I
UNITED STATES ENWRONMENW PROTECTION AGENCY REOlOM 8 8 S~REEI~ sum 300 DENVER, CO 80202-2466
-
oea 1
http:Ilwww.ep.g~ovl~ion~~
111
Ref: 8EfR-EP
1
RECEIVED
Mrudr T.m,Director Water Quality Control Division Colorado Department of Public Health and Environment 4300 C b n y Creek Drive South Denver, Colorado 80246- 153 0
Re:
L Approvds -Ammonia
BdZdm Creek (COPB009) BO&I& Creek (COSPBOIO) Sai $ Yrain Creek (COSPsY03)
Q
Dear Mr. P a w :
We have completed our review ofthe total your office for the St. Vrain River &ahage (Tri-Bash).
daily load (TMDL) as submitted by
[TheTMDL is included in the document
entitled f l
-
to Coal Creek Sement 9, Boulder Creek Cod Creek tb St. Vrain Creek S e m m t 10. St. Vrain Creek Hygiene Rd to S . Plane River - Sement 3 ~ouldkrand Weld Counties. Colorado, ~ 0 1 0 d Department o ofPublic Health and ~ n v i r o n m 4May, 2003). This d o m e n t was submitted to us for review and approval in compondenqe dated May 28,2003 and signed by you. In accordance with the CIean Water Act (33 U.S.C. 125 1 e t seq.), we approve all aspects ofthe TMDL as developed for the water quality limited waterbody as descnied in Section 303(d)(l). Enclosure 1 to this letter provides a the elements ofthe m L and EncIosure 2 provides details of our review of the TMDL.
Based on our review, we fkelthe separate TMDL !elementslisted in Enclosure 2 adequately address the pollutant of concern, taking into consideration seasonal variation and a margin of safety. In approving this TMDL, EPA affumshat the TMDL has been established at a level necessary to attain and maintain the applicable wat standards and has the necessary componmts of an agprovable TMDL,
TRank you for your submittal. If you have any qukFdons concerning this approval, feel free to contact Kathryn Hernandez of my staffat 303/31216101.
Assistant R$md Administrator. Office of E systems Protection and Remediation
Tri-basin TMDL – Ammonia
Final
TOTAL MAXIMUM DAILY LOAD ASSESSMENT
Ammonia Boulder Creek, South Boulder Creek to Coal Creek - Segment 09 Boulder Creek, Coal Creek to St. Vrain Creek - Segment 10 St.Vrain Creek, Hygiene Rd to S. Platte River - Segment 03 Boulder and Weld Counties, Colorado May, 2003 TMDL SUMMARY Waterbody Name/Segment Number
Mainstem of Boulder Creek from South Boulder Creek to Coal Creek, Coal Creek to St. Vrain Creek, St. Vrain Creek from Hygiene Road to the South Platte River COSPBO09, COSPBO10 & COSPSV03
Pollutant/Condition Addressed
Ammonia (NH3) (Protection of aquatic life use)
Affected Portion of Segment
Boulder Creek from South Boulder Creek to Coal Creek, Coal Creek and St. Vrain Creek from Left Hand Creek to I25
Use Classification/Waterbody Designation
COSPBO09, COSPBO10: Water Supply, Agriculture Aquatic life Warm 1, Recreation 1a COSPSV03: Agriculture Aquatic life Warm 1, Recreation 1a
Waterbody Antidegradation Designation
COSPBO09: Reviewable COSPBO10, COSPSV03: Use Protected
Water Quality Target
Assure the un-ionized ammonia concentrations do not exceed 0.06 mg/L in-stream through implementation of controls on discharges of ammonia
TMDL Goal
Attain Colorado water quality un-ionized ammonia standards
I.
EXECUTIVE SUMMARY
Segments 9 and 10 of Boulder Creek and segment 3 of Saint Vrain Creek have been identified as water-quality limited for ammonia based on water quality data. Low-flow modeling indicates that municipal wastewater treatment facilities are the primary sources of ammonia in these segments. This TMDL derives wasteload allocations for ammonia that are necessary to ensure attainment of Colorado water quality standards for ammonia in these streams. Stormwater runoff from nonpoint sources and localized groundwater do not contribute significantly to the ammonia impairment. Water quality monitoring will be necessary to verify that the TMDL requirements result in attainment of the standards. This TMDL should be reviewed when municipal dischargers propose plant expansions beyond the limits utilized in this modeling effort, new wastewater treatment plants are proposed on the main stem or tributaries, or when assumptions JLB C:Boulder\TMDL Boulder.doc
1
5/8/03
Tri-basin TMDL – Ammonia
Final
included in this TMDL assessment are shown to be no longer appropriate. II.
INTRODUCTION
Section 303(d) of the federal Clean Water Act requires states to identify waterbodies or stream segments that are not meeting water-quality standards or have the potential to exceed the standards. Water-quality limited segments currently identified in Colorado are included on the 1998 303(d) list. Water-quality limited segments are those in which one or more classification or standard is not, or may not be fully achieved. The State is required to develop a TMDL assessment for every segment and parameter that is listed. The TMDL describes those changes that will ensure attainment of the stream standard. The TMDL document includes quantification of the amount of pollutants that a segment can assimilate without exceeding water quality standards. The TMDL document also apportions that allowable pollutant load among multiple pollutant sources. The maximum allowable pollutant load is referred to as the Total Maximum Daily Load (TMDL). The TMDL is comprised of the Load Allocation (LA), which is that portion of the pollutant load attributed to background or to nonpoint sources, the Waste Load Allocation (WLA) which is that portion of the pollutant load associated with point-source discharges, and the Margin of Safety (MOS). The St.Vrain Creek and Boulder Creek ammonia TMDLs were assigned a ‘medium’ priority for completion on the 1998 303(d) listing. Completion of the TMDL at this time is consistent with that prioritization. The Colorado Environmental Coalition and Biodiversity Legal Foundation filed a complaint against EPA in 1997 alleging that EPA had failed to assure that Colorado had established a reasonable schedule for completion of TMDLs. The complaint was subsequently amended to address completion of TMDLs for those waters included on the 1998 303(d) list. Colorado intervened in the lawsuit and was signatory to a Settlement Agreement filed in August 1999. The Settlement Agreement stipulated a revised schedule for TMDL completion and included a commitment on the State’s part to complete the St.Vrain Basin ammonia TMDLs by June 30, 2002. The TMDL development process was driven, in large part, by the efforts of a stakeholder group made up of the potentially affected municipalities. Boulder and St.Vrain Creeks are part of the South Platte Basin - Hydrologic Unit Codes (HUC) 10190005. Both Creeks have their headwaters in Boulder County. See Figure 1. South Boulder Creek flows into Boulder Creek on the eastern edge of the City of Boulder. Further east, Rock Creek flows into Coal Creek, which then flows into Boulder Creek downstream of Erie, Colorado. Several miles below the confluence with Rock Creek, Boulder Creek flows into St.Vrain Creek in Weld County. St.Vrain Creek enters the South Platte River near Fort Saint Vrain. This drainage is referred to herein as the ‘Tri-basin’ area. These streams have their headwaters in the Indian Peaks Wilderness Area and in Rocky Mountain National Park. They flow through alpine tundra, sub alpine forests, montane, and prairie ecosystems and feed wetlands and habitat for waterfowl. As they flow out of the mountains through the foothills and across the plains, numerous ditches, providing water for JLB C:Boulder\TMDL Boulder.doc
2
5/8/03
Tri-basin TMDL – Ammonia
Final
agricultural and domestic water supply uses, divert the streams. The creeks become slower and shallower as they flow through the urban, suburban and rural environment. Boulder Creek and St.Vrain Creek receive some water from transbasin diversions. The Moffat tunnel brings west slope water to Gross Reservoir via South Boulder Creek and the Big Thompson/Windy Gap Project provides water to Boulder Reservoir via the St. Vrain supply canal. Rock Creek, with its headwaters on Rocky Flats, flows through an increasingly suburban environment, the Rock Creek subdivision development. Rock Creek flows into Coal Creek downstream of Louisville/Superior and Lafayette. Coal Creek then flows into Boulder Creek in a rural area downstream of Erie. From here Boulder Creek flows several miles before entering St. Vrain Creek downstream of Longmont. Water withdrawals take place at numerous locations along Boulder and St.Vrain Creeks and at several locations along Coal Creek, as well. See Figures 2 and 3. Land use in the St.Vrain watershed is open space, residential, commercial and industrial, and agricultural. Industrial and commercial areas are adjacent to Boulder Creek, particularly where the river flows through the eastern edge of the City of Boulder. There are some industrial and commercial areas adjacent to St.Vrain Creek, as well, as it flows through Longmont. Bike paths, parks and open space, border many of the Creeks, particularly as they flow through urban and suburban areas. In addition, there are rural influences, especially in the lower portions of Boulder Creek and St.Vrain Creek. The entire St. Vrain basin, especially in the foothills and plains, is becoming progressively more influenced by human intervention. This TMDL assessment is accomplished through the use of a water-quality model. The modeling includes the St. Vrain Basin and its major tributary, Boulder Creek, from the edge of the foothills to the confluence with the South Platte. South Boulder Creek and Coal Creek are the major tributaries to Boulder Creek. Although only segments 9 and 10 of Boulder Creek and segment 3 of St. Vrain Creek are impaired, all or portions of Boulder Creek segments 2, 5, 7a, 7b, 8, 11, 12 and St.Vrain Creek segments 2, 5, and 6 are included in the model. The contributions of ammonia to the listed stream segments from the upstream segments are very small or negligible. The primary sources of ammonia are the municipal wastewater treatment facilities located in the listed segments. Lyons, Longmont, Niwot, Boulder, Erie, Lafayette, Louisville, and Superior, as well as several much smaller discharging facilities, are located in this basin. The St. Vrain Sanitation District, which discharges to St.Vrain Creek east of I25, services unincorporated housing developments in the lower portion of the basin. This TMDL relies on the modeling work performed by Drs.William Lewis and James Saunders from the University of Colorado – Center for Limnology, as reported in “Modeling and Analysis of Ammonia for Total Maximum Daily Load in the St. Vrain Creek Drainage, including Boulder Creek and Coal Creek” dated December 31, 2002, henceforth referred to as “the TMDL Modeling Report”. This report is available under separate cover.
JLB C:Boulder\TMDL Boulder.doc
3
5/8/03
Tri-basin TMDL – Ammonia III.
Final
WATER QUALITY STANDARDS
The Colorado Water Quality Control Commission (“WQCC”) has adopted a chronic waterquality standard for un-ionized ammonia of 0.06 mg/L (30 day average) for segments 9 and 10 of Boulder Creek and for segment 3 of St. Vrain Creek. The acute standard for un-ionized ammonia is the Table Value Standard (TVS), which is based upon the receiving stream pH and temperature. Both acute and chronic standards are in non-attainment.
Figure 1. Map of Saint Vrain Creek and it's primary tributaries.
JLB C:Boulder\TMDL Boulder.doc
4
5/8/03
Tri-basin TMDL – Ammonia
Final
Figure 2. Line diagram of Boulder Creek and St. Vrain Creek, Boulder and Weld Counties, Colorado.
JLB C:Boulder\TMDL Boulder.doc
5
5/8/03
Tri-basin TMDL – Ammonia
Final
Figure 3. Line diagram of Coal Creek, Boulder and Weld Counties, Colorado.
JLB C:Boulder\TMDL Boulder.doc
6
5/8/03
Tri-basin TMDL – Ammonia IV.
Final
PROBLEM IDENTIFICATION
Exceedences of both the acute and chronic un-ionized ammonia standards assigned to Boulder Creek have been documented since the late 1980s. EPA reports that a TMDL addressing ammonia was first developed when the City of Boulder prepared an application for renewal of the 75th Street wastewater treatment facility discharge permit in 1985. Although the facility maintained compliance with the ammonia limits included in the permit, instream monitoring showed the chronic un-ionized ammonia standard was violated some five to nine miles downstream (dependant upon temperature and pH conditions). Subsequent investigation indicated that riparian habitat was degraded to the point that those instream conditions favoring conversion of ammonia to its toxic, un-ionized form was exacerbated. A riparian habitat restoration project was developed and implemented in the early 1990s. Ammonia continued to be a concern in the Boulder Creek drainage even as treatment plant improvements and riparian habitat restoration was proceeding. Boulder Creek from its confluence with South Boulder Creek to Coal Creek, segment COSPBO09, and the following segment, Boulder Creek from Coal Creek to St. Vrain Creek (COSPBO10) were included on the 1992 303(d) List. These segments were identified as “partially supporting” their assigned aquatic life use classifications. The parameter identified as the basis for the listings was un-ionized ammonia. These segments have appeared on subsequent 303(d) Lists promulgated in 1994, 1996 and 1998 as monitoring continued to indicate non-attainment of the assigned ammonia standards. A similar problem has existed in St. Vrain Creek. St. Vrain Creek, segment COSPSV03, has also been included on 303(d) lists promulgated in 1992, 1994, 1996 and 1998. Segment COSPSV03 is described as St. Vrain Creek from Hygiene Road in Longmont to its confluence with the South Platte River. Boulder Creek intersects St. Vrain Creek in segment COSPSV03, just downstream of Longmont. The problems in St. Vrain Creek are similar to those in Boulder Creek. Both have a large domestic wastewater treatment facility discharging at the upper end of the segment. Both Boulder Creek and St. Vrain Creek are also subject to instream and riparian conditions which promote the formation of un-ionized ammonia. A review of the data for the St. Vrain basin was performed for the most recent South Platte River Basin standards hearing in 2000. Data for Boulder Creek segment 9 from the City of Boulder and the WQCD indicate an 85th percentile ambient un-ionized ammonia level at 0.062 mg/l from 244 data points, with 20 of those data points exceeding the acute ammonia standard. For segment 10, the 85th percentile of the data from 34 data points was 0.075 mg/l, with several exceedances of the acute standard as well. There are a number of other municipal wastewater treatment plants located in the St. Vrain drainage. The Cities of Louisville, Lafayette and Erie all discharge to Coal Creek, a tributary to Boulder Creek. Additional smaller facilities include a number of municipalities and trailer courts. The facilities that represent significant ammonia point sources are listed in Table 1. Because instream pH and temperature conditions dictate the conversion of ammonia to its unionized state, and because this process may occur over a considerable length of stream, it is necessary to consider the cumulative impact of this large number of facilities in order to fully assess the ammonia issues in the St. Vrain watershed. JLB C:Boulder\TMDL Boulder.doc
7
5/8/03
Tri-basin TMDL – Ammonia Wastewater Treatment Facility
Final Design Capacity (MGD)
Receiving Water
San Lazaro MHP
0.110
Boulder Cr, SPBO02
Lyons WWTF
0.375
St. Vrain, just above gage
Niwot WWTF
1.18
Dry Cr, SPSV06
Erie WWTF
1.2
Coal Cr, SPBO07b
Louisville WWTF
4.2
Coal Cr, SPBO07b
Lafayette WWTF
4.4
Coal Cr, SPBO07b
Boulder WWTF
25
Boulder Cr, SPBO09
Longmont WWTF
14.0
Red Lion Inn Boulder Mountain Lodge (Orodell, Inc.) St. Vrain WWTF
0.009
Boulder Cr, SPBO02
0.0045
Fourmile Cr, SPBO02
4.5
B&B Mobile and RV Park
0.015
Rock Creek WWTF
St. Vrain, SPSV03
2.4
St. Vrain, SPSV03 Boulder Cr, SPBO10 Rock Cr, SPBO08
Table 1. Design Capacities for major point-source dischargers. The work documented by the TMDL Modeling Report was based on an expansion of the Colorado Ammonia Model (CAM) to accommodate the thirteen dischargers in the affected portion of the basin and to account for the complex hydrology of the basin. There are many diversions and significant return flows/seepage along the length of these streams from the foothills to the South Platte River. St. Vrain Creek from Lyons to Longmont, in particular, is rife with many small diversions and significant seepage/return flows. The Colorado ammonia standards are based on the un-ionized fraction of the total ammonia. The un-ionized fraction is that portion that is most toxic to aquatic life. The amount of ammonia that is un-ionized is based on pH and temperature, which vary throughout the day as well as throughout the seasons. Higher instream pH and temperature favors the formation of the unionized ammonia fraction. In addition, some of the ammonia may be removed by nitrification and denitrification in stream, i.e. the “ammonia loss”. The CAM accounts for these complex chemical and biological processes that affect the concentration of un-ionized ammonia in stream below an outfall. JLB C:Boulder\TMDL Boulder.doc
8
5/8/03
Tri-basin TMDL – Ammonia
Final
The modeling work performed for this TMDL is the first such comprehensive assessment for the Tri-basin area. It is based on more extensive and current data and considers the effects of multiple dischargers on the receiving streams. This modeling work has a much better scientific basis than previous assessments performed to facilitate issuance of individual discharge permits. The 1998 303(d) listing for the three segments was modified by the addition of aquatic life as a condition subject to impairment. Fish community investigations performed by the Colorado Division of Wildlife indicated that there has been a loss of fish community diversity and abundance in the drainage over time. Some native warm water species have disappeared entirely. It is likely that instream levels of un-ionized ammonia are a factor in the reported changes in the aquatic community. It is also likely that riparian and instream habitat degradation is also a factor. Developments of TMDLs addressing the aquatic life issue are currently scheduled for completion in 2004. V.
WATER-QUALITY GOALS
The goal of this TMDL assessment is to assure attainment of the Colorado acute and chronic water quality standards for ammonia. Process controls at the affected municipal wastewater treatment facilities, enforced through permit effluent limits and compliance schedules, should result in attainment of standards. VI.
POLLUTANT SOURCE ANALYSIS
As part of the TMDL assessment, all pollutant sources must be considered and contributions from significant sources must be quantified. From this information, the TMDL model can show how much pollutant loading must be reduced in order to meet water quality standards under specific conditions. Pollutant sources include discrete discharges of a pollutant from a pipe or other discharge structure (point-sources), ungaged surface discharges through a pipe (stormwater discharges), or diffuse discharges across a broad reach of stream (nonpoint sources). Identification of Sources: Ammonia enters the St.Vrain Basin primarily from municipal wastewater discharges. Some of these treatment plant effluents are not nitrified or not fully nitrified. Nitrification refers to treatment processes intended to convert ammonia into less toxic nitrate. Significant Point-Source Dischargers – There are thirteen wastewater treatment facilities discharging to this portion of the St. Vrain watershed. These are the primary sources of ammonia to the basin. Boulder and Longmont are the major dischargers. Louisville, Lafayette, Erie, Niwot, Lyons, Superior (Rock Creek), and St. Vrain Sanitation District also discharge to the basin, as well as the smaller effluent sources of Red Lion Inn, Orodell Inc. (Boulder Mountain Lodge), San Lazaro Mobile Home Park, and B&B Mobile and RV Park. The current design capacities for the point-source dischargers are listed in Table 1. Other Permitted Dischargers – Several other facilities are authorized to discharge to this watershed, but they are not significant sources of ammonia. The Ft. St.Vrain Excel Energy plant discharges just upstream of the mouth of the St.Vrain; although this facility is permitted to discharge 3.2 mgd to the lower portion of the St Vrain, this discharge is very infrequent. Xcel Energy -Valmont station also dischargers to the basin; this discharge essentially comprises the JLB C:Boulder\TMDL Boulder.doc
9
5/8/03
Tri-basin TMDL – Ammonia
Final
flow of South Boulder Creek. Although there are a couple groundwater remediation facilities located in the basin, the flows are insignificant and they are not likely sources of ammonia. Likewise, although several water treatment plants have discharge permits for backwash water, this water is not a significant volume nor is it anticipated to be a source of ammonia. Contributions from Dischargers outside the Tri-basin area – The main contributors of ammonia are the basin wastewater treatment facilities. The upstream physical boundaries for the model are the “boundary reservoirs” of Gross Reservoir on South Boulder Creek, Barker Reservoir on Boulder Creek and Buttonrock Reservoir on St. Vrain Creek. There are several small wastewater treatment facilities upstream of these reservoirs. However, indications are that these facilities, located in the mountains upstream of the reservoirs, contribute negligible amounts of ammonia to the impacted portions of the 303 (d) listed segments located on the plains. See ammonia discussion in the TMDL Modeling Report. Other sources The contribution from non-point sources is minimal. In most cases, non-effluent water sources provide dilution of ammonia. There were six locations where there were adequate records to indicate that non-effluent sources are contributing some (very small, less than 1 mg/l) total ammonia to the stream. Although this data was used in the model as background concentrations, no specific sources could be identified for this ammonia. VII.
TECHNICAL ANALYSIS
The TMDL assessment accounts for all influences on loads and concentrations of ammonia based on monitoring data, effluent characteristics, or other pertinent information. The goal of the modeling is to identify the point source allocations that would ensure attainment of the ammonia standards in-stream. Ammonia concentrations, flows, temperature and pH must be evaluated for each source. Time of travel, nitrification (ammonia loss rate) and denitrification (nitrate loss rate) must be evaluated for the river itself. Much of the technical information detailed in this section is extracted from the TMDL Modeling Report. Data Sources: See the TMDL Modeling Report for details of protocols used in establishing model inputs. The following is a summary of the sources of data used in the model. Flow The complex hydrology of the basin resulted in a significant amount of resources being used to compile and analyze the flow data to provide appropriate inputs to the flow model. Data from eleven USGS gaging stations, reservoir outlet flow measurements from the Colorado Decision Support System (CDSS) and from Xcel Energy, in addition to diversion records (primarily CDSS) and treatment facility records were used to establish flow inputs. Flow residuals were calculated by dividing the basin into reaches with adequate flow records, bracketing the flows and evaluating, based on mass balances at gaged locations, the contribution from these ungaged sources.
JLB C:Boulder\TMDL Boulder.doc
10
5/8/03
Tri-basin TMDL – Ammonia
Final
The last 10 years of flow records (water years 1991-2000) were used if sufficient data was available for that period. CDPHE provided historical flow records for most of the smaller facilities located in the basin. Gaps and errors in the hydrologic data sets were usually resolved by using values from adjacent data sets. See Table 5 of the TMDL Modeling Report and further discussion in the Hydrology section of the Technical Analysis portion of this report. Chemistry Channel geometry, source water chemistry and effluent chemistry were needed in order to run the model. Recent chemistry data, primarily from 1995 forward, was used in the model. When available, characteristic values were used for model inputs. Where data was limited, chemistry values had to be estimated. Judgment was used to estimate values based on analogous situations or knowledge of specific conditions. In other cases, the values were set so that they did not alter the conditions of the receiving water. In some cases special data handling was required because dilution or ammonia removal would change the effluent contribution before it was added to the modeled reach. Each of these cases is discussed in detail in the TMDL Modeling Report. Site-specific velocity equations were used for estimating travel time in the model. The equations were derived from data for seven gaging stations (SEO and USGS data) and were applied to stream reaches with similar channel geometry and substrate. (See Table 40 in the TMDL Modeling Report.) Using the available data, and professional judgment where data did not exist, the source water inputs to the model were developed. At six sites where total ammonia data existed for source waters, median concentrations were computed for each month. See Table 2. The six locations with available data were Lyons and Left Hand Creek at the mouth, Boulder Creek and Fourmile Creek at Orodell, South Boulder Creek at the mouth and Coal Creek above Louisville. USGS, J. McCutchan, the Cities of Boulder, Longmont, and Louisville and Xcel Energy provided data used in the model. After reviewing reservoir data and information on each specific reservoir, including information on stratification, distance from receiving waters, etc. it was concluded that reservoir releases in this basin were not likely to be sources of ammonia. In addition, there was no evidence that the ungaged flows, including the agricultural return flows, were a source of ammonia. (See pages 46-54 of the TMDL Modeling Report.) For the discharging facilities with more complete effluent chemistry records, those records were used as input to the model. When paired pH and temperature data was available, that data was used to calculate the percent un-ionized ammonia on each date. The median percent un-ionized ammonia was then determined for each month from the individual values. This median percent un-ionized ammonia and median temperature were then used to back-calculate the “characteristic” pH. This approach yields separate pH and temperature values while preserving the computational consistency in calculating the median percent un-ionized ammonia. For small and/or infrequent dischargers, where sampling records were usually minimal, some judgment had to be used to estimate the effluent chemistry. Where only limited data was available, every attempt was made to obtain median pH and temperature values separately. (See Tables 27, 28, and 29 of the TMDL Modeling Report for the monthly median pH, temperature and percent unionized ammonia for effluents used in the model. Also see Table 30 for characteristic pH data for Longmont, Louisville and Erie, which were used in the model. When available, model inputs were characteristic values.) JLB C:Boulder\TMDL Boulder.doc
11
5/8/03
Tri-basin TMDL – Ammonia
Site St. Vrain at Lyons St. Vrain Supply Canal1 Foothills Reservoir release1 Left Hand Creek at mouth Union Reservoir release1 Howlett Gulch1 Boulder Creek at Orodell Fourmile Creek at Orodell South Boulder Creek at mouth Boulder Supply Canal1 Dry Creek Carrier1 Panama Reservoir release1 Coal Creek above Louisville 1
Final
Jan 0.08
Feb 0.08
Mar 0.02
Apr 0.02
May 0.01
Jun 0.01
Jul 0.01
Aug 0.02
Sep 0.01
Oct 0.02
Nov 0.08
Dec 0.11
0.20
0.20
0.20
0.20
0.20
0.20
0.34
0.34
0.34
0.34
0.20
0.20
0.08 0.04 0.03
0.10 0.04 0.03
0.05 0.04 0.04
0.10 0.04 0.03
0.01 0.04 0.03
0.04 0.04 0.03
0.00 0.04 0.03
0.07 0.04 0.03
0.02 0.04 0.03
0.09 0.04 0.06
0.00 0.04 0.04
0.01 0.04 0.07
0.03
0.05
0.03
0.10
0.03
0.03
0.02
0.03
0.04
0.03
0.04
0.05
Set to 0.0 mg/L.
Table 2. Medians of measured total ammonia concentrations (mg N/L) for the Tri-Basin TMDL model. See model for explanation of sources and procedures for estimation.
JLB C:Boulder\TMDL Boulder.doc
12
5/8/03
Tri-basin TMDL – Ammonia
Final
Where site-specific amplitudes and times of maxima were available, they were used in the model. Where data was limited, judgment was used in determining whether the model default values or other methods should be used to derive appropriate values. For un-monitored waters (seepage, return flows, etc) entering the streams at various locations pH, temperature and total ammonia data was not available. For most of these the flow was minimal. The ammonia was set to 0.0 mg/l and the pH and temperature values were set such that they did not change the pH or temperature of the receiving waters. The Colorado Ammonia Model (CAM) is normally run with the conservative assumption that low flows occur at extreme conditions of pH and temperature. However, a specific analysis of the setpoint conditions for the St.Vrain model showed that there was no correlation between low flows and these extreme conditions. (See Appendix I of the TMDL Modeling Report.) Therefore, the recurrence analysis in the CAM used median pH and temperature to produce setpoint conditions. Boulder Creek ammonia loss rates were evaluated based on a mass balance using data from three reaches where adequate data was available. The three reaches stretched from below Boulder WWTF to Coal Creek. This resulted in estimation from the first reach, which was used as the removal rate for warm-water reaches. Because of lack of confidence in the predicted removal rates for the lower two reaches and because there was only good correspondence between the observed and predicted loss rates for the first reach when the model was run with the validation data set, the rate from the first reach was assigned to all warm-water reaches. The default ammonia removal rate in CAM was used for the cold-water reaches. Hydrology Low-Flow Analysis Stream hydrology is an essential ingredient in modeling ammonia. Because of the numerous dischargers and diverters to and from the Tri-basin area in addition to significant seepage and agricultural return flows in portions of the basin, the hydrology of the area is very complex. The model inputs were based on the most accurate representation of historical low-flow conditions in each stream. The Water Quality Control Division (“Division”) uses DFLOW4 software to develop critical low-flow estimates for permitting purposes. The acute critical low flow is the empirically based 1-day low flow with an average 1-in-3 year recurrence interval. DFLOW4 was used to develop low-flow estimates for the WWTP discharge points. For each location, the monthly acute DFLOW values were obtained using the gage records for the last ten years of record (1991 – 2000), if available, or the most complete and current data judged to be most representative of the last 10 years of flows at that location. The low-flow values for the tributaries and diversions were estimated by the method of differences, to match the DFLOW values in the mainstem in terms of water balance. Tables 3 and 4 show the chronic and acute low-flow values developed for this analysis.
JLB C:Boulder\TMDL Boulder.doc
13
5/8/03
Tri-basin TMDL – Ammonia
Final Chronic Flow, cfs
Site St. Vrain at Lyons Lyons WWTP St. Vrain above Supply Canal Supply Canal
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
16.0
16.0
16.0
21.0
28.0
118.0
65.0
32.0
21.0
20.0
18.0
16.0
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
17.2
17.2
17.0
22.2
29.9
119.7
68.3
35.2
23.9
21.7
19.2
17.2
0.0
0.0
0.0
0.0
2.0
23.0
115.0
24.0
6.0
0.0
0.0
0.0
St. Vrain above Foothill Release
4.9
5.5
3.3
5.6
9.6
3.7
21.8
19.0
18.0
10.2
7.2
5.0
Foothill Release
2.4
2.4
2.6
2.6
3.6
25.7
31.0
8.0
0.0
0.0
0.1
2.2
St. Vrain above Left Hand
14.2
14.6
4.1
3.2
20.9
2.1
40.6
39.7
46.5
18.6
8.4
14.9
Left Hand Creek
-0.2
-0.6
10.2
10.1
-7.9
33.8
1.2
-4.8
-11.6
0.3
4.8
-1.6
St. Vrain above Longmont WWTP
15.0
15.0
15.0
14.2
14.2
37.0
43.0
36.0
36.0
20.0
14.2
14.2
Longmont WWTP
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
St. Vrain above Dry Creek
39.9
39.9
38.9
38.9
39.7
62.5
68.5
61.5
61.5
45.5
39.3
38.9
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
42.3
42.2
41.1
41.2
42.1
64.9
70.9
63.9
63.9
47.9
41.7
41.2
Dry Creek St. Vrain above Union release Union Reservoir release St. Vrain above Boulder Creek Boulder Creek St. Vrain above St. Vrain SD Oxbow Lake Release St. Vrain above Howlett Gulch Howlett Gulch Mouth of St. Vrain Boulder Creek at Orodell
0.0
0.0
0.0
0.0
0.0
0.0
35.0
11.0
0.0
0.0
0.0
0.0
45.8
45.7
43.6
44.6
46.8
70.1
111.1
80.4
68.9
52.0
45.4
44.7
71.8
72.7
67.1
2.0
7.0
33.7
15.3
21.1
9.8
10.6
36.8
66.0
130.7
130.1
117.0
48.4
71.1
98.5
122.8
121.1
82.3
43.3
95.4
125.4
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
146.8
145.4
132.0
64.9
98.6
134.6
158.6
163.5
114.7
58.3
111.5
142.7
3.0
3.0
3.5
3.5
3.5
1.0
6.0
1.0
0.0
0.0
1.0
1.0
165.1
161.3
148.8
84.3
136.2
184.0
207.8
219.6
157.1
71.7
127.7
160.9
10.5
10.5
10.5
16.0
21.0
67.0
43.0
31.0
17.0
12.0
12.0
10.5
Red Lion
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Four Mile Creek
0.5
0.5
0.6
1.1
2.6
0.6
0.1
0.1
0.1
0.1
0.6
0.6
Boulder Creek above San Lazaro
4.8
2.3
3.3
9.6
11.5
1.8
14.2
1.3
5.5
3.8
3.4
5.5
San Lazaro
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
South Boulder Creek
0.0
0.2
0.2
0.2
0.0
16.0
3.5
0.2
0.2
0.2
0.0
1.1
Boulder Creek above Boulder Supply
7.3
4.6
6.1
13.7
16.5
11.8
13.1
2.0
7.0
5.6
4.1
8.8
Boulder Supply Canal
1.9
4.7
2.6
-6.5
-9.3
27.7
36.4
19.8
-0.2
1.0
2.7
-0.9
Boulder Creek above Boulder WWTP
9.4
9.4
9.0
7.7
7.8
40.0
50.0
22.0
7.2
7.0
7.0
8.1
Boulder WWTP
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
Boulder Creek above New Dry Carrier
49.2
49.3
48.1
46.4
50.3
70.3
58.1
58.7
39.3
41.5
41.3
38.2
0.0
0.0
0.0
0.0
0.0
0.0
2.0
1.0
0.0
0.0
0.0
0.0
Boulder Creek above Coal Creek
51.1
51.6
48.6
17.1
9.1
27.5
9.3
34.2
18.5
23.1
30.5
42.1
Coal Creek
18.3
18.3
18.3
9.4
5.3
4.8
2.9
3.7
3.9
8.8
18.7
18.3
B&B MHP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Panama Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
71.8
72.7
67.1
2.0
7.0
33.7
15.3
21.1
9.8
10.6
36.8
66.0
New Dry Carrier
Boulder Creek at mouth Coal Creek above Louisville
1.0
0.8
0.8
1.1
1.5
1.1
0.6
0.6
0.6
1.5
1.4
1.0
Louisville WWTP
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
Coal Creek above Rock Creek
6.3
6.3
6.3
6.4
6.8
6.4
6.3
6.3
6.3
6.8
6.7
6.3
Rock Creek
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
Coal Creek above Lafayette WWTP
9.7
9.7
9.7
9.8
10.2
9.8
9.7
9.7
9.7
10.2
10.1
9.7
Lafayette WWTP
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
Coal Creek above Erie WWTP
16.5
16.5
16.5
7.6
3.5
3.0
1.0
1.9
2.1
7.0
16.9
16.5
Erie WWTP Coal Creek at mouth
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
18.3
18.3
18.3
9.4
5.3
4.8
2.9
3.7
3.9
8.8
18.7
18.3
C:Boulder\TMDL Boulder.doc Table 3 JLB Continued on next page.
14
5/8/03
Tri-basin TMDL – Ammonia
Final Chronic Concentration, mg/L
Site St. Vrain at Lyons
Jan 0.08
Feb 0.08
Mar 0.02
Apr 0.02
May 0.01
Jun 0.01
Jul 0.08
Aug 0.02
Sep 0.01
Oct 0.02
Nov 0.08
Dec 0.11
Lyons WWTP
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
St. Vrain above Supply Canal
0.9
0.9
0.8
0.6
0.5
0.1
0.9
0.4
0.6
0.6
0.8
0.9
Supply Canal
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Foothill Release
0.1
0.1
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.1
Foothill Release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Left Hand
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Left Hand Creek
0.0
0.0
0.2
0.2
0.0
0.2
0.0
0.0
0.0
0.3
0.2
0.0
St. Vrain above Longmont WWTP
0.0
0.0
0.1
0.1
0.0
0.2
0.0
0.0
0.0
0.0
0.1
0.0
Longmont WWTP
25.0
25.0
20.3
11.2
6.1
9.1
25.0
10.7
6.8
10.6
14.9
23.0
St. Vrain above Dry Creek
10.9
10.8
8.5
4.5
2.1
2.1
10.9
2.6
1.6
3.6
6.3
10.1 9.4
9.9
9.3
6.5
4.1
2.1
1.6
9.9
1.1
1.1
2.5
7.2
10.4
10.3
8.0
4.3
1.9
1.9
10.4
2.4
1.4
3.3
6.0
9.6
Union Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Boulder Creek
7.9
7.7
5.8
2.9
1.0
1.2
7.9
1.3
0.9
2.3
4.3
7.2
Boulder Creek
1.9
1.3
0.4
0.0
0.0
0.0
1.9
0.0
0.0
0.0
0.4
1.5
St. Vrain above St. Vrain SD
2.6
2.3
1.2
1.0
0.2
0.1
2.6
0.2
0.1
0.5
1.1
2.2
Oxbow Lake Release
3.8
3.8
1.4
2.0
0.3
0.1
3.8
0.1
0.2
0.6
1.3
3.3
St. Vrain above Howlett Gulch
1.8
1.7
0.7
0.6
0.1
0.0
1.8
0.1
0.0
0.2
0.6
1.5
Howlett Gulch
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Mouth of St. Vrain
1.0
0.9
0.3
0.2
0.0
0.0
1.0
0.0
0.0
0.1
0.2
0.8
Dry Creek St. Vrain above Union release
0.1
0.1
0.1
0.1
0.0
0.0
0.1
0.1
0.0
0.1
0.0
0.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
Four Mile Creek
0.3
0.3
0.3
0.2
0.1
0.3
0.3
1.6
1.6
1.6
0.3
0.3
Boulder Creek above San Lazaro
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
South Boulder Creek
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek above Boulder Supply
0.4
0.5
0.3
0.2
0.1
0.0
0.4
0.0
0.1
0.2
0.3
0.3
Boulder Supply Canal
0.0
0.0
0.0
0.2
0.1
0.0
0.0
0.0
0.1
0.0
0.0
0.3
Boulder Creek above Boulder WWTP
0.3
0.2
0.2
0.1
0.1
0.0
0.3
0.0
0.1
0.1
0.1
0.3
18.5
10.9
5.3
9.7
11.3
23.5
18.5
15.1
15.1
15.1
21.6
24.8
Boulder Creek above New Dry Carrier
8.7
5.1
2.3
4.2
4.2
4.7
8.7
2.9
3.8
4.6
7.8
10.7
New Dry Carrier
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek above Coal Creek
3.7
2.1
0.8
0.8
0.3
0.1
3.7
0.1
0.2
0.5
1.4
3.5
Coal Creek
3.2
3.3
2.0
2.2
2.3
1.0
3.2
0.8
0.9
1.4
1.5
2.6
B&B MHP
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
Panama Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek at mouth
1.9
1.3
0.4
0.0
0.0
0.0
1.9
0.0
0.0
0.0
0.4
1.5 0.1
Boulder Creek at Orodell Red Lion
San Lazaro
Boulder WWTP
0.0
0.1
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
10.3
12.0
7.6
12.0
11.7
8.8
10.3
8.0
9.2
11.0
7.5
9.0
Coal Creek above Rock Creek
2.8
3.2
1.9
2.6
2.1
1.2
2.8
0.8
0.7
1.3
1.5
2.2
Rock Creek
3.9
2.3
1.8
2.5
1.5
0.9
3.9
0.5
0.6
1.7
2.3
3.6
Coal Creek above Lafayette WWTP
1.8
1.6
1.0
1.3
0.9
0.5
1.8
0.2
0.2
0.6
0.9
1.5
10.3
12.0
7.6
12.0
12.5
9.9
10.3
10.6
11.0
10.5
7.5
9.0
2.0
2.2
1.3
1.3
0.5
0.1
2.0
0.0
0.0
0.5
1.0
1.6
25.0
25.0
15.2
13.9
16.4
9.9
25.0
8.9
9.3
12.8
13.5
21.7
3.2
3.3
2.0
2.2
2.3
1.0
3.2
0.8
0.9
1.4
1.5
2.6
Coal Creek above Louisville Louisville WWTP
Lafayette WWTP Coal Creek above Erie WWTP Erie WWTP Coal Creek at mouth
Table 3. Continued on next page. JLB C:Boulder\TMDL Boulder.doc
15
5/8/03
Tri-basin TMDL – Ammonia
Final Chronic Load, kg/d
Site
Jan 3
Feb 3
Mar 1
Apr 1
May 1
Jun 3
Jul 2
Aug 2
Sep 1
Oct 1
Nov 4
Dec 4
Lyons WWTP
35
35
35
35
35
35
35
35
35
35
35
35
St. Vrain above Supply Canal
37
36
34
34
34
37
35
33
32
34
37
38
Supply Canal
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Foothill Release
1
1
0
0
1
0
0
1
1
1
1
1
Foothill Release
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Left Hand
0
0
0
0
0
0
0
0
0
0
0
0
Left Hand Creek
0
0
5
5
0
17
1
0
0
0
2
0
St. Vrain above Longmont WWTP
0
0
5
5
0
15
1
0
0
0
2
0
Longmont WWTP
1325
1325
1076
594
323
482
578
567
360
562
790
1219
St. Vrain above Dry Creek
1068
1056
811
432
203
323
377
386
235
401
604
965
St. Vrain at Lyons
Dry Creek St. Vrain above Union release
44
42
29
18
9
7
3
5
5
11
32
42
1077
1060
805
429
196
308
356
369
225
393
611
973
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Boulder Creek
887
864
615
316
116
199
247
264
150
290
479
788
Boulder Creek
338
227
69
0
0
0
0
0
0
1
34
244
St. Vrain above St. Vrain SD
819
727
346
113
28
31
45
64
28
58
261
665
Union Reservoir release
Oxbow Lake Release St. Vrain above Howlett Gulch
65
65
24
34
5
1
0
1
3
11
22
56
657
588
231
91
15
12
14
25
13
33
172
522
0
0
0
0
0
0
0
0
0
0
0
0
404
359
107
42
4
3
2
5
3
10
76
308
Boulder Creek at Orodell
2
3
1
4
1
7
0
5
1
3
0
0
Red Lion
1
1
1
1
1
1
1
1
1
1
1
1
Four Mile Creek
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above San Lazaro
0
0
0
0
0
0
0
0
0
0
0
0
10
10
10
10
10
10
10
10
10
10
10
10
South Boulder Creek
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above Boulder Supply
6
5
5
5
5
1
1
0
2
2
3
7
Boulder Supply Canal
0
0
0
-2
-3
0
0
0
0
0
0
-1
Howlett Gulch Mouth of St. Vrain
San Lazaro
6
5
5
3
2
1
1
0
2
2
2
6
Boulder WWTP
1751
1031
502
918
1069
2224
1845
1429
1429
1429
2044
2347
Boulder Creek above New Dry Carrier
1051
609
270
471
519
814
461
413
363
467
791
1006
Boulder Creek above Boulder WWTP
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above Coal Creek
469
265
91
34
6
6
0
9
7
31
104
364
Coal Creek
143
149
88
51
30
12
4
7
9
30
71
117
B&B MHP
1
1
1
1
1
1
1
1
1
1
1
1
Panama Reservoir release
0
0
0
0
0
0
0
0
0
0
0
0
338
227
69
0
0
0
0
0
0
1
34
244
New Dry Carrier
Boulder Creek at mouth
0
0
0
0
0
0
0
0
0
0
0
0
133
154
98
154
151
113
109
103
118
142
97
116
Coal Creek above Rock Creek
42
49
29
40
35
19
14
12
11
22
24
34
Rock Creek
33
19
15
20
12
7
3
4
5
14
19
30
Coal Creek above Lafayette WWTP
43
39
24
32
23
11
6
5
5
14
22
35
172
200
127
200
208
165
172
177
183
175
125
150 64
Coal Creek above Louisville Louisville WWTP
Lafayette WWTP Coal Creek above Erie WWTP Erie WWTP Coal Creek at mouth
81
89
53
23
4
1
0
0
0
8
43
114
114
69
63
74
45
34
40
42
58
61
99
143
149
88
51
30
12
4
7
9
30
71
117
Table 3. Low Flows, chronic total ammonia and total ammonia loads derived from the TMDL analysis. JLB C:Boulder\TMDL Boulder.doc
16
5/8/03
Tri-basin TMDL – Ammonia
Final Acute Flow, cfs
Site St. Vrain at Lyons
Jan 11.7
Feb 14.0
Mar 11.7
Apr 15.0
May 36.0
Jun 130.0
Jul 84.0
Aug 43.0
Sep 20.0
Oct 13.0
Nov 14.0
Dec 13.0
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
12.9
15.2
12.7
16.2
37.9
131.7
87.3
46.2
22.9
14.7
15.2
14.2
Supply Canal
0.0
0.0
0.0
0.0
17.0
44.0
90.0
15.0
0.0
0.0
2.0
0.0
St. Vrain above Foothill Release
3.2
3.2
2.4
2.4
4.6
3.7
7.5
7.3
6.5
5.0
5.0
3.4
Foothill Release
0.0
0.6
0.0
0.6
1.6
16.2
25.3
12.6
0.0
0.0
0.0
0.0
12.5
12.3
4.1
3.2
4.6
2.1
9.1
16.1
19.2
12.7
6.9
13.0
0.5
0.8
10.2
5.6
3.4
31.8
21.8
-7.3
-1.3
10.2
2.1
-4.8
St. Vrain above Longmont WWTP
14.0
14.0
15.0
9.7
9.1
35.0
32.0
9.9
19.0
24.0
10.0
9.1
Longmont WWTP
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
St. Vrain above Dry Creek
38.9
38.9
38.9
34.4
34.6
60.5
57.5
35.4
44.5
49.5
35.1
33.8
Lyons WWTP St. Vrain above Supply Canal
St. Vrain above Left Hand Left Hand Creek
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
41.3
41.2
41.1
36.7
37.0
62.9
59.9
37.8
46.9
51.9
37.5
36.1
Dry Creek St. Vrain above Union release
0.0
0.0
0.0
0.0
0.0
0.0
15.0
34.0
11.0
0.0
0.0
0.0
44.8
44.7
43.6
40.1
41.7
68.1
80.1
77.3
62.9
56.0
41.2
39.6
Union Reservoir release St. Vrain above Boulder Creek Boulder Creek St. Vrain above St. Vrain SD
65.9
68.1
63.4
2.0
7.0
32.2
15.3
19.1
9.8
7.2
23.6
61.5
123.8
124.5
113.3
43.9
51.2
92.9
91.8
104.1
76.3
43.9
78.0
115.8
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
139.9
139.8
128.3
60.4
78.7
128.9
127.6
146.5
108.7
58.8
94.1
133.1
Oxbow Lake Release St. Vrain above Howlett Gulch
3.0
4.0
3.5
3.5
3.5
2.0
6.0
8.0
0.0
0.0
1.0
1.0
158.1
156.7
145.1
79.8
116.2
176.4
176.8
209.6
151.1
72.2
110.3
151.3
Boulder Creek at Orodell
3.1
3.9
4.9
6.3
18.0
97.0
48.0
33.0
17.0
12.0
3.1
7.9
Red Lion
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Four Mile Creek
0.5
0.5
0.9
1.1
5.0
2.8
0.1
0.1
0.1
0.3
0.5
0.6
Boulder Creek above San Lazaro
3.1
2.0
2.6
4.9
9.5
1.8
4.3
1.3
1.6
2.4
3.4
3.2
San Lazaro
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
South Boulder Creek
0.1
0.0
0.0
0.0
1.0
2.4
4.1
0.6
0.1
0.0
0.0
0.1
Boulder Creek above Boulder Supply
5.6
4.1
5.3
8.8
15.5
4.2
4.2
2.0
3.2
4.0
4.0
5.5
-2.4
0.5
-0.2
-7.0
-7.0
29.3
30.2
12.8
1.5
-2.2
-1.8
-2.3
Howlett Gulch Mouth of St. Vrain
Boulder Supply Canal
3.5
4.8
5.3
2.2
9.1
34.0
35.0
15.0
5.1
2.2
2.4
3.4
Boulder WWTP
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
Boulder Creek above New Dry Carrier
43.3
44.7
44.4
40.9
31.6
53.3
43.1
38.7
35.0
19.8
28.5
33.5
Boulder Creek above Boulder WWTP
0.0
0.0
0.0
0.0
0.0
0.0
7.0
3.0
0.0
0.0
0.0
0.0
Boulder Creek above Coal Creek
45.2
47.0
44.9
5.3
1.1
18.4
6.6
13.0
4.7
1.4
17.7
37.4
Coal Creek
18.3
18.3
18.3
9.7
4.8
2.9
2.9
3.7
4.1
8.3
18.3
18.5
B&B MHP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Panama Reservoir release
0.0
0.0
0.0
0.0
0.0
11.3
0.0
0.0
0.0
0.0
0.0
0.0
65.9
68.1
63.4
2.0
7.0
32.2
15.3
19.1
9.8
7.2
23.6
61.5
Coal Creek above Louisville
0.9
0.7
0.5
1.4
0.9
0.5
0.7
0.3
1.2
1.0
1.0
1.2
Louisville WWTP
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
Coal Creek above Rock Creek
6.3
6.3
6.3
6.7
6.3
6.3
6.3
6.3
6.5
6.3
6.3
6.5
Rock Creek
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
Coal Creek above Lafayette WWTP
9.7
9.7
9.7
10.1
9.7
9.7
9.7
9.7
9.9
9.7
9.7
9.9
Lafayette WWTP
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
16.5
16.5
16.5
7.9
3.0
1.0
1.0
1.9
2.3
6.5
16.5
16.7
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
18.3
18.3
18.3
9.7
4.8
2.9
2.9
3.7
4.1
8.3
18.3
18.5
New Dry Carrier
Boulder Creek at mouth
Coal Creek above Erie WWTP Erie WWTP Coal Creek at mouth
Table 4. Continued on next page JLB C:Boulder\TMDL Boulder.doc
17
5/8/03
Tri-basin TMDL – Ammonia
Final Acute Concentration, mg/L
Site St. Vrain at Lyons
Jan 0.08
Feb 0.08
Mar 0.02
Apr 0.02
May 0.01
Jun 0.01
Jul 0.01
Aug 0.02
Sep 0.01
Oct 0.02
Nov 0.08
Dec 0.11
Lyons WWTP
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
St. Vrain above Supply Canal
1.3
1.1
1.3
1.0
0.4
0.1
0.2
0.4
0.7
1.1
1.1
1.2
Supply Canal
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Foothill Release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Foothill Release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Left Hand
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Left Hand Creek
0.2
0.2
0.2
0.2
0.2
0.2
0.3
0.0
0.0
0.3
0.2
0.0
St. Vrain above Longmont WWTP
0.0
0.0
0.1
0.1
0.1
0.2
0.2
0.0
0.0
0.1
0.0
0.0
Longmont WWTP
30.0
30.0
30.0
24.8
23.5
30.0
30.0
24.2
26.2
30.0
27.3
30.0
St. Vrain above Dry Creek
13.4
13.3
12.6
11.1
9.0
7.0
6.8
8.5
7.8
9.5
12.6
14.8
Dry Creek
11.7
11.2
7.8
5.7
3.5
2.3
0.9
1.3
1.3
4.4
8.6
11.3
St. Vrain above Union release
12.7
12.6
11.7
10.1
7.9
6.3
6.1
7.4
6.9
8.7
11.7
13.9
Union Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Boulder Creek
9.6
9.4
8.4
6.8
4.1
3.8
2.9
2.4
3.4
6.0
8.2
10.0
Boulder Creek
2.9
2.3
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.5
1.8
St. Vrain above St. Vrain SD
3.3
3.1
1.9
2.1
0.5
0.4
0.3
0.3
0.5
1.5
2.2
2.8
Oxbow Lake Release
4.6
4.6
1.7
2.4
0.3
0.1
0.0
0.1
0.2
0.7
1.6
4.0
St. Vrain above Howlett Gulch
2.4
2.2
1.1
1.1
0.1
0.1
0.1
0.1
0.1
0.6
1.1
1.9
Howlett Gulch
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Mouth of St. Vrain
1.3
1.2
0.5
0.4
0.0
0.0
0.0
0.0
0.0
0.2
0.4
1.0
0.1
0.1
0.1
0.1
0.0
0.0
0.0
0.1
0.0
0.1
0.0
0.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
Four Mile Creek
0.4
0.4
0.2
0.2
0.0
0.1
2.0
2.0
2.0
0.7
0.4
0.3
Boulder Creek above San Lazaro
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
South Boulder Creek
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek above Boulder Supply
0.5
0.6
0.4
0.2
0.2
0.0
0.0
0.0
0.1
0.2
0.3
0.5
Boulder Supply Canal
0.5
0.0
0.4
0.2
0.2
0.0
0.0
0.0
0.0
0.2
0.3
0.5
Boulder Creek above Boulder WWTP
0.4
0.5
0.4
0.2
0.1
0.0
0.0
0.0
0.0
0.1
0.3
0.5
Boulder WWTP
30.0
22.3
15.3
28.0
30.0
30.0
30.0
30.0
30.0
28.3
29.5
30.0
Boulder Creek above New Dry Carrier
14.9
10.7
6.7
12.4
10.4
6.0
5.3
5.5
7.3
7.4
10.2
13.0
New Dry Carrier
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek above Coal Creek
5.6
4.1
2.1
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.8
3.6
Coal Creek
5.7
5.8
4.0
4.2
4.3
2.8
1.9
2.6
3.0
3.4
4.0
5.0
B&B MHP
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
Panama Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek at mouth
2.9
2.3
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.5
1.8
Boulder Creek at Orodell Red Lion
San Lazaro
0.0
0.1
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
28.6
30.0
20.3
25.3
26.6
24.9
25.1
24.9
27.1
26.9
24.5
26.3
Coal Creek above Rock Creek
7.7
8.0
5.1
5.5
4.8
4.0
2.6
2.5
2.0
3.2
4.8
6.6
Rock Creek
3.6
2.3
2.3
3.9
2.7
2.2
0.9
1.3
1.1
3.1
3.4
3.7
Coal Creek above Lafayette WWTP
3.6
3.4
2.3
2.7
1.9
1.4
0.7
0.7
0.5
1.2
2.1
3.1
28.6
30.0
20.3
25.3
30.0
30.0
30.0
30.0
30.0
30.0
24.5
26.3
Coal Creek above Louisville Louisville WWTP
Lafayette WWTP Coal Creek above Erie WWTP Erie WWTP Coal Creek at mouth
5.2
5.4
3.4
2.7
0.9
0.0
0.0
0.1
0.1
1.2
3.2
4.4
30.0
30.0
24.4
25.4
30.0
27.0
27.0
30.0
30.0
30.0
30.0
30.0
5.7
5.8
4.0
4.2
4.3
2.8
1.9
2.6
3.0
3.4
4.0
5.0
Table 4. Continued on next page JLB C:Boulder\TMDL Boulder.doc
18
5/8/03
Tri-basin TMDL – Ammonia
Final Acute Load, kg/d
Site
Jan 2
Feb 3
Mar 1
Apr 1
May 1
Jun 3
Jul 2
Aug 2
Sep 0
Oct 1
Nov 3
Dec 3
Lyons WWTP
43
43
43
43
43
43
43
43
43
43
43
43
St. Vrain above Supply Canal
42
43
40
40
41
44
42
40
39
39
42
43
Supply Canal
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Foothill Release
0
0
0
0
0
0
0
0
0
0
0
0
Foothill Release
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Left Hand
0
0
0
0
0
0
0
0
0
0
0
0
Left Hand Creek
0
0
5
3
2
16
18
0
0
8
1
0
St. Vrain above Longmont WWTP
0
0
5
2
1
14
16
0
0
8
1
0
Longmont WWTP
1590
1590
1590
1314
1245
1590
1590
1282
1388
1590
1447
1590
St. Vrain above Dry Creek
1277
1261
1197
933
761
1037
960
736
848
1145
1080
1223
St. Vrain at Lyons
Dry Creek St. Vrain above Union release Union Reservoir release St. Vrain above Boulder Creek Boulder Creek St. Vrain above St. Vrain SD Oxbow Lake Release St. Vrain above Howlett Gulch
52
50
35
25
16
10
4
6
6
20
39
51
1286
1267
1180
911
715
977
892
683
793
1108
1071
1225
0
0
0
0
0
0
0
0
0
0
0
0
1055
1028
902
663
416
628
574
448
518
819
824
968
466
386
153
0
0
0
0
0
0
0
26
267
1012
939
530
225
60
91
74
73
90
164
415
788
78
78
28
41
6
1
0
1
3
13
26
67
807
753
348
163
28
35
21
28
39
86
263
616
0
0
0
0
0
0
0
0
0
0
0
0
494
457
161
75
8
7
3
6
10
27
113
360
Boulder Creek at Orodell
1
1
1
2
0
9
0
6
1
3
0
0
Red Lion
1
1
1
1
1
1
1
1
1
1
1
1
Four Mile Creek
1
1
1
1
1
1
0
0
0
0
1
1
Boulder Creek above San Lazaro
0
0
0
0
0
0
0
0
0
0
0
0
12
12
12
12
12
12
12
12
12
12
12
12
South Boulder Creek
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above Boulder Supply
7
6
5
5
6
0
0
0
0
2
3
7
-3
0
0
-4
-3
0
0
0
0
-1
-1
-3
Howlett Gulch Mouth of St. Vrain
San Lazaro
Boulder Supply Canal
4
6
5
1
3
0
0
0
0
1
1
4
Boulder WWTP
2839
2110
1448
2650
2839
2839
2839
2839
2839
2678
2792
2839
Boulder Creek above New Dry Carrier
1574
1175
732
1243
805
781
559
519
627
359
710
1068
Boulder Creek above Boulder WWTP
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above Coal Creek
624
469
226
11
0
0
0
1
0
0
36
326
Coal Creek
256
258
179
100
51
20
13
24
30
69
181
228
B&B MHP
2
2
2
2
2
2
2
2
2
2
2
2
Panama Reservoir release
0
0
0
0
0
0
0
0
0
0
0
0
466
386
153
0
0
0
0
0
0
0
26
267
New Dry Carrier
Boulder Creek at mouth
0
0
0
0
0
0
0
0
0
0
0
0
Louisville WWTP
368
386
261
326
342
320
323
320
349
346
315
338
Coal Creek above Rock Creek
118
123
78
89
73
62
40
38
32
49
74
104
Rock Creek
30
19
19
32
23
18
7
11
9
26
28
31
Coal Creek above Lafayette WWTP
85
81
53
66
46
34
16
16
12
27
50
74
Lafayette WWTP
476
500
338
421
500
500
500
500
500
500
408
438
Coal Creek above Erie WWTP
211
216
138
52
7
0
0
0
1
19
131
180
Erie WWTP
136
136
111
115
136
123
123
136
136
136
136
136
Coal Creek at mouth
256
258
179
100
51
20
13
24
30
69
181
228
Coal Creek above Louisville
Table 4. Low flows, acute total ammonia and total ammonia loads derived form the TMDL analysis. JLB C:Boulder\TMDL Boulder.doc
19
5/8/03
Tri-basin TMDL – Ammonia
Final
Data from gauging stations, numerous diversions records and wastewater treatment facility records were compiled and analyzed to establish flow inputs. Flow residuals were significant in some portions of the basin. These ungaged flows, due to surface flows and seepage, were accounted for in estimating low flows. Flows were evaluated by dividing the basin into four reaches. Each of these reaches was bounded by stream locations with adequate flow records. The four reaches are: Ordell to 75th Street to the confluence with ST. Vrain including Coal Creek, St. Vrain at Lyons to the confluence with Boulder Creek including Left Hand Creek, and St. Vrain Creek from Boulder Creek to the South Platte River. The Coal Creek residual could not be established separately due to the lack of flow data; it was incorporated in calculations for the lower Boulder Creek reach. Flow data was generally restricted to the last 10 years of data. Monthly values of residual flows were calculated for each reach. For the reach from Lyons to Longmont, agricultural return flows have been accounted for separately from seepage. Information from this analysis of residuals was used, along with daily records of diversion and additions, to calculate the daily flows under low-flow conditions at points along the reach. From this information, a low flow profile of the stream was constructed. See Table 5. Flow Accumulation/Reset Specialized calculations were used to maintain historical DFLOW values above the Longmont and Boulder WWTP outfalls. See the TMDL Modeling Report for details. CDPS permits are issued using the assumption that the facility will be discharging at design capacity. This assumption complicates the model when running it for multiple dischargers. The model uses the low flows above the WWTPs calculated from the closest gage. For Boulder and Longmont the differences in flows calculated by the reach model and those calculated from the nearest downstream gage had to be reconciled. For Longmont, the St. Vrain flows just above the Longmont outfall were adjusted by using the Left Hand Creek flows. The Creek flows were adjusted so that the historical DFLOW values in the St.Vrain above Longmont’s outfall are used in the model, so called “DFLOW by difference” based on historical conditions. Appropriate chemistry adjustments were also made. For Boulder Creek, the flows above the WWTP outfall were adjusted for the low flow values by using the Boulder Supply ditch as the adjusting mechanism. The assumption of historical low flows above each discharger was incorporated into the model by resetting the flow of the river to the historical low flow above each discharger, thus removing the increment between historical low flow and capacity discharges just above each new discharge. Table 5 summarizes flows for each of the four reaches modeled, lists the threshold flows and provides the residual flows per mile for each reach. See the model report for a catalog of the features of each reach.
JLB C:Boulder\TMDL Boulder.doc
20
5/8/03
Tri-basin TMDL – Ammonia
Reach
Final
Boulder Creek* Orodell to 75th 75th to mouth
Threshold, cfs Length, mi
100 11.40
100 14.40
Lyons to Longmont, seepage 200 15.96
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0.86 0.71 0.93 1.72 2.11 *2.15 2.18 0.66 1.51 1.22 0.83 0.76
0.34 0.41 0.46 0.22 1.28 6.67 3.07 3.87 2.01 1.31 1.25 0.81
1.66 1.63 1.14 1.55 1.96 *1.96 1.96 1.96 1.96 1.96 1.75 1.56
St. Vrain Lyons to Longmont, ag returns* 200 11.23
Longmont to mouth 300 14.84
0.00 0.00 0.00 0.00 1.54 *1.13 5.43 5.19 4.33 1.03 0.00 0.00
2.19 1.99 1.92 2.30 4.90 6.95 6.91 8.48 6.08 1.92 2.19 2.46
Table 5. Residual flows (cfs/mi) for each month in each of the four reaches defined for the analysis. In some months (denoted with an *), too few dates had upstream flows less than the threshold, and residuals were estimated by interpolation. Special considerations: Orodell Inc. discharges to Four Mile Creek (COSPBO02) just upstream of its confluence with Boulder Creek. The low flows calculated for the gage at Fourmile were used to define the dilution flow available each month to dilute the effluent ammonia before it reaches Boulder Creek. Because St. Vrain Sanitation District discharges to Oxbow Lake (COSPSV06), which flows into segment 3 of the St. Vrain, special methods were used to calculate the concentration of ammonia entering St.Vrain Creek from this facility. These are described in the TMDL Modeling Report. The low flow is zero in Dry Creek and Rock Creek, the receiving streams for Niwot and Superior, respectively. When there are flows, the flow in each of these streams is mostly effluent. The ammonia standards of 0.1 mg/l for chronic and TVS for acute apply. The contribution from these waterbodies (COSPSV06 and COSPBO08 ) to the ammonia load in Boulder Creek was considered in the model. Using reasonable assumptions to facilitate modeling of these two small tributaries, the reach model approach was used to develop discharge limits based on an assessment of critical point conditions below the respective outfalls and to predict the amount of ammonia delivered at the mouth of each stream. See Table 6 for chronic and acute total ammonia effluent limits. (See pages 83 and 84 of the TMDL Modeling Report for assumptions used in the model.)
JLB C:Boulder\TMDL Boulder.doc
21
5/8/03
Tri-basin TMDL – Ammonia
Final
Modeling Methodology For this TMDL, the Colorado Ammonia Model was modified to accommodate thirteen dischargers. The CAM determines the allowable concentration of total ammonia in the effluents that will allow the un-ionized ammonia standards to be maintained in the receiving streams. CAM uses instream data, effluent data and many variables (for example – stream velocity, ammonia loss rates, pH amplitude and time of maximum, etc.) to calculate the allowable ammonia for each discharger. Instream flow values were determined from the low flow analysis. The design capacities of each facility were used for the effluent flows. The basin was divided into three modeling reaches: Coal Creek, Boulder Creek and the St. Vrain mainstem. Initially, the chronic effluent limits were set to 25 mg/l total ammonia and the acute limits were set to 30 mg/l total ammonia in all months for all discharges. The model was then run, adjusting these concentrations downward to achieve compliance with stream standards. The limits were lowered uniformly for all the dischargers within each of the three modeled reaches. The model was run again, starting at the bottom of each reach and relaxing each discharger’s effluent concentration until the assimilative capacity of the stream was exhausted. In this way each successive discharger downstream in a reach has limits equal to or greater than those of the next closest upstream discharger. This approach maintains an element of equitability among the various dischargers. If the upstream discharge is isolated from the next discharge downstream, this incremental procedure was not necessary. In some months, the critical point below the Boulder WWTP is below the confluence with Coal Creek and, in that case, the Coal Creek discharge limits were adjusted to achieve parity with Boulder. Also in some months, the critical point below the Longmont WWTP is below the confluence of the St. Vrain with Boulder Creek. In that case, the limits for discharges on Boulder and Coal Creek were adjusted to achieve parity with Longmont. The permit limits that result from this highly iterative process are shown on Table 6. Please refer to Tables 3 and 4 (six pages) for the complete picture of allowable chronic and acute concentrations and loads, as well as the chronic and acute flows used in the model. All sources of ammonia were categorized for the purpose of modeling. Some sources had no significant effect on instream ammonia concentrations, and so were deemed to be non-significant sources for the purposes of this TMDL. Point-Source Dischargers - The municipal dischargers are treated as independent variables. The other point-source dischargers were not included in the model because their ammonia loads and/or the quantity of the discharge were minimal. Again, these sources are considered nonsignificant in terms of their effect on instream ammonia concentrations. Sensitivity analysis showed that these sources did not affect the final in-stream ammonia levels.
JLB C:Boulder\TMDL Boulder.doc
22
5/8/03
Tri-basin TMDL – Ammonia
Final
Contributions from Tributaries and Upstream - Modeling requires assignment of ammonia concentrations at low flow for all water sources, including tributaries. Typical total ammonia data was extracted from data provided by USGS, City of Longmont, City of Boulder, Orodell, Inc., Xcel Energy and the City of Louisville. Specific estimation procedures are included in the model. The background values used in the model are listed separately in Table 2 and reflected at appropriate locations on Tables 3 and 4. Seepage – Ungaged flows from groundwater seepage were accounted for by calculation of flow residuals between adjacent gages. An analysis of discharge at paired gages, taking into account diversions, point-source discharges, and tributary flows support an estimate of the contributions from ungaged flows in distinct reaches of the river. Because of the instability of residual values associated with the higher flows, a threshold flow was established for each reach and the calculation of residuals was restricted to dates below this threshold. In order to provide more realistic seepage estimates, some of the residual flow was assigned to Left Hand Creek and a seasonal component was added for agricultural return flows. The agricultural return flows were a particularly important factor for the reach from Lyons to Longmont. In this reach, the return flows were estimated by subtracting the contribution of Left Hand Creek from the irrigation season flow residuals that had already been reduced by the winter baseline values. See the model report for details of special treatment of the Lefthand Creek and Lyons to Longmont flows. The Coal Creek residual was incorporated into the lower Boulder Creek residual because gaged flows were not available to establish a separate residual for this Creek. See pages 13 through 29 of the TMDL Model Report for a discussion of the estimation procedures for ungaged flows and the analyses procedures used in the model. VIII. TMDL ALLOCATION Allocation Methodology The sources contributing to the ammonia load in the Boulder/St. Vrain Basin have been identified and categorized for the purpose of load allocation. Wastewater treatment plants are the primary sources and a reduction in ammonia discharged by municipal wastewater treatment plants will have to occur in order to assure attainment of the ammonia standard throughout the streams. Modeling established the basis for discussions among the municipal dischargers that were expected to have allocations (Boulder, Lyons, Longmont, Niwot, Louisville, Superior and Lafayette). Allocations to these dischargers were based on a strategy developed in coordination with the dischargers and the State, considering the impacts of each discharger on the stream individually and in concert with the influence of their neighboring facilities. When modeling indicated that a facility or facilities would have an impact on another discharger, the model was adjusted to set the allowable concentrations equal to each other for these facilities. As noted previously, the default condition in the Colorado Ammonia Model is that low flow coincides with extreme stream conditions. Based on a statistical study of the Tri-Basin area that documented no meaningful relationship between low flow and extreme conditions of percent JLB C:Boulder\TMDL Boulder.doc
23
5/8/03
Tri-basin TMDL – Ammonia
Final
un-ionized ammonia (based on pH and temperature), this default assumption is unnecessarily conservative in this case. Therefore this assumption was not used in the modeling for the Tribasin area. TMDL and Allocations Table 6 presents a summary of the modeled concentration of total ammonia (in mg/L) at selected locations that reflects the maximum allowable level that would attain the ammonia standard in stream throughout the Tri-basin area. The Load Allocations have not been summarized in a specific table in this TMDL. At most locations the background concentration of ammonia is 0.0 mg/l. However that is not the case for the six locations listed in Table 2. Because of the complexity of the model, the specific Load Allocation was not sorted out as a separate entity in relation to the Waste Load Allocation. The loads at the stream locations immediately upstream of the discharge points reflect the remaining loads from upstream sources, whether that load is due to background levels or to upstream dischargers. Table 6 is a summary of the allowable effluent limit concentrations for total ammonia in mg/l for the thirteen discharge point locations included in the St. Vrain Basin ammonia model. Tables 3 and 4 reflect the total load allowable at each of these locations. On tables 3 and 4 the TMDL for Niwot, Orodell, Inc and Superior are actually allowable in-stream loads for that node, located at Dry Creek, Four Mile Creek and Rock Creek, respectively. CHRONIC TOTAL AMMONIA, mg/L WWTP Lyons WWTP Longmont WWTP Dry Cr (Niwot WWTP) St. Vrain SD Red Lion Fourmile (Orodell, Inc) San Lazaro Boulder WWTP B&B Louisville WWTP Rock Creek (Superior) Lafayette WWTP Erie WWTP
JAN 25.0 25.0 25.0 25.0 25.0 25.0 25.0 18.5 25.0 10.3 9.5 10.3 25.0
FEB 25.0 25.0 25.0 25.0 25.0 25.0 25.0 10.9 25.0 12.0 6.0 12.0 25.0
MAR APR MAY JUN JUL AUG SEP OCT 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 20.3 11.2 6.1 9.1 10.9 10.7 6.8 10.6 25.0 21.7 17.9 19.5 25.0 25.0 25.0 17.2 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 5.3 9.7 11.3 23.5 19.5 15.1 15.1 15.1 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 7.6 12.0 11.7 8.8 8.5 8.0 9.2 11.0 6.4 12.0 11.3 9.9 10.3 10.6 11.0 10.5 7.6 12.0 12.5 9.9 10.3 10.6 11.0 10.5 15.2 13.9 16.4 9.9 7.4 8.9 9.3 12.8 ACUTE TOTAL AMMONIA, mg/L
NOV DEC 25.0 25.0 14.9 23.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 21.6 24.8 25.0 25.0 7.5 9.0 7.5 9.0 7.5 9.0 13.5 21.7
WWTP Lyons WWTP Longmont WWTP Dry Cr (Niwot WWTP) St. Vrain SD Red Lion Fourmile (Orodell, Inc) San Lazaro Boulder WWTP B&B Louisville WWTP Rock Creek (Superior) Lafayette WWTP Erie WWTP
JAN 30.0 30.0 29.5 30.0 30.0 30.0 30.0 30.0 30.0 28.6 8.7 * 28.6 30.0
FEB 30.0 30.0 30.0 30.0 30.0 30.0 30.0 22.3 30.0 30.0 5.8 * 30.0 30.0
MAR APR 30.0 30.0 30.0 24.8 29.9 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 15.3 28.0 30.0 30.0 20.3 25.3 8.3 18.9 20.3 25.3 24.4 25.4
NOV DEC 30.0 30.0 27.3 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 29.5 30.0 30.0 30.0 24.5 26.3 11.1 9.3 24.5 26.3 30.0 30.0
MAY JUN JUL 30.0 30.0 30.0 23.5 30.0 30.0 29.9 28.5 28.8 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 26.6 24.9 25.1 20.5 24.3 24.0 30.0 30.0 30.0 30.0 27.0 27.0
AUG SEP OCT 30.0 30.0 30.0 24.2 26.2 30.0 29.0 29.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 28.3 30.0 30.0 30.0 24.9 27.1 26.9 26.2 22.4 19.2 30.0 30.0 30.0 30.0 30.0 30.0
*Acute is less than chronic; set acute to chronic.
Table 6. Summary of allowable effluent limit concentrations of total ammonia in mg/l as predicted by the TMDL model to attain standards for un-ionized ammonia in the St. Vrain basin. JLB C:Boulder\TMDL Boulder.doc
24
5/8/03
Tri-basin TMDL – Ammonia
Final
Waste Load Allocation for Sources to be Controlled The model indicates that restrictive concentrations of total ammonia for the Boulder and Longmont WWTPs will have the greatest impact in terms of meeting instream ammonia standards. Establishing allocations for Boulder and Longmont are necessary to attain the ammonia stream standards downstream. The most restrictive concentration for Boulder occurs in March with a concentration allocation of 5.3 mg/l total ammonia (30 day average). This results in a monthly total ammonia allocation of 502 kg/day. The most restrictive concentration for Longmont occurs in May with an allocation of 6.1 mg/l (30 day average) of total ammonia, which results in a monthly allowable load of 323 kg/day. The model output for Louisville, Lafayette and Erie indicates the most restrictive concentrations for these three are greater than 7.0 mg/l total ammonia. The model also shows minimal effect of total ammonia from the Lyons, Orodell, Red Lion Inn, San Lazaro, B&B MHP and St. Vrain Sanitation District facilities. Rock Creek’s most restrictive concentration limits are similar to those of Boulder and Longmont, with a value of 6.0 mg/l (30 day average) total ammonia occurring in February. Table 6 presents the total ammonia permit limits that would implement the waste load allocations. These limits for Boulder and Longmont, in conjunction with limits on the other dischargers, would ensure continued attainment of standards in the Tri-basin area. Load Reduction The TMDLs listed on Table 3 were compared with the effluent concentrations from these facilities for the most recent 5 years of discharge. The largest load reductions are required in different months from the two major facilities located in the tri-basin area. As Table 7 indicates, significant reduction in total ammonia concentration will have to occur in the Boulder WWTP effluent in February and March and in the Longmont WWTP effluent in May, September and October. The load reductions, where known to be necessary, are highlighted in Table 7 below. Boulder TOTAL AMMONIA, mg/L TMDL (chronic) Current concentration*
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 18.5 10.9 5.3 9.7 11.3 23.5 19.5 15.1 15.1 15.1 21.6 24.8 14.2 13.0 11.4 8.9 8.7 6.8 6.9 9.1 10.5 11.1 9.7 10.0 Longmont TOTAL AMMONIA, mg/L
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC TMDL (chronic) 25.0 25.0 20.3 11.2 6.1 9.1 10.9 10.7 6.8 10.6 14.9 23.0 Current 8.6 8.0 9.1 10.5 11.3 9.0 8.6 9.2 9.2 9.1 7.4 10.9 concentration** *Average of the monthly maximum concentration for 10/30/95 – 12/31/01. ** Average of the monthly maximum concentration for 1/31/96 – 12/31/01
Table 7. Comparison of current discharge concentrations versus TMDL allowable concentrations.
JLB C:Boulder\TMDL Boulder.doc
25
5/8/03
Tri-basin TMDL – Ammonia
Final
Load Allocation Among Sources Although the monthly load allocation for ungaged surface water sources and upstream sources is accounted for in this TMDL model for six locations where data was available, no reduction for these minimal and unidentified sources is included in this TMDL. Margin of Safety The margin of safety is the TMDL component that accounts for unknowns or uncertainties in the development of the TMDL. The margin of safety may be explicit (a separate value in the TMDL) or implicit (included in factors determining the TMDL). The MOS in this TMDL is implicit. The magnitude of the margin of safety is reduced by much of the work performed in preparation of this TMDL. The calibrated ammonia model used in this assessment uses implicit mechanisms for managing the margin of safety. The CAM is well accepted, calibrated and validated. Therefore the degree of uncertainty is substantially diminished. Additionally, the model incorporates several worst case and conservative assumptions, including: •
That all facilities are discharging at maximum design flow Because service population growth occurs at different rates for different treatment facilities, the assumption of discharges occurring at design capacities at all times is a very conservative one. It is unlikely that all facilities will be discharging at full capacity at the same time.
•
That all facilities are discharging at maximum permitted concentration It is unlikely that all dischargers in the basin will be discharging at their highest concentrations of total ammonia at the same time.
•
That there is concurrent critical low flow and maximum discharge The facilities discharging at their maximum flow does not usually coincide with conditions of low flow instream.
In addition to the conservative assumptions used in the CAM, this TMDL is based on an improved analysis of the hydrology of the basin, an exhaustive review of the most current stream data and the use of characteristic data whenever available. In addition to these conservatisms, the degree of uncertainty is minimized by evaluating each data source and model variable independently, and by confirming the appropriateness of data used. This results in the most accurate and comprehensive ammonia model for the tri-basin area. The resulting analysis of the allowable loads for ammonia from the various dischargers in the basin includes a significant implicit margin of safety. The above conservative assumptions and operating realities result in a substantial implicit margin of safety assuming implementation of the new discharge permit limits based on this TMDL assessment.
JLB C:Boulder\TMDL Boulder.doc
26
5/8/03
Tri-basin TMDL – Ammonia
Final
Post-Implementation Monitoring Water-quality monitoring will determine the extent to which the implementation plans achieve the compliance goals. Monthly sampling at critical locations on St. Vrain Creek and its tributaries will demonstrate the effectiveness of the wasteload allocations. Monitoring provides an additional element relative to Margin of Safety in that the water quality data generated will allow the efficacy of control measures to be evaluated. Should water quality monitoring indicate there continue to be instream ammonia exceedances, the TMDL will be adjusted as necessary. Implementation Implementation of the TMDL may require that load reduction be apportioned among all significant pollutant sources. The allocations described above will be considered when discharge permits are developed. At the time of permit issuance, the reasonable potential for each source to exceed the allocation will be evaluated and appropriate limitations and monitoring requirements will be imposed. IX.
PUBLIC PARTICIPATION
The issue of ammonia impairment in the Tri-Basin area has been discussed with interested members of the community at various forums over the past several years. In 1999, the Tri-Basin Watershed TMDL Group was formed. The modeling effort discussed in this report was part of the TMDL process initiated under the Tri-basin TMDL Steering Committee. The group included representatives from the discharging facilities, Boulder County Health Department, and the Division, which directed and oversaw the data analysis and modeling efforts conducted by personnel from the Center for Limnology at the University of Colorado (Bill Lewis, Jim Saunders and Marylee Murphy). Members of the TMDL group included the cities of Boulder, Longmont, Louisville, Erie, Lyons, Boulder County Health Department, Lafayette, St. Vrain Sanitation District, and Niwot Sanitation District, most of who represent wastewater utility providers. The ammonia modeling work has been presented to and discussed with this group at numerous meetings held primarily during 2001 and 2002. The draft Tri-basin TMDL was public noticed on January 24, 2002 and open for public comment through March 5, 2003. Comments were received from Jeffrey J. Kahn on behalf of the New Consolidated Lower Boulder Reservoir and Ditch Company. The comments raised concerns about detrimental effects of nitrogen to agricultural crops. In addition there was a comment about participation in the TMDL process and several comments on the hydrology. With respect to the nitrogen issue and the participation issue, the Ditch Company was refered to the standards hearing process through the Water Quality Control Commission. The meetings, discussions and modeling work was highly focused on ammonia, not nitrogen. The comments on hydrology pointed out a typographical error, which in combination with the use of two streams as adjustment mechanisms in the model, led to some confusion about the modeling work. The public notice draft of this document, as well as the TMDL Modeling Report, incorrectly refered to the Boulder Supply Canal as the Boulder Feeder Canal. This typgographical error has been corrected in this Final document and in the Final Modeling Report. The commenters were refered to the TMDL Modeling Report for a detailed explanation of the hydrology and the method of adjusting the flows immediately upstream of the two major outfalls to provide consistency with historical low flows at these locations. JLB C:Boulder\TMDL Boulder.doc
27
5/8/03
Tri-basin TMDL – Ammonia X.
Final
REFERENCES
WQCC 1998a: Colorado Department of Public Health and Environment, Water Quality Control Commission, 1998 303(d) List of Impaired Waters, 1998. WQCC 2001: Colorado Department of Public Health and Environment, Water Quality Control Commission, Classifications and Numeric Standards for South Platte River Basin, Laramie River Basin, Republican River Basin, Smoky Hill River Basin, Regulation No. 38, Effective: October 30, 2001. WQCC 2001: Colorado Department of Public Health and Environment, Water Quality Control Commission, The Basic Standards for Methodologies for Surface Water, Regulation No. 31, 5 CCR 1002-31, Effective March 20, 2001. WQCD 1991: Colorado Department of Public Health and Environment, Water Quality Control Division, Colorado Total Maximum Daily Load and Wasteload Allocation Guidance, Prepared by the WQCD, Groundwater and Standards unit, Revised November 1991. William M. Lewis, Jr. and James F. Saunders, III: Modeling and Analysis of Ammonia for Total Maximum Daily Load in the St. Vrain Creek Drainage, including Boulder Creek and Coal Creek, RPT 164, Revised May 14, 2003
JLB C:Boulder\TMDL Boulder.doc
28
5/8/03
Tri-basin TMDL – Ammonia
Final
Appendix A ACRONYMS ac ch CAM CDSS CDPHE cfs EPA HUC LA MOS ug/l mg/l MGD #/day SEO TVS TMDL USGS WLA WWTF WWTP WQCC WQCD
acute chronic Colorado Ammonia Model Colorado Decision Support System Colorado Department of Public Health and Environment cubic feet per second Environmental Protection Agency Hydrologic Unit Code Load Allocation Margin of safety micrograms per liter milligrams per liter Million gallons per day pounds per day State Engineer’s Office Table Value Standard Total Maximum Daily Load United States Geological Survey Waste Load Allocation Wastewater treatment facility Wastewater treatment plant Water Quality Control Commission Water Quality Control Division
JLB C:Boulder\TMDL Boulder.doc
29
5/8/03
Modelling and Analysis of Ammonia for Total Maximum Daily Load in the St. Vrain Creek Drainage, including Boulder Creek and Coal Creek
Prepared by: William M. Lewis, Jr. James F. Saunders, III Date of preparation: December 31, 2002 Revised: May 14, 2003 R/164
Table of Contents Executive Summary
iii
Introduction
1
St. Vrain Study Area
4
St. Vrain TMDL Hydrology
9 9
Boundaries in Space and Time Sources of Flow Data
12
Estimation of Ungaged Flows
13
Analysis of Ungaged Flows
23
DFLOW Concept and Implementation
29
Low Flows for Future Conditions
39
Special Handling of Low Flows above Major WWTPs
40
Special Handling for the Two Xcel Energy Facilities
41
Special Problems at Acute Low Flows
42
Overview of Hydrologic Conditions for Modelling
43
Chemical Conditions for Modelling of Ammonia
43
Chemistry of Non-Effluent Surface Flows
43
Effluent Chemistry
54 60
Setting Total Ammonia, Temperature, and pH for Unmonitored Water Sources Setpoint and Rebound
62
Site-Specific Amplitudes and Times of Maxima
63
Recurrence Analysis
67
i
Application of Setpoint Conditions
68
Travel Time
72
Estimation of Ammonia Loss Rate: Calibration and Validation
75
Structure of the Model
83
Special Calculations for Niwot and Rock Creek
83
Special Calculations for Orodel and St. Vrain
85
Setting Permit Limits with the Model
85
Future Considerations
90
Appendix I: A Study of Statistical Relationships between Flow, Percent Unionized Ammonia, and pH in the Drainage of the St. Vrain Creek, Including Boulder Creek and Coal Creek
98
ii
Executive Summary
1. Data analysis and modelling were conducted during 2001-2002 in support of an analysis of Total Maximum Daily Load (TMDL) for the St. Vrain watershed, including Boulder Creek and Coal Creek. The project was directed and overseen by a TMDL Steering Committee, which included representatives of all pointsource dischargers and the Colorado Department of Public Health and Environment's Water Quality Control Division. The data analysis and TMDL modelling are to be submitted to CDPHE, which will be preparing the final TMDL document. The TMDL was developed specifically for ammonia, which is considered to be the water-quality issue of greatest concern and greatest complexity in the St. Vrain basin at the present time. The underlying basis for the ammonia TMDL will, however, facilitate the preparation of other TMDL analyses and modelling as needed in the future. 2. Hydrologic analysis for chronic and acute low-flow conditions was an essential precursor to the TMDL modelling. While use of the DFLOW algorithm, which is applied by the state to the issuance of individual permits, is relatively easy for individual gage records, the establishment of acute and chronic low-flow conditions at all points in the entire system is a greater challenge but is essential for the TMDL analysis. The low flows were established by a complete hydrologic accounting of tributary flows and ungaged flows (including seepage) as well as ditch withdrawals for acute and chronic low-flow conditions. The hydrologic analysis resulted in quantitative estimates of system-wide low flows
iii
under recurrence intervals corresponding to acute and chronic conditions used for water-quality protection by the state of Colorado. The flows derived for the TMDL analysis are internally consistent hydrologically in the sense that they account for all sources and losses of water under acute and chronic flow conditions for each month of the year. For future projections, the chronic and acute low flows were augmented with design-capacity wastewater flows from all wastewater treatment facilities. 3. The basis for the TMDL software is the Colorado Ammonia Model (CAM). For TMDL purposes, CAM software was adapted to the geography of the St. Vrain basin and was expanded so that it could accommodate up to 13 dischargers. 4. The TMDL software, through CAM, requires estimates of monthly low flows for acute and chronic conditions, setpoint pH and temperature, rebound rate for pH and temperature below points of discharge, amplitude and time of maxima for pH and temperature, and rates of biologically-mediated ammonia loss (primarily nitrification). Where possible, all of these variables were estimated on the basis of monitoring data from the basin. Varying amounts of information were available for each of the data requirements, but considerable information was generally available for the most important areas of the drainage (below the major discharges). Where no information was available, default values were used or values from analogous reaches were used. 5. Future permit limits were established by incremental adjustment of key variables in the TMDL software for the basin. Three portions of the basin were adjusted separately at first and then overlapping effects were taken into account. The three
iv
reaches, each of which contains one or more major dischargers, are Coal Creek, Boulder Creek, and the St. Vrain main stem. Overlapping effects occur where a critical point (location of maximum unionized ammonia concentration) occurs below the junction of two reaches. Within reaches, dischargers were assigned equal concentration limits for acute and chronic conditions, and these limits were set so as to achieve compliance with acute and chronic standards within or below the reach. Higher allowances were made for specific dischargers when higher allowances did not impair the concentrations that could be assigned to other dischargers. Concentrations of total ammonia in effluents were capped at 25 mg/L for chronic conditions and 30 mg/L for acute conditions, even when assimilation capacity was available for higher concentrations. The result of these procedures is a monthly chronic and monthly acute total ammonia allowance for every discharger consistent with compliance with standards for unionized ammonia throughout the drainage. Final suggested permit limits derived in this way are as shown in Table A, below.
v
CHRONIC TOTAL AMMONIA, mg/L WWTP Lyons WWTP Longmont WWTP Dry Cr (Niwot WWTP) St. Vrain SD Red Lion Fourmile (Orodel, Inc) San Lazaro Boulder WWTP B&B Louisville WWTP Rock Creek (Superior) Lafayette WWTP Erie WWTP
JAN 25.0 25.0 25.0 25.0 25.0 25.0 25.0 18.5 25.0 10.3 9.5 10.3 25.0
FEB 25.0 25.0 25.0 25.0 25.0 25.0 25.0 10.9 25.0 12.0 6.0 12.0 25.0
MAR APR 25.0 25.0 20.3 11.2 25.0 21.7 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 5.3 9.7 25.0 25.0 7.6 12.0 6.4 12.0 7.6 12.0 15.2 13.9
WWTP Lyons WWTP Longmont WWTP Dry Cr (Niwot WWTP) St. Vrain SD Red Lion Fourmile (Orodel, Inc) San Lazaro Boulder WWTP B&B Louisville WWTP Rock Creek (Superior) Lafayette WWTP Erie WWTP
JAN 30.0 30.0 29.5 30.0 30.0 30.0 30.0 30.0 30.0 28.6 8.7 * 28.6 30.0
FEB 30.0 30.0 30.0 30.0 30.0 30.0 30.0 22.3 30.0 30.0 5.8 * 30.0 30.0
MAR APR 30.0 30.0 30.0 24.8 29.9 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 15.3 28.0 30.0 30.0 20.3 25.3 8.3 18.9 20.3 25.3 24.4 25.4
MAY JUN JUL AUG SEP 25.0 25.0 25.0 25.0 25.0 6.1 9.1 10.9 10.7 6.8 17.9 19.5 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 11.3 23.5 19.5 15.1 15.1 25.0 25.0 25.0 25.0 25.0 11.7 8.8 8.5 8.0 9.2 11.3 9.9 10.3 10.6 11.0 12.5 9.9 10.3 10.6 11.0 16.4 9.9 7.4 8.9 9.3 ACUTE TOTAL AMMONIA, mg/L MAY JUN JUL 30.0 30.0 30.0 23.5 30.0 30.0 29.9 28.5 28.8 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 26.6 24.9 25.1 20.5 24.3 24.0 30.0 30.0 30.0 30.0 27.0 27.0
AUG SEP 30.0 30.0 24.2 26.2 29.0 29.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 24.9 27.1 26.2 22.4 30.0 30.0 30.0 30.0
OCT 25.0 10.6 17.2 25.0 25.0 25.0 25.0 15.1 25.0 11.0 10.5 10.5 12.8
NOV 25.0 14.9 25.0 25.0 25.0 25.0 25.0 21.6 25.0 7.5 7.5 7.5 13.5
DEC 25.0 23.0 25.0 25.0 25.0 25.0 25.0 24.8 25.0 9.0 9.0 9.0 21.7
OCT 30.0 30.0 30.0 30.0 30.0 30.0 30.0 28.3 30.0 26.9 19.2 30.0 30.0
NOV 30.0 27.3 30.0 30.0 30.0 30.0 30.0 29.5 30.0 24.5 11.1 24.5 30.0
DEC 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 26.3 9.3 26.3 30.0
*Acute is less than chronic; set acute to chronic. Table A. Effluent limits consistent with standards for unionized ammonia in the St. Vrain basin.
vi
Introduction
The St. Vrain Creek drainage extends from the Continental Divide to the confluence of the St. Vrain with the South Platte River near Platteville (Figure 1). The St. Vrain and its tributaries, the largest of which are Boulder Creek and Coal Creek, receive permitted wastewater discharges from the cities of Boulder, Longmont, Louisville, Lafayette, and Erie as well as a number of other, smaller municipal sources. Because the effluents from these sources either are not nitrified or not fully nitrified, monitoring and analysis of ammonia loads and concentrations in the St. Vrain drainage is essential, especially for the protection of aquatic life. The presence of multiple sources complicates analysis because the sources may have interactive effects. The purpose of this report is to present the results of a TMDL analysis for ammonia in the St. Vrain drainage, including a consideration of the interaction of multiple sources of ammonia. The analysis includes not only municipal discharges but also non-point sources and water-management practices that affect the dilution of ammonia within the basin. Modelling is used as the basis for TMDL analysis of ammonia in the St. Vrain Creek basin. Modelling is essential for this purpose not only as a means of taking into account the interaction of multiple discharges, non-point sources, and water management practices, but also for dealing with complexities of the ammonia standard for protection of aquatic life. Although the chronic standard for ammonia, which typically is the more challenging of the two ammonia standards for dischargers, is a fixed value for any given segment, compliance with the standard must be judged on the basis of both pH and temperature, which affect the percentage of total ammonia that is unionized. Because
1
St. Vrain Supply Canal (from Carter Lake)
South Platte River
LYONS
rai n
Foothills Reservoir
Union Reservoir
Cr .
Cr. y Dr
St. V
Cr.
rain
Cr.
uld
d Han Left
LONGMONT
er Cr .
.V
NIWOT
Bo
St
Panama Reservoir
Boulder Reservoir
BOULDER
Boulder Cr.
Cr .
LAFAYETTE
er ld ou B S.
North
Coa l Cr .
Cr .
ERIE
Valmont Reservoir
D ry Cr .
Fo ur mi le
Baseline Reservoir LOUISVILLE
Discharge Facilities
r. lC Coa
Gaging Stations
r. kC Roc
0
1
2
3
4
5 miles
Figure 1. Map of the study area.
pH and temperature typically change downstream of a point of discharge, the amount of unionized ammonia may change downstream from a discharge as well. In judging or projecting compliance with standards or setting appropriate effluent limits for compliance with a particular standard, changes in pH and temperature, and thus of percent unionized ammonia, must be estimated as a function of distance downstream from any discharge, and the effect of the mixing of discharges with each other also must be anticipated. This complex estimation can only be accomplished by modelling. In addition, because
2
ammonia is removed biologically (mostly it is converted to nitrate by nitrification) in oxygenated streams, the rate of conversion must be estimated and considered as one of the important factors governing the amount of total ammonia, and thus the amount of unionized ammonia, that will be in the stream at any given point. Rates of conversion vary seasonally and spatially. The TMDL analysis described here is achieved through the use of a customized reach model derived from the Colorado Ammonia Model (CAM). It is prepared in spreadsheet form and is capable of taking into account the addition and removal of water through water management practices or natural processes, all sources of ammonia, influences on pH and temperature, and loss of ammonia through biological conversion processes. The model is calibrated for known hydrologic and water-quality conditions as documented by monitoring programs and then is used to predict conditions for the future corresponding to regulatory low flows that are used by the State of Colorado in NPDES permitting (biologically-based low flows). The assumption for modelling future conditions is that all point sources of discharge will be operating at rated capacity. Projections of future conditions involve the determination, through modelling, of total ammonia concentrations from each wastewater discharge that will be compatible with the stream standards for unionized ammonia. The model considers all hydrologic and waterquality influences simultaneously. Because background and non-point sources typically are not controllable by permit, the ammonia content of municipal discharges is adjusted as necessary to bring the entire system into compliance with the standards. This adjustment involves sharing (allocation) of assimilative capacity for total ammonia among dischargers. The allocation process that leads to division of assimilative capacity
3
across dischargers may be either simple or complex, depending on the degree of interaction between dischargers. Where allocation processes are difficult, the final allocation agreement typically involves negotiations among dischargers with oversight from the Colorado Department of Public Health and Environment's Water Quality Control Division.
The St. Vrain TMDL Study Area
The watershed of St. Vrain Creek encompasses 976 mi2 extending from the Continental Divide to the South Platte on the plains. The St. Vrain and its principal tributary, Boulder Creek, are roughly equal in area above their confluence. Most of the population and most of the discharge permits are located in the plains portion of the basin east of the foothills (Table 1). The St. Vrain watershed has been divided into 23 segments by the Colorado Water Quality Control Commission. These fall into three categories with respect to ammonia standards (Table 2). Streams with a cold-water classification have a chronic standard of 0.02 mg N/L of unionized ammonia, warm-water class 1 streams have a chronic standard of 0.06 mg N/L, and warm-water class 2 streams have a chronic standard of 0.06 or 0.10 mg N/L. The portion of the basin included in the TMDL model extends from the mouth of the St. Vrain upstream to the edge of the foothills, as described in the section on hydrology. The connection between the montane and plains parts of the basin is
4
Permit #
Name
Owner
0020184 0020877 0021695 0021831
San Lazaro MHP Lyons WWTF Niwot WWTF Erie WWTF
5005 Properties Town of Lyons Niwot Sanitation District Town of Erie
0023078
Louisville WWTF
City of Louisville
0023124 0024147 0026671 0027260 0040819 0041700 0041882 0043010
Lafayette WWTF Boulder WWTF Longmont WWTF Red Lion Inn Boulder Mountain Lodge St. Vrain WWTF B&B Mobile and RV Park Rock Creek WWTF
City of Lafayette City of Boulder City of Longmont C. Mueller & H. Mueller Orodel, Inc St. Vrain Sanitation District Orville W. and Billie J. Smith Superior Metro District #1
Location SW¼, S22,T1NR70W SE¼,SE ¼, S18,T3NR71W NE¼,SW ¼, S29,T2NR69W NE¼,NW ¼, S18,T1NR68W N½, SW¼ & S ½, NW¼, S9, T1SR69W SE¼,SE ¼, S36,T1NR69W SW¼, S13,T1NR70W NW¼, S11,T2NR69W SW¼,NW ¼, S34,T1NR71W SW¼,SE ¼, S27,T1NR71W NW¼,SE ¼, S31,T3NR67W SW¼,SW ¼, S31,T2NR65W SE¼,NW ¼, S29,T1SR69W
Design Capacity, mgd 0.110 0.375 1.18 1.2 3.4 4.4 25 14.0 0.009 0.0045 4.5 0.015 2.4
Receiving Water Segment Boulder Cr, SPBO02 St. Vrain, just above gage Dry Cr, SPSV06 Coal Cr, SPBO07b Coal Cr, SPBO07b Coal Cr, SPBO07b Boulder Cr, SPBO09 St. Vrain, SPSV03 Boulder Cr, SPBO02 Fourmile Cr, SPBO02 St. Vrain, SPSV03 Boulder Cr, SPBO10 Rock Cr, SPBO08
Table 1. List of NPDES permits in the St. Vrain watershed included in the TMDL modelling. General permits (640000) have been omitted, as have selected other permits not relevant to TMDL modelling for ammonia (see text for explanation).
5
Code SPBO01
Description All tributaries to Boulder Creek, including all lakes, reservoirs and wetlands, within the Indian Peaks Wilderness Area
SPBO02
Mainstem of Boulder Creek, including all tributaries, lakes, reservoirs and wetlands, from the boundary of the Indian Peaks Wilderness Area to a point immediately above the confluence with South Boulder Creek, except for the specific listings in SPBO03 and SPBO12
SPBO03
Mainstem Middle Boulder Creek, including all tributaries, lakes, reservoirs and wetlands, from source to outlet of Barker Reservoir except for specific listings in SPBO01
SPBO04a
Mainstem South Boulder Creek, including all tributaries, lakes, reservoirs and wetlands, from source to outlet of Gross Reservoir
SPBO04b
Mainstem South Boulder Creek, including all tributaries, lakes, reservoirs and wetlands, from outlet of Gross Reservoir to South Boulder Road except for specific listings under SPBO04c & d
SPBO04c UP
Mainstem of Cowdrey drainage from the source below Cowdrey Reservoir #2 to the Davidson Ditch
SPBO04d UP
Mainstem of Cowdrey drainage from immediately downstream of the Davidson Ditch to the confluence with South Boulder Creek
SPBO05 UP
Mainstem South Boulder Cr from South Boulder Road to the confluence with Boulder Creek
SPBO06 UP
Mainstem of Coal Creek, including all tributaries, lakes, reservoirs and wetlands, from the source to highway 93
6
Classification Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Warm 2 Recreation 1a Water Supply Agriculture Aquatic Life Warm 2 Recreation 1a Water Supply Agriculture Aquatic Life Warm 1 Recreation 1a Water Supply Agriculture Aquatic Life Cold 2 Recreation 1a Water Supply
Dischargers
Red Lion Betasso WTP Boulder Mountain Lodge Pinebrook Water District San Lazaro MHP Cross and Caribou Mines Nederland Lake Eldora
Chronic 0.02
0.02
0.02
0.02
San Souci MHP
0.02
0.06
Valmont Station
0.06
0.02
Code
Description
SPBO07a UP
Mainstem of Coal Creek from highway 93 to highway 36 (Boulder Turnpike)
SPBO07b UP
Mainstem of Coal Creek from highway 36 to the confluence with Boulder Creek
SPBO08 UP
All tributaries to South Boulder Creek, including all lakes, reservoirs and wetlands, from South Boulder Road to the confluence with Boulder Creek and all tributaries to Coal Creek, including all lakes, reservoirs and wetlands, from Highway 93 to the confluence with Boulder Creek Mainstem of Boulder Creek from a point immediately above the confluence with South Boulder Creek to the confluence with Coal Creek
SPBO09
SPBO10 UP
Mainstem of Boulder Creek from the confluence with Coal Creek to the confluence with St. Vrain Creek
SPBO11 UP
All tributaries to Boulder Creek, including all lakes, reservoirs and wetlands, from a point immediately above the confluence with South Boulder Creek to the confluence with St. Vrain Creek, except for specific listings in SPBO05 & 7
SPBO12
Boulder Reservoir and Coot Lake
SPSV01
All tributaries to St. Vrain Creek, including all lakes, reservoirs and wetlands, which are within the Indian Peaks Wilderness Area and Rocky Mountain National Park to Hygiene Road
SPSV02
Mainstem of St. Vrain Creek, including all tributaries, lakes, reservoirs and wetlands, from the boundary of the Indian Peaks Wilderness Area and Rocky Mountain National Park to Hygiene Road
SPSV03
Mainstem of St. Vrain Creek from Hygiene Rd to the confluence with the South
7
Classification Agriculture Aquatic Life Warm 1 Recreation 1a Agriculture Aquatic Life Warm 2 Recreation 1a Agriculture Aquatic Life Warm 2 Recreation 1a Agriculture Aquatic Life Warm 1 Recreation 1a Water Supply Agriculture Aquatic Life Warm 1 Recreation 1a Water Supply Agriculture Aquatic Life Warm 2 Recreation 1a Water Supply Agriculture Aquatic Life Warm 1 Recreation 1a Water Supply Agriculture Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Warm1
Dischargers
Chronic 0.06
Louisville WWTP Lafayette WWTP Erie WWTP Rock Creek
0.06
Boulder WWTP
0.06
B&B Mobile and RV Park
0.06
0.10
0.10
0.06
0.02
Lyons WTP Lyons WWTP Glacier View Ranch
0.02
Longmont WWTP
0.06
Code UP
Description Platte River and Barbour Ponds
SPSV04a
Mainstem of Left Hand Creek, including all tributaries, lakes, reservoirs and wetlands, from the source to highway 36 except for specific listings in SPSV04b
SPSV04b
Mainstem of James Creek, including all tributaries, lakes, reservoirs and wetlands, from the source to the confluence with Left Hand Creek
SPSV05 UP
Mainstem of Left Hand Creek, including all tributaries, lakes, reservoirs and wetlands, from highway 36 to the confluence with St. Vrain Creek
SPSV06 UP
All tributaries to St. Vrain Creek, including all tributaries, lakes, reservoirs and wetlands, from Hygiene Road to the confluence with the South Platte River, except for specific listings in the Boulder Creek subbasin and in SPSV04a&b and SPSV05
Classification Recreation 1a Agriculture Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Cold 1 Recreation 1a Water Supply Agriculture Aquatic Life Warm 2 Recreation 1a Water Supply Agriculture Aquatic Life Warm 2 Recreation 1a Agriculture
Dischargers St. Vrain WSD Fort St. Vrain
Chronic 0.02
0.02
0.10
Boulder WTP Dover Industries Niwot Weld County Tri-Area Raytheon Lyons Cement Plant
0.10
Table 2. Stream classifications and chronic standards for unionized ammonia (mg N/L) for segments in the Boulder Creek and St. Vrain Creek watersheds. Table value standards (TVS) apply for chronic standards in all segments. Dischargers are listed for each segment. Use-protected (UP) segments are noted in the first column.
8
effectively broken, from the perspective of water-quality modelling, by the presence of mainstem reservoirs. Gross Reservoir on South Boulder Creek, Barker Reservoir on Middle Boulder Creek, and Buttonrock Reservoir on the St. Vrain are logical endpoints for the processing of ammonia carried in mountain streams, and can be used as convenient starting points for modelling ammonia in the rest of the basin.
St. Vrain TMDL Hydrology
A detailed understanding of flows in the basin is essential for water-quality modelling. The general aim of the analysis of flows is accurate representation of the lowflow regime throughout the basin on the basis of historical conditions. This analysis does not seek a full representation of water rights, but rather is a documentation of the outcomes that water rights have had on stream flows in the recent past.
Boundaries in Space and Time An operational definition of the boundaries for modelling is based on the location of gaging stations (Table 3). The Orodell gage on Boulder Creek and the Lyons gage on the St. Vrain mark the western edge of the portion of the basin that will be included in the TMDL model. These are the gages nearest to the reservoir releases that constitute the origins of the streams for modelling purposes. Fourmile Creek, a small gaged tributary near Orodell, also defines part of the western boundary of the model. The flow of South Boulder Creek is virtually all diverted into the lakes comprising the cooling system for Xcel Energy’s Valmont Station power plant. In
9
Location St. Vrain at Lyons St. Vrain below Longmont St. Vrain at Mouth Boulder Creek at Orodell Boulder Creek at 75th Street Boulder Creek at Mouth Fourmile Creek Coal Creek near Plainview Coal Creek near Louisville Rock Creek at Hwy 128
Code 06724000 06725450 06731000 06727000 06730200 06730500 06727500 06730300 06730400 395452105113800
Area, mi2 212 424 976 102 304 439 24.1 15.1 27.3 N/A
Record Used in Notes the TMDL Analysis WY1991-2000 WY1991-2000 WY1991-2000 WY1986-2000 WY1991-2000 WY1992-2000 WY1986-1995 Discontinued in 1995 WY1997-2000 WY1997-2000 Started in 1997 WY1996 Operated for 1 year
Table 3. Description of gages used in the TMDL analysis. See text for further explanation of differences in the periods of record among the gages.
essence, South Boulder Creek arises from the Leggett outlet, at least in terms of its effect on water quality in Boulder Creek. Flow measurements are available for the Leggett outlet. Coal Creek is a significant component of the analysis largely because there are four major discharges in its basin. The starting point for modelling on Coal Creek is the gage at Louisville, which is upstream of the discharges. A small, ungaged tributary, Rock Creek, is included because it carries effluent from the Rock Creek WWTP. A few smaller streams also are included in the hydrologic analysis because they may contribute significant flow to the major streams or because they serve as conduits for wastewater effluent. Left Hand Creek is tributary to the St. Vrain near Longmont and may cause significant flow in the St. Vrain even in dry weather. Nearby Dry Creek carries effluent from the Niwot WWTP.
10
Twenty-three permitted facilities fall within the geographic boundaries described above. Several are excluded from the TMDL model because they are not expected to have any effect on the ammonia load or assimilative capacity of modeled reaches. Raytheon is excluded; it is a groundwater remediation facility that discharges to an unnamed tributary of Dry Creek upstream of Boulder Reservoir. The discharge is small and unlikely to contain ammonia. Dover/Dietriech, another groundwater remediation facility, has been excluded because it has low flow and is not expected to discharge ammonia. The Weld Tri-Area Sanitation District fits the profile of facilities that are included in the TMDL, and its flow is included in the hydrologic analyses for the TMDL, but it is not treated as a source of ammonia for classified waters. The wastewater is discharged to a ditch and must travel a considerable distance before reaching the St. Vrain via the Last Chance return. The route is long and circuitous, and involves dilution. It is unlikely that it would have any effect on the ammonia load of the St. Vrain, and prospects for modelling any minor contributions from it are poor. Several permits associated with water treatment facilities also have been excluded (Betasso, Lyons, Boulder Reservoir, Pine Creek). Their discharge is chiefly backwash that flows irregularly and is unlikely to contain ammonia. Lyons Cement Plant (Southdown) also has been excluded because the discharge is small, intermittent, and unlikely to contain ammonia. Analysis of flow records generally is restricted to the last 10 years; this is consistent with CDPHE policy regarding low-flow determination for permitting. Thus, WY1991 – WY2000 is the primary period of record (WY = water year). More recent
11
records (2001-2002) are not available for all sources, or may be provisional. Exceptions requiring data outside of the primary period of record are as described below. Sources of Flow Data Daily flow records are available for all gaging stations and most of the diversions in the St. Vrain basin (Tables 3 and 4). Some of the larger treatment facilities provided Type of Record
Source
Typical Frequency Stream gages USGS Daily Stream gages CDSS Daily Diversions CDSS Daily Reservoir releases CDSS Daily Upper Wellman Xcel Daily Leggett Outlet Xcel Daily Boulder WWTP Discharger Daily Longmont WWTP Discharger Daily Louisville WWTP Discharger Daily Lafayette WWTP Discharger Daily Niwot WWTP Discharger Daily Other dischargers Discharger or CDPHE Monthly Goosequill Pump Station Xcel Monthly Fort St. Vrain 002 release Xcel Monthly Howlett Gulch Xcel Charts
Notes On-line historical and recent (provisional) Supplement to USGS Primary source, but see exceptions in text
Estimated from charts
Table 4. Primary data sources for flows used in calculating residuals and as the basis for hydrologic modelling. CDSS = Colorado Decision Support System.
daily records, but smaller facilities reported only monthly averages and maxima. Older records for most of the small facilities were obtained from the CDPHE data base. Xcel Energy, which operates the Valmont and Fort St. Vrain power stations, provided daily records.
12
Item
Problem
Resolution
Chapman McCaslin Ditch, 7/1/93, 33 cfs Davis Downing, 7/2/94, 133.19 Davis Downing, 9/2/00, 44.87 Peck, 6/13/94, 131.86 True Webster, 8/6/97, 92 Lower Boulder Ditch, 8/30/95, 585.6 Release to Lower Boulder Ditch, 8/30/95, 563.5 Erie and Coal Creek Ditch Erie and Coal Creek Ditch Boulder WWTP, 12/17/94 Boulder WWTP, 4/6/96 Boulder WWTP, 5/28/96 Boulder WWTP, 6/28-29/99 Boulder Supply Canal, Mar 94 Leggett Release, 4/30/93 Longmont WWTP, 12/18-31/90 Longmont WWTP, 2/23/99 Niwot WWTP, Oct-Dec, 1990 San Lazaro WWTP, Oct 90 – Apr 96
Greatly exceeds capacity (est. 8 cfs) and adjacent dates Greatly exceeds capacity (est. 28 cfs) and adjacent dates Greatly exceeds capacity (est. 28 cfs) and adjacent dates Greatly exceeds capacity (est. 35 cfs) and adjacent dates Greatly exceeds capacity (est. 7 cfs) and adjacent dates Greatly exceeds capacity (est. 230 cfs) and adjacent dates Greatly exceeds capacity 5/14-6/5/97, 20 cfs, exceeds capacity (est. 10 cfs) 5/11/00, 15 cfs, exceeds capacity (est. 10 cfs) Missing Missing Missing Missing Missing Missing Missing Missing Missing Not available
San Lazaro WWTP, Oct 96 San Lazaro WWTP, Aug-Sep 97 San Lazaro WWTP, Sep 2000 San Lazaro WWTP, Dec 2000 Rock Creek WWTP, Oct 91 – Mar 96 Rock Creek WWTP, Aug 97 Rock Creek WWTP, Sep 97 Rock Creek WWTP, Aug – Sep 2000 Tri-Area WWTP, Oct 91 – May 95 Tri-Area WWTP, Jun 96 – Dec 97 St. Vrain WWTP, Oct 91 – Dec 93
Missing Missing Missing Missing Not available Missing Missing Missing Not available Not available Not available
Set to 3 cfs Set to 13.19 cfs Set to 4.87 cfs Set to 11.86 cfs Set to 0.92 cfs Set to 85.6 cfs Set to 63.5 cfs Set to 10 Set to 10 Set to 13.07 mgd Set to 17.15 mgd Set to 24.4 mgd Set to 17.35 mgd Set to zero on all days Set to 1.53 cfs Set to 6.5 mgd Set to 4.685 mgd Set to 0.36 mgd Set with sine function to yield 0.05 mgd in Jan and 0.10 mgd in Jul Set to 0.055 mgd Set to 0.08 mgd Set to 0.115 mgd Set to 0.135 mgd Set to 0.20 mgd Set to 0.27 mgd Set to 0.22 mgd Set to 0.30 mgd Set to 0.5 mgd Set to 0.5 mgd Set to 0.20 mgd
Table 5. Catalog of problems encountered in the hydrologic data sets. A correction or substitution is listed in each case, typically based on values recorded for adjacent dates.
In any large data set there will be missing information, transcription errors, and other problems that must be acknowledged and corrected, if possible. Brief descriptions of problems and corrections are shown in Table 5.
Estimation of Ungaged Flows In much of the South Platte basin, streams gain water from return flows, groundwater discharge, or small, ungaged tributaries. These flows can contribute significantly to the total flow in the stream. Calculation of flow residuals (flow unaccounted for by known sources) between adjacent gages provides an estimate of this 13
ungaged flow, most of which in dry weather probably travels through the alluvium and enters the stream as a distributed source (seepage). The calculation of flow residuals assumes that there may be a seasonal pattern (hence the need for monthly values), and that the seasonal pattern is repeated each year. A 10-y period of record is ideal for analysis of residuals, but a shorter record was used when there was no alternative. The basin can be divided into four reaches for the purpose of estimating ungaged flows. Each reach is bounded by gages with adequate records (Figure 2). The first reach, on Boulder Creek, is bounded by gages at Orodell and 75th Street. The major inflows are Fourmile Creek, South Boulder Creek (Leggett outlet), the Boulder Supply Canal, and the City’s WWTP (Table 6, Figure 2). Several ditches withdraw water. The record for Fourmile Creek, which ended with March 1995, was completed by use of historical monthly median flows (record = 10 years beginning April 1986) for the missing daily flows. Two very small discharges (Red Lion Inn and Boulder Mountain Lodge) located near the Orodell gage were not included in the flow residual calculations because of their small volume and infrequent direct discharge to the river. The second reach extends on Boulder Creek from the 75th Street gage to the St. Vrain confluence (Table 7, Figure 2). Major inflows include release from Baseline Reservoir (via the New Dry Creek Carrier), Coal Creek, and release from Panama Reservoir. Fourteen ditches divert flow from this part of Boulder Creek. Daily records were not available in all years for some of the small ditches; some estimation was required. A minor facility, B&B Mobile Home & RV Park, is not included in the hydrologic analysis because the flow is quite small (0.015 mgd). The absence of data for
14
South
Boulder Creek - Orodell North
South Lyons WWTP
St. Vrain - Lyons
Fourmile Creek Silver Lake Swede Smead Foothills Inlet South Branch
Anderson Farmers
Upper Wellman
Boulder White Rock Smith Goss Boulder Left Hand North Boulder Farmers
Foothills Release
San Lazaro WWTP Hager Meadow Niwot Northwest Mut. Ins. Co.
Leggett Outlet (South Boulder Creek)
North
St. Vrain Supply Canal Highland Rough and Ready St. Vrain Palmerton
Longmont Supply Chapman McCaslin Oligarchy Denio Taylor Runyon Zweck Turner
South Flat James Mason Cushman Beckwith, Island Bonus Left Hand Creek
Butte Mill Green Boulder Supply Canal
Longmont WWTP Dry Creek
Boulder WWTP
Union Res Release
Boulder Creek - 75th Street
St. Vrain Creek - Longmont
Boulder Creek - 75th Street Dry Creek Carrier
Boulder Creek Mouth
Leggett
St. Vrain Creek - Longmont
Lower Boulder Boulder and Weld County Last Chance
Coal Creek Howell
Weld Tri Area WWTP
Armstead
Panama Reservoir Release Last Chance Return Godding Daily and Plumb Idaho Creek Carrier Houck #2 Carr Tyler Smith Emmons Tom Delehant Highland and South Side
St. Vrain SD WWTP
Rural Goosequill Pump Station Boulder Creek - Mouth St. Vrain Creek Longmont
Fort St. Vrain Discharge Slough (002) St. Vrain Creek - Plattesville
Figure 2. Line diagram of Boulder and St. Vrain Creeks.
15
South Platte River
River Gages Mile Orodell to 75th Street 0.00 Orodell 1.07 1.90 2.80 2.95 4.14 4.14 4.14 4.14 4.14 5.31 7.82 8.20 8.50 9.41 10.89 11.15 11.40 75th Street
Sources
Withdrawals
Fourmile Creek Silver Lake Anderson Ext Farmers Boulder Left Hand Boulder White Rock McCarty North Boulder Farmers Smith Goss Upper Wellman Butte Mill San Lazaro WWTP Leggett Outlet Green Boulder Supply Canal Boulder WWTP
Table 6. Catalog of hydrologic features included in the analysis of flow residuals for the reach of Boulder Creek between the Orodell and 75th Street gages.
WY1991 at the mouth of Boulder Creek reduces the analysis to a nine-year record. A separate residual would be useful for Coal Creek (Table 7, Figure 3), but there is no gage at the mouth. The gage on Coal Creek near Louisville shows the contribution from upstream sources, but the period of record is brief (July 1997 to present). Another gage upstream at Plainview has a long period of record, but many winter values are missing. A comparison of the two gages (Figure 4) suggests that Coal Creek gains flow above the Louisville gage.
16
River Mile Gages Boulder Creek 11.40 75th Street 13.12 13.96 14.80 16.18 19.44 19.54 20.25 20.27 20.37 20.39 21.31 21.31 21.31 21.31 21.71 25.80 At mouth Coal Creek 0.00 0.01 1.10 1.90 3.49 5.91 7.76 11.99 14.40
Sources
Withdrawals Leggett Ditch
New Dry Carrier Lower Boulder Boulder Weld County Howell Coal Creek B&B MHP Panama Reservoir Armstead Godding Dailey Plumb Houck 2 Carr Tyler Smith Emmons Highland South Side Rural
Louisville Louisville WWTP Willis Harris Rock Creek (including WWTP) Lafayette WWTP Erie Coal Creek Erie WWTP Confluence
Table 7. Catalog of hydrologic features included in the analysis of flow residuals for the reach of Boulder Creek between the 75th Street gage and the gage at the mouth and for Coal Creek.
In addition to the three major treatment facilities on Coal Creek (Louisville, Lafayette, and Erie), there is also a facility on Rock Creek, which joins Coal Creek downstream of the Louisville gage. There is essentially no flow in Rock Creek above
17
Superior according to the brief record of a USGS gage at Highway 128 (Figure 5). In addition, there are a few small ditches that remove water from Coal Creek. Data from Boulder’s stream monitoring program provide a basis for direct assessment of flow residuals on Coal Creek for a limited number of dates (Table 8). Residuals are usually very small and can be accounted for by ungaged flow above the Louisville WWTP. Thus, ungaged flow probably does not play a role in the flow of Coal Creek below the Louisville gage. Consequently, all residuals between the two gages on Boulder Creek (75th and mouth) are assumed to reflect ungaged flow directly to the 14.4-mile reach of Boulder Creek. The third and most complex reach begins on the St. Vrain just above the Lyons gage and extends to the gage just above the confluence with Boulder Creek (Table 9, Figure 2). There are several sources for the reach, the most important of which are the St. Vrain Supply Canal, Left Hand Creek, Longmont WWTP, Dry Creek, and releases from Union Reservoir. Water is removed by 33 ditches, many of which are small. Chief uncertainties in this reach involve contributions from Dry Creek and Left Hand Creek. The flow in Dry Creek consists of effluent from the Niwot WWTP plus releases from Boulder Reservoir as measured at the Dry Creek gage. The record for the Dry Creek gage, which was available only through WY99, shows that reservoir releases often were zero and rarely exceeded a few cfs. For modelling, it is assumed that there is no release from Boulder Reservoir. In the past, there have been diversions from Dry Creek, but the CDSS record includes no withdrawals for the 10-year period of record used in the TMDL analysis. There are a few very small discharges in the upper part of the Dry Creek basin, but these are not included in the flow analysis. They include the Fairways
18
Metro facility, which is permitted for land application and does not release surface flows, although it could, and the Raytheon and Dover facilities, which are very small.
Coal Creek - Louisville South
North
Louisville WWTP
TN Willis
Harris
Lafayette WWTP
Superior WWTP
Rock Creek
Erie Coal Cr
Erie WWTP
Boulder Creek
Figure 3. Line diagram of Coal Creek.
19
WWTP effluent, mgd Measured Louisville Date at Mouth Gage Louisville Lafayette 1/21/1993 5.0 1.659 1.331 4/6/1993 5.5 1.713 1.335 5/11/1993 6.9 1.775 1.346 6/8/1993 10.4 1.933 1.416 8/10/1993 3.6 1.902 1.524 9/8/1993 3.6 1.569 1.532 10/13/1993 4.3 2.088 1.466 11/3/1993 24.4 1.940 1.505 12/8/1993 5.4 1.876 1.421 1/5/1994 4.0 1.779 1.419 3/28/1994 3.7 1.831 1.441 4/20/1994 7.3 1.964 1.416 6/16/1994 2.5 1.769 1.442 7/21/1994 5.9 1.702 1.464 8/30/1994 5.7 1.401 1.487 9/21/1994 3.4 1.188 1.502 11/16/1994 4.8 1.720 1.505 12/13/1994 3.7 1.723 1.407 3/14/1995 4.4 1.416 1.413 4/11/1995 7.2 2.127 1.479 10/17/1995 8.0 1.695 1.646 12/12/1995 13.3 1.867 1.556 2/13/1996 12.7 1.787 1.557 4/8/1997 9.9 2.119 1.751 2/10/1998 13.7 2.2 1.902 1.873 7/13/1999 7.8 1.6 2.191 2.124 11/9/1999 11.9 2.6 2.017 2.086 6/13/2000 10.7 2.9 1.997 2.048 Table 8.
Erie 0.103 0.131 0.117 0.099 0.130 0.144 0.105 0.125 0.112 0.069 0.115 0.112 0.127 0.111 0.111 0.131 0.125 0.132 0.096 0.087 0.133 0.137 0.13 0.125 0.188 0.501 0.493 0.604
Rock Creek Erie Coal (est.) Cr Ditch Residual 0.200 0.0 0.1 0.200 0.0 -0.3 0.200 0.0 -1.6 0.200 9.1 -13.8 0.200 0.0 2.2 0.200 0.0 1.7 0.200 0.0 1.7 0.200 0.0 -18.5 0.200 0.0 0.1 0.200 0.0 1.3 0.200 0.0 1.8 0.200 0.0 -1.6 0.200 2.7 0.4 0.200 0.0 -0.5 0.200 0.0 -0.8 0.200 0.0 1.3 0.200 0.0 0.7 0.200 0.0 1.7 0.200 0.0 0.4 0.200 0.0 -1.1 0.200 0.0 -2.4 0.200 0.0 -7.5 0.200 0.0 -7.0 0.280 0.0 -3.3 0.280 0.0 -5.0 0.040 0.0 1.3 0.560 0.0 -1.3 0.010 0.0 -0.6
Estimation of flow residuals at the mouth of Coal Creek using instantaneous discharge measurements taken by the City of Boulder at the mouth of Coal Creek.
20
Coal Creek 100
Flow at Louisville, cfs
Figure 5. Flow (cfs) on Rock Creek at Highway 128 during WY1996. The flow axis has been truncated at 1 cfs to emphasize typical flows. The maximum recorded flow was 4.8 cfs. 10
1
0.1 0.1
1
Flow at Plainview, cfs
10
100
Figure 4. Relationship between concurrent daily flows at the Louisville and Plainview gages on Coal Creek (log scales). Rock Creek
1.0 0.9 0.8
Flow, cfs
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 Oct-95
Nov-95
Jan-96
Feb-96
Apr-96
Jun-96
Jul-96
Sep-96
Figure 5. Flow (cfs) on Rock Creek at Highway 128 during WY 1996. The flow axis has been truncated at 1 cfs. The maximum recorded flow is 4.8 cfs.
21
River Mile Gage Lyons to Longmont 0.05 0.30 Lyons 0.37 0.48 0.97 0.97 1.08 1.08 1.30 2.03 2.30 2.30 2.30 2.30 2.30 2.30 2.30 2.30 2.30 2.30 2.93 2.93 3.81 3.81 4.72 4.72 4.72 4.72 6.03 6.03 6.03 6.59 6.59 6.59 6.94 6.94 8.15 8.15 8.15 11.17 11.53 12.11 14.07 14.39 16.26 Longmont 16.47
Sources
Withdrawals
Lyons WWTP St. Vrain Supply Highland Rough and Ready Palmerton Swede Smead Montgomery Pvt Foothills Inlet Goss Pvt 1 Goss Pvt 2 Clough Pvt Clough True Webster McCaslin True Webster Weese Pvt James Davis Downing Baker and Weese Longmont Supply Chapman McCaslin Foothills Reservoir Oligarchy Denio Taylor Runyon Zweck Turner Dickens Pvt Peck Pella Clover Basin Hager Meadow Niwot NW Mutual Ins Co South Flat James Mason Cushman Island Beckwith Bonus Left Hand Creek Longmont WWTP Dry Creek Union Reservoir Boulder Creek
Table 9. Catalog of hydrologic features included in the analysis of flow residuals for the reach of St. Vrain Creek between Lyons and the mouth of Boulder Creek.
22
Left Hand Creek joins the St. Vrain upstream of Longmont’s wastewater treatment plant, but the flow has not been gaged since WY1956. There is probably some flow at the mouth in all months; records from the drought of the 1930s show at least 1 or 2 cfs. Observations by the present water commissioner suggest typically 10-25 cfs in summer. The routing and management of water in Left Hand Creek are complex, and there is not enough information to allow calculation of flow at the mouth. For modelling purposes, the flow of Left Hand Creek is estimated from instantaneous current-meter measurements made by the City of Longmont, as explained below. The final reach in the analysis begins at the confluence of the St. Vrain and Boulder creeks (represented as the sum of the two gaged flows) and extends to the mouth of the St. Vrain near Platteville (Figure 2, Table 10). The St. Vrain Sanitation District discharges to a small lake, but it is so close to the St. Vrain that it is assumed to be relevant to the flow balance of the St. Vrain. The Weld County Tri-Area WWTP discharges to a ditch some distance from the St. Vrain, but for hydrologic analysis it is assumed that the flow reaches the St. Vrain. The period of record for this analysis is restricted to WY1992-2000 due to the absence of data for WY1991 at the mouth of Boulder Creek. The Fort St. Vrain facility can withdraw water from the St. Vrain at a pump station that is upstream of the USGS gage. The facility also can discharge to the St. Vrain (discharge point 002), but discharge at that location has been very infrequent in the past, and is assumed to be zero for modelling purposes. Analysis of Ungaged Flows Flow residuals were calculated for each day of the relevant period of record for
23
River Mile Gage Longmont to South Platte 16.26 Longmont 16.47 18.29 22.45 25.26 25.91 26.63 29.40 30.80 31.80 mouth
Sources
Withdrawals
Boulder Last Chance St. Vrain SD Last Chance Return Goosequill (inactive) Howlett Gulch Goosequill Pump Station Fort St. Vrain Drainage Slough (002)
Table 10. Catalog of hydrologic features included in the analysis of flow residuals for the reach of St. Vrain Creek between the confluence with Boulder Creek and the gage at the mouth. The gage joins the South Platte at river mile 33.60.
the four reaches mentioned above and shown in Figure 2, but not all of the daily residuals were used in estimating ungaged flow. When plotted against daily flow, daily residuals show a substantial increase in variance toward higher flows. Estimation of ungaged flows at higher channel flows is hampered by the addition of ungaged storm flows, which are not relevant to low-flow modelling, and by higher absolute error inherent in estimating higher absolute flows. Therefore, a threshold is chosen for each reach; below this threshold the residuals are quite stable and are representative of ungaged flows under low-flow conditions. The calculation of monthly median residuals is restricted to dates below the threshold. Residuals provide the basis for estimating all ungaged contributions to stream flows. In most cases for the Front Range, ungaged flows are accounted for by groundwater seepage that is distributed evenly along the stream channel. In some cases, ungaged tributaries or agricultural returns also may be important, as appears to be the case for the St. Vrain from Lyons to Longmont (see below).
24
Seasonal variation in ungaged flows is much greater for the St. Vrain than for other streams for which data are available (Table 11). For most streams of low elevation in the Front Range, winter residuals are on the scale of 1-3 cfs/mi and summer residuals are about two times greater. For the St. Vrain, however, summer residuals are about five times greater than the winter residuals. It seems unlikely that seepage would vary seasonally to this degree unless the alluvial aquifer were very small and of especially high transmissivity. A more likely explanation is that summer residuals are strongly affected by agricultural surface returns or ungaged tributaries. In many situations, it is sufficient to lump ungaged surface sources with seepage for ease of calculation. It is best not to lump contributions together in this case because the spatial distribution of flows would not be represented realistically through lumping. The need for a realistic spatial distribution of residuals in the St. Vrain above Boulder Creek became evident through efforts to reconstruct stream flows above the Longmont WWTP. The nearest gage is about 4 miles downstream, just above the confluence of Boulder Creek with the St. Vrain. Between the WWTP and the gage, there are no diversions and only a few additions to flow (chiefly releases from Union Reservoir). Flow above the treatment plant can be calculated starting with the gage and subtracting the additions, which include ungaged flow and effluent flow. On a surprising number of dates, however, the flow computed in this way was negative; experienced observers and limited field data suggest that the flow should not be less than 10 cfs. Estimates of ungaged flow can be made more realistic by assigning some of the residual flow to Left Hand Creek and by creating a seasonal component for agricultural returns.
25
Upper South Platte Segment 14 Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Top 2.34 1.99 2.10 2.24 2.93 3.21 3.62 3.84 3.92 3.65 3.09 2.62
Middle Bottom 3.95 3.50 3.52 4.10 3.33 3.02 2.10 3.02 3.57 4.60 4.90 4.64
1.65 1.70 1.53 1.71 1.99 2.64 2.87 3.23 2.89 2.48 1.78 1.65
Upper South Platte Segment 15 Upper
Lower
Boulder Cr to 75th Street
2.56 2.71 2.30 3.23 3.49 5.16 4.42 3.32 4.09 3.60 3.64 2.97
2.34 1.23 1.60 2.55 4.72 3.16 3.92 2.42 2.52 3.12 2.56 2.12
0.86 0.71 0.93 1.72 2.11 2.15* 2.18 0.66 1.51 1.22 0.83 0.76
St. Vrain upper 1.94 1.98 1.53 2.13 3.90 5.49* 7.09 6.65 5.90 3.48 2.34 1.91
*Calculated by interpolation because too few dates were below the threshold for calculation of residuals. Table 11. Seasonal variation in flow residuals (cfs/mi) for five reaches in the South Platte basin (from TMDL studies), and of Upper Boulder Creek, as included in this study. The residuals include ungaged tributaries and seepage. The residuals are for comparison with those of the St. Vrain above Longmont (last column), which are very high in warm months.
Accurate assignment of flows to Left Hand Creek is difficult because there is no gage near the mouth and direct measurements are scarce. Water managers were not able to recommend a scheme for calculating the flow from available records. The St. Vrain water commissioner indicated that 25 cfs is a typical flow for Left Hand Creek in late summer. The City of Longmont has measured flow at the mouth of Left Hand Creek on 30 dates beginning in November 1997. The data show a temporal pattern in which flows increase during the irrigation season (Figure 6). The flows are large enough that they
26
Left Hand Creek
80 70
Flow, cfs
60 50 40 30 20 10 0 1-Jan
20-Feb
10-Apr
30-May
19-Jul
7-Sep
27-Oct
16-Dec
Day of Year
Figure 6.
Instantaneous flow measurements made by the City of Longmont at the mouth of Left Hand Creek from November 1997 through March 2001. Data have been aggregated across years to show the seasonal pattern.
should be considered explicitly in the water-quality model. For modelling purposes, median values have been calculated for each month (Table 12), and these were subtracted from total ungaged flows on the main stem. When the seasonal increase in ungaged flows is small or the spatial distribution of ditches is uniform, a complete understanding of the source of seasonal variation may be unnecessary. In the St. Vrain, however, seasonal variation is large and the ditches are concentrated in a portion of the reach. The large seasonal increase in ungaged flows is evidence that agricultural surface returns make a significant contribution to stream flow. This contribution should be segregated spatially and temporally from the seepage component of the ungaged flows to the St. Vrain.
27
Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Median Flow, cfs 4.4 5.5 6.3 9.2 13.6 43.8 20.9 16.6 14.4 12.7 9.5 5.6
Notes
One measured value plus one marked “runoff”
Only one measurement available
Table 12. Monthly median flows at the mouth of Left Hand Creek estimated from 30 instantaneous flow measurements made by the City of Longmont.
For the St. Vrain from Lyons to Longmont, it is assumed for present purposes that agricultural returns are restricted to points upstream of the confluence with Left Hand Creek, but are distributed uniformly within that region. The amount of return flow is estimated as the increase in residual above a winter baseline, less the contribution of Left Hand Creek. To facilitate the parsing of residuals, seepage is assumed to be constant during the irrigation season (May through October for this calculation), and it is assumed that the seepage rate is equal to the median of the winter residuals (November through April), which should not contain agricultural returns (Table 13). In the TMDL model all three components of the ungaged flow (Left Hand, agricultural, surface flow, seepage) are shown explicitly, but agricultural returns are added to seepage in the appropriate reaches for the purpose of simplifying mass-balance calculations. At present, there are no data on the water quality of agricultural returns.
28
Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Total Ungaged
Left Hand
Seepage
30.9 31.6 24.4 33.9 62.2 87.7 113.2 106.2 94.2 55.5 37.4 30.5
4.4 5.5 6.3 9.2 13.6 43.8 20.9 16.6 14.4 12.7 9.5 5.6
26.5 26.1 18.2 24.7 31.3 31.3 31.3 31.3 31.3 31.3 27.9 24.9
Ag Surface Returns 0 0 0 0 17.3 12.6 61.0 58.3 48.6 11.5 0 0
Table 13. Separation of flow into components representing the typical contributions of Left Hand Creek, groundwater seepage, and agricultural returns. All flows are given as total cfs for the entire reach. Seepage during the irrigation season (May – October) is set equal to the median seepage rate (31.3) in winter months (November - April).
Table 14 summarizes the ungaged flow for the four reaches and also includes the estimated agricultural surface flows for the reach between Lyons and Longmont. The values are given as cfs per mile, which allows the estimation of seepage contributions at any particular point on any of the reaches.
DFLOW Concept and Implementation Biologically-based low flows can be defined at any location for which daily flows are available. Analysis is straightforward where gage records exist, and can be extended to other locations along a stream provided that daily flows can be calculated. Combining estimates of ungaged flow with daily records of diversions and additions makes it
29
Reach Threshold, cfs Length, mi Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Boulder Creek Orodell to 75th 75th to mouth 100 11.40 0.86 0.71 0.93 1.72 2.11 2.15 * 2.18 0.66 1.51 1.22 0.83 0.76
100 14.40
Lyons to Longmont, seepage 200 15.96
St. Vrain Lyons to Longmont, ag returns 200 11.23
0.34 0.41 0.46 0.22 1.28 6.67 3.07 3.87 1.89 1.31 1.25 0.81
1.66 1.63 1.14 1.55 1.96 1.96 * 1.96 1.96 1.96 1.96 1.75 1.56
0.00 0.00 0.00 0.00 1.54 1.13 * 5.43 5.19 4.33 1.03 0.00 0.00
Longmont to mouth
Table 14. Ungaged flows (cfs/mi) for each month in each of the four reaches defined for the analysis. In some months (denoted with an *), too few dates had upstream flows less than the threshold, and ungaged flows were estimated by interpolation from adjacent months. The ungaged flows for the lower part of Boulder Creek do not include Coal Creek, for which ungaged flows are set to zero. For the reach from Lyons to Longmont, seepage and surface agricultural returns reflect adjustment for a contribution from Left Hand Creek (see Table 13). The threshold for the final reach (Longmont to mouth) is based on the sum of the Boulder Creek and St. Vrain gages.
possible to calculate daily flows under low-flow conditions at any point between adjacent gaging stations. The objective of the low-flow analysis for the TMDL is the construction of a DFLOW-based, low-flow profile for each mainstem reach. The analysis starts with monthly acute and chronic low flows for each reach, obtained by use of DFLOW4, at the western boundary of the watershed. A low-flow regime in the main stem then is developed with appropriate adjustments for additions and withdrawals as well as
30
300 14.84 2.19 1.99 1.92 2.30 4.90 6.95 6.91 8.48 6.08 1.92 2.19 2.46
contributions from ungaged flows. The procedure for estimating adjustments for additions and withdrawals involves calculation of historical daily flows above and below each point of addition or withdrawal. For example, the adjustment for withdrawals from Boulder Creek to the Silver Lake Ditch involves estimation of the daily flow above the ditch (Orodell gage + Fourmile gage + ungaged flow contribution over 1.90 miles) and below the ditch (the flow calculated upstream less the measured diversion) for the same period of record, and processing both data sets with DFLOW4 software (as used by the state in permitting). Each low flow obtained in this way is consistent with the historical record. The difference between the upstream and downstream DFLOW value for a point of addition or withdrawal is the amount of water that must be added or removed in order to maintain the observed historical low-flow regime in the main stem. The practice of calculating low-flow additions or withdrawals by difference is necessary because the amount of an addition or a withdrawal is not necessarily correlated with stream flows. Thus, it does not make sense to calculate low flows separately for tributaries (e.g., Fourmile Creek) and expect the sum of the tributary and main-stem low flows to provide a realistic representation of main-stem low flows below their confluence. Calculation of separate low flows leads to an underestimate of the combined flow because of the implicit assumption that minima for the two flows coincide exactly in time, which they typically do not in a managed system. Some ditch diversions have been lumped to simplify the analysis. In most cases, lumping is dictated by use of a common headgate (e.g., Idaho Creek diversions from lower Boulder Creek and South Branch diversions from the St. Vrain), but proximity also
31
justifies lumping in some cases, especially where small ditches are involved. A list of ditches combined for the DFLOW calculations is given in Table 15. Stream (River Mile) Boulder Creek (4.14)
Primary Ditch Boulder White Rock
Boulder Creek (21.31)
Idaho Creek group
St. Vrain (0.97) St. Vrain (1.08)
Rough & Ready Swede
St. Vrain (2.30)
South Branch group
St. Vrain (2.93) St. Vrain (4.72)
Longmont Supply Denio Taylor
St. Vrain (6.03)
Clover Basin
St. Vrain (6.59)
Niwot
St. Vrain (6.94) St. Vrain (8.15)
South Flat Beckwith
Ditches Combined Boulder Left Hand McCarty North Boulder Farmers Smith Goss Houck 2 Carr Tyler Smith Emmons Highland South Side St. Vrain Palmerton Smead Montgomery Pvt Goss Pvt 1 Goss Pvt 2 Clough Pvt Clough True Webster McCaslin True Webster James Davis Downing Baker and Weese Chapman McCaslin Runyon Zweck Turner Dickens Pvt Peck Pella NW Mutual Hagers Meadow James Mason Cushman Island
Table 15. Grouping of ditches to facilitate calculation of the low-flow regime in the main stems of Boulder Creek and St. Vrain Creek.
In some instances, recorded diversions exceed the flow calculated for the stream at a point of diversion. The result of such a combination is negative estimated flow in the stream. Closer examination of the daily records often shows that the dates with negative estimated flows coincided with atypical flow patterns, indicating the possibility of an error in the data set. Based on careful examination of flow records in conjunction with spatial patterns of flows calculated along stream reaches, several rules were adopted in order to eliminate negative or unreasonably low flows. These rules were used to
32
maintain enough flow in the stream to satisfy the recorded demands (diversions). The resulting adjusted stream flows were used to estimate low-flow conditions above and below diversions and tributaries. The most troublesome reach for flow calculations was the St. Vrain from Lyons to Longmont. There are many diversion ditches in this reach and the recorded diversions exceeded the gaged supplies on a surprising number of dates. Three factors could contribute to the apparent insufficiencies of flow. One bias or random error is in gaging. There is no indication of gross error in gaging, but even small errors could cause negative residuals at low flow. A second issue concerns the timing of flow measurements. Daily average flows at gages are based on readings beginning at midnight, whereas the recording interval for the Supply Canal begins at 10 AM. Travel time between gages could be as much as a day. In addition, readings for ditches may be daily or less frequent, and diversion flows may be affected by river stage. Thus, when the river flow is not in steady state, the passage of a pulse through the system, for example, might not occur on the same day for two adjacent gages. A third issue concerns the timing and spatial distribution of agricultural return flows. At present it is possible to represent agricultural returns only as a set of monthly values distributed evenly across a long distance. On average, this may be reasonable; for a single day, it may be an inaccurate representation of the location and timing of returns. The objective of the flow adjustment procedure for the St. Vrain above Longmont was to insure that calculated flows were not less than 1 cfs below any diversion. The mechanics of the procedure involved supplementing the flow at points of water addition (e.g., St. Vrain Supply Canal, Foothills Reservoir release). These locations were chosen
33
for computational convenience and not because of specific suspicions about the accuracy of flow records. The practical effect of the flow adjustments is to facilitate the estimation of diversions using the DFLOW-by-difference approach described previously. Actual diversion amounts are preserved in the analysis rather than truncating the withdrawal when calculations suggest that streamflow was insufficient to meet demand. The supplemental flows (i.e., flow adjustments) applied to the St. Vrain tended to come in clusters, and the need for supplemental flows diminishes sharply after 1997. The biggest clusters for supplemental flows introduced via the St. Vrain Supply Canal were in spring 1994, spring 1996, and winter 1997. Clusters for flows introduced via the Foothills Reservoir release occurred in the summer of most years from 1991-1997. On the lower part of Boulder Creek, there were dates on which reported ditch diversions exceeded the flow calculated in the stream for some dates. Relatively large, late-season diversions to the Leggett ditch in late October 1991 were responsible for a few of the negative flows. Because the Leggett can dry up the river, and because the negative estimates are rare and small, a floor of 1 cfs was set on Boulder Creek below the Leggett ditch. Only five dates in October 1991 were affected. Negative calculated flows associated with the Lower Boulder ditch could be the result of underestimated delivery calculated for the Dry Creek Carrier. Flows are not measured at the mouth of the Carrier, but have been reconstructed for present purposes based on accounting for diversions to the Lower Boulder and Boulder and Weld ditches, as reported in the CDSS system. Records of releases from Baseline Reservoir to the Carrier also are available, but may not provide a complete picture of flows in the Carrier
34
because Boulder Creek or South Boulder Creek can put water directly into the Carrier, and because there are some small ditches that withdraw water directly from the Carrier. A review of CDSS data relevant to the Dry Creek Carrier showed some interesting patterns. Beginning about 1996, release from Baseline Reservoir matched or exceeded the total reported by the two ditches. Prior to 1996, flows were reported either as releases from Baseline or Baseline water taken by the ditches, but not both. The flow added by the Carrier to Boulder Creek was estimated by use of two rules. First, flow in the Carrier must be at least as large as the larger of two flows available from CDSS (Baseline release or the sum of Baseline water reported by the Lower Boulder and Boulder & Weld ditches). The rationale for this rule is influenced by the apparent change in accounting practices in WY96. Application of the rule assumes that all releases from the reservoir will reach Boulder Creek, as indicated by records since 1996. The second rule is that supplemental flow will be added to the Carrier as needed to ensure that there is at least 1 cfs in Boulder Creek after the Boulder & Weld ditch diverts the recorded amount. This rule is based on the likelihood that the Carrier at times contains water that did not pass through Baseline Reservoir. This rule is especially important for 1996, when CDSS records show that diversions to the two ditches greatly exceeded the flow in Boulder Creek below the Leggett ditch plus the release from Baseline Reservoir. Every day in August 1996 required a large supplemental flow (average about 50 cfs). This rule is important for mass-balance calculations because Carrier flows probably are very low in total ammonia. Application of the rule requires an iterative approach to setting ungaged flows because the estimation of supplemental flows
35
depends on an initial estimation of ungaged flow that is revised when the supplemental flows are added. Introduction of supplemental flows in the Carrier reduced but did not eliminate problems in calculation of downstream flows. Present procedure starts with the gage at the mouth of Boulder Creek as the basis for calculating flows in Boulder Creek below the Coal Creek confluence. In months when median ungaged flows are large, the estimated ungaged flow between the Rural Ditch and the gage often was more than the flow measured at the gage (i.e., subtraction of ungaged flow from the gage would produce a small negative flow just below the Rural Ditch). This problem was rectified by setting a floor of 1 cfs on flows calculated below the Rural Ditch (about 200 days, chiefly in July and August of the 9-y record, were affected by this change). A similar problem occurred for Boulder Creek below the Idaho Creek group of ditches, but only 12 dates were affected. The negative flows were small (less than 5 cfs) and a similar solution was applied. This practice was continued for Boulder Creek below the Godding, Daily & Plumb Ditch (4 dates). Calculated flows above and below the Panama Reservoir release were positive on all dates. Calculation of flows in Boulder Creek between Orodell and 75th presents problems similar to those described above. In an effort to reduce error accumulated through sequential calculations, the reach above the Green Ditch was divided. Calculations for the upper subreach begin with the Orodell gage, and calculations for the lower subreach begin with the gage at 75th Street. In the upper subreach, the first problems appear at the Farmers Ditch diversion about 3 miles downstream of the gage. In September 1991, there were six consecutive days on which the recorded diversion
36
exceeded estimated stream flow. The calculations are simple and there is little or no opportunity for other sources to add flow downstream of the gage, yet the shortage is as much as 27 cfs. The diversions listed for the Farmers Ditch on the dates in question (9/38/91) represent a large increase over the previous days, but are still within the bounds of normal operation in other years. The same period of time causes problems with calculations at downstream diversions, too. If the diversion records are correct, flow must have been added below the Orodell gage. The source of the flow is not known, but is treated as a supplemental flow added at the Orodell gage. In order to maintain positive calculated flows downstream, supplemental flows of 2-35 cfs were required. The target is at least 1 cfs downstream of the Butte Mill diversion. A similar cluster of negative calculations occurred for October 1993 and March 1999; these affected ditches further downstream. Supplemental flows in the range of 1-8 cfs were sufficient to leave at least 1 cfs below each headgate. There were 30 other dates, appearing singly or in small clusters, on which diversions exceeded known sources. Supplemental flows were used to maintain calculated flows above a 1 cfs threshold below each headgate. Supplemental flows were in the range of 1-21 cfs. In the subreach from the Green Ditch to the gage at 75th Street, flow calculations begin with the gage and move upstream. Problems begin immediately because the daily flow recorded for the WWTP effluent occasionally exceeded the average flow recorded at the gage. A floor of 2.1 cfs was estimated by means described in the section on DFLOW calculations. Problems also were associated with calculations for the Boulder Supply Canal. Negative values above the Boulder Supply Canal occurred mainly in August 1992 and August 1994. The only way to adjust the calculations is by reducing arbitrarily the
37
amount of flow contributed by the Supply Canal. Adjustments of 1-32 cfs were made on 37 dates. As a check on the reasonableness of the results obtained from the two subreaches, flows calculated below South Boulder Creek were compared with those calculated above the Green Ditch (Figure 7). Agreement is generally good at low flow, but at higher flows there is a tendency for the value below South Boulder Creek to underestimate the value above the Green Ditch. This is expected because surface runoff or other ungaged contributions are likely to be more important during periods of high flow. Fortunately, modelling focuses on low flows. Considering the opportunities for cumulation of error through sequential calculations, the agreement is very good.
Flow Computed above the Green Ditch, cfs
Boulder Creek 1600 1400 1200 1000 800 600 400 200 0 0
100
200
300
400
500
600
700
800
900
Flow computed below South Boulder Creek, cfs
Figure 7. Comparison of flows calculated above the Green Ditch with those calculated below South Boulder Creek. The line of equivalence is shown for perspective.
38
The reach from Lyons to Longmont presents a difficult challenge for the calculation of mainstem flows because there are so many diversions, and because the flow residuals are complex. The approach is similar to that employed on Boulder Creek, where the objective was maintenance of at least 1 cfs below all headgates. Supplemental flow was introduced via the St. Vrain Supply Canal. In the upper portion of the reach, most of the problems occurred in November 1994 and November 1995. A few additional dates required supplemental flows to achieve the target flow below each successive ditch downstream. The diversion for Foothills Reservoir required more adjustment than most other sites; about 100 dates requiring adjustment tended to be in small clusters, but also included large blocks in April 1994, February and April 1996, and January-February 1997. Beginning at the Oligarchy Ditch, supplemental flows are introduced via the Foothills Reservoir release. The intent is not to suggest that there has been a mistake in the records, but to acknowledge that ditches are taking more water than can be accounted for by the known sources. The water could have been introduced via the St. Vrain Supply Canal, but this would have raised instream flows in the St. Vrain above the Foothills release more than necessary to ensure that the river contained 1 cfs below each headgate. Based on historical records, the maximum release from Foothills Reservoir is about 35 cfs.
Low Flows for Future Conditions
The low-flow analysis, ungaged flows, and design capacities for treatment plants provide the basis for estimating acute and chronic low-flows in each month for the
39
streams in the St. Vrain watershed under future conditions for TMDL modelling. Some special handling has been necessary for some situations, as described below.
Special Handling of Low Flows above Major WWTPs The TMDL model constructs stream flows beginning upstream and moving downstream with adjustments for additions, withdrawals, and ungaged flows. On the St. Vrain, for example, calculations begin with the gage at Lyons and proceed sequentially to the confluence with the South Platte. A sequential approach is unavoidable in a reach model designed for mass-balance calculations. In contrast, low flows for a single WWTP could be calculated most easily from the closest gage. Thus, flows calculated by the reach model could differ from those calculated from the nearest downstream gage. The two locations of greatest concern for discrepancies are above the outfalls for the Boulder and Longmont WWTPs. Each facility is more than 10 miles from the upstream end of its reach. If reach modelling were not required, low flows for each facility could be calculated from the nearest gage rather than being brought down sequentially from upstream. Ideally, low flows calculated from the reach model would match the low flows calculated from the nearest gage for historical conditions. Unfortunately, there are differences in the outcome of the two procedures for determining low flows above the Boulder and Longmont outfalls (Table 16), and the differences are too large to ignore. The differences in the two estimation procedures can be reconciled through use of upstream tributaries to make an adjustment to flow in the main stem. The estimates from the reach model were adjusted by a change in tributary flows sufficient to cause a match with the flows at the nearest gage.
40
Boulder Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Acute Reach Nearest Model Gage 5.4 3.5 5.2 4.8 5.8 5.3 12.3 2.2 18.6 9.1 27.2 34.0 37.2 35.0 14.3 15.0 4.0 5.1 10.3 2.2 5.4 2.4 8.8 3.4
Longmont
Chronic Reach Nearest Model Gage 7.5 9.4 4.7 9.4 6.6 9.0 14.2 7.7 17.1 7.8 29.4 40.0 58.2 50.0 16.9 22.0 8.8 7.2 6.8 7.0 5.1 7.0 9.0 8.1
Acute Reach Nearest Model Gage 20.5 14 20.7 14 17.6 15 15.7 9.7 25.5 9.1 49.9 35 58.5 32 57.7 9.9 50.2 19 32.8 24 25.8 10 21.8 9.1
Chronic Reach Nearest Model Gage 19.6 15 21.1 15 10.3 15 13.1 14.2 35.7 14.2 46.1 37 62.7 43 57.4 36 62.0 36 32.4 20 18.9 14.2 21.4 14.2
Table 16. Comparison of low flows derived by two different procedures (see text) for Boulder Creek above the Boulder WWTP and St. Vrain Creek above the Longmont WWTP.
For Boulder Creek, the Boulder Supply Canal was used for adjustments; for the St. Vrain, Left Hand Creek provided the mechanism for adjusting flow. Special Handling for the Two Xcel Energy Facilities The Valmont and Fort St. Vrain facilities of Xcel are unlikely to add ammonia to the system or to have an ammonia limit in their NPDES permits. It does not make sense, therefore, to model these facilities at design capacity. In the absence of a permit limit for ammonia, these dischargers are providing dilution flow; the contributions should be shown in a manner that maintains the low-flow regime of the stream (i.e., DFLOW by difference). This is easy to do with the Valmont Station, where daily flows are recorded (flows comprise the entire low flow of South Boulder Creek at the point of discharge). Fort St. Vrain is more problematic because discharge to the St. Vrain has occurred very
41
infrequently. Although the pattern of operation could change, yielding more consistent flows in the future, a conservative assumption, used here for the ammonia TMDL, is to show no discharge from Fort St. Vrain.
Special Problems at Acute Low Flows The amount diverted to maintain historical low-flow regimes is constrained by the availability of water. The constraint may be more severe for acute than for chronic low flows. The end result is that the amount diverted under acute conditions may be less than that for chronic conditions. While this makes sense in terms of historical conditions, it causes problems for predictions where effluent flows are set to design capacities (future conditions). The symptom that appears in the modelling is acute low flows exceeding chronic low flows at the downstream end of main-stem reaches. The solution is to reset acute diversions so that they must be greater than or equal to chronic diversions. This has been done in the model. One other anomaly related to acute and chronic flows was detected for the release of water from Union Reservoir. Analysis of historical conditions in the St. Vrain above and below the Union Reservoir release showed that the acute low flow in June could only be sustained by a release of 20 cfs. In contrast, maintenance of chronic low flows in June required no release from Union Reservoir. The explanation for this peculiar situation is not apparent, but it does require an adjustment. In order to avoid a situation in which acute low flows in the main stem of the St. Vrain exceed chronic low flows in June, the release from Union Reservoir under acute conditions is set to zero.
42
Overview of Hydrologic Conditions for Modelling Table 17 summarizes the hydrologic conditions for modelling future conditions. The model includes design-capacity flows for all point sources, as listed in Table 17, for modelling future conditions.
Chemical Conditions for Modelling of Ammonia
The hydrologic information presented in the previous section of this report is useful for any TMDL application that might become relevant to in the future. Present modelling is for ammonia. Therefore, the goal of the present report is to use the hydrologic information to model the source and fate of ammonia throughout the basin with the constraint that the concentrations of unionized ammonia are consistent with acute and chronic standards. A basic requirement of the modelling is ammonia, pH, and temperature information for all water sources.
Chemistry of Non-Effluent Surface Flows Modelling for the St. Vrain TMDL on ammonia requires monthly pH, temperature, and total ammonia for each of 13 non-effluent water sources (Table 18), as well as ungaged flows. Extensive data are available for only a few of these sources; chemistry for most of the sources must be estimated from small amounts of data or no
43
CHRONIC FLOWS, cfs SITE St Vrain Headwater Boulder Headwater Coal Creek Headwater Lyons WWTP Supply Canal Highland ditch Rough & Ready Swede Foothills Inlet South Branch Longmont Supply Foothills Res Oligarchy Denio Taylor Clover Basin Niwot South Flat Cushman Bonus Left Hand Longmont WWTP Dry Cr (Niwot WWTP) Union Res. Last Chance St Vrain SD Howlett Gulch Goosequill Pump Station Ft St Vrain Red Lion Fourmile (Orodel, Inc) Silver Lake Anderson Farmers Boulder White Rock Upper Wellman Butte Mill San Lazaro South Boulder Green Boulder Supply Canal Boulder WWTP Leggett New Dry Cr Carrier Lower Boulder Boulder Weld B&B Howell Panama Res Armstead Godding Idaho Creek Rural Louisville WWTP Harris & Willis Rock Creek (Superior) Lafayette WWTP Erie Coal Cr Erie WWTP
Design Capacity, mgd
0.375
14 1.18 4.5 3.2 0.009 0.0045
0.11
25
0.015
3.4 2.2 4.4 1.2
JAN 16 10.5 1
FEB 16 10.5 0.8
MAR APR MAY 16 21 28 10.5 16 21 0.8 1.1 1.5
JUN 118 67 1.1
JUL 65 43 0.6
AUG 32 31 0.6
SEP 21 17 0.6
OCT 20 12 1.5
0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.0 0.0 0.0 0.0 2.0 23.0 115.0 24.0 6.0 13.5 13.4 13.4 17.7 11.0 39.0 112.0 26.0 21.5 0.0 0.1 0.6 0.6 13.4 49.0 35.0 16.0 2.7 0.0 0.0 0.0 0.0 0.0 13.0 8.0 1.0 0.0 3.4 2.7 2.5 2.4 8.5 37.9 0.0 0.0 2.9 0.8 0.8 0.8 0.9 1.0 2.8 24.9 15.3 1.6 0.3 0.3 0.3 0.3 0.4 10.1 7.0 6.5 4.8 2.4 2.4 2.6 2.6 3.6 25.7 31 8 0 4.6 4.6 4.8 4.8 5.5 30.5 24.0 10.4 3.7 1.2 1.2 1.3 1.3 1.5 1.4 12.2 4.2 3.9 0.0 0.0 0.0 0.0 0.0 0.0 4.0 2.0 0.7 0.1 0.1 2.2 3.2 3.2 9.4 8.2 8.1 5.9 0.0 0.0 1.0 1.7 1.7 4.6 2.5 1.7 1.7 0.0 0.0 1.1 3.1 3.1 3.3 3.1 3.1 3.2 0.0 0.0 1.0 3.0 4.3 12.3 15.3 13.0 1.0 Used as a flow regulator; see text 21.7 21.7 21.7 21.7 21.7 21.7 21.7 21.7 21.7 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 0.0 0.0 0.0 0.0 0.0 0.0 35.0 11.0 0.0 0 0.2 5.2 12 12 46.8 45 31 32.8 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 3 3 3.5 3.5 3.5 1 6 1 0 0.0 1.0 0.1 0.1 0.0 0.0 5.0 4.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.5 0.6 1.1 2.6 0.6 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 1.0 4.0 2.0 0.0 0.8 0.8 0.8 6.0 4.0 6.0 4.0 5.0 4.0 0.0 0.0 0.0 0.0 0.0 38.0 19.0 11.0 0.0 4.1 8.7 11.1 12.0 23.8 41.1 16.8 15.3 16.0 7.9 4.0 2.5 2.5 1.7 0.0 0.0 0.0 0.9 0.4 1.0 1.1 1.1 0.0 11.0 3.0 2.9 3.1 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.0 0.2 0.2 0.2 0.0 16.0 3.5 0.2 0.2 0.0 0.0 0.0 0.9 0.8 12.0 10.7 8.7 2.9 Used as a flow regulator; see text 38.7 38.7 38.7 38.7 38.7 38.7 38.7 38.7 38.7 0 0 1 1 0 26 39 12 11.8 0.0 0.0 0.0 0.0 0.0 0.0 2.0 1.0 0.0 0 0 2 12 25 55 71.3 35.3 18 0 0 0 18.5 18.8 20.5 9.8 9.8 9.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 0 0 4.5 4.5 2 2 3.5 0 0 0 0 0 0 0 0 0 Exclude this small ditch because the CDSS record is incomplete 0 0 0 14.9 14.9 15.9 8.8 13.6 13.6 0 0 0 9 13.5 13.6 26.6 14.6 14.4 0 0 3 9 16 14.9 15.2 15.2 15.9 5.3 5.3 5.3 5.3 5.3 5.3 5.3 5.3 5.3 These two small ditches excluded because DFLOW-by-difference not feasible. 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 0.0 0.0 0.0 4.3 5.6 3.9 2.0 0.0 0.0 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1.9
Table 17. Summary of low flows for future conditions. See text for explanation.
44
NOV 18 12 1.4
DEC 16 10.5 1
0.6 0.0 17.9 1.1 0.0 1.4 0.0 1.4 0 2.7 1.0 0.7 5.9 1.3 3.0 0.0
0.6 0.0 15.9 0.7 0.0 1.5 0.0 0.0 0.1 3.0 0.0 0.0 3.1 2.8 3.5 0.0
0.6 0.0 13.6 0.7 0.0 2.5 0.7 0.0 2.2 4.4 0.0 0.0 0.0 0.0 0.0 0.0
21.7 1.8 0.0 30.8 7.0 0 0.0 0.0 0.0 0.1 0.0 2.2 0.0 12.0 1.8 2.3 0.2 0.2 1.8
21.7 1.8 0.0 0 7.0 1 0.0 0.0 0.0 0.6 0.0 2.2 0.0 12.0 1.8 0.0 0.2 0.0 1.8
21.7 1.8 0.0 0 7.0 1 0.0 0.0 0.0 0.6 0.0 0.8 0.0 4.1 6.9 0.0 0.2 1.1 0.0
38.7 7.8 0.0 16.7 4.6 0.0 4.5 0
38.7 7.8 0.0 16.7 1 0.0 0 0
38.7 10.8 0.0 0.7 0 0.0 0 0
0 10 20.3 5.3
0 0 0 0 21 0 5.3 5.3
3.4 6.8 0.0 1.9
3.4 6.8 0.0 1.9
3.4 6.8 0.0 1.9
ACUTE FLOWS, cfs SITE St Vrain Headwater Boulder Headwater Coal Headwater Lyons WWTP Supply Canal Highland ditch Rough & Ready Swede Foothills Inlet South Branch Longmont Supply Foothills Res Oligarchy Denio Taylor Clover Basin Niwot South Flat Cushman Bonus Left Hand Longmont WWTP Dry Cr (Niwot WWTP) Union Res. Last Chance St Vrain SD Howlett Gulch Goosequill Pump Station Ft St Vrain Red Lion Fourmile (Orodel, Inc) Silver Lake Anderson Farmers Boulder White Rock Upper Wellman Butte Mill San Lazaro South Boulder Green Boulder Supply Canal Boulder WWTP Leggett New Dry Cr Carrier Lower Boulder Boulder Weld B&B Howell Panama Res Armstead Godding Idaho Creek Rural Louisville WWTP Harris & Willis Rock Creek (Superior) Lafayette WWTP Erie Coal Cr Erie WWTP
Design Capacity, mgd
0.375
14 1.18 4.5 3.2 0.009 0.0045
0.11
25
0.015
3.4 2.2 4.4 1.2
JAN 11.7 3.1 0.9
FEB 14 3.9 0.7
MAR APR MAY 11.7 15 36 4.9 6.3 18 0.5 1.4 0.9
JUN 130 97 0.5
JUL 84 48 0.7
AUG 43 33 0.3
SEP 20 17 1.2
0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.0 0.0 0.0 0.0 17.0 44.0 90.0 15.0 0.0 13.5 13.4 13.4 17.7 11.0 95.0 112.0 32.0 21.5 0.0 1.0 0.6 9.9 27.0 49.0 57.0 16.0 8.2 0.0 0.0 0.0 0.0 6.0 21.0 8.0 6.0 0.5 4.4 2.7 3.0 2.4 23.9 37.9 0.0 7.0 2.9 0.8 0.8 0.8 0.9 1.0 12.3 24.9 15.3 5.6 0.3 0.3 0.3 0.9 1.7 10.1 7.0 6.5 4.8 0.0 0.6 0.0 0.6 1.6 16.2 25.3 12.6 0.0 4.6 4.6 4.8 4.8 5.5 30.5 24.0 10.4 3.7 1.2 1.2 1.3 1.4 1.7 10.4 13.6 12.7 3.9 0.0 0.0 0.0 0.5 0.0 2.1 4.4 2.0 1.5 0.1 0.1 2.2 3.2 3.9 9.4 8.2 8.1 5.9 0.0 0.0 1.0 1.7 1.7 4.6 2.5 1.7 1.7 0.0 0.0 1.1 3.1 3.1 3.3 3.1 3.1 3.2 0.0 0.0 1.0 3.0 13.0 14.7 25.0 21.0 16.0 Used as a flow regulator; see text 21.7 21.7 21.7 21.7 21.7 21.7 21.7 21.7 21.7 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 1.8 0 0 0 0 0 0 15 34 11 0 0.2 5.2 12.0 26.8 49 45.0 43 32.8 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 7.0 3.0 4 3.5 3.5 3.5 2 6.0 8 0 0.1 1.0 0.1 0.1 0.1 3.0 5.0 4.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.5 0.9 1.1 5 2.8 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 1.0 4.0 4.0 3.0 0.8 0.8 2.4 6.0 4.0 10.0 4.0 5.0 5.0 0.0 0.0 0.0 0.0 0.0 38.0 34.0 19.2 6.3 4.1 8.7 11.1 15.9 25.2 77.9 16.8 15.3 16.0 8.2 7.2 2.5 2.5 1.7 0.0 0.0 0.3 1.7 0.4 1.0 1.1 1.1 0.0 11.0 5.6 2.9 3.9 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.1 0.0 0.0 0.0 1.0 2.4 4.1 0.6 0.1 0.0 0.0 0.0 0.9 0.8 12.0 10.7 8.7 5.3 Used as a flow regulator; see text 38.7 38.7 38.7 38.7 38.7 38.7 38.7 38.7 38.7 0.0 0.0 1.0 1.0 20.0 37.0 39.0 25.0 14.0 0.0 0.0 0.0 0.0 0.0 0.0 7.0 3.0 0.0 0 0 2.0 18.3 25.0 55.0 71.3 35.3 26 0 0 0 18.5 18.8 20.5 9.8 9.8 9.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 0 0 4.5 5 4.7 5.2 5 0 0 0 0 0 11.3 0 0 0 Exclude this small ditch because the CDSS record is incomplete 0 0 0 14.9 14.9 15.9 8.8 13.6 13.6 0 0 0 9.0 13.5 28.9 26.6 14.6 14.4 0.0 0.0 3.0 9.0 16.0 14.9 15.2 15.2 15.9 5.3 5.3 5.3 5.3 5.3 5.3 5.3 5.3 5.3 These two small ditches excluded because DFLOW-by-difference not feasible. 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 6.8 0 0 0 5 5.6 3.9 3.4 0 0 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1.9 1.9
Table 17 (continued). Summary of low flows for future conditions. See text for explanation.
45
OCT 13 12 1
NOV 14 3.1 1
DEC 13 7.9 1.2
0.6 0.0 17.9 1.1 0.0 3.4 0.0 1.4 0.0 2.9 1.6 0.7 5.9 1.3 3.0 0.0
0.6 2.0 15.9 0.9 0.0 1.5 0.0 0.0 0.0 3.2 0.0 0.0 3.1 2.8 3.5 0.0
0.6 0.0 13.6 0.7 0.0 2.5 0.7 0.0 0.0 4.4 0.0 0.0 0.0 0.0 0.0 0.0
21.7 1.8 0 30.8 7.0 0 0.0 0.0 0.0 0.3 0.0 3.0 0.0 12.8 1.8 2.3 0.2 0.0 1.8
21.7 1.8 0 0 7.0 1.0 0.0 0.0 0.0 0.5 0.0 2.2 0.0 12.0 3.2 0.0 0.2 0.0 1.8
21.7 1.8 0 0 7.0 1.0 0.0 0.0 0.0 0.6 0.0 1.8 0.0 4.1 6.9 0.0 0.2 0.1 0.0
38.7 24.7 0.0 16.7 4.6 0.0 4.5 0
38.7 16.0 0.0 16.7 1.0 0.0 0 0
38.7 10.8 0.0 0.7 0 0.0 0 0
0 10.0 20.3 5.3
0 0 0 0 21.0 0.0 5.3 5.3
3.4 6.8 0 1.9
3.4 3.4 6.8 6.8 0 0 1.9 1.9
St. Vrain at Lyons St. Vrain Supply Canal (Carter Lake Release) Foothills Reservoir Release Left Hand Creek at Mouth Union Reservoir Release Howlett Gulch Boulder Creek at Orodell Fourmile Creek at Orodell South Boulder Creek at Mouth (Valmont Release) Boulder Supply Canal Dry Creek Carrier (Baseline Reservoir Release) Panama Reservoir Release Coal Creek above Louisville Table 18. Non-effluent surface water sources for the St. Vrain TMDL model.
data. For grab-sample data on pH and temperature, the CAM recurrence model was used to translate grab sample values into daily average pH and temperature. The goal is to produce typical values for each constituent. An overview of the available data sources is given in Table 19. For sites with an adequate set of recent measurements of total ammonia (e.g., Boulder Creek at Orodell), median concentrations were computed for each month. Unfortunately, this approach can be applied to relatively few sites (Tables 19, 20). In other cases, some assumptions are required. For example, eight of the sources are reservoir releases, for which ammonia concentrations are likely to be low unless water is drawn from an oxygen-depleted bottom layer, in which case ammonia could be moderately high. Recent data are available only for Boulder Reservoir, for which ammonia concentrations are not elevated even when oxygen is depleted in late summer.
46
Monitoring indicates that ammonia in the Boulder Supply Canal will be undetectable (<0.1 mg/L) in all months. For Carter Lake (source of the St. Vrain Supply Canal), the STORET system contains very limited data from 1990. A bottom sample from mid-September showed little depletion of oxygen and only 0.14 mg/L for total ammonia. Ammonia is treated as undetectable in the water that reaches the St. Vrain from this source. For the smaller reservoirs (Foothills, Panama, Union, Thomas, Elliott), no data are available. The likelihood of summer stratification in these lakes varies with their mean depth. If stratification is unlikely, the release likely will contain little or no ammonia. If stratification occurs, there is potential for elevated ammonia concentration in late summer. Table 21 shows mean depth of reservoirs calculated on the basis of normal capacity and surface area; late summer mean depth is likely to be less (often by 50% or more). For perspective, Boulder Reservoir, with a mean depth of 25 feet, stratifies in the summer, whereas Cherry Creek Reservoir (not shown), with a mean depth of about 16 ft, shows unstable stratification. Union Reservoir, which has a mean depth of about 16 ft, probably does not stratify (Cal Youngberg, personal communication). Panama Reservoir, which has a mean depth of only 13 feet, probably is too shallow to generate elevated ammonia concentrations in the outflow to Boulder Creek.
47
Site St. Vrain at Lyons St. Vrain Supply Canal Foothills Reservoir release Left Hand Creek at mouth Union Reservoir release Howlett Gulch Boulder Creek at Orodell Fourmile Creek at Orodell South Boulder Creek at mouth Boulder Supply Canal
Period of Record 1976-81 1997-2001
1993-2002 1998 1994-2002
0 0 90 City of Boulder 3 Orodel, Inc. permit rationale 88 Xcel Energy 0
Dry Creek Carrier Panama Reservoir release Coal Creek above Louisville
Number of Source Measurements 56 USGS 0 0 31 City of Longmont
0
1996-2000
0 48 City of Louisville
Comments Carter Lake data show low concentrations Aggregated months to improve N (Nov-Mar, AprJun, Jul-Oct) Bottom release, but probably no stratification Small contribution of little consequence Set to average 0.04 mg/L in all months Most values below detection limits Boulder Reservoir data suggest very low concentrations Bottom release from Baseline Reservoir may have ammonia but travels several miles to Boulder Creek Bottom release, but probably no stratification
Table 19. Sources of data on total ammonia in each non-effluent surface water source for the St. Vrain TMDL model.
48
Site St. Vrain at Lyons St. Vrain Supply Canal1 Foothills Reservoir release1 Left Hand Creek at mouth Union Reservoir release1 Howlett Gulch1 Boulder Creek at Orodell Fourmile Creek at Orodell South Boulder Creek at mouth Boulder Supply Canal1 Dry Creek Carrier1 Panama Reservoir release1 Coal Creek above Louisville 1
Jan 0.08
Feb 0.08
Mar 0.02
Apr 0.02
May 0.01
Jun 0.01
Jul 0.01
Aug 0.02
Sep 0.01
Oct 0.02
Nov 0.08
Dec 0.11
0.20
0.20
0.20
0.20
0.20
0.20
0.34
0.34
0.34
0.34
0.20
0.20
0.08 0.04 0.03
0.10 0.04 0.03
0.05 0.04 0.04
0.10 0.04 0.03
0.01 0.04 0.03
0.04 0.04 0.03
0.00 0.04 0.03
0.07 0.04 0.03
0.02 0.04 0.03
0.09 0.04 0.06
0.00 0.04 0.04
0.01 0.04 0.07
0.03
0.05
0.03
0.10
0.03
0.03
0.02
0.03
0.04
0.03
0.04
0.05
Set to 0.0 mg/L.
Table 20. Medians of measured total ammonia concentrations (mg N/L) for releases from reservoirs. See text for explanation.
49
Reservoir Carter Lake Foothills Reservoir Union Reservoir Lake Thomas Boulder Reservoir Baseline Reservoir Panama Reservoir
Capacity, Surface Area, Mean Depth, AF acres feet 112200 1161 96.6 4239 159 26.7 12739 780 16.3 3728 182 20.5 13300 530 25.1 5300 240 22.1 4691 370 12.7
Conclusion Stratified Stratified Unstable Stratified? Stratified Stratified Unstable
Table 21. Morphometric characteristics of reservoirs providing water to the St. Vrain drainage. Mean depth is calculated as the ratio of capacity to surface area, and is the basis for judging the potential for summer stratification. Anecdotal evidence (J. Nuttle, CDPHE) of recent taste and odor problems below Lake Thomas suggests probable oxygen depletion and a potential for elevated ammonia concentrations. Lake Thomas is not directly relevant, however, because the flow passes through Elliott Reservoir (no data available) before entering the St. Vrain via Howlett Gulch. Thus, there is some potential for elevated ammonia in Howlett Gulch. The level of concern about this source is small, however, because the flow from Howlett Gulch is small and ammonia is not an issue for the only discharger downstream, Fort St. Vrain. Foothills Reservoir and Baseline Reservoir are both deep enough to show oxygen depletion in late summer, but there are no data to confirm stratification. Moreover, even if stratification does occur, there is no guarantee that ammonia concentrations will be elevated (cf. Boulder Reservoir). The position of Foothills Reservoir in the basin and the relatively small size of the releases suggest that there is not much risk if ammonia
50
concentrations are specified incorrectly. Baseline Reservoir is a different situation because the amount of water released from it can be large and the timing of the release (mainly in August) coincides with the time when oxygen depletion is most likely. Concentrations would have to be high in the reservoir release to have a measurable effect on Boulder Creek several miles away, however. Temperature has been measured only for a few of the water sources (Tables 22, 23). Where suitable measurements exist, they must be adjusted to a daily average temperature on each sampling date before a characteristic value is determined for each month. The best temperature record is an hourly data set from the St. Vrain at Lyons. In this case, no adjustment is required for converting a grab sample temperature to a daily average temperature. Instead, the median of all observations in each month characterizes the typical temperature of the stream. These values also will be applied to the nearby St. Vrain Supply Canal, for which no data are available. Adequate temperature records are available from Boulder Creek at Orodell and Coal Creek at Louisville. In both cases, daily average temperature values are available for each grab sample date from the CAM recurrence analysis. From the set of daily averages, a median is calculated for each month. Additional data are available for Fourmile Creek from the USGS. Medians of the unadjusted values were used for this site. Only a few measurements are available for temperature in Left Hand Creek. The general pattern of the temperatures suggests that
51
Site St. Vrain at Lyons St. Vrain Supply Canal Foothills Reservoir release Left Hand Creek at mouth
Period of Record 9/96-6/97
Number of Source Measurements 450 days J. McCutchan Longmont
Union Reservoir release Howlett Gulch Boulder Creek at Orodell Fourmile Creek at Orodell
1/98-11/01 11/71-2/95
41 City of Boulder 111 USGS
South Boulder Creek at mouth Boulder Supply Canal Dry Creek Carrier Panama Reservoir release Coal Creek above Louisville
1/97-12/00
183 Louisville
Comments Hourly record; Set equal to St. Vrain at Lyons Brief record with no data for determining daily average. Close to values for Coal Creek at Louisville. Use them.
No basis for adjustment to daily average
Table 22. Sources of data for characterizing typical temperatures for the St. Vrain TMDL model.
52
Site St. Vrain at Lyons St. Vrain Supply Canal1 Foothills Reservoir release2 Left Hand Creek at mouth3 Union Reservoir release2 Howlett Gulch2 Boulder Creek at Orodell Fourmile Creek at Orodell South Boulder Creek at mouth2 Boulder Supply Canal2 Dry Creek Carrier2 Panama Reservoir release2 Coal Creek above Louisville
Jan 0.1 0.1
Feb 0.7 0.7
Mar 3.6 3.6
Apr 5.6 5.6
May 8.7 8.7
Jun 10.7 10.7
Jul 15.3 15.3
Aug 16.0 16.0
Sep
Oct 5.5 5.5
Nov 2.8 2.8
Dec 0.4 0.4
2.4
3.3
7.3
10.1
13.3
15.5
19.6
18.3
18.3
11.9
6.3
3.1
0.9 0.0
1.7 2.0
3.0 5.8
6.2 6.3
8.1 10.8
12.6 12.3
14.9 15.8
16.9 17.0
13.1 14.0
9.4 9.5
4.2 1.3
0.5 0.0
2.4
3.3
7.3
10.1
13.3
15.5
19.6
18.3
18.3
11.9
6.3
3.1
1
Set equal to St. Vrain at Lyons
2
Set to match end of upstream reach
3
Set equal to Coal Creek at Louisville
Table 23. Medians of measured temperatures for water sources. See text for explanation.
53
they are similar to those of Coal Creek at Louisville. Consequently, the medians from Coal Creek are used for Left Hand Creek. For all other tributaries, the temperature is set so that it does not alter the temperature of the receiving water. The data for pH are sparse. Good records are available only for Boulder Creek at Orodell and Coal Creek at Louisville (Tables 24, 25). Data also are available for the St. Vrain at Lyons, but the data are old (nothing more recent than 1980). A few measurements are available for the mouth of Left Hand Creek, but these are insufficient for constructing a set of monthly values. The small contribution of Fourmile Creek will be set to the pH of the nearby Boulder Creek site at Orodell. In most cases, there is little risk in setting the input so that it will not change the pH of the receiving water.
Effluent Chemistry For each WWTP included in the St. Vrain TMDL study, temperature and pH must be set for modelling. Ammonia concentrations are not set at this stage because they are adjusted in the modelling process to meet regulatory requirements. For the most part, attention is focused on data obtained during or after 1995. The restriction is partly because older data are scarce, but also because pH is sensitive to changes in WWTP operations. The addition of effluent to a stream alters the fraction of ammonia in the unionized form by changing the pH and temperature of the stream. The effect of pH and
54
Site St. Vrain at Lyons
Period of Record 2/5812/80
Number of Source Measurements 63 USGS
St. Vrain Supply Canal Foothills Reservoir release Left Hand Creek at mouth Union Reservoir release Howlett Gulch Boulder Creek at Orodell Fourmile Creek at Orodell South Boulder Creek at mouth Boulder Supply Canal Dry Creek Carrier Panama Reservoir release Coal Creek above Louisville
Longmont
1/98-11/01
Comments No basis for adjustment to daily average
Brief record with no data for determining daily average. Close to values for Coal Creek at Louisville. Use them.
41 City of Boulder No data; set equal to Boulder Creek at Orodell
1/97-12/00
183 Louisville
Table 24. Sources of data for characterizing typical pH for the St. Vrain TMDL model.
55
Site St. Vrain at Lyons St. Vrain Supply Canal1 Foothills Reservoir release2 Left Hand Creek at mouth2 Union Reservoir release2 Howlett Gulch2 Boulder Creek at Orodell Fourmile Creek at Orodell2 South Boulder Creek at mouth2 Boulder Supply Canal2 Dry Creek Carrier2 Panama Reservoir release2 Coal Creek above Louisville 1
Set equal to St. Vrain at Lyons
2
Set to match end of upstream pH
Jan 7.40
Feb 7.40
Mar 7.60
Apr 7.30
May 7.10
Jun 7.10
Jul 7.20
Aug 7.20
Sep 7.45
Oct 7.30
Nov 7.55
Dec 7.45
6.86
7.46
8.05
7.56
7.82
7.31
7.69
7.37
7.59
7.35
7.11
7.45
8.15
8.25
8.39
8.26
8.08
7.99
8.24
7.99
7.97
8.11
8.28
8.17
Table 25. Characteristic pH corresponding to median % unionized ammonia and median temperature for water sources. See text for explanation.
56
temperature is joint and nonlinear; the monthly median (or average) for the unionized fraction is not the same as the unionized fraction calculated from the medians (or averages) of temperature and pH. Thus, it is desirable to preserve the linkage between pH and temperature measurements. When concurrent measurements of pH and temperature are available, the fraction of unionized ammonia (expressed as % unionized) is calculated on each date, and a median percent unionized is determined for each month. Median temperature is calculated separately. A "characteristic" pH then is back-calculated from the medians for percent unionized and temperature. This procedure, although complex, yields both pH and temperature for mass-balance calculations while also retaining computational consistency with the median percent unionized. The treatment facilities in the basin have varied requirements for monitoring pH and temperature (Table 26). Differences are so great in terms of frequency and coverage that multiple strategies must be used to obtain the best estimates of effluent characteristics for the dischargers. The most complete records are from Longmont and Louisville, which measure and report daily pH and temperature. At the other extreme, some of the small facilities report monthly minimum and maximum pH, but do not report temperature. The Red Lion Inn has a monthly reporting requirement when it discharges, but discharge is such a rare event that only 10 DMRs (Discharge Monitoring Reports) from the period of record have pH data. DMRs were not available for two of the smallest facilities; the average value listed in the permit rationales is used for these. Except where paired measurements were available from grab samples, medians were obtained separately for pH and temperature.
57
Discharger
Period of record 1/96-3/01 11/01-7/02 1/95-3/01
Frequency of reporting Monthly Daily Daily
Number of records 59 263 2252
Lyons Niwot Longmont Weld Tri-area St. Vrain Fort St. Vrain Red Lion Orodel San Lazaro Boulder
1/95-3/01
Monthly
2/96-7/99 6/99-6/01 1/96-12/00 1/95-12/99
Occasional Monthly Monthly Daily
10 1 60 1875
Rock Creek Louisville Lafayette Erie B&B
4/96-12/00 1/95-12/00 1/96-11/00 1/99-12/00 1/99-12/00
Monthly Daily Monthly Daily Monthly
52 2192 59 694 1
75
Comments DMR max-min pH; no temperature
DMR max-min pH and avg temperature DMR max-min pH; no temperature Avg in permit rationale; DMRs not available DMR max-min pH; no temperature Monthly temperatures from stream monitoring data sets; 1993-2002 DMR max-min pH; no temperature DMR max-min pH; no temperature Avg in permit rationale; DMRs not available
Table 26. Overview of the data sets available for characterization of pH and temperature in the effluents of discharges included in the St. Vrain TMDL.
The pH of Niwot's effluent has changed recently in response to modifications of its facility. Niwot made available daily pH measurements that began in November 2001. The pH values were steady for the first four months, but increased slightly in March coincident with a change in time of day for monitoring. The pH has been steady since March. The pH values from recent monitoring are generally lower by a few tenths than those used previously. Medians were calculated for each month for which data were available (November-July), and the July value was applied to the remaining months (August-October). For dischargers that report pH monthly, a minimum and a maximum value are given in the DMR. The average of the minimum and the maximum is used to represent pH for the month.
58
The effluents vary widely in pH, but there is little indication of strong seasonal variation (Table 27). Temperature, on the other hand, shows a strong seasonal pattern, especially in the smaller facilities (Table 28). For the few facilities with concurrent grab sample data for pH and temperature, monthly medians are calculated for % unionized (Table 29), and these are used to derive characteristic pH values (Table 30). Where it
Facility Lyons Niwot1 Longmont St. Vrain Fort St. Vrain Red Lion Orodel San Lazaro Boulder Rock Creek Louisville Lafayette1 Erie B&B 1 2
Jan 6.98 6.69 7.15 7.43
Feb 6.77 6.71 7.15 7.45
Mar 6.82 6.92 7.15 7.68
Apr 6.80 6.98 7.29 7.99
May 6.75 6.95 7.42 7.58
Jun 6.69 7.00 7.43 7.70
Jul 6.71 6.95 7.43 7.89
Aug 6.79 6.952 7.44 7.91
Sep 6.68 6.952 7.41 7.75
Oct 6.74 6.952 7.40 7.66
Nov 6.72 6.65 7.31 7.71
Dec 6.71 6.69 7.21 7.41
8.00 7.10 7.45 7.02 7.50 7.06 7.20 7.14 7.00
8.00 7.10 7.40 7.11 7.64 7.06 7.20 7.07 7.00
8.00 7.10 7.31 7.04 7.47 7.11 7.20 7.09 7.00
8.00 7.10 7.35 7.09 7.35 7.15 7.20 7.11 7.00
8.00 7.10 7.35 7.12 7.39 7.22 7.20 7.27 7.00
8.00 7.10 7.44 6.97 7.21 7.28 7.20 7.32 7.00
8.00 7.10 7.45 6.95 7.20 7.26 7.20 7.34 7.00
8.00 7.10 7.30 6.99 7.10 7.27 7.20 7.35 7.00
8.00 7.10 7.35 7.00 7.28 7.21 7.20 7.30 7.00
8.00 7.10 7.33 7.04 7.15 7.20 7.20 7.28 7.00
8.00 7.10 7.35 7.01 7.38 7.13 7.20 7.28 7.00
8.00 7.10 7.40 6.98 6.99 7.11 7.20 7.17 7.00
Based on recent and expected performance. No suitable data; set equal to preceding month.
Table 27. Monthly median pH values for effluents included in the St. Vrain TMDL. Number of cases available for calculating each median varies greatly; see text for further explanation.
Facility Lyons Longmont St. Vrain Fort St. Vrain Boulder Louisville Erie
Jan 12.7 10.6 3.7
Feb 13.0 11.0 4.1
Mar 13.8 12.2 8.4
Apr 15.2 13.9 11.6
May 16.2 16.3 17.1
Jun 19.8 18.9 20.5
Jul 22.2 20.8 24.2
Aug 23.4 21.2 22.8
Sep 21.6 19.9 20.0
Oct 18.3 16.7 13.4
Nov 16.1 13.5 7.2
Dec 13.5 11.5 4.8
12.8 15.3 14.7
13.3 15.3 15.2
14.7 15.8 15.9
14.8 16.9 16.2
16.3 18.2 17.2
18.5 20.0 18.6
21.3 21.5 20.3
22.0 22.4 20.9
21.3 22.3 20.3
19.6 20.4 18.5
17.2 18.2 17.7
14.6 15.9 16.3
Table 28. Monthly median temperature values for effluents included in the St. Vrain TMDL. Number of cases available for calculating each median varies greatly; see text for further explanation.
59
Facility Longmont Louisville Erie
Jan 0.27 0.32 0.37
Feb 0.29 0.32 0.33
Mar 0.31 0.36 0.35
Apr 0.47 0.43 0.37
May 0.82 0.58 0.61
Jun 0.94 0.75 0.78
Jul 1.11 0.80 0.86
Aug 1.19 0.87 0.92
Sep 1.01 0.75 0.83
Oct 0.75 0.64 0.69
Nov 0.52 0.46 0.60
Dec 0.34 0.39 0.44
Table 29. Monthly median % unionized for effluents included in the St. Vrain TMDL. Facility Longmont Louisville Erie
Jan 7.15 7.06 7.14
Feb 7.16 7.07 7.08
Mar 7.14 7.10 7.08
Apr 7.27 7.14 7.09
May 7.44 7.22 7.28
Jun 7.41 7.28 7.34
Jul 7.43 7.26 7.33
Aug 7.44 7.27 7.35
Sep 7.42 7.21 7.31
Oct 7.39 7.20 7.29
Nov 7.33 7.13 7.26
Dec 7.22 7.12 7.16
Table 30. Characteristic pH for effluents from facilities reporting concurrent measurements of pH and temperature. See text for explanation of calculations. Characteristic pH is used in preference to median pH in modelling. is available, characteristic pH will supersede median pH for modelling; a comparison of values in Tables 27 and 30 shows little difference in characteristic and median pH for these discharges, however. Temperature data are not available for all discharges. The protocol for assigning monthly values of temperature to each facility is based on the size of the facility. Thus, temperatures from the Louisville facility will be used for Niwot and Lafayette, and temperatures from the Lyons facility will be used for the smaller WWTPs. Although Boulder does not measure effluent temperature as a part of routine effluent monitoring, monthly measurements of effluent temperature are available from the Boulder stream monitoring program. Setting Total Ammonia, Temperature, and pH for Unmonitored Water Sources Total ammonia, temperature, and pH data are not available for a number of water sources. In most cases, these add little flow. A default assumption is that these sources
60
have the temperature and pH of the receiving water, and that total ammonia is 0.0 mg/L. Sources handled according to these assumptions are listed in Table 31, and the relevant cells in the model are marked with a comment. Diversions are assumed to have the temperature, pH, and ammonia concentrations of the stream near the point of withdrawal.
Source Ungaged Flow Foothills Reservoir Release Left Hand Creek Union Reservoir Release SVSD (Oxbow Lake) Howlett Gulch Red Lion Fourmile Creek San Lazaro South Boulder Creek Boulder Supply Canal New Dry Creek Carrier B&B MHP Panama Reservoir Release
Temperature + + + + + + + + + + + + +
pH + + + + + + + + + + +
Table 31. Conventions for setting temperature and pH of sources for which no, or too few, field measurements were available. A symbol (+) indicates that temperature or pH for a particular source was set equal to that of the receiving stream.
For supplementary flows used in flow adjustments (see hydrology section), the pH is set equal to that of the reach upstream. Formulas governing temperature and ammonia are superseded by upstream conditions if the flow is negative, as is necessary on some dates (e.g., the Left Hand Creek node is used to remove water from the stream). The same approach is applied to Boulder Creek above the WWTP by using the Boulder Supply Canal as the regulator of flow.
61
Setpoint and Rebound
The setpoint for CAM analysis of ammonia is the downstream target for pH and temperature below each discharger. The trajectories of pH and temperature determine the downstream profile of the fraction of ammonia that will be unionized. Setpoint conditions are determined with CAM, which uses information about temporal patterns of variation to convert grab sample data on pH and temperature into daily average or characteristic values. CAM also defines the rebound rates for pH and temperature below a point of wastewater discharge and the conditions in the stream for each month that are consistent with the once-in-three-year recurrence assumption of current regulatory practice. Although the CAM analysis can be performed with default conditions set by CAM software, it is preferable to use site-specific patterns of variation where suitable data exist. Site-specific data are scarce for the rebound rate, which determines the distance over which pH and temperature move toward the setpoint target. No existing studies on the St. Vrain system are well suited for this analysis. CAM contains default rates for rebound of temperature and pH that could be used to determine the longitudinal trajectory of pH and temperature. Use of these default rates is common when there is a single discharger. When there are multiple dischargers, however, an alternate approach is preferable. When multiple discharges enter a stream or when data are available at several stations, setpoint conditions can be assumed to coincide with each sampling station. The longitudinal sequence of setpoint values then
62
acts by interpolation as a guide for rebound. This approach works for Boulder Creek and Coal Creek, and is used here. The default rates of CAM are required for the St. Vrain main stem, where sampling sites are spread more thinly.
Site-Specific Amplitudes and Times of Maxima
Setpoint analysis in CAM is designed to encourage use of site-specific data for defining the diel characteristics (24-hour cycle) of variation in pH and temperature. The amplitude of variation and the time of the daily maximum are key components of the procedure used to calculate daily means from grab sample data. The characteristics of variation must be defined for all months, either from site-specific data or by acceptance of default values. The strategy for defining site-specific variation depends in part on the availability of data (Table 32). For the St. Vrain at Lyons, there is a set of hourly temperatures from consecutive days spanning more than a year (with one gap of several weeks). In this unique case, the typical amplitude of variation in temperature in each month is easily calculated as half the median of the observed daily range. Interpolation is used to fill in missing values, and the time of the daily maximum can be determined from plots of observed values. The next most complete data record is a set of data for 38 dates obtained by the City of Boulder in 1986-87 for two stations (Table 32). Although the data are older than is ideal, the characteristics of variation are not likely to be as susceptible to change as the means. Each study consists of a 24-h record of hourly pH and temperature in Boulder Creek and Coal Creek at points just above their confluence. The number of sample sets
63
in a given month (2 or 3) is too small to produce a meaningful statistical description of typical values in any month. Instead, a curve-fitting approach was used to define seasonal patterns of daily maximum and daily minimum temperatures (Figure 8). A nonlinear curve-fitting program was used to fit a sine function to the maxima and minima; the daily amplitude for temperature then was calculated as the difference between the two curves at mid-month (Table 33). Similar but smaller (N=25) data sets were available just above Louisville and Lafayette (Table 32); these were analyzed by the same method as used for the City of Boulder's data. Time of maximum temperature for the six documented locations
B o u ld e r C re e k 28
o
Temperature, C
23
18
13
8
3
-2 1 -Ja n
2 0 -F e b P re d ic te d M a x
Figure 8.
1 0 -A p r
3 0 -M a y
1 9 -Ju l
O b s e rv e d M a x
7 -S e p P re d ic te d M in
2 7 -O ct
1 6 -D e c
O b s e rv e d M in
Daily maximum and minimum temperatures recorded during each of 38 diel studies on Boulder Creek above the confluence with Coal Creek. A sine curve has been fitted to each set of points.
64
was determined by analysis of frequency distributions or simple inspection. Amplitude and time of maximum temperature for any given location were determined by proximity to the documented sites of Tables 33-34. Above Boulder WWTP, CAM default values were used. Because the amplitude of daily pH variation is less consistent than that of temperature, it must be derived differently. The daily minimum pH is relatively constant across months, but the maximum varies in a manner that is difficult to characterize with a simple function. The basic pattern shows higher amplitude of variation in the summer than in the winter, as expected, because algal demand for carbon is higher in the summer. A complication occurs during spring runoff, when the amplitude of variation in pH variation is depressed. Because the seasonal pattern of variation is more complex than that of temperature, the amplitude of variation in each month was set to the median of the observed values for that month rather than being derived by curve fitting (Table 35). Times of pH maxima were determined by inspection of field data and some smoothing based on judgment (Table 36).
Site St. Vrain at Lyons Boulder Cr above Coal Creek Coal Cr at mouth Coal Cr at Louisville St. Vrain above Boulder Creek Boulder Creek at mouth
Period 9/96-6/97 3/86-11/87
Number of 24h events 450 38
Comments
3/86-11/87 4/00-4/02 3/86-9/87
38 25 13
Temperature and pH from City of Boulder Temperature and pH from City of Louisville Temperature and pH from City of Longmont
3/86-9/87
13
Temperature and pH from City of Longmont
Temperature data from J. McCutchan Temperature and pH from City of Boulder
Table 32. Data sets available for characterizing diel variation in pH and temperature. The number of events represents the opportunities to obtain daily maxima and minima.
65
Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Coal Cr at Louisville 2.02 2.64 3.44 4.38 5.10 5.44 5.30 4.70 3.81 2.90 2.18 1.86
Coal Cr at mouth 1.90 2.21 2.73 3.44 4.08 4.50 4.58 4.29 3.71 3.03 2.37 1.97
Boulder Cr above Coal Cr 2.78 3.10 3.50 3.97 4.31 4.47 4.38 4.07 3.62 3.18 2.83 2.68
St. Vrain at Lyons 0.27 1.28 3.08 2.86 2.21 1.63 2.46 2.47 2.10 * 1.80 1.58 0.41
St. Vrain abv Boulder Cr 2.78 3.09 3.47 3.91 4.22 4.36 4.27 3.97 3.54 3.12 2.80 2.68
Boulder Cr at mouth 2.35 2.74 3.31 4.01 4.58 4.91 4.88 4.50 3.88 3.20 2.62 2.31
*Interpolated Table 33. Amplitude of diel temperature variation for each month. Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Coal Cr at Louisville 15:00 15:00 15:00 16:00 16:00 15:00 15:00 15:00 17:00 17:00 14:30 14:30
Coal Cr at mouth 15:00 15:00 15:00 15:00 16:00 16:00 15:00 15:00 15:00 15:00 15:00 15:00
Boulder Cr above Coal Cr 15:00 15:00 15:00 15:00 16:00 17:00 15:00 15:00 15:00 15:00 15:00 15:00
St. Vrain at Lyons 13:00 16:00 17:00 17:00 17:00 15:00 17:00 17:00 16:00 16:00 16:00 16:00
St. Vrain abv Boulder Cr
Boulder Cr at mouth
15:00
15:00
18:00 18:00 16:00 15:00 16:00 16:00 15:00
16:00 18:00 16:00 16:00 16:00 16:00 14:00
Table 34. Time of daily temperature maximum. Month Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Coal Cr at Louisville 0.22 0.25 0.39 0.21 0.23 0.21 0.55 0.21 0.23 0.28 0.23 0.18
Coal Cr at mouth 0.48 0.65 0.45 0.23 0.28 0.20 0.30 0.25 0.23 0.35 0.15 0.33
Boulder Cr above Coal Cr 0.60 0.65 0.78 0.70 0.79 0.40 0.75 0.90 1.10 0.75 0.58 0.53
St. Vrain abv Boulder Cr 0.54
Boulder Cr at mouth 0.54
0.32
0.64
0.40 0.25 0.24 0.23 0.49 0.44 0.38
0.80 0.35 0.68 0.61 0.75 0.53 0.61
Table 35. Amplitude of diel pH variation for each month.
66
Month
Coal Cr at Louisville 16:00 16:00 16:00 15:00 15:00 15:00 14:00 14:00 14:00 15:00 15:00 16:00
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Coal Cr at mouth 14:00 14:00 15:00 16:00 15:00 15:00 15:00 14:00 14:00 14:00 15:00 14:00
Boulder Cr above Coal Cr 14:00 15:00 15:00 14:00 15:00 15:00 14:00 14:00 16:00 14:00 13:00 14:00
St. Vrain abv Boulder Cr
Boulder Cr at mouth 14:00
15:00
15:00
16:00
15:00 14:00 16:00 16:00 16:00 16:00 14:00
16:00 16:00 16:00 15:00 15:00
Table 36. Time of daily pH maximum.
Recurrence Analysis The characteristics of variation in pH and temperature documented in the preceding section provide the foundation for the recurrence analysis through which CAM produces setpoint conditions. Normally, these setpoint conditions would be used directly in the next step of modelling. A recent analysis (Appendix I) of setpoint conditions has shown, however, that one of the assumptions of CAM is not valid for the St. Vrain system. There is no correlation between low flows and extremes of pH and temperature, contrary to the usual assumption for CAM. Under these circumstances, the logical alternative is the use of median pH and temperature. The full procedure involves use of adjusted grab sample data to compute monthly median temperature and fraction unionized, from which the characteristic pH is back-calculated. When setpoints are defined with monthly medians, there is no distinction between chronic and acute setpoint conditions.
67
Application of Setpoint Conditions
Setpoint conditions are defined in CAM on the basis of analysis of stream monitoring data. Ideally, the grab sample data record would consist of at least 3 years of collection at a weekly to monthly frequency. Several data sets of this type are available from the St. Vrain watershed (Table 37). Almost all monitoring sites have been included in the analysis even though the sampling interval exceeds the recommended maximum length (30 days) on numerous occasions at a few sites. Site St. Vrain at Lyons St. Vrain above Left Hand St. Vrain at County Line Boulder Cr at Orodell Boulder Cr at Eben Fine Boulder Cr at 61st Boulder Cr abv WWTP Boulder Cr at 75th Boulder Cr at 95th Boulder Cr abv Dry Cr Boulder Cr at 107th Boulder Cr abv Coal Boulder Cr blw Coal Coal Creek abv Louisville Coal Creek blw Louisville Coal Creek abv Lafayette Coal Creek blw Lafayette Coal Creek at mouth
Record
Frequency Source
CAM conditions
2/98-3/01 2/98-3/01 1/98-11/01 1/98-4/02 1/98-4/02 1/98-4/02 1/98-4/02 1/98-4/02 1/98-2/02 1/98-4/02 1/98-4/02 1/98-4/02 1/97-12/00 1/97-12/00 1/96-4/02 1/98-4/02 1/98-4/02
Monthly Monthly Monthly Monthly Monthly Monthly Monthly Monthly Monthly Monthly Monthly Monthly Weekly Weekly Monthly Monthly Monthly
St. Vrain abv Boulder St. Vrain abv Boulder Default, medium Default, medium Default, medium Boulder Cr abv Coal Boulder Cr abv Coal Boulder Cr abv Coal Boulder Cr abv Coal Boulder Cr abv Coal Boulder Cr abv Coal Boulder Cr mouth Louisville gage Louisville gage Louisville gage Louisville gage Coal Cr mouth
Longmont Longmont Boulder Boulder Boulder Boulder Boulder Boulder Boulder Boulder Boulder Boulder Louisville Louisville Lafayette Lafayette Boulder
Table 37. Data records used in setpoint analysis for CAM. The column for CAM conditions identifies the site from which characteristics of variation were taken.
As explained above, characteristics of variation were available for only 6 locations. Setpoint analysis for a particular monitoring location was based on
68
South Platte River
St. Vrain Supply Canal (from Carter Lake)
*St. Vrain at Lyons
St .V r ai nC r.
Cr. rain St. V
Dr
yC
r.
below Coal Creek
above Coal Creek
Cr .
at 75th Street
at 107th Street at 95th Street
Cr .
mi le
above Dry Creek
Dr y
Fo ur
above WWTP
Boulder Cr.
Cr .
above Lafayette
S.
de ul Bo
above Louisville
Coal Creek Mouth
Sampling Sites Discharge Facilities Gaging Stations
below Lafayette
Setpoint Data Coal Creek at Louisville Coal Creek at Mouth
below Louisville
r
l Coa
North
er C
r.
Bo uld
dC Han
Coal Cr.
Left
r.
at County Line Road
above Left Hand Creek
Boulder Creek at Mouth
Cr.
kC Roc
Boulder Creek above Coal Creek St. Vrain above Boulder Creek
r.
St. Vrain at Lyons
0
1
2
3
4
5 miles
Figure 9. Map of division of study area into reaches.
characteristics of variation at a nearby location as indicated in the final column of Table 37 and in Figure 9. CAM analysis was used only to provide daily average temperature and pH, and not to estimate recurrence values of temperature and pH, which were set to median values. A summary of setpoint conditions for pH and temperature appears in Tables 3839. For perspective on the longitudinal patterns that emerge from the data sets, Figure 10
69
Site St. Vrain at Lyons1 St. Vrain above Left Hand St. Vrain at County Line Boulder Creek at Orodell Boulder Creek at Eben Fine Boulder Creek at 61st Boulder Creek above WWTP Boulder Creek at 75th Boulder Creek above Dry Creek Boulder Creek at 95th Boulder Creek at 107th Boulder Creek above Coal Creek Boulder Creek below Coal Creek Coal Creek at Louisville Coal Creek below Louisville Coal Creek above Lafayette Coal Creek below Lafayette Coal Creek at mouth 1
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
0.7 2.3 0.9 0.7 2.2 0.8 10.6 7.0 5.3 3.3 3.5 3.3 2.4 11.5 2.9 4.1 2.8
1.9 4.1 1.7 1.1 2.2 0.8 11.1 7.3 6.0 5.1 4.0 3.3 3.6 11.7 2.2 2.4 3.6
7.4 8.4 3.0 3.0 6.1 4.9 12.3 10.5 9.7 9.2 10.6 8.9 9.0 12.6 5.4 5.9 8.1
6.0 10.2 6.2 6.2 9.2 7.2 12.3 8.8 10.6 10.3 12.6 7.1 11.5 13.1 8.7 10.7 5.6
17.3 17.2 8.1 8.3 10.1 9.5 12.4 11.8 12.4 13.8 14.9 14.7 14.2 15.2 12.4 13.2 12.7
16.5 16.4 12.6 12.8 14.3 15.2 16.7 15.3 15.5 16.4 17.9 18.1 16.1 16.1 16.5 16.9 16.8
15.8 15.6 14.9 15.2 17.4 17.6 18.4 19.2 19.0 20.5 20.9 21.7 19.7 20.0 19.4 19.6 19.9
12.3 11.8 16.9 16.0 19.2 20.4 20.0 19.5 19.3 19.3 19.5 18.6 18.0 20.6 17.5 19.1 17.8
15.2 15.3 13.1 11.9 15.6 13.7 18.5 18.0 17.3 17.3 18.2 16.1 17.9 22.1 17.7 18.7 16.7
8.6 9.7 9.4 8.9 11.2 10.4 17.6 14.5 14.4 13.4 11.7 12.1 13.2 19.2 12.7 14.0 12.2
2.1 4.1 4.2 3.1 7.6 6.6 14.2 12.3 11.5 10.3 10.0 9.2 8.9 14.9 5.8 8.1 7.7
1.4 3.2 0.5 0.4 1.0 0.5 11.4 7.0 5.4 3.6 3.0 4.3 5.0 12.7 2.6 4.1 1.9
Set equal to Bounder Creek at Orodell
Table 38. Setpoint temperature.
70
Site St. Vrain at Lyons1 St. Vrain above Left Hand St. Vrain at County Line Boulder Creek at Orodell Boulder Creek at Eben Fine Boulder Creek at 61st Boulder Creek above WWTP Boulder Creek at 75th Boulder Creek above Dry Creek Boulder Creek at 95th Boulder Creek at 107th Boulder Creek above Coal Creek Boulder Creek below Coal Creek Coal Creek at Louisville Coal Creek below Louisville Coal Creek above Lafayette Coal Creek below Lafayette Coal Creek at mouth 1
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
7.66 7.10 6.86 7.08 7.59 7.40 6.84 7.20 7.45 7.77 7.99 7.89 8.09 7.31 8.23 8.18 7.98
7.32 7.60 7.47 7.28 7.79 7.78 6.83 7.38 7.63 8.03 8.26 7.96 8.22 7.40 8.43 8.29 7.88
7.66 7.56 8.05 7.65 8.37 8.02 7.22 7.61 8.28 8.34 8.55 8.18 8.21 7.33 8.40 8.39 8.27
7.80 7.79 7.53 7.57 8.09 7.84 7.21 7.52 7.83 8.10 8.38 8.39 8.05 7.42 7.90 7.61 8.45
8.07 7.97 7.74 7.65 8.44 7.94 7.62 7.74 7.82 7.97 8.41 8.26 7.92 7.53 7.88 7.78 7.94
8.02 7.87 7.36 7.55 7.36 7.42 7.09 7.72 8.01 8.40 9.14 8.78 7.95 7.53 8.29 8.18 8.22
7.95 7.72 7.68 7.71 7.88 7.77 7.16 7.84 8.11 8.49 8.90 8.81 8.18 7.35 8.39 8.25 8.28
7.90 7.82 7.38 7.28 7.89 7.89 7.05 7.52 7.54 7.88 8.51 8.32 8.00 7.38 8.29 8.02 7.94
8.16 8.20 7.54 7.54 8.29 8.45 7.46 7.68 7.74 7.95 8.42 7.94 7.90 7.29 8.31 7.99 8.26
8.06 7.80 7.38 7.40 8.43 8.22 7.18 7.40 7.83 7.97 8.61 8.37 7.98 7.37 8.29 8.16 8.18
7.66 7.74 7.14 7.09 7.55 7.52 6.66 7.17 7.45 7.91 8.44 8.04 8.11 7.38 8.21 8.20 8.24
7.66 7.55 7.46 7.63 7.85 7.74 6.87 7.17 7.34 7.60 8.01 7.87 8.07 7.42 8.39 8.30 8.07
Set equal to Boulder Creek at Orodell
Table 39. Setpoint pH.
71
shows chronic and acute pH values for Boulder Creek in selected months. There is a strong pattern of pH change across months. This pattern sets a sequence of steps through which rebound will occur in the reach model.
Travel Time
Prediction of changes in ammonia concentration over distance requires estimates of travel time, which are derived from water velocity. Default equations for velocity are available in the Colorado Ammonia Model (CAM), but site-specific information is preferred. The relationship between discharge and velocity in streams of the St. Vrain watershed can be obtained from stream discharge rating data for the seven gaging stations (Table 40). In most cases, numerous data points are available encompassing a wide range of flows because the field measurements are taken once or twice per month. Included in Location St. Vrain at Lyons St. Vrain at Longmont St. Vrain near Platteville Boulder Cr at Orodell Boulder Cr at 75th Boulder Cr at mouth Coal Cr at Louisville
Source SEO USGS
Equation V = 0.0788Q0.6405 V = 0.5580Q0.2665
R2 0.989 0.394
Apply to Above Left Hand Left Hand to RM21
SEO
V = 0.6384Q0.2150
0.839
RM21 to mouth
SEO USGS SEO USGS
V = 0.1875Q0.4870 V = 0.2086Q0.4488 V = 0.3454Q0.3561 V = 0.2824Q0.5665
0.853 0.750 0.537 0.895
Above South Boulder South Boulder to Coal Cr Coal Cr to mouth Coal Cr
Table 40. Equations defining stream velocity (V, ft/s) as a function of stream discharge (Q, cfs). Equations will be applied to stream reaches based on best judgment concerning substrate and channel configuration.
the present analysis are 30-50 of the most recent measurements; older data are less desirable because stream cross sections may change through time. The relationship from
72
Boulder Creek 9.5
9.0
8.5 Setpoint pH
March July August October 8.0
7.5
lC r
r
oa C w bl
C oa
lC
St v ab
10
7t h
St th 95
D ry
C r
St v ab
th 75
W W
TP
St ab
v
st 61
en Eb
O ro d
el
l
Fi ne
7.0
Figure 10. Longitudinal pattern of pH defining chronic setpoint conditions for a series of monitoring stations along Boulder Creek. Results for four months are shown; see Table 39 for all months.
Coal Creek at Louisville is shown as an example of the method used to derive velocity equations (Figure 11). A minimal amount of data screening was required. The data set for the St. Vrain at Lyons contained one apparent error on 10/13/1999: velocity was recorded as 1.77 ft/s,
73
C o a l C r e e k a t L o u is v ille 10
Velocity, ft/s
y = 0.2858x 0 .5 6 3 7 R 2 = 0.9007 1
0.1 0.1
1
10
100
1000
D isch arg e , cfs
Figure 11. Relationship between velocity and discharge based on USGS field calibration records for the gage on Coal Creek at Louisville. but the ratio of discharge to area (V=Q/A) was 0.77; the value of 0.77 was taken as correct. No other errors were detected. The relationship between velocity and discharge in Boulder Creek at 75th Street appears to have been altered by channel changes due to high flows during runoff in 1995. Consequently, analysis is restricted to measurements beginning in WY1996 for this site and for the next station downstream on Boulder Creek (at the mouth). A similar situation occurred with the data set for the St. Vrain below Longmont, although the fit of the function to the observed values was not improved by restricting analysis to a subset of the data. The strength of the relationship between velocity and discharge in Coal Creek at Louisville was weakened substantially by three points for which the measured width was 74
much less than that recorded on most other dates (3-6 ft vs. 10-16 ft). These dates were excluded on the assumption that the measurements were taken on a transect that had different channel characteristics than the primary transect.
Estimation of Ammonia Loss Rate: Calibration and Validation
Ammonia is removed from streams by biological processes, the most important of which is typically nitrification. In general, it is not practical to model the effect of each process separately; instead, the loss rate lumps all processes. Ammonia loss in CAM is modeled with first-order kinetics, and the default rate is 6 d-1 at 20oC. Where suitable data exist, determination of a site-specific rate is preferred. Rates can be estimated by mass-balance calculations in which any change in concentration not attributable to other causes is assumed to be the result of biological processes. In order to be useful for rate estimation, a study must include ammonia concentration and temperature at each end of the study reach, and travel time through the reach. If there are tributaries, diversions, or seepage, their effect on mass balance also must be included. Seepage poses a special problem because the contribution is distributed over the entire reach. The best approach to rate estimation is to create a reach model of the same structure as the one used in CAM but with a larger number of short reaches that can be used in producing a finite approximation by iteration. Rates are always standardized to 20oC, but when applied to a specific place and time in CAM, the rate of 20oC is adjusted for ambient temperature.
75
Suitable site-specific data are available only for a few reaches in the St. Vrain watershed. The most comprehensive data set comes from the City of Boulder’s routine monitoring program. All data from 1993-2002 were included on the assumption that any changes in WWTP processes or effluent quality since 1993 would not have affected the transformations occurring in Boulder Creek. Data were available for three reaches of Boulder Creek. The first reach extends from 75th Street to a site above the New Dry Creek Carrier, a distance of about 2.5 miles (Table 41) in which the only major hydrologic influence on this is seasonal diversion by the Leggett ditch. The next reach begins at 95th Street, downstream of the New Dry Creek Carrier, and extends to 107th Street. This reach is influenced by two ditches. The final reach on Boulder Creek extends from 107th to the confluence with Coal Creek. All reaches receive seepage, which is assumed to have no measurable ammonia. Sites upstream of the WWTP are not included in the analysis because ammonia concentrations are too low to allow removal rates to be estimated with any precision. Stream
Start
End
Distance, Source mi 2.5 Boulder 2.0 Boulder
Boulder Boulder
75th Street 95th Street
Abv Dry Creek 107th Street
Boulder St. Vrain St. Vrain
107th Street Hwy 66 Abv Left Hand
Abv Coal Creek Gage at mouth Longmont gage
2.9 6.6 4.7
Boulder WQCD WQCD
Coal Coal
Louisville gage Abv Erie
Abv Rock Creek At mouth
3.5 2.4
WQCD WQCD
Hydrologic Influences Leggett Ditch Lower Boulder, Boulder Weld Left Hand, Union, Longmont WWTP Louisville WWTP Erie WWTP
Table 41. Stream reaches with data potentially suitable for calculating ammonia removal rates.
76
The Boulder data set consists of 94 sampling events. One-third of the dates were set aside by use of a random number generator for use in validation (Table 42). Not all sampling dates could be used for calibration or validation because for some the ammonia concentrations were missing or were too low to allow resolution of the rate of ammonia loss (<1 mg/L at the upstream end of the reach) to provide adequate resolution of the removal rate. In the few cases where temperature data were missing, a value was calculated based on a relationship with an adjacent station (see example in Figure 12). This approach was not taken for missing ammonia data; the set was omitted if either the upstream or downstream ammonia concentration was missing. Estimated ammonia removal rates in Boulder Creek below the WWTP expressed on a common basis (20oC) are higher than the default rate for CAM. Confidence (as shown by standard error) in the rates derived for the three reaches declines sharply with distance downstream, for reasons that are not apparent. Furthermore, the middle reach has a higher median rate than the other two reaches, which is peculiar and not attributable to known factors. In view of the relatively broad confidence intervals for rates in reaches 2 and 3 and the anomalous pattern over distance, the rate from reach 1 is assigned to all three reaches. Averaging of the medians for all three reaches would give the nearly the same result as using the value from reach 1. Date 21-Jan-93 11-Feb-93 6-Apr-93 11-May-93 8-Jun-93 7-Jul-93 10-Aug-93 8-Sep-93 13-Oct-93 3-Nov-93
Set 1 2 3 4 5 6 7 8 9 10
Reach 1 6.4 Missing 11.0 7.1 8.0 Missing Low 2.5 Low
Reach 2 21.9 30.5 5.3 17.5 6.4 Low Missing 12.4 0.9
77
Reach 3
Use 3.8
Low 0.4 Low Low Low Missing Low
Validation Validation 5.4 70.0
Date 8-Dec-93 5-Jan-94 1-Feb-94 24-Mar-94 20-Apr-94 25-May-94 16-Jun-94 21-Jul-94 30-Aug-94 26-Sep-94 26-Oct-94 16-Nov-94 13-Dec-94 14-Feb-95 14-Mar-95 11-Apr-95 8-May-95 13-Jun-95 11-Jul-95 8-Aug-95 12-Sep-95 17-Oct-95 7-Nov-95 12-Dec-95 13-Feb-96 9-Apr-96 14-May-96 11-Jun-96 20-Aug-96 24-Oct-96 11-Feb-97 8-Apr-97 13-May-97 10-Jun-97 8-Jul-97 14-Oct-97 16-Dec-97 20-Jan-98 10-Feb-98 21-Apr-98 16-Jun-98 14-Jul-98 10-Sep-98 13-Oct-98 17-Nov-98 12-Jan-99 16-Feb-99 13-Apr-99 11-May-99 22-Jun-99 13-Jul-99 10-Aug-99 14-Sep-99 12-Oct-99
Set 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
Reach 1 Missing 8.0 2.8
Reach 2 Low 12.9 4.4
Reach 3 Low Low 17.4
Use Validation Validation
5.5 16.3 10.6 Low
24.0 Low Low Low
6.6
35.0 Low Low
Low Low Low Low Low Low Low
Validation Validation Validation Validation Validation Validation
Low
5.1 Low Low Low Low 11.2 7.0
8.0 Low Low Low Low Low Low 9.2 0.0 11.9 4.9
Low 12.0 12.8 10.8 10.4 14.0 9.8 0.0 Low
Low Low Low 29.0 2.7 Low Low
12.3 11.2 9.3 6.0 17.0 9.0 8.0 11.9 4.0 Missing Temperature 34.0 31.5
33.0 27.3 10.8 16.6
Validation Validation Validation 0.1
Low 14.7 Low Low Low Low 4.2 Validation Validation Validation Validation
Low Low Low 5.8 87.0 23.1 Low Low Low Low
Low Low Low 0.0 Low Low Missing Temperature 4.3 Low Low
5.2 2.1
Low Low Low Low Low Low Low Low
55.0 10.4
78
Validation Validation 3.7 32.0
Low Low Missing Temperature 140.0 Low Low Low Low Missing
Validation
Validation Validation
Date 9-Nov-99 14-Dec-99 11-Jan-00 8-Feb-00 14-Mar-00 11-Apr-00 9-May-00 13-Jun-00 11-Jul-00 8-Aug-00 12-Sep-00 10-Oct-00 14-Nov-00 12-Dec-00 9-Jan-01 13-Feb-01 13-Mar-01 10-Apr-01 8-May-01 12-Jun-01 10-Jul-01 14-Aug-01 11-Sep-01 9-Oct-01 13-Nov-01 11-Dec-01 8-Jan-02 12-Feb-02 12-Mar-02 9-Apr-02 Median Count Std Error
Table 42.
Set 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94
Reach 1 7.8 4.5 2.0 6.0 5.2
Reach 2 36.0 19.0 38.5 26.4 9.5
Reach 3 Low 20.6 Low Missing 7.3
Use
Validation Validation 13.6 9.3 8.7
Low Low Low Low
0.6 0.0
Low Low Low Low 4.6 13.7
Validation 5.4 2.5 Validation Validation Validation
4.0 Flow Flow Flow Flow Flow Flow Flow
7.2 Flow Flow Low Low Flow Flow Flow
0.0 Flow Low Low Low Low Flow Low
Validation Validation
6.2 0.9
0.3 0.4
19.5 2.2
0.0 2.4
0.0 6.9
0.0
Validation Flow
Low 7.8 49 0.95
10.4 38 2.16
Missing Low 6.6 22 7.40
Summary of ammonia removal rate estimates (adjusted to 20oC) based on data from the City of Boulder. Estimates could not be made on all dates. Absence of ammonia data (“Missing”) at either end of a reach, concentrations below 1 mg/L (“Low”) at the upper end of a reach, or incomplete flow records (“Flow”) were grounds for omitting the rate for a reach on a particular date. If a date was selected for validation, a rate is not shown in the table; all other dates were used for calibration. Medians and standard errors for the calibration set are shown at the bottom of the table.
Median rates from the calibration sets were applied to the validation sets; a graphical comparison of observed and predicted ammonia concentrations at the end of
79
each reach is shown in Figures 13-15. Only in reach 1 was the correspondence good between observed and predicted concentrations at the bottom of the reach. The WQCD has conducted 4 synoptic studies covering the TMDL study area. Several reaches for which ammonia concentrations were high enough and hydrologic influences were represented adequately were used in checking rates derived here for Boulder Creek (Table 43). This limited examination to a few reaches downstream of the major treatment facilities. An additional reach on Coal Creek between the Louisville gage and the confluence with Rock Creek was also examined, but starting concentrations were too low to support the analysis. In general, the small sample size (N=4) and low starting concentrations yielded highly variable estimates of the removal rate (Table 43). The greatest precision was achieved with the reach of the St. Vrain from Highway 66 to the gage at the mouth. The median removal rate for that reach was essentially the same as the default rate in CAM. Median rates for the other two reaches were less than the default rate, but precision is so poor that the difference cannot be taken as significant.
Stream Coal St. Vrain St. Vrain
Start Abv Erie Abv Longmont Hwy 66
End Mouth Gage Gage
4/14/98 0.58 0 6.56
Ammonia Removal Rates, d-1 6/9/98 10/20/98 5/8/01 Median 0 12.00 0.79 0.69 0 18.30 6.62 2.86 7.44 5.95 1.99 6.25
Table 43. Estimates of ammonia removal rates for selected reaches based on data from WQCD synoptic studies.
80
s.e. 3.39 4.85 1.20
Temperature 35 30 y = 1.2064x - 3.0524 R2 = 0.9616
107th
25 20 15 10 5 0 0
5
10
15
20
25
30
95th
Figure 12.
Statistical relationship between temperatures observed in Boulder Creek at 107th Street and 95th Street. The regression equation was used to estimate temperature at 107th on dates when temperature had not been recorded there. A m m o n ia in B o u ld e r C r e e k a b o v e D r y C r e e k
16
Observed Concentration, mg/L
14 y = 1 .7 2 8 3 x - 1 .6 2 0 4 R 2 = 0 .8 4 2 6
12 10 8 6 4 2 0 0
1
2
3
4
5
6
7
8
9
P r e d ic te d C o n c e n tr a tio n , m g /L
Figure 13. Comparison of observed and predicted ammonia concentrations in Boulder Creek above the Dry Creek Carrier. The median ammonia removal rate from all calibration data sets was applied to each of the validation data sets after correction for ambient temperature. 81
A m m o n ia in B o u ld e r C r e e k a t 1 0 7 t h S t 6 .0
Observed Concentration, mg/L
5 .0
4 .0 y = 0 .7 0 1 8 x + 0 .7 9 2 6 R 2 = 0 .6 5 3 4
3 .0
2 .0
1 .0
0 .0 0 .0
1 .0
2 .0
3 .0
4 .0
5 .0
6 .0
7 .0
P r e d ic t e d C o n c e n t r a t io n , m g /L
Figure 14. Comparison of observed and predicted ammonia concentrations in Boulder Creek at 107th Street. The median ammonia removal rate from all calibration data sets was applied to each of the validation data sets after correction for ambient temperature. Am m onia in Boulde r Cr e e k a bove Coa l Cr e e k 7.0
Observed Concentration, mg/L
6.0
5.0 y = 0.9242x + 0.777 R 2 = 0.3286
4.0
3.0
2.0
1.0
0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Pr e d ict e d C o n ce n t r at io n , m g /L
Figure 15. Comparison of observed and predicted ammonia concentrations in Boulder Creek above Coal Creek. The median ammonia removal rate from all calibration data sets was applied to each of the validation data sets after correction for ambient temperature. 82
For the St. Vrain ammonia TMDL model, the default ammonia removal rate (6.0 d-1) is used for all cold-water reaches, and the rate from Boulder Creek (7.8 d-1) is used for all warm-water reaches.
Structure of the Model The TMDL model has been constructed to handle 13 discharges in the St. Vrain watershed. The model has parallel pages for acute and chronic conditions, and separate pages for the main stem and the two major tributaries (Boulder Creek, Coal Creek).
Special Calculations for Niwot and Rock Creek The model incorporates customized calculations for the two small facilities (Niwot and Rock Creek) that discharge to small tributaries (Dry Creek, Rock Creek). Each facility discharges to a small tributary in which the low-flow condition is zero in every month. Recent changes in stream classifications have resulted in the imposition of ammonia standards on Dry Creek and Rock Creek. Water quality in these streams must meet the applicable ammonia standards (0.1 mg/L for chronic and table values for acute) even though the flow may be mainly or entirely effluent. Compliance can be predicted only if a reach model approach is used for these two streams. Calculations for each of these two discharges have been customized to produce limits based on an assessment of critical-point conditions below the outfall and to predict the amount of ammonia that will be delivered at the mouth. Several assumptions have been applied to facilitate modelling of the two small tributaries: 1) no dilution flow is available, 2) stream velocity is 1 fps at all times, 3)
83
warm-water ammonia removal rate is 7.8 d-1 at 20oC, 4) temperature is constant for the length of the tributary and equal to the monthly temperature in Coal Creek above Louisville (for Rock Creek) or Left Hand Creek at the mouth (for Dry Creek), 5) effluent pH shows default rebound rates, and 6) setpoint pH is taken from Coal Creek above Lafayette (for Rock Creek) or St. Vrain above Left Hand Creek (for Dry Creek). Calculations for chronic limits start with the premise that the limiting condition for the concentration of unionized ammonia occurs where the product of the unionized fraction (a function of pH and temperature) and the fraction of total ammonia remaining (a function of travel time and the removal rate, adjusted for temperature) is at its maximum (target maximum). Locating the target maximum requires predictions of pH, temperature, and ammonia removal rate between the outfall and the mouth of the stream (0.2-mile increments were used). The target maximum is divided into the chronic standard (0.1 mg/L in this case) to make the effluent limit consistent with the stream standard. Calculations for acute limits in the two streams are more complicated because the standard is a function of pH and temperature and thus is specific to a location. It was necessary to calculate the acute standard for each 0.2-mile increment between the outfall and the mouth. The product of the unionized fraction and the fraction of total ammonia remaining (target maximum) then was divided by the acute standard to create a ratio for each increment. The maximum value of the ratio established the location of the limiting condition, at which the acute standard was calculated. The acute limit was calculated as the acute standard applicable to the location with the limiting condition divided by the target maximum at the same location.
84
Special Calculation for Orodel and St. Vrain Special calculations also were developed for two other small facilities (Orodel and St. Vrain). St. Vrain Sanitation District discharges to Oxbow Lake, which spills into St. Vrain Creek. A small amount of dilution flow also enters Oxbow Lake, and there is likely to be some processing of effluent ammonia before the flow leaves the lake and joins the St. Vrain. Unfortunately, there is no information on the amount of the dilution flow or the residence time of flow in the lake. The reduction in ammonia concentration within the lake is modeled with the dilution ratios given in the existing permit as 19:1 for chronic and 1:1 for acute. These ratios are used to calculate the concentration of ammonia that reaches the St. Vrain, but the flow is represented simply as the design capacity (i.e., no dilution flow is added). The design capacity in the existing permit is 0.5 mgd, whereas present modelling is based on 4.5 mgd, the proposed hydraulic capacity; the dilution ratio probably should be reconsidered in the future. Fourmile Creek joins Boulder Creek a short distance downstream of the gage at Orodell. One small facility (Orodel, Inc.) discharges to Fourmile Creek just upstream of its confluence with Boulder Creek. The distance is short enough to preclude measurable removal of ammonia, but there is enough flow in Fourmile Creek to dilute effluent ammonia before it reaches Boulder Creek. Low flows calculated for the gage on Fourmile Creek, which is just upstream of the outfall, define the dilution flow available in each month.
Setting Permit Limits with the Model The TMDL model has been set up to handle 13 dischargers, although more could
85
be added if necessary. As a starting point, effluent limits were set to 25 mg/L for chronic and 30 mg/L for acute in all months for all discharges. Concentrations were adjusted downward as needed to achieve compliance with the stream standards (Table 44).
CHRONIC TOTAL AMMONIA, mg/L WWTP Lyons WWTP Longmont WWTP Dry Cr (Niwot WWTP) St. Vrain SD Red Lion Fourmile (Orodel, Inc) San Lazaro Boulder WWTP B&B Louisville WWTP Rock Creek (Superior) Lafayette WWTP Erie WWTP
JAN 25.0 25.0 25.0 25.0 25.0 25.0 25.0 18.5 25.0 10.3 9.5 10.3 25.0
FEB 25.0 25.0 25.0 25.0 25.0 25.0 25.0 10.9 25.0 12.0 6.0 12.0 25.0
MAR APR 25.0 25.0 20.3 11.2 25.0 21.7 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 5.3 9.7 25.0 25.0 7.6 12.0 6.4 12.0 7.6 12.0 15.2 13.9
WWTP Lyons WWTP Longmont WWTP Dry Cr (Niwot WWTP) St. Vrain SD Red Lion Fourmile (Orodel, Inc) San Lazaro Boulder WWTP B&B Louisville WWTP Rock Creek (Superior) Lafayette WWTP Erie WWTP
JAN 30.0 30.0 29.5 30.0 30.0 30.0 30.0 30.0 30.0 28.6 8.7 * 28.6 30.0
FEB 30.0 30.0 30.0 30.0 30.0 30.0 30.0 22.3 30.0 30.0 5.8 * 30.0 30.0
MAR APR 30.0 30.0 30.0 24.8 29.9 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 15.3 28.0 30.0 30.0 20.3 25.3 8.3 18.9 20.3 25.3 24.4 25.4
MAY JUN JUL AUG SEP 25.0 25.0 25.0 25.0 25.0 6.1 9.1 10.9 10.7 6.8 17.9 19.5 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 25.0 11.3 23.5 19.5 15.1 15.1 25.0 25.0 25.0 25.0 25.0 11.7 8.8 8.5 8.0 9.2 11.3 9.9 10.3 10.6 11.0 12.5 9.9 10.3 10.6 11.0 16.4 9.9 7.4 8.9 9.3 ACUTE TOTAL AMMONIA, mg/L MAY JUN JUL 30.0 30.0 30.0 23.5 30.0 30.0 29.9 28.5 28.8 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 26.6 24.9 25.1 20.5 24.3 24.0 30.0 30.0 30.0 30.0 27.0 27.0
AUG SEP 30.0 30.0 24.2 26.2 29.0 29.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 24.9 27.1 26.2 22.4 30.0 30.0 30.0 30.0
OCT 25.0 10.6 17.2 25.0 25.0 25.0 25.0 15.1 25.0 11.0 10.5 10.5 12.8
NOV 25.0 14.9 25.0 25.0 25.0 25.0 25.0 21.6 25.0 7.5 7.5 7.5 13.5
DEC 25.0 23.0 25.0 25.0 25.0 25.0 25.0 24.8 25.0 9.0 9.0 9.0 21.7
OCT 30.0 30.0 30.0 30.0 30.0 30.0 30.0 28.3 30.0 26.9 19.2 30.0 30.0
NOV 30.0 27.3 30.0 30.0 30.0 30.0 30.0 29.5 30.0 24.5 11.1 24.5 30.0
DEC 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 30.0 26.3 9.3 26.3 30.0
*Acute is less than chronic; set acute to chronic. Table 44. Effluent limits consistent with standards for unionized ammonia in the St. Vrain basin.
The starting point for all dischargers in the assessment is a fixed concentration close to what would be expected with secondary treatment alone. Thus, chronic limits of 25 mg/L and acute limits of 30 mg/L were applied initially for all months at all facilities. Minor permits were not considered further unless there were specific concerns about compliance below the outfall (e.g., Rock Creek, Niwot, and San Lazaro).
86
0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00
Total Ammonia, mg/L
10.0 8.0 6.0 4.0 2.0 0.0 0
3
6
9
12
Unionized Ammonia, mg/L
Coal Creek Chronic for January
15
Distance, mi Total
Unionized
Chronic Std
0.07
16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0
0.06 0.05 0.04 0.03 0.02 0.01
Unionized Ammonia, mg/L
Total Ammonia, mg/L
Boulder Creek Chronic for January
0.00 0
3
6
9
12
15
18
21
24
27
Distance, mi Total
Unionized
Chronic Std
16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0
0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 0
5
10
15
20
25
30
Unionized Ammonia, mg/L
Total Ammonia, mg/L
St Vrain Chronic for January
35
Distance, mi Total
Unionized
Chronic Std
Figure 16. Model output for January (chronic conditions) showing longitudinal changes in ammonia.
87
As a second step, limits were lowered uniformly for all major discharges within each of three modelling reaches: Coal Creek, Boulder Creek, St. Vrain main stem. The parity assumption then was relaxed for downstream dischargers, beginning with the last in a given reach, until the assimilative capacity was exhausted. The result for multiple dischargers along a reach would be that each successive discharger downstream in the reach has limits equal to or greater than those of the closest one upstream. Departure from the incremental procedure described above is allowed if it can be shown that an upstream discharger is isolated from (i.e., has no detectable effect on) the next discharger downstream. Isolation is most likely in the summer months when the temperature-adjusted ammonia removal rate is high. Isolation could also occur through replacement of the water mass (e.g., where water with high ammonia concentration is diverted from the stream and replaced by seepage flows containing no ammonia). The procedure was applied first to Coal Creek. Chronic limits for three of the four dischargers (Louisville, Lafayette, and Erie) were set initially to 25 mg/L in all months. Rock Creek was set as described previously, but was constrained not to exceed the limit set for Lafayette, the next discharger downstream. Limits were reduced equally for the three major dischargers until compliance was predicted at all points along Coal Creek. Conditions in each month were examined for opportunities to relax the parity requirement. In most months, it was possible to increase the limits for Erie to take advantage of increases in assimilative capacity downstream of Lafayette (Figure 16). In a few months, limits for Lafayette exerted no effect on Erie and thus could be set higher than those for Erie; these were the warm summer months (July-September), when the
88
model predicts that virtually all ammonia will disappear from Coal Creek between Lafayette and Erie because of nitrification. The procedure was repeated for acute limits. The procedure continues with Boulder Creek, which has only one major discharger, the City of Boulder WWTP. San Lazaro was adjusted as necessary to achieve compliance with the stream standard (Figure 16). The discharge point for San Lazaro is located at the bottom of Segment 2 of Boulder Creek, which has a cold-water classification. Because the outfall is so close to the top of Segment 9, however, compliance in the TMDL model is judged with respect to the warm-water standards that apply in Segment 9. The limit for the Boulder WWTP then was adjusted without regard to the effect of San Lazaro, which is a very small discharge. For months when the critical point (i.e., the location of maximum concentration of unionized ammonia) below the Boulder outfall is downstream of the Coal Creek confluence, one or more of the Coal Creek dischargers must be considered for adjustment to achieve parity with Boulder. Setting limits in the TMDL model concludes with the St Vrain (Figure 16). Niwot is set according to the procedure described above, and the other dischargers, except for Longmont, are too small to have a noticeable effect on ammonia in the main stem. Limits are set for Longmont, and the cumulative effect of Boulder Creek and Coal Creek dischargers is reviewed for months when the critical point for Longmont falls downstream of the confluence with Boulder Creek. The procedure is tedious because of the iterations required to reach a final endpoint. Also, there may be some opportunities for trading that would alter the basis for decisions about limits. Trades are possible where one discharger constrains the limits of
89
another discharger at parity, if parity were voluntarily waived. There are some opportunities for trading on Coal Creek, but probably not elsewhere in the basin. Application of the decision rules to the TMDL model yields a set of permit limits that are shown in the model and in Table 44. No trades have been proposed at this time. Estimates of the load (mass) of ammonia delivered to and carried by the main stem reaches of the St Vrain basin are an essential part of the TMDL analysis. A page has been added to the model on which flow, concentration, and transport are shown for selected nodes (see Tables 45, 46). Nodes include headwater contributions, tributaries, WWTPs, reservoir releases, the main stem above each water source to the main stem, and the mouth of the stream. The tables must be revised if modelling conditions are altered.
Future Considerations
It is not unusual in a system such as the St. Vrain, where water is highly managed, for the operation of the river to be changed through exchanges, changes in use, or relocation of structures. Changes in the operation of the river can influence water quality in ways that can be incorporated into the TMDL model, provided that enough information is available to characterize future conditions. One example in the St. Vrain basin is a proposal to change the physical locations of a diversion and an outfall. The cities of Lafayette and Boulder have reached an agreement by which Lafayette would relocate its diversion on Boulder Creek to a point just east of 75th Street and Boulder would move its WWTP outfall about 100 feet east of Lafayette’s intake. Lafayette would not reduce flows in Boulder Creek below threshold
90
values, which would be set seasonally. The schedule for completion of the project is not known at this time, however. Approvals must be obtained for various steps in the process, and a construction timetable must be created. In view of uncertainties about the schedule for and operation of this project, it is premature to include it in the TMDL model.
91
Chronic Flow, cfs Site St. Vrain at Lyons Lyons WWTP St. Vrain above Supply Canal
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
16.0
16.0
16.0
21.0
28.0
118.0
65.0
32.0
21.0
20.0
18.0
16.0
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
17.2
17.2
17.0
22.2
29.9
119.7
68.3
35.2
23.9
21.7
19.2
17.2
Supply Canal
0.0
0.0
0.0
0.0
2.0
23.0
115.0
24.0
6.0
0.0
0.0
0.0
St. Vrain above Foothill Release
4.9
5.5
3.3
5.6
9.6
3.7
21.8
19.0
18.0
10.2
7.2
5.0
2.4
2.4
2.6
2.6
3.6
25.7
31.0
8.0
0.0
0.0
0.1
2.2
14.2
14.6
4.1
3.2
20.9
2.1
40.6
39.7
46.5
18.6
8.4
14.9
Foothill Release St. Vrain above Left Hand Left Hand Creek
-0.2
-0.6
10.2
10.1
-7.9
33.8
1.2
-4.8
-11.6
0.3
4.8
-1.6
St. Vrain above Longmont WWTP
15.0
15.0
15.0
14.2
14.2
37.0
43.0
36.0
36.0
20.0
14.2
14.2
Longmont WWTP
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
St. Vrain above Dry Creek
39.9
39.9
38.9
38.9
39.7
62.5
68.5
61.5
61.5
45.5
39.3
38.9
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
42.3
42.2
41.1
41.2
42.1
64.9
70.9
63.9
63.9
47.9
41.7
41.2
Dry Creek St. Vrain above Union release Union Reservoir release St. Vrain above Boulder Creek Boulder Creek St. Vrain above St. Vrain SD Oxbow Lake Release St. Vrain above Howlett Gulch Howlett Gulch Mouth of St. Vrain Boulder Creek at Orodell
0.0
0.0
0.0
0.0
0.0
0.0
35.0
11.0
0.0
0.0
0.0
0.0
45.8
45.7
43.6
44.6
46.8
70.1
111.1
80.4
68.9
52.0
45.4
44.7
71.8
72.7
67.1
2.0
7.0
33.7
15.3
21.1
9.8
10.6
36.8
66.0
130.7
130.1
117.0
48.4
71.1
98.5
122.8
121.1
82.3
43.3
95.4
125.4
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
146.8
145.4
132.0
64.9
98.6
134.6
158.6
163.5
114.7
58.3
111.5
142.7
3.0
3.0
3.5
3.5
3.5
1.0
6.0
1.0
0.0
0.0
1.0
1.0
165.1
161.3
148.8
84.3
136.2
184.0
207.8
219.6
157.1
71.7
127.7
160.9
10.5
10.5
10.5
16.0
21.0
67.0
43.0
31.0
17.0
12.0
12.0
10.5
Red Lion
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Four Mile Creek
0.5
0.5
0.6
1.1
2.6
0.6
0.1
0.1
0.1
0.1
0.6
0.6
Boulder Creek above San Lazaro
4.8
2.3
3.3
9.6
11.5
1.8
14.2
1.3
5.5
3.8
3.4
5.5
San Lazaro
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
South Boulder Creek
0.0
0.2
0.2
0.2
0.0
16.0
3.5
0.2
0.2
0.2
0.0
1.1
Boulder Creek above Boulder Supply
7.3
4.6
6.1
13.7
16.5
11.8
13.1
2.0
7.0
5.6
4.1
8.8
Boulder Supply Canal
1.9
4.7
2.6
-6.5
-9.3
27.7
36.4
19.8
-0.2
1.0
2.7
-0.9
Boulder Creek above Boulder WWTP
9.4
9.4
9.0
7.7
7.8
40.0
50.0
22.0
7.2
7.0
7.0
8.1
Boulder WWTP
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
Boulder Creek above New Dry Carrier
49.2
49.3
48.1
46.4
50.3
70.3
58.1
58.7
39.3
41.5
41.3
38.2
0.0
0.0
0.0
0.0
0.0
0.0
2.0
1.0
0.0
0.0
0.0
0.0
Boulder Creek above Coal Creek
51.1
51.6
48.6
17.1
9.1
27.5
9.3
34.2
18.5
23.1
30.5
42.1
Coal Creek
18.3
18.3
18.3
9.4
5.3
4.8
2.9
3.7
3.9
8.8
18.7
18.3
B&B MHP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
New Dry Carrier
Panama Reservoir release Boulder Creek at mouth
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
71.8
72.7
67.1
2.0
7.0
33.7
15.3
21.1
9.8
10.6
36.8
66.0
Coal Creek above Louisville
1.0
0.8
0.8
1.1
1.5
1.1
0.6
0.6
0.6
1.5
1.4
1.0
Louisville WWTP
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
Coal Creek above Rock Creek
6.3
6.3
6.3
6.4
6.8
6.4
6.3
6.3
6.3
6.8
6.7
6.3
Rock Creek
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
Coal Creek above Lafayette WWTP
9.7
9.7
9.7
9.8
10.2
9.8
9.7
9.7
9.7
10.2
10.1
9.7
Lafayette WWTP
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
Coal Creek above Erie WWTP
16.5
16.5
16.5
7.6
3.5
3.0
1.0
1.9
2.1
7.0
16.9
16.5
Erie WWTP Coal Creek at mouth
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
18.3
18.3
18.3
9.4
5.3
4.8
2.9
3.7
3.9
8.8
18.7
18.3
Table 45. Continued on next page. 92
Chronic Concentration, mg/L Site St. Vrain at Lyons
Jan 0.08
Feb 0.08
Mar 0.02
Apr 0.02
May 0.01
Jun 0.01
Jul 0.08
Aug 0.02
Sep 0.01
Oct 0.02
Nov 0.08
Dec 0.11
Lyons WWTP
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
St. Vrain above Supply Canal
0.9
0.9
0.8
0.6
0.5
0.1
0.9
0.4
0.6
0.6
0.8
0.9
Supply Canal
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Foothill Release
0.1
0.1
0.0
0.0
0.0
0.0
0.1
0.0
0.0
0.0
0.0
0.1
Foothill Release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Left Hand
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Left Hand Creek
0.0
0.0
0.2
0.2
0.0
0.2
0.0
0.0
0.0
0.3
0.2
0.0
St. Vrain above Longmont WWTP
0.0
0.0
0.1
0.1
0.0
0.2
0.0
0.0
0.0
0.0
0.1
0.0
Longmont WWTP
25.0
25.0
20.3
11.2
6.1
9.1
25.0
10.7
6.8
10.6
14.9
23.0
St. Vrain above Dry Creek
10.9
10.8
8.5
4.5
2.1
2.1
10.9
2.6
1.6
3.6
6.3
10.1 9.4
9.9
9.3
6.5
4.1
2.1
1.6
9.9
1.1
1.1
2.5
7.2
10.4
10.3
8.0
4.3
1.9
1.9
10.4
2.4
1.4
3.3
6.0
9.6
Union Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Boulder Creek
7.9
7.7
5.8
2.9
1.0
1.2
7.9
1.3
0.9
2.3
4.3
7.2
Boulder Creek
1.9
1.3
0.4
0.0
0.0
0.0
1.9
0.0
0.0
0.0
0.4
1.5
St. Vrain above St. Vrain SD
2.6
2.3
1.2
1.0
0.2
0.1
2.6
0.2
0.1
0.5
1.1
2.2
Oxbow Lake Release
3.8
3.8
1.4
2.0
0.3
0.1
3.8
0.1
0.2
0.6
1.3
3.3
St. Vrain above Howlett Gulch
1.8
1.7
0.7
0.6
0.1
0.0
1.8
0.1
0.0
0.2
0.6
1.5
Howlett Gulch
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Mouth of St. Vrain
1.0
0.9
0.3
0.2
0.0
0.0
1.0
0.0
0.0
0.1
0.2
0.8
Dry Creek St. Vrain above Union release
Boulder Creek at Orodell Red Lion Four Mile Creek
0.1
0.1
0.1
0.1
0.0
0.0
0.1
0.1
0.0
0.1
0.0
0.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
0.3
0.3
0.3
0.2
0.1
0.3
0.3
1.6
1.6
1.6
0.3
0.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
South Boulder Creek
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek above Boulder Supply
0.4
0.5
0.3
0.2
0.1
0.0
0.4
0.0
0.1
0.2
0.3
0.3
Boulder Supply Canal
0.0
0.0
0.0
0.2
0.1
0.0
0.0
0.0
0.1
0.0
0.0
0.3
Boulder Creek above Boulder WWTP
0.3
0.2
0.2
0.1
0.1
0.0
0.3
0.0
0.1
0.1
0.1
0.3
18.5
10.9
5.3
9.7
11.3
23.5
18.5
15.1
15.1
15.1
21.6
24.8
Boulder Creek above New Dry Carrier
8.7
5.1
2.3
4.2
4.2
4.7
8.7
2.9
3.8
4.6
7.8
10.7
New Dry Carrier
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek above Coal Creek
3.7
2.1
0.8
0.8
0.3
0.1
3.7
0.1
0.2
0.5
1.4
3.5
Coal Creek
3.2
3.3
2.0
2.2
2.3
1.0
3.2
0.8
0.9
1.4
1.5
2.6
B&B MHP
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
25.0
Panama Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek at mouth
1.9
1.3
0.4
0.0
0.0
0.0
1.9
0.0
0.0
0.0
0.4
1.5
Coal Creek above Louisville
0.0
0.1
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
10.3
12.0
7.6
12.0
11.7
8.8
10.3
8.0
9.2
11.0
7.5
9.0
Coal Creek above Rock Creek
2.8
3.2
1.9
2.6
2.1
1.2
2.8
0.8
0.7
1.3
1.5
2.2
Rock Creek
3.9
2.3
1.8
2.5
1.5
0.9
3.9
0.5
0.6
1.7
2.3
3.6
Coal Creek above Lafayette WWTP
1.8
1.6
1.0
1.3
0.9
0.5
1.8
0.2
0.2
0.6
0.9
1.5
10.3
12.0
7.6
12.0
12.5
9.9
10.3
10.6
11.0
10.5
7.5
9.0
2.0
2.2
1.3
1.3
0.5
0.1
2.0
0.0
0.0
0.5
1.0
1.6
25.0
25.0
15.2
13.9
16.4
9.9
25.0
8.9
9.3
12.8
13.5
21.7
3.2
3.3
2.0
2.2
2.3
1.0
3.2
0.8
0.9
1.4
1.5
2.6
Boulder Creek above San Lazaro San Lazaro
Boulder WWTP
Louisville WWTP
Lafayette WWTP Coal Creek above Erie WWTP Erie WWTP Coal Creek at mouth
Table 45. Continued on next page.
93
Chronic Load, kg/d Site
Jan 3
Feb 3
Mar 1
Apr 1
May 1
Jun 3
Jul 2
Aug 2
Sep 1
Oct 1
Nov 4
Dec 4
Lyons WWTP
35
35
35
35
35
35
35
35
35
35
35
35
St. Vrain above Supply Canal
37
36
34
34
34
37
35
33
32
34
37
38
Supply Canal
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Foothill Release
1
1
0
0
1
0
0
1
1
1
1
1
Foothill Release
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Left Hand
0
0
0
0
0
0
0
0
0
0
0
0
Left Hand Creek
0
0
5
5
0
17
1
0
0
0
2
0
St. Vrain above Longmont WWTP
0
0
5
5
0
15
1
0
0
0
2
0
Longmont WWTP
1325
1325
1076
594
323
482
578
567
360
562
790
1219
St. Vrain above Dry Creek
1068
1056
811
432
203
323
377
386
235
401
604
965
44
42
29
18
9
7
3
5
5
11
32
42
1077
1060
805
429
196
308
356
369
225
393
611
973
St. Vrain at Lyons
Dry Creek St. Vrain above Union release
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Boulder Creek
887
864
615
316
116
199
247
264
150
290
479
788
Boulder Creek
338
227
69
0
0
0
0
0
0
1
34
244
St. Vrain above St. Vrain SD
819
727
346
113
28
31
45
64
28
58
261
665
65
65
24
34
5
1
0
1
3
11
22
56
657
588
231
91
15
12
14
25
13
33
172
522
0
0
0
0
0
0
0
0
0
0
0
0
404
359
107
42
4
3
2
5
3
10
76
308
Boulder Creek at Orodell
2
3
1
4
1
7
0
5
1
3
0
0
Red Lion
1
1
1
1
1
1
1
1
1
1
1
1
Four Mile Creek
0
0
0
0
0
0
0
0
0
0
0
0
Union Reservoir release
Oxbow Lake Release St. Vrain above Howlett Gulch Howlett Gulch Mouth of St. Vrain
0
0
0
0
0
0
0
0
0
0
0
0
10
10
10
10
10
10
10
10
10
10
10
10
South Boulder Creek
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above Boulder Supply
6
5
5
5
5
1
1
0
2
2
3
7
Boulder Supply Canal
0
0
0
-2
-3
0
0
0
0
0
0
-1
Boulder Creek above Boulder WWTP
6
5
5
3
2
1
1
0
2
2
2
6
Boulder WWTP
1751
1031
502
918
1069
2224
1845
1429
1429
1429
2044
2347
Boulder Creek above New Dry Carrier
1051
609
270
471
519
814
461
413
363
467
791
1006
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above Coal Creek
469
265
91
34
6
6
0
9
7
31
104
364
Coal Creek
143
149
88
51
30
12
4
7
9
30
71
117
B&B MHP
1
1
1
1
1
1
1
1
1
1
1
1
Panama Reservoir release
0
0
0
0
0
0
0
0
0
0
0
0
338
227
69
0
0
0
0
0
0
1
34
244
0
0
0
0
0
0
0
0
0
0
0
0
133
154
98
154
151
113
109
103
118
142
97
116
Coal Creek above Rock Creek
42
49
29
40
35
19
14
12
11
22
24
34
Rock Creek
33
19
15
20
12
7
3
4
5
14
19
30
Boulder Creek above San Lazaro San Lazaro
New Dry Carrier
Boulder Creek at mouth Coal Creek above Louisville Louisville WWTP
Coal Creek above Lafayette WWTP Lafayette WWTP Coal Creek above Erie WWTP Erie WWTP Coal Creek at mouth
43
39
24
32
23
11
6
5
5
14
22
35
172
200
127
200
208
165
172
177
183
175
125
150
81
89
53
23
4
1
0
0
0
8
43
64
114
114
69
63
74
45
34
40
42
58
61
99
143
149
88
51
30
12
4
7
9
30
71
117
Table 45. Chronic low flows and total ammonia concentrations and loads derived from the TMDL analysis. 94
Acute Flow, cfs Site St. Vrain at Lyons
Jan 11.7
Feb 14.0
Mar 11.7
Apr 15.0
May 36.0
Jun 130.0
Jul 84.0
Aug 43.0
Sep 20.0
Oct 13.0
Nov 14.0
Dec 13.0
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.6
12.9
15.2
12.7
16.2
37.9
131.7
87.3
46.2
22.9
14.7
15.2
14.2
Supply Canal
0.0
0.0
0.0
0.0
17.0
44.0
90.0
15.0
0.0
0.0
2.0
0.0
St. Vrain above Foothill Release
3.2
3.2
2.4
2.4
4.6
3.7
7.5
7.3
6.5
5.0
5.0
3.4
Foothill Release
0.0
0.6
0.0
0.6
1.6
16.2
25.3
12.6
0.0
0.0
0.0
0.0
12.5
12.3
4.1
3.2
4.6
2.1
9.1
16.1
19.2
12.7
6.9
13.0
Lyons WWTP St. Vrain above Supply Canal
St. Vrain above Left Hand
0.5
0.8
10.2
5.6
3.4
31.8
21.8
-7.3
-1.3
10.2
2.1
-4.8
St. Vrain above Longmont WWTP
14.0
14.0
15.0
9.7
9.1
35.0
32.0
9.9
19.0
24.0
10.0
9.1
Longmont WWTP
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
21.7
St. Vrain above Dry Creek
38.9
38.9
38.9
34.4
34.6
60.5
57.5
35.4
44.5
49.5
35.1
33.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
1.8
41.3
41.2
41.1
36.7
37.0
62.9
59.9
37.8
46.9
51.9
37.5
36.1
Left Hand Creek
Dry Creek St. Vrain above Union release Union Reservoir release St. Vrain above Boulder Creek Boulder Creek St. Vrain above St. Vrain SD
0.0
0.0
0.0
0.0
0.0
0.0
15.0
34.0
11.0
0.0
0.0
0.0
44.8
44.7
43.6
40.1
41.7
68.1
80.1
77.3
62.9
56.0
41.2
39.6
65.9
68.1
63.4
2.0
7.0
32.2
15.3
19.1
9.8
7.2
23.6
61.5
123.8
124.5
113.3
43.9
51.2
92.9
91.8
104.1
76.3
43.9
78.0
115.8
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
139.9
139.8
128.3
60.4
78.7
128.9
127.6
146.5
108.7
58.8
94.1
133.1
3.0
4.0
3.5
3.5
3.5
2.0
6.0
8.0
0.0
0.0
1.0
1.0
158.1
156.7
145.1
79.8
116.2
176.4
176.8
209.6
151.1
72.2
110.3
151.3
Boulder Creek at Orodell
3.1
3.9
4.9
6.3
18.0
97.0
48.0
33.0
17.0
12.0
3.1
7.9
Red Lion
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Four Mile Creek
0.5
0.5
0.9
1.1
5.0
2.8
0.1
0.1
0.1
0.3
0.5
0.6
Boulder Creek above San Lazaro
3.1
2.0
2.6
4.9
9.5
1.8
4.3
1.3
1.6
2.4
3.4
3.2
San Lazaro
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
South Boulder Creek
0.1
0.0
0.0
0.0
1.0
2.4
4.1
0.6
0.1
0.0
0.0
0.1
Boulder Creek above Boulder Supply
5.6
4.1
5.3
8.8
15.5
4.2
4.2
2.0
3.2
4.0
4.0
5.5
-2.4
0.5
-0.2
-7.0
-7.0
29.3
30.2
12.8
1.5
-2.2
-1.8
-2.3
3.5
4.8
5.3
2.2
9.1
34.0
35.0
15.0
5.1
2.2
2.4
3.4
Boulder WWTP
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
38.7
Boulder Creek above New Dry Carrier
43.3
44.7
44.4
40.9
31.6
53.3
43.1
38.7
35.0
19.8
28.5
33.5
0.0
0.0
0.0
0.0
0.0
0.0
7.0
3.0
0.0
0.0
0.0
0.0
Boulder Creek above Coal Creek
45.2
47.0
44.9
5.3
1.1
18.4
6.6
13.0
4.7
1.4
17.7
37.4
Coal Creek
18.3
18.3
18.3
9.7
4.8
2.9
2.9
3.7
4.1
8.3
18.3
18.5
B&B MHP
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Panama Reservoir release
0.0
0.0
0.0
0.0
0.0
11.3
0.0
0.0
0.0
0.0
0.0
0.0
65.9
68.1
63.4
2.0
7.0
32.2
15.3
19.1
9.8
7.2
23.6
61.5
Coal Creek above Louisville
0.9
0.7
0.5
1.4
0.9
0.5
0.7
0.3
1.2
1.0
1.0
1.2
Louisville WWTP
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
5.3
Coal Creek above Rock Creek
6.3
6.3
6.3
6.7
6.3
6.3
6.3
6.3
6.5
6.3
6.3
6.5
Rock Creek
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
3.4
Coal Creek above Lafayette WWTP
9.7
9.7
9.7
10.1
9.7
9.7
9.7
9.7
9.9
9.7
9.7
9.9
Lafayette WWTP
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
6.8
Coal Creek above Erie WWTP
16.5
16.5
16.5
7.9
3.0
1.0
1.0
1.9
2.3
6.5
16.5
16.7
Erie WWTP Coal Creek at mouth
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
1.9
18.3
18.3
18.3
9.7
4.8
2.9
2.9
3.7
4.1
8.3
18.3
18.5
Oxbow Lake Release St. Vrain above Howlett Gulch Howlett Gulch Mouth of St. Vrain
Boulder Supply Canal Boulder Creek above Boulder WWTP
New Dry Carrier
Boulder Creek at mouth
Table 46. Continued on next page. 95
Acute Concentration, mg/L Site St. Vrain at Lyons
Jan 0.08
Feb 0.08
Mar 0.02
Apr 0.02
May 0.01
Jun 0.01
Jul 0.01
Aug 0.02
Sep 0.01
Oct 0.02
Nov 0.08
Dec 0.11
Lyons WWTP
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
St. Vrain above Supply Canal
1.3
1.1
1.3
1.0
0.4
0.1
0.2
0.4
0.7
1.1
1.1
1.2
Supply Canal
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Foothill Release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Foothill Release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Left Hand
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Left Hand Creek
0.2
0.2
0.2
0.2
0.2
0.2
0.3
0.0
0.0
0.3
0.2
0.0
St. Vrain above Longmont WWTP
0.0
0.0
0.1
0.1
0.1
0.2
0.2
0.0
0.0
0.1
0.0
0.0
Longmont WWTP
30.0
30.0
30.0
24.8
23.5
30.0
30.0
24.2
26.2
30.0
27.3
30.0
St. Vrain above Dry Creek
13.4
13.3
12.6
11.1
9.0
7.0
6.8
8.5
7.8
9.5
12.6
14.8
Dry Creek
11.7
11.2
7.8
5.7
3.5
2.3
0.9
1.3
1.3
4.4
8.6
11.3
St. Vrain above Union release
12.7
12.6
11.7
10.1
7.9
6.3
6.1
7.4
6.9
8.7
11.7
13.9
Union Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
St. Vrain above Boulder Creek
9.6
9.4
8.4
6.8
4.1
3.8
2.9
2.4
3.4
6.0
8.2
10.0
Boulder Creek
2.9
2.3
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.5
1.8
St. Vrain above St. Vrain SD
3.3
3.1
1.9
2.1
0.5
0.4
0.3
0.3
0.5
1.5
2.2
2.8
Oxbow Lake Release
4.6
4.6
1.7
2.4
0.3
0.1
0.0
0.1
0.2
0.7
1.6
4.0
St. Vrain above Howlett Gulch
2.4
2.2
1.1
1.1
0.1
0.1
0.1
0.1
0.1
0.6
1.1
1.9
Howlett Gulch
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Mouth of St. Vrain
1.3
1.2
0.5
0.4
0.0
0.0
0.0
0.0
0.0
0.2
0.4
1.0
Boulder Creek at Orodell Red Lion Four Mile Creek
0.1
0.1
0.1
0.1
0.0
0.0
0.0
0.1
0.0
0.1
0.0
0.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
0.4
0.4
0.2
0.2
0.0
0.1
2.0
2.0
2.0
0.7
0.4
0.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
South Boulder Creek
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek above Boulder Supply
0.5
0.6
0.4
0.2
0.2
0.0
0.0
0.0
0.1
0.2
0.3
0.5
Boulder Supply Canal
0.5
0.0
0.4
0.2
0.2
0.0
0.0
0.0
0.0
0.2
0.3
0.5
Boulder Creek above Boulder WWTP
0.4
0.5
0.4
0.2
0.1
0.0
0.0
0.0
0.0
0.1
0.3
0.5
Boulder WWTP
30.0
22.3
15.3
28.0
30.0
30.0
30.0
30.0
30.0
28.3
29.5
30.0
Boulder Creek above New Dry Carrier
14.9
10.7
6.7
12.4
10.4
6.0
5.3
5.5
7.3
7.4
10.2
13.0
New Dry Carrier
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek above Coal Creek
5.6
4.1
2.1
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.8
3.6
Coal Creek
5.7
5.8
4.0
4.2
4.3
2.8
1.9
2.6
3.0
3.4
4.0
5.0
B&B MHP
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
30.0
Panama Reservoir release
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Boulder Creek at mouth
2.9
2.3
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.5
1.8
Coal Creek above Louisville
0.0
0.1
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
28.6
30.0
20.3
25.3
26.6
24.9
25.1
24.9
27.1
26.9
24.5
26.3
Coal Creek above Rock Creek
7.7
8.0
5.1
5.5
4.8
4.0
2.6
2.5
2.0
3.2
4.8
6.6
Rock Creek
3.6
2.3
2.3
3.9
2.7
2.2
0.9
1.3
1.1
3.1
3.4
3.7
Coal Creek above Lafayette WWTP
3.6
3.4
2.3
2.7
1.9
1.4
0.7
0.7
0.5
1.2
2.1
3.1
28.6
30.0
20.3
25.3
30.0
30.0
30.0
30.0
30.0
30.0
24.5
26.3
5.2
5.4
3.4
2.7
0.9
0.0
0.0
0.1
0.1
1.2
3.2
4.4
30.0
30.0
24.4
25.4
30.0
27.0
27.0
30.0
30.0
30.0
30.0
30.0
5.7
5.8
4.0
4.2
4.3
2.8
1.9
2.6
3.0
3.4
4.0
5.0
Boulder Creek above San Lazaro San Lazaro
Louisville WWTP
Lafayette WWTP Coal Creek above Erie WWTP Erie WWTP Coal Creek at mouth
Table 46. Continued on next page. 96
Acute Load, kg/d Site
Jan 2
Feb 3
Mar 1
Apr 1
May 1
Jun 3
Jul 2
Aug 2
Sep 0
Oct 1
Nov 3
Dec 3
Lyons WWTP
43
43
43
43
43
43
43
43
43
43
43
43
St. Vrain above Supply Canal
42
43
40
40
41
44
42
40
39
39
42
43
Supply Canal
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Foothill Release
0
0
0
0
0
0
0
0
0
0
0
0
Foothill Release
0
0
0
0
0
0
0
0
0
0
0
0
St. Vrain above Left Hand
0
0
0
0
0
0
0
0
0
0
0
0
Left Hand Creek
0
0
5
3
2
16
18
0
0
8
1
0
St. Vrain above Longmont WWTP
0
0
5
2
1
14
16
0
0
8
1
0
Longmont WWTP
1590
1590
1590
1314
1245
1590
1590
1282
1388
1590
1447
1590
St. Vrain above Dry Creek
1277
1261
1197
933
761
1037
960
736
848
1145
1080
1223
52
50
35
25
16
10
4
6
6
20
39
51
1286
1267
1180
911
715
977
892
683
793
1108
1071
1225
St. Vrain at Lyons
Dry Creek St. Vrain above Union release Union Reservoir release St. Vrain above Boulder Creek
0
0
0
0
0
0
0
0
0
0
0
0
1055
1028
902
663
416
628
574
448
518
819
824
968
466
386
153
0
0
0
0
0
0
0
26
267
1012
939
530
225
60
91
74
73
90
164
415
788
78
78
28
41
6
1
0
1
3
13
26
67
807
753
348
163
28
35
21
28
39
86
263
616
0
0
0
0
0
0
0
0
0
0
0
0
494
457
161
75
8
7
3
6
10
27
113
360
Boulder Creek at Orodell
1
1
1
2
0
9
0
6
1
3
0
0
Red Lion
1
1
1
1
1
1
1
1
1
1
1
1
Four Mile Creek
1
1
1
1
1
1
0
0
0
0
1
1
Boulder Creek St. Vrain above St. Vrain SD Oxbow Lake Release St. Vrain above Howlett Gulch Howlett Gulch Mouth of St. Vrain
0
0
0
0
0
0
0
0
0
0
0
0
12
12
12
12
12
12
12
12
12
12
12
12
South Boulder Creek
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above Boulder Supply
7
6
5
5
6
0
0
0
0
2
3
7
-3
0
0
-4
-3
0
0
0
0
-1
-1
-3
4
6
5
1
3
0
0
0
0
1
1
4
Boulder WWTP
2839
2110
1448
2650
2839
2839
2839
2839
2839
2678
2792
2839
Boulder Creek above New Dry Carrier
1574
1175
732
1243
805
781
559
519
627
359
710
1068
0
0
0
0
0
0
0
0
0
0
0
0
Boulder Creek above Coal Creek
624
469
226
11
0
0
0
1
0
0
36
326
Coal Creek
256
258
179
100
51
20
13
24
30
69
181
228
B&B MHP
2
2
2
2
2
2
2
2
2
2
2
2
Panama Reservoir release
0
0
0
0
0
0
0
0
0
0
0
0
466
386
153
0
0
0
0
0
0
0
26
267
0
0
0
0
0
0
0
0
0
0
0
0
Louisville WWTP
368
386
261
326
342
320
323
320
349
346
315
338
Coal Creek above Rock Creek
118
123
78
89
73
62
40
38
32
49
74
104
Rock Creek
30
19
19
32
23
18
7
11
9
26
28
31
Coal Creek above Lafayette WWTP
85
81
53
66
46
34
16
16
12
27
50
74
Lafayette WWTP
476
500
338
421
500
500
500
500
500
500
408
438
Coal Creek above Erie WWTP
211
216
138
52
7
0
0
0
1
19
131
180
Erie WWTP Coal Creek at mouth
136
136
111
115
136
123
123
136
136
136
136
136
256
258
179
100
51
20
13
24
30
69
181
228
Boulder Creek above San Lazaro San Lazaro
Boulder Supply Canal Boulder Creek above Boulder WWTP
New Dry Carrier
Boulder Creek at mouth Coal Creek above Louisville
Table 46. Acute low flows and total ammonia concentrations and loads derived from the TMDL analysis. 97
Appendix I
A Study of Statistical Relationships between Flow, Percent Unionized Ammonia, and pH in the Drainage of the St. Vrain Creek, Including Boulder Creek and Coal Creek
98
In preparing permits for discharge of ammonia, the State of Colorado uses the Colorado Ammonia Model (CAM). CAM software incorporates the assumption that the extreme conditions of percent unionized ammonia (as determined mainly by pH but also to some extent by temperature) coincide with biologically-based low flows for acute and chronic conditions. At some locations this assumption may be incorrect. The assumption is conservative in that it leads to requirements for dischargers that are stricter than would be the case if there were no statistical relationship between low flow and extremes of percent unionized ammonia or pH. Where sufficient evidence is available, the validity of the assumption can be checked. Because extensive monitoring information is available on the St. Vrain drainage, a check was made of the assumption for key locations relevant to TMDL modelling for ammonia in the St. Vrain basin, including Boulder Creek and Coal Creek. The procedure for checking relationships between flow and percent unionized ammonia or pH involved first adjusting, by use of the recurrence module of CAM, the instantaneous pH and percent unionized ammonia as observed at the time of sampling to a daily mean pH or daily mean percent unionized ammonia. This was accomplished by use of the recurrence module of CAM. For a given site, pH and percent unionized ammonia then were plotted against flow, and a test was made for a linear relationship between the two. The results are shown in the attached figures. The lines on the figures appear curvilinear because the scale for flow is shown logarithmically as a means of accommodating the wide variation in flows. As shown by the graphs, there is no indication of any meaningful statistical relationship between flow and either percent unionized ammonia or pH. Furthermore, the
99
extreme values show no indication of alignment with the lowest flows. Thus, it appears that the default assumption of CAM with regard to the coincidence of extreme conditions of pH and percent unionized ammonia with flow may be unnecessarily conservative for setting of permit limits in the St. Vrain basin, including Boulder Creek and Coal Creek.
100
St Vrain below Longmont
Daily Average % Un-ionized
12.0% 10.0%
y = 7E-05x + 0.012 R2 = 0.0255
8.0% 6.0% 4.0% 2.0% 0.0% 10
100
1000
Flow , cfs
St Vrain below Longmont 9.5
Daily Average pH
9.0 8.5 8.0 7.5 7.0
y = 0.0031x + 7.4465 R2 = 0.0694
6.5 6.0 10
100 Flow , cfs
101
1000
Coal Creek at Louisville 16.0% y = 5E-05x + 0.0294 R2 = 0.0003
Daily Average % Un-ionized
14.0% 12.0% 10.0% 8.0% 6.0% 4.0% 2.0% 0.0% 0.1
1
10
100
Flow , cfs
Coal Creek at Louisville 9.0
Daily Average pH
8.5
8.0
7.5
7.0 y = -0.0013x + 8.0574 2 R = 0.001
6.5
6.0 0.1
1
10 Flow, cfs
102
100
Boulder Creek at 75th St
Daily Average % Un-ionized
2.5% y = 2E-06x + 0.005 R2 = 0.0031
2.0%
1.5%
1.0%
0.5%
0.0% 10
100
1000
10000
Flow, cfs
Boulder Creek at 75th St 8.0 7.8
Daily Average pH
7.6 7.4 7.2 7.0 6.8 6.6
y = 0.0003x + 7.1498 2 R = 0.0164
6.4 6.2 6.0 10
100
1000 Flow, cfs
103
10000
Boulder Creek above WWTP
Daily Average % Un-ionized
25.0%
20.0%
15.0% y = -3E-05x + 0.0271 2 R = 0.0146 10.0%
5.0%
0.0% 1
10
100
1000
10000
Flow, cfs
Boulder Creek above WWTP 9.5 9.0
y = -0.0013x + 7.8525 2 R = 0.038
Daily Average pH
8.5 8.0 7.5 7.0 6.5 6.0 1
10
100 Flow, cfs
104
1000
10000
Boulder Creek at Orodell
Daily Average % Un-ionized
35% 30% 25%
y = 3E-05x + 0.0369 R2 = 0.0068
20% 15% 10% 5% 0% 1
10
100
1000
Flow, cfs
Boulder Creek at Orodell 9.5
Daily Average pH
9.0 8.5 8.0 7.5 y = -0.0009x + 8.217 R2 = 0.0894
7.0 6.5 6.0 1
10
100 Flow, cfs
105
1000
