phosphorus data. Date 7/22/97 9/16/97 7/15/98 7/29/98 8/11/98 8/26/98 9/16/98 8/17/99 7/10/00 Average (Jul-Sept) 1998 average (JulSept)

Chlorophyll a ug/L 45.5 245.2 60.4 86.0 90.2 59.7 88.8 33.6 88.7 77.0

31

Total Phosphorus (ug/L) 60 290 60 340 110 70 50 140 60 131 126

Final TMDL, October 2012

Fruitgrowers Reservoir is a warmwater reservoir in an agriculturally rich basin. Currently, Fruitgrowers Reservoir is hypereutrophic (TSI around 75) with chlorophyll concentrations ranging from 34 ug/L to 245 ug/L in the summer with an average of 89 ug/L. Based on Best Professional Judgement (BPJ), an appropriate trophic state for the reservoir could be in the mid-ranges of eutrophic conditions for chlorophyll and total phosphorus. This would represent a drastic change from current conditions and would support a healthy warmwater fishery. The observed values for chlorophyll and total phosphorus are quite high when assessed on a trophic state index. Chlorophyll and total phosphorus are used as indicators of the trophic state of a water body. Trophic state is a system to classify lakes according to their biological productivity, based on the amount of algal biomass, typically as related to nutrient enrichment or eutrophication. A commonly used system is the Carlson’s Trophic State Index (TSI) (Carlson, 1977). Carlson’s index used the classic terms of oligotrophy, mesotrophy, and eutrophy in their original context of the amount of algae, as represented by chlorophyll a concentrations, in the water and is the context used for this report. The index is not based on hypolimnetic oxygen concentrations, so it is possible for an oligotrophic lake to have no hypolimnetic oxygen. Each trophic state represents a range of chlorophyll concentrations. Table 11 describes attributes associated with each trophic state, as well as characteristics, such as hypolimnetic oxygen, that may be associated with each trophic state (Http://dipin.kent.edu/tsi.htm). Using the Carlson TSI, the chlorophyll TSI for Fruitgrowers is 75. Fruitgrowers Reservoir is in the hypereutrophic range. (Table Carlson TSI). Table 11. Carlson’s TSI. A list of possible changes that might be expected in a north temperate lake as the amount of algae changes along the trophic state gradient. TSI

Chl (ug/L)

SD (m)

TP (ug/L)

Attributes

Water Supply

<30

<0.95

>8

<6

Oligotrophy: Clear water, oxygen throughout the year in the hypolimnion

30-40

0.952.6

8-4

6-12

Hypolimnia of shallower lakes may become anoxic

4-2

Iron, Mesotrophy: Water manganese, moderately clear; taste, and increasing odor 12-24 probability of problems hypolimnetic anoxia worsen. Raw during summer water turbidity

40-50 2.6-7.3

32

Fisheries & Recreation

Water may be Salmonid suitable for fisheries an unfiltered dominate water supply. Salmonid fisheries in deep lakes only Hypolimnetic anoxia results in loss of salmonids. Walleye may predominate

Final TMDL, October 2012 requires filtration. 50-60 7.3-20

Warm-water fisheries only. Bass may dominate.

0.5-1

Blue-green algae dominate, algal 48-96 scums and macrophyte problems

Nuisance macrophytes, algal scums, and low transparency may discourage swimming and boating.

0.250.5

Hypereutrophy: (light limited 96-192 productivity). Dense algae and macrophytes

2-1

60-70 20-56

70-80 56-155

>80

Eutrophy: Anoxic hypolimnia, 24-48 macrophyte problems possible Episodes of severe taste and odor possible.

Rough fish dominate; summer fish kills possible

Algal scums, few <0.25 192-384 macrophytes

>155

A Trophic State Index (TSI) is an absolute scale that describes the biological condition of a waterbody. While trophic state may be related to water quality, the concepts are not interchangeable or synonymous. Water quality connotes “good” or “bad” and is primarily determined by the use classification and the attitude of the user.

Watershed Data: The study design was discussed in Section V. Data collected from the watershed during the WQCD study were used to estimate the annual external phosphorus load to FGR, and to characterize phosphorus sources to FGR (Table 12). Table 12. Fruitgrowers Reservoir watershed sources (WY1998). Sampling Sites

Flow (Ac-Ft) Annual Total

Surface Creek diversion to Alfalfa Ditch

9248

Town of Cedaredge WWTP

209

Cedar Mesa Ditch

880

33

Annual Average Total Phosphorus (mg/L) Mean (range) 0.05 (0.030.19) 3.79 (3.264.87) 0.345 (0-0.78)

N

10 10 6

Final TMDL, October 2012 Transfer Ditch Burgess Ditch

430 433

Alfalfa Ditch at Western (upstream from Fruitgrowers Reservoir Alfalfa Run downstream from Fruitgrowers Reservoir Ungaged Inflows

7337 12979

0.68 .095 (0.020.37) 0.235 (0.090.83) 0.125 (0.060.27)

2 4 10 7

4382

Modeling Fruitgrowers Reservoir: Modeling FGR was completed in a step-wise method. First, a mass-balance approach was employed to determine external loads. Then a common steady-state model was applied to determine the internal load. The same model was used to determine the loading capacity or Total Maximum Daily Load (TMDL) of the reservoir, as well as potential reductions of loads. External (Observed) Loads Existing (or observed) loads were estimated from ambient conditions as identified in the table above. A mass-balance approach was used to determine the annual external phosphorus load for the 1998 water year (November 1997-October 1998). Phosphorus loads entering and leaving the reservoir were calculated using the water budget flow values, as determined in the Hydrology section, and phosphorus concentrations for each source as determined from the watershed and lake monitoring program conducted by the EPA and the WQCD. Monthly and annual phosphorus loads were calculated for each source (Surface Creek, Alfalfa Ditch, Transfer Ditch, Cedar Mesa Ditch, Cedaredge WWTP, ungaged inflows, and reservoir outflow). Annual loads were summed to determine total annual phosphorus loading into and out of Fruitgrowers Reservoir. No chemistry data were available for the ungaged inflows. Therefore, the total phosphorus concentrations from the Surface Creek diversion were applied to the ungaged inflows to represent that portion of the load. The 1998 water year annual load of total phosphorus to the reservoir was 3464 kg. The phosphorus loads calculated for Alfalfa Ditch included contributions from the WWTP, Cedar Mesa Ditch and Surface Creek and non-point sources. The mass-balance approach enabled the separation of these sources. Loads that could not be attributed to Surface Creek, Cedar Mesa Ditch and Cedaredge WWTP were considered to be from non-point sources. Ungaged inflow loads were lumped with non-point source loads (Table 13). Table 11 does not include a load for Alfalfa Ditch because that load was separated into Surface Creek, Cedar Mesa Ditch, Cedaredge WWTP and nonpoint sources. Sources contributing to the total external phosphorus load included Surface Creek, Town of Cedaredge WWTP, Cedar Mesa Ditch, non-point sources and Transfer Ditch. Figure 16 illustrates the relative contributions of phosphorus from each watershed source. Although Cedaredge contributed only 2% of the total water load to the reservoir, Cedaredge represents the largest portion of the watershed phosphorus load at 28%. 34

Final TMDL, October 2012 Nonpoint sources and Surface Creek contribute slightly more than 20 % each, while Transfer Ditch and Cedar Mesa Ditch contributes approximately 10%. Burgess Ditch represents the smallest portion of the load at 3%. Alfalfa Ditch conveys phosphorus loads from Surface Creek, Cedar Mesa Ditch and the Town of Cedaredge WWTP. The combined loads in Alfalfa Ditch make up over 60 percent of the watershed load to the reservoir. Cedaredge makes up nearly 50% of the load carried in Alfalfa Ditch. Table 13. Watershed (external) phosphorus loads to FGR. Source Load (kg/yr) Surface Creek 766 Town of Cedaredge 1028 Cedar Mesa Ditch 375 Transfer Ditch 361 Burgess Ditch 114 Non-point Sources 820 Total 3464

35

Final TMDL, October 2012 Figure 16. Relative load contributions of external sources to FGR. The smaller pie on the right represents the sources of the 63% phosphorus load carried in Alfalfa Ditch.

Internal Loads To provide a complete estimate of phosphorus loads impacting the ecology of the reservoir, internal loads were estimated, in addition to external watershed loads. Internal phosphorus loading results from the release of sediment-bound phosphorus from the lake bottom into the water column during anoxic conditions. Internal release of phosphorus is expected in Fruitgrowers Reservoir due to the observed hypoxic (low dissolved oxygen) and anoxic (without or zero dissolved oxygen) conditions that triggered this TMDL. Internal loads primarily are the legacy effect of excess external loads. Some of the phosphorus, as particulates in the forms of sediment-attached phosphorus, as well as living and dead biomass and detritus, in the water column may settle and be retained in the lake. A portion of this phosphorus is then released either later in the season or in the following year when the lake exhibits conditions of low dissolved oxygen. Internal phosphorus releases exacerbate the trophic response of the reservoir beyond what would be expected from external loads alone. A model developed by Nurnberg (1984) is a common method used to determine the internal phosphorus load for a lake. This model is appropriate for determining annual internal loads for lakes that exhibit anoxic hypolimnia and that show retention of phosphorus, such as FGR. The model separates the external and internal loads. The Nurnberg model uses in-lake phosphorus concentrations, annual areal water load, phosphorus retention, and external phosphorus load information to estimate internal loads.

36

Final TMDL, October 2012 The Nurnberg model can be used to calculate internal loading based on observed external loads and modeled in-lake phosphorus concentrations. It also can be used to calculate expected in-lake phosphorus concentrations based on observed or expected internal and external phosphorus loading. Other models were also tested to estimate the expected in-lake total phosphorus concentrations based on observed loads (Vollenweider, 1975; Dillon-Rigler, 1974; Larsen and Mercier, 1976). Although these models yielded results that were within reasonable ranges of observed in-lake concentrations, the Nurnberg model was the only model that also accounted for internal loads and therefore was the most comprehensive model. Therefore, the Nurnberg model was the preferred model for this TMDL. The FGR values used to parameterize this model are found in Table 14.

The equations used in the Nurnberg model are: P=LExt/ qs (1-R) + LInt / qs (equation 1) R= 15/(18+ qs) (equation 2) The terms in the model are: P = annual average total phosphorus concentration, ug/l LExt = areal annual external P load, mg/m2/yr, qs = areal water load, m/yr R = phosphorus retention coefficient LInt = areal internal P load, mg/m2/yr

37

Final TMDL, October 2012

Table 14. Fruitgrowers Reservoir values used to parameterize Nurnberg model. Term Unit FGR System Comment/Description Observed Value P (observed) ug/L 126 1998 Growing Season average Lext kg/yr 3464 Lext mg/m2/yr 2078.6 Areal external P load LInt mg/m2/yr 213.5 areal internal P load, modeled LInt kg/yr 475 areal internal Pad, modeled A qs RPred

m2 m/yr

RMeas Total Load

1770923.4 9.04 0.55 0.30

kg/yr

3939

Lake Area, 437 acres areal water load Retention, predicted using equation 2 Retention, calculated from inflow and outflow loads Lext + LInt

The external load for water year 1998 to FGR determined by mass-balance was 3464 kg/year. The Nurnberg model estimated 475 kg/yr of internal loading to FGR. Thus, the total load to FGR combining external and internal loads was 3939 kg/yr.

Figure 17 illustrates the 1998 relative contributions of total phosphorus to the reservoir from each source. These results demonstrate that although the Cedaredge WWTP contributes only approximately 2% of the total inflow to the reservoir (see Hydrology section), it contributes 26% of the Total Phosphorus load to the reservoir.

38

Final TMDL, October 2012 Figure 17. Total phosphorus load to FGR. The smaller pie on the right illustrates the relative contributions of the phosphorus load sources carried in Alfalfa Ditch.

The sources of phosphorus contribution in order from largest to smallest were: the Town of Cedaredge WWTP, non-point sources, Surface Creek, Transfer Ditch, Cedar Mesa Ditch and Internal Load. The WWTP contributed 1028 kg annually. This load was approximately 26% of the total load to the reservoir. The WWTP makes up nearly half of the load carried in Alfalfa Ditch and Alfalfa Ditch makes up over half of the total load to Fruitgrowers Reservoir. Cedar Mesa Ditch, Transfer Ditch and Burgess Ditch do not have continuous flow throughout the year, but if used, may serve as a significant source of phosphorus to the reservoir. WQCD was notified by Mel Shroeder (Ditch Rider-OCID) that Transfer Ditch was opened for a short time, as requested by the EPA, so that nutrient and flow data could be collected. However, according to Mr. Shroeder, the ditch had not been used during the previous 4-5 years, and is assumed to continue to be used infrequently.

Reservoir Phosphorus Loading Capacity The reservoir loading capacity is the total annual phosphorus load that FGR can receive, yet attain the dissolved oxygen standard to protect aquatic life uses. Attainment of the dissolved oxygen standard is implemented through attain the TMDL target of 43 ug/L total phosphorus in-lake. The Nurnberg model described above was used to estimate the phosphorus loading capacity of the reservoir. The model equation was rearranged and solved for expected external load based on the following assumptions and goals. The areal water load is assumed to remain the same. The in-lake phosphorus concentration was set to the TMDL target of 43 ug/L. (The median TMDL target was 39

Final TMDL, October 2012 used in this modeling scenario because the model requires the summer average total phosphorus value. In this case, it is assumed that the median or typical summer average is the best choice for parameterizing the model. The internal load was set to 10% of the existing internal load or 47.5 kg/yr. Best professional judgment suggests that this is a reasonable expectation for remediated internal loads. Internal loads are expected to decrease over time (10-20 years) after external loads are reduced. Alternatively, after implementation of external controls, internal loads may be reduced by phosphorus removal methods (which could accelerate steady state conditions). The external loading capacity for FGR, based on the assumptions above is 1439 kg/yr. Table 15 presents the expected loading capacity for FGR based on the assumptions described above. The loading capacity is 42% of the observed external load.

Table 15. Expected FGR annual phosphorus mass balance. (TP target=43 ug/L) Parameter Units Observed Expected Comments % Reduction External kg/yr 3464 1439 External Load 58% Load Capacity Internal Load kg/yr 475 47.5 Internal Load 90% Capacity Total Load Kg/yr 3939 1486 External+Internal 62% In-lake 66% Phosphorus ug/L 126 43 Target in-lake P

40

Final TMDL, October 2012 TMDL Allocation A TMDL is comprised of the Load Allocation (LA), which is that portion of the pollutant load attributed to natural background and/or the nonpoint sources, the Waste Load Allocation (WLA), which is that portion of the pollutant load associated with point source discharges, and a Margin of Safety (MOS). 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)” There is one permitted point source in this reservoir basin: Town of Cedaredge WWTP. Adequate flow and total phosphorus data for the effluent are available to establish that the load from this source has been consistent, with no increasing trends. The permit will be renewed in 2010. Currently the permit is based on zero low-flow. In this TMDL, the Town of Cedaredge will be given a wasteload allocation. Load Allocations “(LA)” Any remaining sources are considered to be background and non-point sources and are accountable to load allocations. For lakes and reservoirs, the internal load also will be included as a Load Allocation. 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 margin of safety may be explicit (a separate value in the TMDL) or implicit (included in factors determining the TMDL). For the FGR TMDL, an implicit margin of safety was used. The targets set for the TMDL are based on median and 80th percentile summer average values, rather than maximum values. This is consistent with other standards in Colorado. These target values were developed using an extensive data set compiled by pooling data from lakes similar to FGR. The TMDL is based on the phosphorus loading capacity as modeled from the target in-lake phosphorus concentration and the observed water loading conditions as described in modeling section of this report. After determining the total internal and external loads to Fruitgrowers Reservoir, the Nurnberg model was used to back-calculate the phosphorus loading capacity of the reservoir which is the load that would meet the inlake total phosphorus target of 43 ug/L as a typical summer average. Applying the Nurnberg model, and assuming a 90% reduction of the internal load, the predicted external load would be 1439 kg/yr. This represents a 62% reduction of the external load 41

Final TMDL, October 2012 and a 67% reduction in all phosphorus loads (internal and external) to the reservoir. The resulting load capacity is modeled as the external load, based on an assumed reduction of the internal load. The result is allocated between the Waste Load Allocation and the Load Allocation. The entire Waste Load Allocation is assigned to the Town of Cedaredge WWTP, a permitted WLA. The Load Allocation includes the watershed background sources, nonpoint sources, and the internal load. The TMDL allocations are determined using several assumptions. The water imported from outside the basin is assumed to be background and was not assigned a reduction. The internal load was assigned at 10% of the existing load as part of the modeling exercise. The remainder of the allowable load was assigned to the other sources using BPJ. The permitted source, as a WLA, is assigned a 99% reduction for several reasons. Reductions of phosphorus from point sources are most likely to be successful. Point sources are the easiest phosphorus source to control, and technologically, this level of reduction is feasible. However, the WWTP makes up only 25% of the total load to the reservoir, so reductions from this source alone are not adequate to meet the TMDL. Therefore, additional reductions are assigned to other nonpoint sources. The reductions assumed for the nonpoint sources are quite aggressive and are likely at the high end of reductions possible for nonpoint sources. Table 16 itemizes the allocations by source.

42

Final TMDL, October 2012

Table 16. TMDL allocation by source. Source Load (kg/yr) TMDL Load observed (kg/yr) Waste Loads Town of 1028 18 Cedaredge Loads Surface Creek 766 766 Cedar Mesa Ditch 375 375 Transfer Ditch 361 0 Burgess Ditch 114 0 Non-point 820 279 Sources Internal Load 475 47.5 External Load 3464 1439 Total Load 3939 1486

% Reduction

TMDL Load (kg/d)

98%

0.049

0% 0% 100% 100% 66%

2.1 1.03 0 0 0.76

90% 58% 62%

0.13 3.94 4.07

Note: as explained in text above, Transfer Ditch and Burgess Ditch are used very little, so 0 contributions are assumed in most years, and applied this to the model. The loads used in the development of this TMDL and the resulting allocations are based on annual loads. This is a typical approach for modeling lake responses to nutrient input. The algal response to nutrient loading in lakes is temporal on an annual basis. Seasonal and/or annual dynamics such as flushing rate (or hydraulic residence time), spring phosphorus loading, periods of stratification, internal phosphorus loading are more determinant of critical periods in lake response than are daily loading events. In lakes, the phosphorus load on a given day has less impact on the biological response of the lake than does annual loading. Therefore, it is appropriate to express lake TMDLs as annual, rather than daily loads. However, to meet the requirements of daily loads for a TMDL, the loads are also expressed as 1/365th of the annual load. Using the TMDL load allocations in Table 14 above, and based on the increased Cedaredge WWTP capacity, the expected WWTP effluent phosphorus concentration would be 0.05 mg/L (Table 17). Based on the 1998 WWTP capacity, the effluent phosphorus concentrations would be 0.07 mg/L.

43

Final TMDL, October 2012

Table 17. Expected Cedaredge WWTP effluent TP concentrations under different discharge volume scenarios. Scenario Annual Discharge Effluent TP Volume (ac-ft) (mg/L) WY98 (Observed) 209 3.85 (median) WY98 (TMDL) 209 0.07 2005 Design Capacity 244 0.06 2010 Design Capacity 301 0.05 Reductions for non-point sources are set at 66%. These reductions will be based on control measures associated with land uses including modification to agricultural livestock practices. Possible reductions from BMPs could be modeled using land-use and/or watershed models.

44

Final TMDL, October 2012

VIII. RESTORATION PLANNING AND IMPLEMENTATION PROCESS The phosphorus loading reductions necessary to meet the TMDL target and thus the TVS standard for dissolved oxygen are identified in Table 16. Only one permitted discharger is identified in the TMDL. The TMDL waste load allocation will be implemented in the facility’s permit during the permit renewal cycle. Although substantial reductions are prescribed for this point source, reductions from other sources also are necessary to meet the TMDL. Additional modeling is needed to identify contributions from specific land uses, as well as effective BMPs to reduce phosphorus from those sources. Subsequent to the Public Notice period of this TMDL, the Division was asked to recommend projects for a new USDA-Natural Resources Conservation Service (NRCS) program named the National Water Quality Initiative (NWQI). Under the NWQI, the NRCS is directing their NRCS-state conservationists to spend 5% of their FYI 12 allotment of Environmental Quality Incentives Program (EQIP) funds to assist farmers in undertaking conservation measures that address high-priority water resources that meet certain criteria, including impaired waters on a state’s 303(d) list. The NWQI included nutrients as an impact to target for the funding. The Division recommended the Fruitgrowers Reservoir TMDL as a candidate for this funding. These funds are expected to be used to develop BMPs to reduce total phosphorus from the nonpoint sources contributing to the impairment of Fruitgrowers Reservoir. Previous Water Quality Improvements in the Watershed The Division has no knowledge of previous water quality improvements that might have been made in the Fruitgrowers Reservoir watershed. Monitoring Additional monitoring of FGR was conducted in 2010 (one sampling event), but the data are not available at this time. No watershed monitoring has occurred since the 1998 study, and none is planned at this time. More detailed watershed chemistry and flows will be needed to select and implement nonpoint source BMPs. When phosphorus reductions are implemented in the watershed, monitoring of FGR and the watershed should be conducted in order to ensure that nonpoint source efforts are effective and that the TMDL is adequately protective of the segment. Conclusion The objective of this TMDL is the attainment of the dissolved oxygen standard within FGR. Substantial loading reductions of total phosphorus are necessary to meet this objective. Improvements in pH, as well as dissolved oxygen would result from this TMDL. The recommended phosphorus loading reductions should result in attainment of the applicable water quality standards. IX.

PUBLIC INVOLVEMENT 45

Final TMDL, October 2012 This segment was included on Colorado’s 303(d) list of impaired segments in 2002, 2004, 2006, 2008, and 2010. The development of the 303(d) list is a public process involving solicitation from the public of candidate waterbodies, formation of a technical review committee comprised of representatives of both the public and private sector, and a public hearing before the Colorado Water Quality Control Commission. Public notice is provided concerning both the solicitation of impaired waterbodies and the public hearing. The TMDL itself is the subject of an independent public process. This TMDL report was made available for public review and comment during a 30 day public notice period in May 2011. The EPA provided minimal comments on the draft TMDL. The Division received four sets of comments on the Fruitgrowers TMDL. Comments ranged from issues with technical approach and data quality, editorial concerns of specific language used, TMDL process and legal process, as well as economic feasibility. The Division has addressed these public notice comments by addressing each individual contributor. Where needed, editorial comments and corrections were inserted in the text of the report. Comments from David Smith: Note that these issues were not necessarily framed as questions, but the Division made efforts to address the relevant points. DS Comment 1: There appears to be significant internal loading from accumulated total phosphorus in the sediment over the years. It seems that over the course of the life of the reservoir, the town would be less than a 20% contributor. David Smith views the internal loading as a source that needs to be controlled in order to affect a healthy reservoir, and only if removal of the internal loading is infeasible should limits beyond non-enhanced biological nutrient removal be set on the sole point discharger. Division Response: Although phosphorus has been added to the sediments each year, for the TMDL, the lake was modeled on an annual basis and the percent contribution for each source is reported as the portion of the annual load. The Town of Cedaredge represents 30 percent of the external load and 26 percent of the total load of phosphorus, annually. Although the phosphorus continues to accumulate in the sediments, it is generally believed that the upper few centimeters to inches of sediment are the most active or important with respect to internal loading, therefore, the estimates of the percent contributions by each source are reasonable for purposes of the TMDL. The endpoint of the TMDL is attainment of the assigned water quality standard. Therefore, the TMDL includes reductions for the internal load as well as the point source discharger. However, the internal load is legacy watershed load and without adequate watershed reductions, removal of the internal load would be temporary, because watershed sources would merely regenerate the internal load. To state this another way, removal of the internal load is infeasible without adequate reductions of the point and nonpoint sources. DS Comment 2: Because of the complexity of the reservoir operation and loading 46

Final TMDL, October 2012 scenarios, and the negative economic and social impacts that may result, David Smith requests an independent peer review of the TMDL. Division Response: The TMDL must address the assigned water quality standard, or in this case the surrogate water quality target that drives attainment of the assigned water quality standard. Economic and social impacts may be considered by the Water Quality Control Commission with respect to an alternate Water Quality Standard, but the TMDL is not usually the forum for these types of analyses. The TMDL is effectively a pollutant „budget‟ that is necessary for the underlying standard to be achieved. In the past, some parties have commissioned third party reviews of TMDL modeling. However, the WQCD does not, as a normal matter, enlist independent reviews of TMDLs. The Public Notice process allows third-parties an opportunity to review the TMDL and analysis of the WQCD. We have received comments from Cedaredge‟s consultants, and their comments did not appear to question the modeling to any great extent. In developing the TMDL, the Division examined several approaches using the work of several professional limnologists, and presented several lines of reasoning for development of the TMDL. Also, the Division had EPA review the characterization of the reservoir operation and loading and they concurred with the TMDL analysis. DS Comment 3: Is setting of targets immaterial in light of the infeasibility of attaining them? Division Response: The TMDL targets are set at a level modeled to achieve the dissolved oxygen standard. The dissolved oxygen standard was promulgated by the WQCC and, as such, is legally enforceable under the Clean Water Act. The Town of Cedaredge has the option of pursuing alternate standards based upon a demonstration that the existing standard is unattainable. Any such demonstration would require a showing that the standard is unattainable due to irreversible human caused conditions 31.7(1)(b)(ii), (Regulation 31 Basic Standards and Methodologies for Surface Water, Assigning Standards). Alternately, after 1/1/2013 the Town may pursue a site-specific variance per section 31.7(4) (Regulation 31 Basic Standards and Methodologies for Surface Water, Granting, Extending and Removing Variances to Numeric Standards (effective October 1, 2013)). There are several permit related options that may also be available, and the WQCD TMDL program understands that there are ongoing discussions regarding permit related options.

47

Final TMDL, October 2012 Comments from Metro Wastewater Reclamation District (MWRD): MWRD Comment 1: MWRD takes issue with the Division’s use of “draft proposed standards” in the Fruitgrowers Reservoir Dissolved Oxygen TMDL. Their concerns are that it may interfere with work in progress with the Nutrient Workgroup; that the “draft proposed standards” are now referred to as “Current interim numeric values”; that the “current interim numeric values” may change over time, and that these “interim numeric values” have not been adopted by the Water Quality Control Commission. MWRD states that “Therefore, it is premature to use these specific interim numeric values as a basis for setting enforceable water quality standards on a site-specific basis. Division Response: As explained in the draft TMDL, the “draft proposed standards” or “interim numeric values” are used in the TMDL as a line of evidence in identifying a range chlorophyll and phosphorus concentration targets that would achieve attainment of the dissolved oxygen standard. The TMDL clearly acknowledges that these are draft values and that they have not been adopted by the Water Quality Control Commission. The TMDL seeks to develop water quality endpoints on a site specific basis. In doing so, it is appropriate to evaluate the available range of modeled endpoints. Moreover, we do not believe that the inclusion of the Division‟s analysis which led to the proposed draft interim standards will in any way influence the WQCCs ultimate decision with respect to nutrient standards. The proposed values have been identified by the Division as protective of aquatic life uses, and this line of reasoning provides an upper bound on a range of potential targets for the TMDL. Ultimately, the Division‟s total phosphorus interim numeric values were not selected as the final target for the Fruitgrowers TMDL. Along this line, in response to this comment, the Division realized that we also should have included a discussion of EPA‟s 304a criteria, for an additional line of reasoning. EPA‟s recommendations were based on analysis of aggregated ecoregions and Level 3 ecoregions. Both aggregate (Table 1) and Level 3 ecoregion (Table 2) criteria applicable to Colorado are presented here. The EPA recommended criteria represent the 25th percentile of measured values. EPA‟s recommended criteria are lower than other targets that were presented in the original Fruitgrowers Reservoir TMDL.

48

Final TMDL, October 2012 Table 1. Aggregate Nutrient Ecoregion II. Western Forested Mountains III. Xeric West IV. Great Plains Grass and Shrublands V. South Central Cultivated Great Plains Table 2. Level 3 Ecoregion 18: Wyoming Basin 21: Southern Rockies 20: Colorado Plateaus 22: Arizona/New Mexico Plateau 25: Western High Plains 26: Southwestern Tablelands

TP, mg/L

Chl a, ug/L

0.009

1.9

0.017 0.020

3.4 2.0

0.033

2.3

TP, mg/L

Chl a, ug/L

0.010 0.015 0.003 0.015

1.4 1.7 1.4 2.0

0.024 0.020

2.4 1.2

Therefore, the Division will add a discussion of the 304a to the TMDL report. Note: Subsequent to the drafting of the Division‟s Response to Comments, the Water Quality Control Commission adopted the proposed interim numeric values in section 31.17 of Regulation 31, The Basic Standards and Methodologies For Surface Water, 2012(effective September 30, 2012). The interim values changed slightly, however, as stated in the TMDL and in the response above, the interim numeric values were not used for the TMDL target. Below is a table of the interim numeric values adopted by the Water Quality Control Commission. Classification

Cold Warm

Domestic Use Water Supply (DUWS)2 Chlorophyll (ug/L) 5 5

Chlorophyll (ug/L)

Aquatic Life1 Total P (ug/L)

Total N (ug/L)

8 20

25 83

426 910

1 – summer (July1-September 30) average in the mixed layer of lakes (median of multiple depths), allowable exceedance frequency 1in-5 years. 2 – March 1-November 30 average chlorophyll a (ug/L) in the mixed layer of lakes (median of multiple depths), allowable exceedance frequency 1-in-5 years.

Colorado interim nutrient values for lakes. (Regulation 31, effective date, September 2012.

MWRD Comment 2: MWRD is concerned about the 99% level of phosphorus reduction for the Cedaredge lagoon system and how a small community may be able to accomplish tertiary treatment technologies. Division Response: 49

Final TMDL, October 2012 The TMDL identifies necessary reductions to attain the water quality standard. As stated in the TMDL, the nonpoint source reductions are at the extremely high end of what we understand to be feasible for nonpoint sources. This was balanced with the reduction for the point source, which was based on what is known to be technologically feasible. The Division believes that balance between point and nonpoint source reductions is appropriate. Several other options do exist, beyond treatment plant upgrades. These include: alternate standards based upon a demonstration that the existing standard is unattainable, a site-specific variance per section 31.7(4) (Regulation 31 Basic Standards and Methodologies for Surface Water, Granting, Extending and Removing Variances to Numeric Standards (effective October 1, 2013)) could also be considered after /1/2013, and several permit related options that may also be available. The WQCD TMDL program understands that there are ongoing discussions with the Town regarding permit related options.

50

Final TMDL, October 2012 MWRD Comment 3: MWRD states that the proposed phosphorus effluent limit is a level beyond even enhanced biological nutrient removal and that it may not be financially feasible. Division Response: The TMDL does not address economic feasibility. Economic feasibility is a factor the WQCC considers during the standard setting process. The TMDL compares the standard that the WQCC has adopted and the resulting water quality improvements necessary to achieve that standard. The Division is aware that the TMDL point source reduction identified would require treatment beyond BNR. However, the level is technologically feasible, and is achieved in other facilities in Colorado including facilities discharging to Cherry Creek, and facilities in Summit County in the Lake Dillon watershed. MWRD Comment 4: MWRD suggests that even with implementation of a WWTF discharging at 0.07 mg/L phosphorus, the reservoir would not attain the proposed water quality targets because the majority of the loading is from nonpoint sources. MWRD suggests that it would be more appropriate to focus on nonpoint source reductions first before requiring-or even considering-drastic changes at the wastewater treatment facility. Division Response: The TMDL includes both point and nonpoint source reductions. The TMDL does not specify nonpoint source implementation or the chronological order of any related reductions. The TMDL already addresses high reductions in the nonpoint source Load Allocation (LA). The Town may pursue a variety of options within either or both rulemaking and permitting frameworks that would allow exploration of non-point source reductions before treatment plant improvements are implemented. JVA Consulting Engineers provided comments on behalf of the Town of Cedaredge. JVA Comment 1: FGR is filled with water imported from Surface Creek via the Alfalfa Ditch diversion and other irrigation diversions. The predominant source of water is imported from other basins and according to the Colorado Division of Water Resources, all the water in FGR is agricultural and private. The FGR was placed on the 303(d) List in 1998 for fecal coliform and ammonia, then subsequently removed. Then in 2010, the FGR was added to the 303(d) list for dissolved oxygen. Please explain the “community” that came together to discuss water quality in FGR as described on page 5 of the TMDL. Does CDPHE or EPA have jurisdiction for regulating private agricultural water and associated impoundments. Division Response: FGR was de-listed for fecal coliform and ammonia in 2002. Also, in 2002, FGR was added to the 303(d) list for dissolved oxygen. Notes from 1999 indicate “The Fruitgrowers Colition is made up of representatives from the Delta Soil Conservation District, Colorado State Soil Conservation Board, Natural 51

Final TMDL, October 2012 Resources Conservation Service, Environmental Protection Agency, Bureau of Reclamation, Colorado Division Of Wildlife, Colorado Department of Public Health and Environment, Delta County Commissioners, Orchard City Irrigation District, and the towns of Cedaredge and Orchard City. CDPHE has jurisdiction for regulating waters of the state under the Colorado Water Quality Control Act. Fruitgrowers Reservoir is a State Water, as defined in Regulation 31. Regulation 31 includes the formal definition at 31.5 “STATE WATERS” means any and all surface and subsurface waters which are contained in or flow in or through this state, but does not include waters in sewage systems, waters in treatment works of disposal systems, waters in potable water distribution systems, and all water withdrawn for use until use and treatment have been completed. Fruitgrowers Reservoir is included in Regulation 35-Classifications and Numeric Standards for Gunnison and Lower Dolores River Basins and is Segment 09 of the Lower Gunnison Basin, Fruitgrowers Reservoir is designated for Agricultural, Aquatic Life and Recreation Uses, with associated numeric standards. Under Colorado Law a water right allows the holder to divert water for beneficial use, it does not provide the water right holder ownership of the water. JVA Comment 2: As indicated in the TMDL, Cedaredge only accounts for 25 percent of the total phosphorus (TP) load to FGR. Therefore, additional reductions were assigned to nonpoint sources and recognized to be on the high end of what is possible. If the 66 percent reduction in non-point source phosphorus is not achievable, how will that affect the TMDL allocation for Cedaredge in the future? Furthermore, please describe what happens if the FGR cannot attain the water quality target for dissolved oxygen by reducing TP. Lastly, please describe the State‟s policy for trading load allocation from point source to non-point source. For example, the Town could invest in eliminating non-point sources in lieu of advanced wastewater treatment. Division Response The TMDL includes phosphorus reductions of both nonpoint source and point source. Although the Division has no regulatory authority over nonpoint sources, the success of the TMDL is evaluated based on attainment of the assigned dissolved oxygen standard in Fruitgrowers Reservoir. Tracking of TMDL implementation will consider both the inreservoir dissolved oxygen concentrations and phosphorus loading reductions. There would be no implications to the TMDL wasteload allocation until the TMDL was fully implemented. To restate, the Division would not revisit the Town‟s wasteload allocation until such time as the phosphorus reductions identified in the TMDL had been achieved and water quality monitoring indicated that such reductions were inadequate to achieve the assigned dissolved oxygen standard The attainment status of waterbodies in Colorado is periodically reviewed and is tracked in the Assessment Database (ADB). The ADB is maintained jointly by the WQCD, EPA and an EPA contractor. This information is reported biennially in the Colorado 52

Final TMDL, October 2012 Integrated Report (aka the 305(b) report). At some point after a TMDL is completed and implemented, if the waterbody is still not attaining standards it would be placed back on the 303(d) List of impaired waters and the TMDL would subsequently be revisited. At that time, wasteload allocation may be revised. The appropriate timeframe for waters to attain standards after TMDL implementation has not been determined. The timeframe is adequate, however, to ensure that the Town has the opportunity to pursue alternate remedies available as provided in the State‟s Basic Standards and Methodologies for Surface Waters (Regulation 31). Trading is allowed and is guided by the Colorado Pollutant Trading Policy (2004): Trading in Impaired Waters after EPA Approval of a TMDL. Trades in impaired waters for which a TMDL has been approved should be consistent with the assumptions and requirements upon which the TMDL was established. This requires assurance that the system used to determine when credits are created and how they may be applied are consistent with the wasteload and load allocations established by the TMDL and any associated implementation plan. Where a TMDL has been approved or established by EPA, the applicable point source waste load allocation or nonpoint source load allocation would establish the baselines for generating credits from that point forward. Although phosphorus trading may be an option, based on the loads and resulting allocations in this TMDL, trading alone without some degree of enhanced phosphorus removal at the WWTP is unlikely to attain adequate load reductions to attain water quality standards. JVA Comment 3. The expected Cedaredge WWTP effluent TP concentrations calculated in the TMDLwere based on discharge volumes. As shown in Table 15 of the TMDL, the 2005 design capacity of 244 acre-feet (0.217 MGD) would result in an effluent TP of 0.06 mg/l and the 2010 design capacity of 301 acre-feet (0.269 MGD) resulting in an effluent of TP of 0.05 mg/l. The Town of Cedaredge is in the process of planning for a WWTP expansion with sufficient capacity to meet the 20-year projections. If the Town constructs a WWTP with a design capacity of 453 acre-feet (0.404 MGD), will the resulting effluent TP be less than 0.05 mg/l or is this the current lowest limit achievable by current technology? Additionally, is the TMDL allocation based on design capacity or actual wastewater discharge flows which can be far less early in the planning period. Furthermore, as long as the wastewater treatment discharge does not exceed the annual TP allocation, can the monthly discharge limit vary thus allowing for seasonal variation in treatment. Division Response: The TMDL presents effluent concentrations under various discharge scenarios to illustrate how loads might be translated in a CDPS Permit. However, according to Permits staff, The practical quantification limit for phosphorous is 10 ug/l. This PQL allows a limitation of 0.01 mg/l and above. 0.05 mg/l is well above the PQL. Therefore, potential limitations below 0.05 mg/l could be applicable. As for the monthly discharge limits, these are details to be worked out in the permitting process. Historically, facilities in Colorado that perform low-level phosphorus reductions have a 30-day average limit in their permits. It is expected that the Cedaredge Permit will be implemented similarly. 53

Final TMDL, October 2012

JVA Comment 4: The TMDL indicates that the waste load allocation will be implemented in the facility’s permit during the permit renewal cycle. The Town of Cedaredge’s existing permit expired on May 31, 2010. A renewed permit has not been received. During the May 5, 2011 meeting, the State indicated that the TP limit would not be included on this renewed permit, but the next permit. Please clarify the timing of implementation of the TP discharge limit.

Division Response: When the TMDL is finalized we will include it in the renewal permit. If we actually stated that it wouldn‟t be included, it must have been under the impression that the renewal would be completed well before the TMDL was finalized. It is likely that a new permit would incorporate a compliance schedule for implementing the WLA, but the actual effluent limitations would not become effective until after the term of the compliance schedule. These details would be developed with Permits staff.

JVA Comment 5: The Town of Cedaredge is in the process of exploring options for alternate discharge locations including Surface Creek and Alfalfa Run (downstream of the FGR). Please verify that Cedaredge will be removed from the TMDL if the WWTP discharge is relocated. Division Response: Only Fruitgrowers Reservoir has been included in the Colorado Section 303(d) List of impaired waters. Neither Alfalfa Run nor Surface Creek have been listed. Therefore, it is likely that the facility will not be subject to the TMDL terms. It should be noted that other segments under consideration should also be looked at in terms of their limiting conditions, if any. If the discharge point is removed from the basin, the TMDL would not be relevant to the Permit. The Permittee would need to work with CDPHE Permits Program regarding alternate discharge locations.

54

Final TMDL, October 2012 Town of Cedaredge Wastewater Treatment Planning Committee Questions Regarding TMDL Assessment of Fuitgrowers Reservoir Questions of Assumptions: 1) How can values that have not yet been adopted by the Water Quality Control Commission as either water quality criteria or water quality standards be enforceable. Division Response: As explained in the TMDL document, the impairment for dissolved oxygen is related to excess nutrients, and is addressed by setting a target for total phosphorus. The WQCC has promulgated dissolved oxygen standards which are applicable to Fruitgrowers Reservoir and which are not presently attained. The TMDL must therefore identify the underlying cause of the impairment and must determine to what extent the underlying cause(s) must be controlled in order to bring the reservoir into attainment of the assigned DO standard. References to draft proposed nutrient criteria were used as one of several lines of reasoning for identifying a TP target. Ultimately, however, a different TP target was selected for the TMDL. Note: Subsequent to the drafting of the Division‟s Response to Comments, the Water Quality Control Commission adopted the proposed interim numeric values. The numbers were slightly modified. See response to MWRD Comment 1. 2) In calculating the relative sources, how does the assessment account for cattle that graze in the upper reservoir and the grazing on north and northeast perimeter of the reservoir, where runoff enters the reservoir before being collected in a measureable draw or ditch? Division Response: The assessment does not account for cattle specifically. Cattle, as well as manure, fertilizer sources, soil erosion, etc, are lumped together in the nonpoint source Load Allocation (LA)The nonpoint source component was calculated as a residual in the mass balance as described in the TMDL. 3) Assessment (Pg 13 last paragraph). Why is it not important to know whether nitrogen or phosphorus is limiting? Please explain how nitrogen limitation can be due to excess phosphorus? It says that phosphorus is the focus because nitrogen mitigation is not effective? Is it saying the reducing nitrogen does not reduce algal growth or that it is not practical to reduce nitrogen? Division Response: With an excess of phosphorus available (and phosphorus is in excess in FGR), reducing nitrogen will not reduce algal growth. Therefore it is not practical to reduce nitrogen alone. It is not important to know the limiting nutrient because currently in FGR, both nutrients are in high enough concentrations to grow excess algae. We are 55

Final TMDL, October 2012 interested in managing nutrients to reduce the amount of algae that can grow. Algae need nitrogen in a fixed form (ammonia or nitrate). If the fixed nitrogen is depleted, nitrogen becomes the limiting nutrient. However elemental nitrogen is abundant in the atmosphere and is dissolved in the water. Some species of algae (bluegreen algae or Cyanobacteria) can fix nitrogen (convert nitrogen to a bioavailable form) and thus alleviate any nitrogen limitation and continue to make use of the excess phosphorus and grow to excess. This is the case in FGR. Therefore, it is not effective or practical to mitigate only nitrogen. As explained in the TMDL, controlling nitrogen encourages growth of Cyanobacteria. The text of the TMDL explains more why excess Cyanobacteria are undesireable. Therefore, from a management perspective it is important to reduce phosphorus to the extent that it becomes the limiting factor and reduces the amount of algae that can grow in the reservoir. 4) What is the basis for the conclusion that Fruitgrowers stratifies “intermittently”? Division Response: This was based on the typical depth in FGR. The persistence and stability of stratification (or inversely, the likelihood of mixing) is a function of depth. Many lake processes and cycles are related to stratification and mixing. Therefore, typical depth and stratification patterns are often used as factors for identifying lakes of similar characteristics. (See Division response to POINTS TO CONSIDER IN FRUITGROWERS RESERVOIR TMDL DEVELOPMENT AND 303(d) LISTING #5 on Page 14.) 5) Page 18, Dillon Rigler model-used many North American Lake. Were they lakes or reservoir? Were they kept full or mostly drained? What was the range of size, depth, volume, temperature? Division Response: The Dillon-Rigler model is not a loading model, so there are no assumptions about filling or draining. Relevant lake information used to develop the model was summer average chlorophyll a concentrations and spring overturn phosphorus concentrations from a variety of lakes. These are important factors for most models. The important point of the Dillon-Rigler model is that it is a simple empirical approach to relating spring in-lake total phosphorus to summer chlorophyll. It is a general model developed using a variety of lakes, but does not represent one type of lake. 6) As noted in the Assessment, most of the calculations are based on one season of data more than a decade ago. If more data is collected in the future, or if the reservoir changes in capacity, will the Assessment be redone? Division Response: The assessment may be redone if the TMDL is implemented and the reservoir does not attain standards.

56

Final TMDL, October 2012 7) At what depth were the samples in Table 8 taken? Division Response: From 0.5 to 2 meters as depth-integrated samples.

57

Final TMDL, October 2012

Questions about load allocations: 1) What value of PO4 concentration was used to get the WWTP load? When we use 209 AF and 3.790 mg/l and we get 979 kg/yr compared to 1028 in the Assessment. Division Response: The WWTP load was calculated using monthly flows and monthly concentrations, then summed to get 1028 kg/yr. The 3.79 mg/L is an annual average. 2) We are curious as to what values do the models other than the 1984 Nurnberg model provide? It says there are other models used to estimate in-lake loading, but only the Nurnberg model accounts for in-lake loads. That appears to be a contradiction. Does the Nurnberg model account for the fact that cattle graze in part of the reservoir and along its banks or that the reservoir is largely drained annually? Division Response: There is no contradiction. The question confuses "Internal Load" (mass) with inlake phosphorus concentrations. The model is used to translate phosphorus loads to in-lake concentrations. While other models can also be used for this load to concentration translation, the other models do not account for internal loads. As stated in the TMDL text, the Nurnberg model separates the external and internal loads. The model uses in-lake phosphorus concentrations, annual areal water load, phosphorus retention, and external phosphorus load information to estimate internal loads. The model also can be used to calculate expected in-lake phosphorus concentrations based on observed or expected internal and external phosphorus loading. Cattle are part of the nonpoint source load. The Nurnberg model does not allocate nonpoint sources. The Nurnberg model uses annual parameters. It is not a dynamic hydrologic model. 3) In what other cases has the Division required a 99% reduction from an aerated lagoon POTW, especially one that discharges to an unclassified water? Division Response: The Division has not previously prepared a TMDL assigning a wasteload requiring a 99 percent reduction to an aerated sanitary lagoon. Several wastewater treatment plants in Colorado (under Control Regulations) reduce phosphorus to the levels required in this TMDL. However, many TMDLs have required 99% reductions in wasteloads. This is because TMDLs must be developed so as to attain assigned water quality standards. The required reductions (wasteloads and load reductions) are based on the differences between existing pollutant concentrations and the water quality standards. 58

Final TMDL, October 2012

Note that although Cedaredge WWTF discharges to an unclassified water, that unclassified water, Alfalfa Ditch, empties into FGR, which has been classified, by the WQCC, for Aquatic Life, Recreation, and Agricultural Uses. The dissolved oxygen standard which is not attained is associated with the reservoir‟s Aquatic Life Use classification. The reservoir therefore represents the point of application of instream standards. This is the same rational which serves as the basis of the Cedaredge WWTF Permit, CDPS CO-0031984.

59

Final TMDL, October 2012 POINTS TO CONSIDER IN FRUITGROWERS RESERVOIR TMDL DEVELOPMENT AND 303(d) LISTING 1) The water in the reservoir is privately owned and distributed for agricultural purposes. There has not been an effort to actually manage the water for recreational or aquatic life use. Division Response: The Reservoir is part of Regulation 35 Classifications and Numeric Standards for Gunnison and Lower Dolores River Basins (amended 1/10/11, effective 6/30/11), and is classified for Aquatic Life Warm 2, Agriculture, and Recreation. At present the reservoir has been classified for both Aquatic Life and Recreation Uses. A water right gives the holder the right to divert water for beneficial use, but not ownership of the water itself. Except in limited cases, diverted water is still considered waters of the state, and thus subject to the Water Quality Control Act. 2) We have no reason to believe the water in the reservoir is “waters of the state” and the process that resulted in a 303(d) listing should be carefully examined. Division Response: FGR is a classified water of the state. Only the Water Quality Control Commission has the authority to classify the beneficial uses of water in the state, with respect to water quality. Currently FGR is included in Regulation 35 Classifications and Numeric Standards for Gunnison and Lower Dolores River Basins (amended 1/10/11, effective 6/30/11) in the Lower Gunnison River Segment 09. The formal definition for “waters of the state”, as it appears in Regulation No. 31 at 31.5(38) is: “STATE WATERS” means any and all surface and subsurface waters which are contained in or flow in or through this state, but does not include waters in sewage systems, waters in treatment works of disposal systems, waters in potable water distribution systems, and all water withdrawn for use until use and treatment have been completed.”

60

Final TMDL, October 2012 3) The Orchard City Irrigation District has been trying to work with the Bureau of Reclamation to increase reservoir capacity, either by dredging or raising dam or the spillway. Will the state have funding to assess the effects of these changes to the reservoir? If not, who will pay for the Assessment update? Will removing sediment decrease the internal loading enough for the reservoir to achieve the water quality target? Will increasing the depth create new areas of stratification and low dissolved oxygen? Division Response: Regarding future assessments, see Division response to JVA Comment 2. Removing sediment is a method of reducing internal loading. However, the internal load results from, or is legacy watershed load (aka external load). Therefore, without adequate reductions to the watershed load, dredging would not have any practical or long-term benefits. Changes in the reservoir related to increasing the depth would depend on how much deeper the reservoir might become. 4) Classifying the reservoir as Aquatic Life Warm 2 may not make sense. While achieving that target may quite literally be out of the Town of Cedaredge‟s control, the current methodology places it squarely in the Town‟s lap, at an unreasonably high cost with a water quality result that is uncertain at best. Division Response: The Town is provided an opportunity to propose removal of the Aquatic Life Use classification through the WQCC rulemaking process. The procedural requirements are outlined in the Colorado Basic Standards and Methodolologies for Surface Waters, Regulation 31, at section 31.6 and in particular 31.6(2)(b). Classification of uses is not part of the TMDL. The TMDL provides an assessment of loads and necessary load reductions to attain existing, relevant water quality standards. In the case of FGR, the TMDL identifies nonpoint source reductions as well as point source reductions to attain the dissolved oxygen standard. 5) The reservoir is manmade and seasonal. Extrapolation of data from other lakes and reservoirs played a major part in TMDL development. While it may be the best available data, they are not necessarily from the same geology or topography, and the biology, chemistry and certainly the manual operation of the body of water could be quite different. If the lakes used for comparison are not adequately comparable, much of the rest of the analysis may be incorrect. Division Response: The chemistry and biology related to the dissolved oxygen impairments in FGR were nutrients and chlorophyll a. The biology of a lake or reservoir (based on 61

Final TMDL, October 2012 chlorophyll or algal productivity) is strongly influenced by nutrient concentrations. We were not looking for comparison lakes with similar nutrient concentrations or similar biology, but rather for lakes with similar processes. Depth affects the stability and persistence of stratification, which in turn is an important factor driving the oxygen, nutrient and biological processes of a lake or reservoir. The important point of comparison is that the comparable lakes and reservoir function similarly because of similar depths It was necessary to evaluate other lakes because the phosphorus concentrations in Fruitgrowers were so high, that conditions in the range of attainment of the DO standard were not observed which was not the case in the extended data set that included other lakes and reservoirs. Geology and topography are factors related to nutrient loading from the watershed, may be addressed as part of land use and or groundwater. The operation of the waterbody is accounted for in the loading assessments, as it was used to calculate loading to the reservoir as well as residence time of the reservoir. Again, it is not important to compare lakes with similar loading, it was important to compare lakes that would have similar oxygen and nutrient processes, the most important selection factor for which is depth. 6) It appears the lakes for comparison were chosen based on depth and how frequently they stratify. Does that alone cause them to have relatable chlorophyll a concentrations? Division Response: Depth influences the persistence and stability of stratification. Stratification is an important factor in processes that influence dissolved oxygen and nutrient cycling. Therefore, the Division considered depth to be the most important characteristic in selecting comparable lakes. Relatable chlorophyll a concentration was not a selection factor. In fact, FGR chlorophyll and phosphorus values are so high that they are not in a range that would exhibit attainment of the DO standard.

62

Final TMDL, October 2012

7) Can any reservoir model account for the variability of water depth, surface area and variable annual availability of water?.....Could a higher concentration of phosphorus, such as 1.5 mg/L, realistically accomplish the same benefits at a dramatically reduced impact to the community? Would the complete removal of our treated wastewater from the basin cause any measureable water quality improvement? With the pool of knowledge that currently exists regarding the complexities of nutrients in the environment, we believe those questions cannot be answered with enough certainty to warrant the cost. Division Response: Some models possibly could account for the factors listed, but the available data were insufficient to parameterize such models. In any case, 1.5 mg/L would not result in attainment of the dissolved oxygen standard. The TMDL concludes that reductions in point and nonpoint sources are required to attain the dissolved oxygen standard. The Division believes the TMDL targets can reasonably be expected to have FGR meet the existing DO standard, and are within ranges that other models would produce. 8) The nutrient concentration of the water used to fill the reservoir is dependent upon the time of year and upon which of the many reservoirs on Grand Mesa is being used to supply the Surface Creek drainage, which may not be the same from year to year. This builds on an argument that the equivalent of one growing season is not enough data to develop a legally binding discharge limit. Division Response: Variability of the water quality for total phosphorus concentrations from Surface Creek are well represented in the dataset. However, the TMDL was developed using medians, and other measures of central tendency. The Division acknowledges that the chemical water quality data from one year probably does not represent the worst case scenario on which to base a TMDL. The Division acknowledges that the current TMDL represents estimates, but believes the TMDL targets will meet the underlying Dissolved Oxygen standard. The targets were determined to be neither the most or nor the least restrictive among the range of targets that were identified. 9) From what we have available to us, it appears almost all of the data used in the calculations are over 10 years old. Is that a reasonable baseline from which to measure improvement? Division Response: Yes, it is reasonable. The Division found no information to suggest that water quality would have changed significantly over the past 10 years, other than 63

Final TMDL, October 2012 potential increases in the quantity of WWTP effluent. The assumption used was that there was no appreciable increase in effluent TP. This may result in an underestimate of the required WL reduction. 10) The reservoir samples that were taken were done so within a relatively short time after an accidental release of 800,000 gallons of partially treated sewage that sparked a great deal of attention to the reservoir and wastewater treatment lagoons. Are samples taken that shortly after the addition of nutrients be representative of the current limnology? Division Response: Yes, samples are representative of existing limnology. The Division would need more documentation on the timing of the spill, but 800,000 gallons is a trivial volume relative to the total inflow volume of the reservoir. It is unlikely that the effects of the spill produced long-term impacts on the reservoir. The TMDL was developed from annual values rather than worst case scenarios. 11) Summarized: The Committee disputes the anecdotal information suggesting the reservoir declined in water quality after the wastewater lagoons started discharging. The Committee notes there is also anecdotal information suggesting that effects of the treatment plant aren't evident, and additional anecdotal claims that the reservoir has been clearer in the last few years. Division Response: The Division has no record of the additional anecdotal information claimed by the Committee, and wonders why this wasn't raised during the many meetings with the Fruitgrowers Coalition. However, the anecdotal information was included in the TMDL as background description of concerns, but as there is no way to quantify anecdotal information, this information is not part of the modeling approach used or resulting TMDL. Furthermore, the impairment for Dissolved Oxygen, cannot be evaluated visually. Therefore, reports of improved clarity do not quantify attainment of the dissolved oxygen standard. 12) The Town has been monitoring quarterly for effluent phosphate since the 2006 permit was issued. Since 2005 the phosphate concentration has averaged about 3.5 mg/l rather than the 3.79 used in the TMDL. Since 2009, the average has been about 3.4 mg/L. This may seem like a minor difference, but the data since 2009 is about a 10% reduction from the valued used in the assessment calculations and would result in a drop of the Town’s overall contribution to the loading in Fruitgrowers Reservoir.

64

Final TMDL, October 2012 Division Response: The TMDL, Wasteload and load allocations and associated reductions were based on watershed and reservoir conditions for the years the study was conducted. The wasteloads were calculated using flows and concentrations reported in the Town‟s discharge monitoring reports (DMRs) from that time period. Note that the TMDL is based on annual phosphorus loads which are calculated using discharge volumes as well as concentrations. Thus, even if the effluent phosphorus concentrations decreased slightly, if the discharge volume increased there may be no significant change in the total annual load. Cedaredge DMR data were examined during development of the TMDL. Reported quarterly values for effluent TP concentrations ranged from 2.2 mg/L to 5.15 mg/l. The 5-year median is 3.4 mg/L. Annual average concentrations calculated using the quarterly values are: 2005=3.74 2006=3.630 2007=3.42 2008=3.618 2009=3.340 2010=3.705 These annual values indicate no significant decrease in effluent concentrations of total phosphorus.

13) It does not appear that the Assessment accounted for the impacts of grazing operations that surround the reservoir during high water and that actually graze within the flood fringe and basin itself as the reservoir is pulled down and periodically drained. There are irrigated, fertilized, and ungrazed lands to the north and northeast of the reservoir that do not flow into alfalfa Ditch before entering the reservoir. This does not appear to be included in Table 10 or accurately reflected in the calculations for the reservoir loading. Division Response: Grazing and other agricultural operations are part of the nonpoint source load. See response to Comment 2 in Questions of Assumptions above. 14) In reviewing the Standard Assessment Summary (SAS), we note that only 2 of 8 profiles were listed as out of attainment. For this and other reasons listed we anticipate recommending to the Cedaredge Board of Trustees that they participate in the collection of updated data to be compared with the old data. We will explain to them that the data could show that things are worse, better or substantially similar. We assume that there is no recent data due to the lack of EPA or CDPHE funding and that the Town of Cedaredge may have to provide 65

Final TMDL, October 2012 funding, and perhaps work with the NRCS or another local entity to ensure sample and data integrity. Regardless, the DO profile results are a significant piece of data and are too important to not know with more certainty, aren‟t they? Division Response: Impairment decisions are determined during the 303(d) listing process according to the most current iteration of the Section 303(d) Listing Methodology.. Should additional data become available it may be assessed in accordance with the State‟s Listing Methodology. This document identifies the procedures to be used for determining the attainment status of a given waterbody. The Listing Methodology is revisited on a biennial basis and the Town may wish to become involved in this process as it is reviewed in 2012. In any event should additional data adequately demonstrate the reservoir to be in attainment the TMDL will become moot and no additional actions will be necessary on the part of the Town.

POTENTIAL ECONOMIC AND SOCIAL IMPACTS; EQUITIBILITY A consulting firm hired by the Town of Cedaredge to update a facility plan is examining the life cycle costs of a facility that could meet the TMDL requirements. Preliminary costs for a facility that goes beyond enhanced biological nutrient removal lead them to believe that implementation would cause substantial and widespread adverse social and economic impacts to the community. The Committee questions the equitability of the TMDL and reiterates their concerns of the costs of implementing the TMDL. Division Response: The TMDL is written to attain the assigned water quality standards for dissolved oxygen. The Division made efforts to balance the point source and nonpoint source reductions based on technological feasibility. The Town has the opportunity to raise the issue of social and economic impacts to the WQCC as provided and subject to any limitations of Regulation 31.

The TMDL targets are set at a level modeled to achieve the dissolved oxygen standard. The dissolved oxygen standard was promulgated by the WQCC and, as such, is legally enforceable under the Clean Water Act. The Town of Cedaredge has the option of pursuing alternate standards based upon a demonstration that the existing standard is unattainable. Any such demonstration would require a showing that the standard is unattainable due to irreversible human caused conditions (31.7(1)(b)(ii)(Regulation 31 Basic Standards and Methodologies for Surface Water, Assigning Standards). Alternately, after 1/1/2013 the Town may pursue a site-specific variance per section 31.7(4) (Regulation 31 Basic Standards and Methodologies for Surface Water, Granting, Extending and Removing Variances to Numeric Standards (effective October 1, 2013)). There are several permit related options that may also be available, and the WQCD 66

Final TMDL, October 2012 TMDL program understands that there are ongoing discussions regarding permit related options. Additional Notes by the Division: Subsequent to the Public Notice period of this TMDL, the Division was asked to recommend projects for a new USDA-Natural Resources Conservation Service (NRCS) program named the National Water Quality Initiative (NWQI). Under the NWQI, the NRCS is directing their NRCS-state conservationists to spend 5% of their FYI 12 allotment of Environmental Quality Incentives Program (EQIP) funds to assist farmers in undertaking conservation measures that address high-priority water resources that meet certain criteria, including impaired waters on a state’s 303(d) list. The NWQI included nutrients as an impact to target for the funding. The Division recommended the Fruitgrowers Reservoir TMDL as a candidate for this funding. These funds are expected to be used to develop BMPs to reduce total phosphorus from the nonpoint sources contributing to the impairment of Fruitgrowers Reservoir.

This language is inserted into Section VIII, Restoration Planning and Implementation Process. Literature Citation Dillon, P. J., and F. H. Rigler. 1974. The phosphorus-chlorophyll relationship in lakes. Limnol. Oceanogr. 19:767-773. EPA, 2001. Ambient Water Quality Criteria Recommendations, Lakes and Reservoirs in Nutrient Ecoregion III. EPA 822-B-01-088. Eutrophication TMDLs. Kansas Department of Health and Environment: http://www.kdheks.gov/tmdl/eutro.htm. Hern, S.C., V.W. Lambou, L.R. Williams, and W.D. Taylor. 1981. Modifications of Models Predicting Trophic State of Lakes. EPA-600/3-81-001. Kirchner, W. B. and P. J. Dillon, 1975. An Empirical Method of Estimating the Retention of Phosphorus in Lakes. Water Resources Research 11(1):182. Larson, D. P. and H. T. Mercier. 1976. Phosphorus retention capacity of lakes. J. Fish. Res. Board Can. 33:1742-1750. Nurnberg, G.K. 1984. The Prediction of internal phosphorus load in lakes with anoxic hypolimnia. Limnol. Oceanogr., 29 (1) 111-124. 67

Final TMDL, October 2012 Nürnberg, Gertrud . 1998. Prediction of annual and seasonal phosphorus concentrations in stratified and polymictic lakes. Limnol. Oceanogr. 43:1544-1552. Ozga, John. 1999. Draft Working Paper: Fruitgrowers Reservoir 1998 Water Budget Evaluation. U.S.B.R. Preliminary Study of Eutrophication of Fruitgrowers Reservoir, Draft, CDPHE, WQCD, October 31, 2002. Thomann, R.V., and J.A. Mueller. 1987. Principles of Surface Water Quality Modeling and Control. Harper & Row, New York, NY. Vollenweider, R. A. (1976) Advances in defining critical loading levels for phosphorus in lake eutrophication. Memorie delllIstituto Italiano di Idrobiologia 3353-83. Vollenweider, R. A., W. Rast, and J. Kerekes. 1980. The phosphorus loading concept and Great Lakes eutrophication. In Phosphorus Management Strategies for Lakes, R. C. Loehr, C. S. Martin, and W. Rast, eds. pp. 207-234. Ann Arbor: Ann Arbor Science. WQCC 2005. Colorado Department of Public Health and Environment. Water Quality Assessment Alfalfa Ditch, Thence to Fruitgrowers Reservoir Cedaredge WWTF. WQCC 2010a. Colorado Department of Public Health and Environment, Water Quality Control Commission, Colorado‟s Section 303(d) List of Impaired Waters and Monitoring and Evaluation List, Effective April 30, 2010. WQCC 2010b. Colorado Department of Public Health and Environment, Water Quality Control Commission, Classification and Numeric Standards for Gunnison and Lower Dolores River Basins, Regulation 35, amended effective June 30, 2010. WQCC 2012. Colorado Department of Public Health and Environment, Water Quality Control Commission, The Basic Standards and Methodologies for Surface Water, Regulation No. 31. Effective 9/30/12.

68