View
3
Download
0
Category
Preview:
Citation preview
ESCHERICHIA COLI TOTAL MAXIMUM DAILY LOADS (TMDLs) CONVERSION WITH EXISTING FECAL COLIFORM TMDLs FOR IMPAIRED STREAMS
DESIGNATED RECREATION USES IN SOUTH DAKOTA
Watershed Protection Program
Division of Financial and Technical Assistance
South Dakota Department of Environment and Natural Resources
Paul Lorenzen and Anine Rosse
NOVEMBER 2020
ii
Table of Contents
INTRODUCTION ............................................................................................................................................. 1
WATER QUALITY STANDARDS AND TMDL TARGETS ..................................................................................... 3
BACTERIA TMDL CONVERSION APPROACH .................................................................................................. 4
NONPOINT SOURCES .................................................................................................................................... 6
POINT SOURCES ............................................................................................................................................ 6
SD-BS-R-BEAVER_02.................................................................................................................................. 6
SD-CH-R-RAPID_04 .................................................................................................................................... 6
SD-BS-R-BRULE_01 .................................................................................................................................... 7
TMDL AND ALLOCATIONS ............................................................................................................................. 8
SD-VM-R-VERMILLION_E_FORK_01 .......................................................................................................... 8
SD-BS-R-BRULE_01 .................................................................................................................................... 9
SD-BS-R-BEAVER_02.................................................................................................................................. 9
SD-CH-R-RAPID_04 .................................................................................................................................. 10
SUMMARY AND FUTURE CONSIDERATIONS ............................................................................................... 11
PUBLIC COMMENT ...................................................................................................................................... 11
LITERATURE CITED ...................................................................................................................................... 12
List of Tables Table 1. Fecal coliform TMDL information and references for stream segments impaired for E. coli. ....... 2
Table 2. Designated recreation uses and bacteria standards for the impaired stream segments. ............. 3
Table 3. Bacteria translation results from South Dakota’s statewide equation ........................................... 4
Table 4. EC/FC ratios applied to SD bacteria standards using Ohio, Virginia and Oregon bacteria
translator equations...................................................................................................................................... 5
Table 5. CAFOs identified in the East Fork Vermillion River Segment 01 watershed. ................................. 8
Table 6. Bacteria standards and ratio for immersion and limited contact recreation uses. ....................... 8
Table 7. Existing Fecal coliform TMDL and allocations for the East Fork Vermillion River-Segment 01
based on GM standard (1000 CFU/100 mL) for limited contact recreation. ................................................ 8
Table 8. E. coli TMDL and Load allocations for the East Fork Vermillion River-Segment 01 based on GM
standard (630 CFU/100 mL) for limited contact recreation. ........................................................................ 9
Table 9. Existing Fecal coliform TMDL and Load allocations for Brule Creek-Segment 01 based on GM
standard (1000 CFU/100 mL) for limited contact recreation. ...................................................................... 9
Table 10. E. coli TMDL and Load allocations for Brule Creek-Segment 01 based on GM standard (630
CFU/100 mL) for limited contact recreation. ................................................................................................ 9
Table 11. Existing Fecal coliform TMDL and Load allocations for Beaver Creek-Segment 01 based on SSM
standard (2000 CFU/100 mL) for limited contact recreation. ...................................................................... 9
iii
Table 12. E. coli TMDL and Load allocations for Beaver Creek-Segment 01 based on SSM standard (1,178
CFU/100 mL) for limited contact recreation. .............................................................................................. 10
Table 13. Existing Fecal coliform TMDL and Load allocations for Rapid Creek-Segment 04 based on SSM
standard (400 CFU/100 mL) for immersion recreation. ............................................................................. 10
Table 14. E. coli TMDL and Load allocations for Rapid Creek-Segment 04 based on SSM standard (235
CFU/100 mL) for immersion recreation. ..................................................................................................... 10
Table 15. Existing Fecal coliform TMDL and Load allocations for Rapid Creek-Segment 04 based on GM
standard (200 CFU/100 mL) for immersion recreation. ............................................................................. 10
Table 16. E. coli TMDL and Load allocations for Rapid Creek-Segment 04 based on GM standard (126
CFU/100 mL) for immersion recreation. ..................................................................................................... 11
Appendices Appendix A. E. coli Numeric TMDL Target Selection Rationale………………………………………………………………13
Appendix B. EPA approval letter and decision document………………………………………………………………………17
1
INTRODUCTION South Dakota adopted EPA recommended national standards and criteria for Escherichia coli (E. coli)
into state water quality standards in 2009 (Administrative Rules of South Dakota 74:51:01: 50:51). E.
coli eventually replaced Fecal coliform as the primary indicator to protect waters designated immersion
and limited contact recreation uses. Fecal coliform initially remained in state water quality standards to
allow NPDES permits to transition to E. coli during the permit renewal process. This allowed DENR time
to build E. coli datasets through various monitoring networks. The transition was complete in 2015
when Fecal coliform was officially removed from state water quality standards. Several Fecal coliform
TMDLs were developed and approved by EPA prior to South Dakota’s transition to E. coli.
South Dakota’s current Long Term 303(d) Vision Strategy includes a systematic process for prioritizing
TMDL development (https://denr.sd.gov/dfta/wp/tmdl/tmdlvision2020.pdf). TMDL prioritization is
significantly focused on waters impaired for immersion and limited contact recreation due to E. coli.
The TMDL priority list contains four stream segments impaired for E. coli with EPA approved Fecal
coliform TMDLs (Table 1). Considerable time and resources were spent on developing independent
Fecal coliform TMDLs for the impaired segments. Converting existing Fecal coliform TMDLs and
allocations to E. coli would eliminate the need to duplicate resource intensive TMDL development
efforts. The relationship between both bacteria has been used for E. coli TMDL development with
translators as the foundation (Limno Tech, 2011, Dila and McLellan, 2016).
The intent of this document is to convert existing Fecal coliform TMDLs and allocations for four stream
segments to E. coli using defensible methods to satisfy 303(d) requirements. The following stream
segments (Assessment Units) were considered impaired for designated recreation uses due to E. coli in
South Dakota’s most recent 2020 Integrated Report (IR) for Surface Water Quality Assessment
https://denr.sd.gov/documents/SD_2020_IR_approved.pdf (Table 1).
https://denr.sd.gov/dfta/wp/tmdl/tmdlvision2020.pdfhttps://denr.sd.gov/documents/SD_2020_IR_approved.pdf
2
Table 1. Fecal coliform TMDL information and references for stream segments impaired for E. coli.
East Fork Vermillion River_01: EAST FORK VERMILLION RIVER FROM MCCOOK/LAKE COUNTY LINE TO ITS CONFLUENCE WITH LITTLE VERMILLION RIVER AUID: SD-VM-R-VERMILLION_E_FORK_01 (formerly SD-VM-R-VERMILLION_EAST_FORK_01) EPA TMDL Approval Date: September 26, 2012 Attains TMDL ID: 42525 http://denr.sd.gov/dfta/wp/tmdl/tmdl_vermillioneastforkfecal0512.pdf
Brule Creek_01: BRULE CREEK FROM THE CONFLUENCE OF EAST AND WEST BRULE CREEK TO THE CONFLUENCE WITH THE BIG SIOUX RIVER
AUID: SD-BS-R-BRULE_01 EPA TMDL Approval Date: June 2, 2011 Attains TMDL ID: 40438 http://denr.sd.gov/dfta/wp/tmdl/tmdl_brulecreekfecal0611.pdf
Beaver Creek_02: BEAVER CREEK - SPLIT ROCK CREEK TO SD-MN BORDER (CENTRAL BIG SIOUX RIVER TMDLS)
AUID: SD-BS-R-BEAVER_02 EPA TMDL Approval Date: May 28, 2008 Attains TMDL ID: 34499 https://denr.sd.gov/dfta/wp/tmdl/tmdl_bigsiouxcentral.pdf
Lower Rapid Creek_04: LOWER RAPID CREEK - FROM RC WWTF TO ABOVE FARMINGDALE
AUID: SD-CH-R-RAPID_04 EPA TMDL Approval Date: September 28, 2010 Attains TMDL ID: 39427 https://denr.sd.gov/dfta/wp/tmdl/tmdl_rapidcreeklowerfecal.pdf
http://denr.sd.gov/dfta/wp/tmdl/tmdl_vermillioneastforkfecal0512.pdfhttp://denr.sd.gov/dfta/wp/tmdl/tmdl_brulecreekfecal0611.pdfhttps://denr.sd.gov/dfta/wp/tmdl/tmdl_bigsiouxcentral.pdfhttps://denr.sd.gov/dfta/wp/tmdl/tmdl_rapidcreeklowerfecal.pdf
3
WATER QUALITY STANDARDS AND TMDL TARGETS South Dakota E. coli standards and criteria for immersion (ARSD 74:51:01:50) and limited contact
recreation (ARSD 74:51:01:51) consist of a single sample maximum (SSM) and a monthly geometric
mean (GM). Criteria for the SSM requires that no single daily sample exceed the standard. The GM is
also considered a “must not exceed” criteria, calculated based on a minimum of 5 samples collected
during separate 24-hr periods over a 30-day period. The standards structure and criteria for Fecal
coliform was identical to E. coli for both recreation uses, respectively. However, actual numeric
standard limits for E. coli are considerably lower than former limits for Fecal coliform (Table 2).
Water quality standards and criteria, while explicit, are used by states as benchmarks to make
impairment decisions in accordance with 303(d) listing methods (SD DENR 2020 IR). When two
standards exist for the same use (i.e. SSM and GM) an impairment assessment is performed on both
standards depending on data availability. Impaired waters require TMDL development based on the
most protective standard regardless of the assessment result. Selecting the most protective numeric
target for TMDL development ensures attainment with both standards. With regards to Fecal coliform
and E. coli the GM standards are considerably lower than the SSM standards for both recreation uses,
respectively. Therefore, it was common practice for many years to use the GM as the numeric target for
TMDL development.
Fecal coliform TMDLs for the four stream segments generally used the GM as the TMDL target (Table 2).
The exception was Beaver Creek_02 which used the SSM as the TMDL target. In order to convert the
Fecal coliform TMDLs to E. coli, it is necessary to establish that the TMDL target will be protective of
both standards. An investigation into the development of the E. coli standards revealed a statistical
linkage between GM and SSM E. coli criteria. It was concluded that the GM and SSM E. coli standards
assigned to both recreation uses are equally protective (Appendix A). As a result, E. coli TMDL
development was based on the same target used to development the Fecal coliform TMDLs.
Table 2. Designated recreation uses and bacteria standards for the impaired stream segments.
Impaired Stream Segment
AUID
Designated Recreation
Use
Fecal Coliform Geomean
CFU/100mL
Fecal Coliform
SSM CFU/100mL
E. coli Geomean
CFU/100mL
E. coli SSM
CFU/100 mL
SD-VM-R-VERMILLION_E_FORK_01
Limited Contact
*< 1,000 < 2,000 *< 630 < 1,178
SD-BS-R-BRULE_01 Limited Contact
*< 1,000 < 2,000 *< 630 < 1,178
SD-BS-R-BEAVER_02 Limited Contact
< 1,000 *< 2,000 < 630 *< 1,178
SD-CH-R-Rapid_04 Immersion *< 200
*< 400 *< 126 *< 235
*Refers to numeric standard used for TMDL development.
http://www.sdlegislature.gov/rules/DisplayRule.aspx?Rule=74:51:01:50http://www.sdlegislature.gov/rules/DisplayRule.aspx?Rule=74:51:01:51
4
BACTERIA TMDL CONVERSION APPROACH E. coli is a Fecal coliform bacteria and both indicators originate from common sources in relatively
consistent proportions. The inherent similarity in both bacteria can be exploited for a variety of water
quality based applications. SD DENR developed regional relationships to support the addition of Fecal
coliform data in E. coli TMDL development (SD DENR, 2010). Some states have developed bacteria
translator equations to convert Fecal coliform TMDLs and load allocations to E. coli (Limno Tech, 2011).
The translator approach was explored as a potential option for converting the bacteria TMDLs.
A statewide translator equation was developed from 10,686 paired bacteria samples collected from
rivers and streams across South Dakota. All applicable bacteria data were logarithmically transformed
(LOG10), and E. coli was plotted as a function of Fecal coliform using simple linear regression. The
analysis yielded a significant (p=
5
Table 4. EC/FC ratios applied to SD bacteria standards using Ohio, Virginia and Oregon bacteria translator equations.
Ohio Bacteria Translator (Statewide excluding NE region)
Fecal coliform X Predicted E. coli Y EC/FC Ratio
200 93 0.47
400 191 0.48
1,000 489 0.49
2,000 997 0.50
Virginia Statewide Bacteria Translator
Fecal coliform X Predicted E. coli Y EC/FC Ratio
200 129 0.64
400 243 0.61
1,000 564 0.56
2,000 1,068 0.53
Oregon Statewide Bacteria Translator
Fecal coliform X Predicted E. coli Y EC/FC Ratio
200 146 0.73
400 304 0.76
1,000 804 0.80
2,000 1,676 0.84
Virginia considered using bacteria standards as the basis for developing a translator equation (Limno
Tech, 2011). This approach involves using the ratio between E. coli and Fecal coliform standards to
predict E. coli. For example, the GM for Fecal coliform and E. coli is 200 (CFU/100mL) and 126
(CFU/100mL), respectively. The resultant EC/FC ratio is calculated at 0.63. The following equation
would be used to predict E. coli concentrations using the Fecal coliform GM standard for immersion
recreation.
[E. coli (Y) = EC/FC ratio* FC (X)] (FC (X)=200; E. coli (Y)=126)
As expected, E. coli equates to 126 (CFU/100mL) consistent with the GM standard for immersion
recreation. Virginia decided to use actual paired bacteria data to develop the translator versus the
standards ratio approach.
South Dakota’s statewide bacteria relationship is robust with regards to the total number of paired
observations and high correlation. The near 1:1 ratio is impressive considering paired samples were
collected over a large temporal and spatial scale from many diverse systems. The statewide translator
equation would result in E. coli TMDLs and load allocations that are slightly over the TMDL targets. It is
possible to improve the equation to achieve desired ratios by screening the data and exploring different
regression analysis techniques (Limno Tech, 2011). This was not considered due to limitations with
obtaining conditional information and potential risk of making biased decisions from such a large
dataset that encompasses a wide range of environmental controls. A decision was made to use the
standards ratio approach to convert Fecal coliform TMDLs to E. coli. This approach ensures that E. coli
6
TMDLs and load allocations are protective of the standards despite variability in bacteria data. The
process is documented in the TMDL and Load Allocation section.
NONPOINT SOURCES The level of information provided in the nonpoint source assessments varies among the different Fecal
Coliform TMDL documents. In general, the predominate land use in the watersheds of the impaired
stream segments residing in eastern South Dakota was documented as cropland and pasture. Lower
Rapid Creek _04 is the only segment located in western South Dakota. This segment begins at the edge
of Rapid City though a fair portion of the watershed is rangeland. The largest source of bacteria
production documented in the impaired watersheds was from livestock (beef, dairy, and hogs) on grass
and in feedlots. Bacteria source estimates are based on literature values for Fecal coliform, which are
considered synonymous with E. coli based on the statewide bacteria relationship. Land use and bacteria
production characteristics in the impaired watersheds are expected to be vastly similar to that
documented during the respective Fecal Coliform TMDL assessments.
POINT SOURCES There are several documented point source discharges within the watersheds of three out of the four
impaired stream segments. This includes five permitted National Pollutant Discharge Elimination
Systems (NPDES) that may directly contribute E. coli bacteria. These potential sources of E. coli bacteria
are documented here to provide a watershed scale account of the systems operational characteristics
(discharge permits etc.), potential impact and Waste Load Allocation consideration.
SD-BS-R-BEAVER_02 The city of Valley Springs Wastewater Treatment Facility consists of a 3 cell pond system and is
authorized (NPDES permit SD0020923) to periodically discharge directly to the impaired segment of
Beaver Creek. Discharge from the facility must comply with effluent limits established for various
pollutants. E. coli concentrations must not exceed the SSM and GM E. coli criteria for limited contact
recreation waters which is consistent with the TMDL target. The E. coli TMDL would not add new
requirements or implementation expectations to the permit. The permit also demonstrates that E. coli
limits meet Fecal coliform TMDL goals.
The Fecal coliform TMDL included a Waste Load Allocation (WLA) for the City of Valley Springs. A Waste
Load Allocation of 4.01X1010 CFU/day was assigned to the Fecal coliform TMDL. The WLA was based on
the average design flow (0.82) multiplied by the SSM Fecal coliform standard (2,000 CFU/100mL) for
limited contact waters, times a unit conversion factor (24465715). The E. coli WLA was derived from
the same calculation substituting the SSM E. coli standard (1,178 CFU/100mL), resulting in a WLA of
2.36EX1010. This facility has only discharged once for a brief period since 2000. The actual E. coli waste
load to Beaver Creek_02 is considered negligible and as long as future discharges from this facility do
not exceed permit effluent limits for E. coli, impact to the TMDL is considered minimal.
SD-CH-R-RAPID_04 The Rapid City Water Reclamation Facility is a mechanical system and is authorized to discharge (NPDES
permit SD0023574) to Rapid Creek Segment 04. Continuous discharge from the facility must comply
with effluent limits established for various pollutants. E. coli concentrations must not exceed the SSM
7
and GM E. coli criteria for immersion recreation waters which is consistent with the most stringent
TMDL target. The E. coli TMDL would not add new requirements or implementation expectations to the
permit. The permit also demonstrates that E. coli limits meet Fecal coliform TMDL goals.
The Fecal coliform TMDL included WLAs for the Rapid City Water Reclamation facility. The WLAs were
based on the average design flow (23.2 cfs) multiplied by the SSM and GM standards (400 CFU/100mL
and 200 CFU/100mL) for immersion recreation waters, times a unit conversion factor (24465715). The E.
coli WLA was derived from the same calculation substituting the SSM (235 CFU/100mL) and GM (126
CFU/100mL) E. coli standards resulting in a WLA of 1.30EX1011 and 7.18EX1011, respectively.
Rapid City Regional Airport Wastewater Facility is permitted (NPDES permit SD0028638) to discharge
wastewater in the event of an emergency, otherwise it is a no discharge system in accordance with
provisions of the permit. Emergency wastewater discharges are conveyed to an unnamed tributary
approximately 2.1 miles upstream of the impaired segment (04) of Rapid Creek. The permit contains E.
coli monitoring provisions for emergency discharges. As long as this system complies with the
requirements of the “no discharge” permit ensuring discharges are unlikely and indirect loading events,
the TMDL assumes E. coli contribution is minimal. A WLA for this facility was not provided in the TMDL.
Rapid Valley Sanitary District is a rural drinking water system covered under a minor water treatment
and distribution general permit (permit SDG86007). This system was not identified as a source of E. coli
bacteria for the impaired segment of Rapid Creek. A Waste Load Allocation (WLA) was not provided for
this system in the TMDL.
SD-BS-R-BRULE_01 L.G. Everist. Inc.-NWIA is permitted (NPDES permit SD0027928) to discharge outfall from the Spink Gravel
Pit (Outfall 001) to the impaired segment of Brule Creek only under emergency conditions. Discharges
from this operation have been infrequent (13 days out of 193) according to recent monitoring reports.
TMDL considerations from the permit suggest this operation is not expected to be a contributor of E. coli.
As a result, a WLA was not provided for this operation in the TMDL. The WLA for SD-BS-R-BRULE_01 is
considered zero.
There were no Concentrated Animal Feeding Operations (CAFOs) identified in the Fecal Coliform TMDL
documents of the four impaired segments. A recent search found that two CAFOs have since been
established in the East Fork Vermillion River_01 watershed (Table 5). Both facilities have permit
coverage under South Dakota’s 2017 general permit. For more information about South Dakota’s CAFO
requirements and general permits visit: http://denr.sd.gov/des/fp/cafo.aspx. As long as these facilities
comply with the general CAFO permit requirements ensuring their discharges are unlikely and indirect
loading events, the TMDL assumes their E. coli contribution is minimal. As a result, the Waste Load
Allocation (WLA) for this segment remains zero. There are no CAFOs in the other impaired watersheds
at this time.
http://denr.sd.gov/des/fp/cafo.aspx
8
Table 5. CAFOs identified in the East Fork Vermillion River Segment 01 watershed.
Name of Facility Type of Operations SD General Permit #
Christopher Katzer Swine Facility swine (housed lot) SDG-100150
Orland Hutterian Brethren, Inc. multiple animals (housed lots) SDG-109100
TMDL AND ALLOCATIONS A load duration curve framework was used to develop the existing Fecal coliform TMDLs for the
impaired stream segments. A standards ratio approach was used to convert existing Fecal coliform
TMDLs to E. coli TMDLs for each flow zone. E. coli TMDLs were calculated by multiplying the existing
Fecal coliform TMDLs by the ratio (EC:FC) between the applicable bacteria standards (Table 6). The E.
coli TMDL allocations (TMDL=WLA+LA=MOS) were based on the same percent contribution as
established for the Fecal coliform TMDL allocations in each flow zone.
The Fecal coliform current load in each flow zone was used to calculate the percent reduction required
to meet the E. coli TMDLs. The E. coli percent reduction was calculated as the current Fecal coliform
load minus the E. coli TMDL divided by the current Fecal coliform load. Using the current Fecal coliform
load to determine the required E. coli load reductions is considered conservative.
Table 6. Bacteria standards and ratio for immersion and limited contact recreation uses.
Fecal coliform standards E. coli standards EC/FC ratio
200 126 0.63
400 235 0.5875
1,000 630 0.63
2,000 1,178 0.589
The E. coli TMDLs are protective of applicable standards assigned to designated recreation uses of the
impaired stream segments. The existing Fecal coliform TMDL documents contain supporting
information for implementing the E. coli TMDLs in accordance with 303(d) requirements. The original
Fecal coliform and converted E. coli TMDLs, allocations and reductions are provided by stream segment
in Tables 7 through 16.
SD-VM-R-VERMILLION_E_FORK_01 Table 7. Existing Fecal coliform TMDL and allocations for the East Fork Vermillion River-Segment 01 based on GM standard (1000 CFU/100 mL) for limited contact recreation.
Flow Zone Fecal TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/day)
MOS (CFU/day)
% Reduction
High 5.55E+13 0.00E+00 5.00E+13 5.55E+12 0.0%
Moist 5.23E+12 0.00E+00 4.71E+12 5.23E+11 0.0%
Mid-Range 8.75E+11 0.00E+00 7.88E+11 8.75E+10 93.8%
Dry 3.03E+11 0.00E+00 2.73E+11 3.03E+10 98.7%
Low 6.71E+10 0.00E+00 6.04E+10 6.71E+09 67.43%
9
Table 8. E. coli TMDL and Load allocations for the East Fork Vermillion River-Segment 01 based on GM standard (630 CFU/100 mL) for limited contact recreation.
Flow Zone E. coli TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High 3.50E+13 0.00E+00 3.15E+13 3.50E+12 0.0%
Moist 3.29E+12 0.00E+00 2.97E+12 3.29E+11 0.0%
Mid-Range 5.51E+11 0.00E+00 4.96E+11 5.51E+10 96.1%
Dry 1.91E+11 0.00E+00 1.72E+11 1.91E+10 99.2%
Low 4.23E+10 0.00E+00 3.80E+10 4.23E+09 79.5%
SD-BS-R-BRULE_01 Table 9. Existing Fecal coliform TMDL and Load allocations for Brule Creek-Segment 01 based on GM standard (1000 CFU/100 mL) for limited contact recreation.
Flow Zone Fecal TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High 3.10E+16 0.00E+00 3.10E+16 2.80E+12 99.9%
Moist 3.10E+14 0.00E+00 3.10E+14 5.40E+11 98%
Mid-Range 2.20E+13 0.00E+00 2.20E+13 1.70E+11 99%
Dry 4.80E+12 0.00E+00 4.70E+12 1.30E+11 7.6%
Low 3.90E+11 0.00E+00 3.40E+11 5.10E+10 79%
Table 10. E. coli TMDL and Load allocations for Brule Creek-Segment 01 based on GM standard (630 CFU/100 mL) for limited contact recreation.
Flow Zone E. coli TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High 1.95E+16 0.00E+00 1.95E+16 1.76E+12 99.9%
Moist 1.95E+14 0.00E+00 1.95E+14 3.40E+11 98.7%
Mid-Range 1.39E+13 0.00E+00 1.38E+13 1.07E+11 0.0%
Dry 3.02E+12 0.00E+00 2.94E+12 8.19E+10 0.0%
Low 2.46E+11 0.00E+00 2.14E+11 3.21E+10 0.0%
SD-BS-R-BEAVER_02 Table 11. Existing Fecal coliform TMDL and Load allocations for Beaver Creek-Segment 01 based on SSM standard (2000 CFU/100 mL) for limited contact recreation.
Flow Zone Fecal TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High/Moist 5.65E+12 4.01E+10 5.04E+12 5.65E+11 84.7%
Mid-Range 1.24E+12 4.01E+10 1.08E+12 1.24E+11 0.0%
Dry/Low 2.96E+11 4.01E+10 2.26E+11 2.96E+10 0.0%
10
Table 12. E. coli TMDL and Load allocations for Beaver Creek-Segment 01 based on SSM standard (1,178 CFU/100 mL) for limited contact recreation.
Flow Zone E. coli TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High/Moist 3.33E+12 2.36E+10 2.97E+12 3.33E+11 91%
Mid-Range 7.29E+11 2.36E+10 6.33E+11 7.29E+10 0.0%
Dry/Low 1.74E+11 2.36E+10 1.33E+11 1.74E+10 0.0%
SD-CH-R-RAPID_04 Table 13. Existing Fecal coliform TMDL and Load allocations for Rapid Creek-Segment 04 based on SSM standard (400 CFU/100 mL) for immersion recreation.
Flow Zone Fecal TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High 6.47E+12 2.27E+11 4.24E+12 2.00E+12 94.7%
Moist 2.15E+12 2.27E+11 1.23E+12 6.95E+11 84.4%
Mid-Range 7.99E+11 2.27E+11 4.45E+11 1.27E+11 0.0%
Dry 5.38E+11 2.27E+11 1.84E+11 1.27E+11 0.0%
Low 3.13E+11 2.27E+11 4.70E+10 3.90E+10 0.0%
Table 14. E. coli TMDL and Load allocations for Rapid Creek-Segment 04 based on SSM standard (235 CFU/100 mL) for immersion recreation.
Flow Zone E. coli TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High 3.80E+12 1.33E+11 2.49E+12 1.17E+12 96.9%
Moist 1.27E+12 1.33E+11 7.24E+11 4.08E+11 90.8%
Mid-Range 4.69E+11 1.33E+11 2.61E+11 7.46E+10 37.2%
Dry 3.16E+11 1.33E+11 1.08E+11 7.46E+10 29.1%
Low 1.84E+11 1.33E+11 2.76E+10 2.29E+10 0.0%
Table 15. Existing Fecal coliform TMDL and Load allocations for Rapid Creek-Segment 04 based on GM standard (200 CFU/100 mL) for immersion recreation.
Flow Zone Fecal TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High 3.23E+12 1.14E+11 2.12E+12 9.98E+11 55.9%
Moist 1.08E+12 1.14E+11 6.16E+11 3.47E+11 0.0%
Mid-Range 3.99E+11 1.14E+11 2.21E+11 6.40E+10 0.0%
Dry 2.69E+11 1.14E+11 9.10E+10 6.40E+10 0.0%
Low 1.57E+11 1.14E+11 2.30E+10 2.00E+10 0.0%
11
Table 16. E. coli TMDL and Load allocations for Rapid Creek-Segment 04 based on GM standard (126 CFU/100 mL) for immersion recreation.
Flow Zone E. coli TMDL (CFU/day)
WLA (CFU/day)
LA (CFU/Day)
MOS (CFU/day)
% Reduction
High 2.04E+12 7.18E+10 1.34E+12 6.29E+11 72.2%
Moist 6.79E+11 7.18E+10 3.88E+11 2.19E+11 27.4%
Mid-Range 2.52E+11 7.18E+10 1.40E+11 4.03E+10 0.0%
Dry 1.69E+11 7.18E+10 5.73E+10 4.03E+10 0.0%
Low 9.86E+10 7.18E+10 1.42E+10 1.26E+10 23%
SUMMARY AND FUTURE CONSIDERATIONS This document provides a framework to convert Fecal coliform TMDLs and allocations to E. coli to
address impaired streams designated recreation uses in South Dakota. A statewide translator equation
was developed from over 10,000 paired bacteria samples. The resulting equation indicates a clear
similarity in both bacteria with a near 1:1 ratio. The statewide equation was not considered a viable
translator option as ratios should range between 0.5875 to 0.63 to be consistent with the ratio between
standards to account for variability in bacteria data. A standards ratio approach was used to convert
existing Fecal coliform TMDLs to E. coli within each of the established flow zones. This approach
accounts for variability in bacteria data and provides assurance that E. coli TMDLs are protective of
applicable standards.
This TMDL conversion process only applies to impaired waters where conditions present during Fecal
coliform TMDL development have remained static. The process should not be followed in instances
where significant changes have occurred in the watershed (source assessments) or new NPDES permits
have been issued requiring a WLA. The translation process and resulting E. coli TMDLs and allocations
were subject to public comment prior to submittal to EPA for review and approval consideration in
accordance with 303(d) requirements. Once approved, the standards ratio approach will be used, when
relevant, to convert Fecal coliform TMDLs to E. coli following an addendum process.
PUBLIC COMMENT A public notice letter was published in several local newspapers within the watersheds of the four
impaired stream segments to announce the availability of this bacteria TMDL conversion document for
review and comment. The document and comment process was made available on the South Dakota
Department of Environment and Natural Resources One Stop Public Notice webpage at:
https://denr.sd.gov/public/default.aspx. The public comment period began September 4, 2020 and
ended on October 8, 2020. No public comments were received during this period.
https://denr.sd.gov/public/default.aspx
12
LITERATURE CITED Dila, DK and McLellan, SL. Revised 2016. Translator Development for Bacterial Indicator TMDLs.
Prepared for CDM Smith. UWM-School of Freshwater Sciences, Milwaukee, WI 53204.
Limno Tech. 2011, June 3. Final Memo Summarizing DC Bacteria Data and Recommending a DC Bacteria
Translator (Task 2). Prepared for U.S. Environmental Protection Agency. Washington, DC, by Limno
Tech, Washington DC.
SD DENR, 2010. Fecal Coliform, Escherichia coli Bacteria Total Maximum Daily Loads (TMDLs) for the
Lower Rapid Creek, Pennington County, South Dakota. South Dakota Department of Environment and
Natural Resources, Pierre, SD. https://denr.sd.gov/dfta/wp/tmdl/tmdl_rapidcreeklowerfecal.pdf.
SD DENR, 2020. The 2020 South Dakota Integrated Report for Surface Water Quality Assessment. South
Dakota Department of Environment and Natural Resources, Pierre, SD.
https://denr.sd.gov/documents/SD_2020_IR_final.pdf.
USEPA (United States Environmental Protection Agency). 1986. Ambient Water Quality Criteria for
Bacteria. EPA 440/5-84-002. Office of Water Regulations and Standards Criteria and Standards Division,
United States Environmental Protection Agency, Washington D.C.
USEPA (United States Environmental Protection Agency). 2001. Protocol for Developing Pathogen
TMDLs. EPA 841-R-00-002. Office of Water, United States Environmental Protection Agency,
Washington D.C.
USEPA (United States Environmental Protection Agency). 2012. Recreational Water Quality Criteria.
EPA 820-F-12-058. Office of Water, United States Environmental Protection Agency, Washington D.C.
https://denr.sd.gov/dfta/wp/tmdl/tmdl_rapidcreeklowerfecal.pdfhttps://denr.sd.gov/documents/SD_2020_IR_final.pdf
13
Appendix A: E. coli Numeric TMDL Target Selection Rationale
TMDLs are required to identify a numeric target to measure whether or not the applicable water
quality standard is attained. A maximum allowable load, or TMDL, is ultimately calculated by
multiplying this target with a flow value and a unit conversion factor. Generally, the pollutant
causing the impairment and the parameter expressed as a numeric water quality criterion are the
same. In these cases, selecting a TMDL target is as simple as applying the numeric criteria.
Occasionally, an impairment is caused by narrative water quality criteria violations or by
parameters that cannot be easily expressed as a load. When this occurs, the narrative criteria
must be translated into a numeric TMDL target (e.g., nuisance aquatic life translated into a total
phosphorus target) or a surrogate target established (e.g., a pH cause addressed through a total
nitrogen target) and a demonstration should show how the chosen target is protective of water
quality standards.
As seen from Table 2 there are two numeric E. coli criteria for TMDL target consideration. When
multiple numeric criteria exist for a single parameter, the most stringent criterion is selected as the
TMDL target. To judge whether one is more protective of the beneficial use, it is necessary to
further elaborate how the criteria were derived.
South Dakota’s E. coli criteria are based on EPA recommendations originally published in 1986
(USEPA, 1986). EPA issued slightly modified recommendations in 2012 that did not substantially
change the underlying analysis or criteria values in South Dakota (USEPA, 2012). As
recommended, SDDENR adopted E. coli criteria that contain two components: a geometric mean
(GM) and a single sample maximum (SSM). The GM was established from epidemiological
studies by comparing average summer exposure to an illness rate of 8:1,000. The SSM component
was computed using the GM value and the corresponding variance observed in the epidemiological
study dataset (i.e., log-standard deviation of 0.4). EPA provided four different SSM values
corresponding to the 75th, 82nd, 90th, and 95th percentiles of the expected water quality sampling
distribution around the GM to account for different recreational use intensities (Figure 2). South
Dakota adopted the most stringent recommendation, the 75th percentile, into state water quality
standard regulations as the SSM protective of designated beaches.
14
Figure 2. Log-Normal Frequency Distribution Used to Establish South Dakota’s Immersion Recreation E. coli
Criteria of 126 (GM) and 235 (SSM) #/100mL (EPA, 1986).
Dual criteria were established to balance the inherent variability of bacteria data and provide
flexibility for handling different sampling routines. Together, the GM and SSM describe a water
quality distribution expected to be protective of immersion contact recreation. The GM and SSM
are equally protective of the beneficial use because they are based on the same illness rate and
differ simply representing different statistical values and sampling timeframes. While this
investigation has revealed the GM and SSM E. coli criteria to be equally protective of the
immersion recreation use, a likewise conclusion can be made for the GM and SSM criteria
associated with the limited contact recreation use since those values were simply derived as five
times the immersion values.
As described in EPA’s Protocol for Developing Pathogen TMDLs, the availability of data may
dictate which criterion should be used as the TMDL target (EPA, 2001). When a geometric mean
of the sampling dataset can be calculated as defined by South Dakota Administrative Rules (i.e.,
at least five samples separated by a minimum of 24-hours over a 30-day period) and compared to
the GM criterion, SDDENR uses the GM criterion as the TMDL target. This establishes a smaller
overall loading capacity and is considered a conservative approach to setting the TMDL.
When a proper GM cannot be calculated SDDENR uses the SSM as the TMDL target. This is
permissible because the SSM is equally protective of the beneficial use as discussed above.
Although this target selection leads to the establishment of a larger allowable load, in some respects
it is more appropriate because timeframes align better (i.e., the SSM is associated with a single
day and TMDLs establish daily loads, versus the 30-day GM). Additionally, certain aspects of
15
SDDENR’s E. coli assessment method, when combined with a SSM TMDL target, result in an
expected dataset GM more protective than the GM criterion. SDDENR uses assessment methods
to define how to interpret and apply water quality standards to 303(d) impairment decisions. These
methods are further discussed in Section 3.4, however for this discussion, it is important to note
that SDDENR allows a 10% exceedance frequency of both the SSM and GM. In other words, as
long as the E. coli dataset meets other age and size requirements, a waterbody is considered
impaired (i.e., not meeting water quality standards) when greater than 10% of samples exceed
either the SSM or GM. Water quality standards are met if the exceedance frequency is 10% or less.
Returning to the original distribution used to establish South Dakota’s Immersion Recreation E.
coli criteria in Figure 2 remember that SDDENR chose to adopt a SSM concentration based on the
most stringent recommendation (75th percentile). According to assessment methods in South
Dakota, however, the SSM concentration is treated as a 90th percentile (i.e., 10% exceedance
frequency). Step #1 in Figure 3 shows how doing so effectively moves the SSM point to the right.
If the original log-normal frequency distribution with a log-standard deviation of 0.4 is
subsequently re-fitted to this new 90th percentile point at 235 #/100mL (red dotted line), the
corresponding 50th percentile (GM) is 72 #/100mL as shown in Step #2 of Figure 3.
Figure 3. The Effective Impact of South Dakota’s E. coli Assessment Method on the Criteria’s Original Log-
Normal Frequency Distribution (Black line = original; red dotted line = shifted)
The GM associated with this shifted distribution is more stringent than the GM of the original
distribution (126 #/100mL), thus this demonstrates that attaining a maximum daily SSM target in
16
a TMDL will also achieve the 30-day GM criterion when following South Dakota’s assessment
method. A similar conclusion was determined by EPA in An Approach for Using Load Duration
Curves in the Development of TMDLs (USEPA, 2007) using Michigan criteria as an example.
Once again, this outcome holds true for South Dakota’s limited contact recreation E. coli criteria
since they were simply derived as five times the immersion values.
Finally, while the SSM is associated with a single day of sampling and the GM is associated with
30 days of sampling, it is not technically appropriate to refer to them as “acute” and “chronic”
criteria. Those terms distinguish timeframes over which harm-to-use impacts develop, not the
sampling or averaging timeframe as with the SSM and GM. Acute refers to an effect that comes
about rapidly over short periods of time. Chronic refers to an effect that can build up over longer
periods, sometimes as long as the lifetime of a subject. In the case of E. coli, gastrointestinal illness
develops within a matter of hours to days. Both the SSM and GM are derived from this same
timeframe and based on the same underlying illness rate, thus treating the SSM as an acute criterion
and assuming it to be less stringent is incorrect. EPA recommends states use the GM and SSM
together, rather than just the GM or just the SSM, to judge whether water quality is protective of
recreational uses. SDDENR follows these guidelines and only relies on one criterion when forced
by data availability.
17
Appendix B
EPA Approval Letter and Decision Document
Ref: 8WD-CWS
SENT VIA EMAILDIGITAL READ RECEIPT REQUESTED
Hunter Roberts, SecretarySouth Dakota Department of Environment & Natural ResourcesHunter.Roberts@state.sd.us
Re: Approval of Escherichia coli Total Maximum Daily Loads (TMDLs) Converted from Existing Fecal Coliform TMDLs in South Dakota
Dear Mr. Roberts,
The U.S. Environmental Protection Agency (EPA) has completed review of the total maximum daily loads (TMDLs) submitted by your office on October 13th, 2020. In accordance with the Clean Water Act (33 U.S.C. §1251 et. seq at 40 C.F.R. Part 130, the EPA hereby approves TMDLs for segments of Beaver Creek, Brule Creek, East Fork Vermillion River and Rapid Creek. The EPA has determined that the separate elements of the TMDLslisted in the enclosure adequately address the pollutant of concern, are designed to attain and maintain applicable water quality standards, consider seasonal variation and includes a margin of safety. The
enclosure.
Thank you for submitting these TMDLs for our review and approval. If you have any questions, please contact Peter Brumm on my staff at (406) 457-5029.
Sincerely,
Judy Bloom, ManagerClean Water Branch
Enclosure:E. coli TMDLs Converted from Existing Fecal Coliform TMDLs in South Dakota EPA Decision Rationale
Cc: Barry McLaury, Administrator, South Dakota Department of Environment and Natural ResourcesPaul Lorenzen, Environmental Scientist III, South Dakota Department of Environment & Natural Resources
UNITED STATES ENVIRONMENTAL PROTECTION AGENCYREGION 8
1595 Wynkoop StreetDenver, CO 80202-1129
Phone 800-227-8917www.epa.gov/region08
EPA TOTAL MAXIMUM DAILY LOAD (TMDL) REVIEW SUMMARY TMDL: E. coli TMDLs Converted from Existing Fecal Coliform TMDLs in South Dakota ATTAINS TMDL ID: R8-SD-2021-01 LOCATION: Minnehaha, McCook, Pennington, and Union Counties, South Dakota IMPAIRMENTS/POLLUTANTS: The TMDL submittal addresses four river segments with an immersion recreation and/or limited contact recreation use that is impaired due to high concentrations of E. coli bacteria. Waterbody/Pollutant Addressed in this TMDL Action Assessment Unit ID Waterbody Description Pollutants Addressed SD-BS-R-BEAVER_02 Beaver Creek (Split Rock Creek to South Dakota-
Minnesota Border) Escherichia coli (E. coli)
SD-BS-R-BRULE_01 Brule Creek (From the Confluence of East and West Brule Creek to the Confluence with the Big Sioux River)
Escherichia coli (E. coli)
SD-CH-R-RAPID_04 Rapid Creek (From RC WWTF to Above Farmingdale)
Escherichia coli (E. coli)
SD-VM-R-VERMILLION_E_FORK_01
East Fork Vermillion River (From McCook/Lake County Line to Its Confluence with Little Vermillion River)
Escherichia coli (E. coli)
BACKGROUND: The South Dakota Department of Environment and Natural Resources (DENR) submitted to EPA final E. coli TMDLs for segments of Beaver Creek, Brule Creek, East Fork Vermillion River and Rapid Creek with a letter requesting review and approval dated October 13th, 2020. The submittal included: Letter requesting EPA’s review and approval of the TMDL Final TMDL report
All four river segments are subject to existing fecal coliform TMDLs approved by EPA separately between 2008 and 2012. During that time period, South Dakota adopted, and was implementing across several programs, EPA’s 2012 Recreational Water Quality Criteria which recommended states establish E. coli criteria after scientific advancements had demonstrated E. coli was a better indicator of fecal contamination and recreational harm than fecal coliform (USEPA, 2012). That process in South Dakota involved adopting new criteria for E. coli, maintaining dual criteria for several years to facilitate the transition and allow for the collection of additional E. coli data, and eventually dropping the fecal coliform criteria. The intent of this submittal is to revisit existing fecal coliform TMDLs, demonstrate through conversion that the TMDLs are protective of newer E. coli criteria, and apply the converted TMDLs to address current E. coli impairments. Much of the data, maps, figures, assumptions and analyses relied upon for this submittal are contained in the original fecal coliform TMDLs which are referenced on page 2.
2
APPROVAL RECOMMENDATIONS: Based on the review presented below, the reviewer recommends approval of the final E. coli TMDLs for segments of Beaver Creek, Brule Creek, East Fork Vermillion River and Rapid Creek. All the required elements of an approvable TMDL have been met.
TMDL Approval Summary Number of TMDLs Approved: 4 Number of Causes Addressed by TMDLs: 4
REVIEWERS: Peter Brumm, EPA The following review summary explains how the TMDL submission meets the statutory and regulatory requirements of TMDLs in accordance with Section 303(d) of the Clean Water Act (CWA), and EPA’s implementing regulations in 40 C.F.R. Part 130.
3
EPA TMDL REVIEW FOR E. COLI TMDLS CONVERTED FROM EXISTING FECAL COLIFORM TMDLS IN SOUTH DAKOTA This TMDL review document includes EPA’s guidelines that summarize the currently effective statutory and regulatory requirements relating to TMDLs (CWA Section 303(d) and 40 C.F.R. Part 130). These TMDL review guidelines are not themselves regulations. Any differences between these guidelines and EPA's regulations should be resolved in favor of the regulations themselves. The italicized sections of this document describe the information generally necessary for EPA to determine if a TMDL submittal fulfills the legal requirements for approval. The sections in regular type reflect EPA's analysis of the State’s compliance with these requirements. Use of the verb “must” below denotes information that is required to be submitted because it relates to elements of the TMDL required by the CWA and by regulation.
1. Identification of Waterbody, Pollutant of Concern, Pollutant Sources, and Priority Ranking
The TMDL submittal must clearly identify (40 C.F.R. §130.7(c)(1)): • the waterbody as it appears on the State’s/Tribe’s 303(d) list; • the pollutant for which the TMDL is being established; and • the priority ranking of the waterbody.
The TMDL submittal must include (40 C.F.R. §130.7(c)(1); 40 C.F.R. §130.2):
• an identification of the point and nonpoint sources of the pollutant of concern, including location of the source(s) and the quantity of the loading (e.g., lbs. per day);
• facility names and NPDES permit numbers for point sources within the watershed; and • a description of the natural background sources, and the magnitude and location of the sources, where
it is possible to separate natural background from nonpoint sources. This information is necessary for EPA’s review of the load and wasteload allocations, which are required by regulation. The TMDL submittal should also contain a description of any important assumptions made in developing the TMDL, such as:
• the spatial extent of the watershed in which the impaired waterbody is located; • the assumed distribution of land use in the watershed (e.g., urban, forested, agriculture); • population characteristics, wildlife resources, and other relevant information affecting the
characterization of the pollutant of concern and its allocation to sources; • present and future growth trends, if taken into consideration in preparing the TMDL (e.g., the TMDL
could include the design capacity of a wastewater treatment facility); and • an explanation and analytical basis for expressing the TMDL through surrogate measures, if
applicable. Surrogate measures are parameters such as percent fines and turbidity for sediment impairments; chlorophyll a and phosphorus loadings for excess algae; length of riparian buffer; or number of acres of best management practices.
Beaver Creek (SD-BS-R-BEAVER_02) is located in eastern South Dakota within the Big Sioux River Basin as shown in Appendix AAA Figure 2 of the original fecal coliform TMDL (DENR, 2004). This segment was first listed as impaired by E. coli on the 2014 303(d) List and was assigned a high priority for TMDL development on the most recent 303(d) list in 2020. DENR addressed a Total Suspended Solids (TSS) impairment on this segment in the same 2004 TMDL report that addressed fecal coliform and there are no other impairments currently identified. According to the original fecal coliform TMDL, which summarized information at the larger Split Rock Creek Watershed, land use in the watershed is
4
predominantly cropland covering 76% of the total area (see page 183) and sources are attributed to agriculture and the City of Valley Springs Wastewater Treatment Facility (NPDES Permit #SD0020923) (see page 73) (DENR, 2004). This submittal further characterizes this point source on page 7. Brule Creek (SD-BS-R-BRULE_01) is located in south-eastern South Dakota within the larger Big Sioux River Basin as shown in Figure 1 of the original fecal coliform TMDL (DENR, 2011). This segment was first listed as impaired by E. coli on the 2014 303(d) List and was assigned a high priority for TMDL development on the most recent 303(d) list in 2020. DENR is working to address another impairment on this waterbody for TSS in a separate TMDL study. According to the original fecal coliform TMDL, land use in the watershed is predominantly row crop which covers 71% of the total area (see Section 1.1) and sources are attributed to nonpoint categories such as agriculture, wildlife, and tributaries (see Section 3.0) (DENR, 2011). This submittal includes the characterization of an additional point source, L.G. Everist, Inc. (NPDES Permit #SD0027928), on page 7. Rapid Creek (SD-CH-R-RAPID_04) is located in western South Dakota within the Cheyenne River Basin as shown in Figure 2 of the original fecal coliform TMDL (DENR, 2010). This segment was first listed as impaired by E. coli on the 2014 303(d) List and was assigned a high priority for TMDL development on the most recent 303(d) list in 2020. There are no other impairments currently identified for this segment. According to the original fecal coliform TMDL, land use in the watershed is predominantly herbaceous rangeland which covers 61% of the total area (see Section 1.4) and sources are attributed to the Rapid City Wastewater Treatment Facility (NPDES Permit #SD0023574) and nonpoint categories such as agriculture, human, and wildlife (see Section 4.2) (DENR, 2010). This submittal includes the characterization of two additional point sources: the Rapid City Regional Airport Wastewater Facility (NPDES Permit #SD0028638) and the Rapid Valley Sanitary District (NPDES Permit #SDG86007) on page 7. East Fork Vermillion River (SD-VM-R-VERMILLION_E_FORK_01), formally assessment unit SD-VM-R-VERMILLION_EAST_FORK_01, is located in eastern South Dakota within the Vermillion River Basin as shown in Figure 1 of the original fecal coliform TMDL (DENR, 2012). This segment was first listed as impaired by E. coli on the 2016 303(d) List and was assigned a high priority for TMDL development on the most recent 303(d) list in 2020. There are no other impairments currently identified for this segment. According to the original fecal coliform TMDL, land use in the watershed is predominantly cultivated crops which cover 67% of the total area (see Table 2) and sources are attributed to nonpoint categories such as agriculture, humans, wildlife, and tributaries (see Section 6.2) (DENR, 2012). This submittal includes the characterization of two additional point sources for Concentrated Animal Feeding Operations (CAFOs) on page 7. Assessment: EPA concludes that DENR adequately identified the impaired waterbodies, the pollutant of concern, the priority ranking, the identification, location and magnitude of the pollutant sources, and the important assumptions and information used to develop the TMDLs.
5
2. Description of the Applicable Water Quality Standards and Numeric Water Quality Target
The TMDL submittal must include: • a description of the applicable State/Tribal water quality standard, including the designated use(s) of
the waterbody, the applicable numeric or narrative water quality criterion, and the antidegradation policy (40 C.F.R. §130.7(c)(1)); and
• a numeric water quality target for each TMDL. If the TMDL is based on a target other than a numeric water quality criterion, then a numeric expression must be developed from a narrative criterion and a description of the process used to derive the target must be included in the submittal (40 C.F.R. §130.2(i)).
EPA needs this information to review the loading capacity determination, and load and wasteload allocations, which are required by regulation.
Page 3 discusses water quality standards and TMDL targets with citations to relevant South Dakota regulations. The designated uses applicable to the four waterbodies subject to TMDLs in this submittal are noted in the table below. Limited contact recreation is impaired by E. coli in all four waterbodies. The immersion contact recreation use is also impaired by E. coli in segment 4 of Rapid Creek. In addition, the warmwater aquatic life use in segment 1 of Brule Creek is considered non-supporting due to TSS which will be addressed by a separate TMDL effort. Assessment Unit ID *Designated Uses Pollutant(s) Causing
Impairment SD-BS-R-BEAVER_02 Warmwater Marginal Fish Life Propagation
Limited Contact Recreation Escherichia coli (E. coli) Fish and Wildlife Propagation, Recreation, and Stock Watering
Irrigation SD-BS-R-BRULE_01 Warmwater Marginal Fish Life Propagation Total Suspended Solids (TSS)
Limited Contact Recreation Escherichia coli (E. coli) Fish and Wildlife Propagation, Recreation, and Stock Watering
Irrigation SD-CH-R-RAPID_04 Warmwater Marginal Fish Life Propagation
Immersion Recreation Escherichia coli (E. coli) Limited Contact Recreation Escherichia coli (E. coli) Fish and Wildlife Propagation, Recreation, and Stock Watering
Irrigation SD-VM-R-VERMILLION_E_FORK_01
Warmwater Marginal Fish Life Propagation Limited Contact Recreation Escherichia coli (E. coli) Fish and Wildlife Propagation, Recreation, and Stock Watering
Irrigation *Shaded rows denote impaired designated uses according to South Dakota’s 2020 303(d) list (DENR, 2020) The numeric criteria applicable to these uses are presented in Table 2 and are comprised of a 30-day geometric mean criterion and a single sample maximum criterion which are seasonally applicable from May 1 to September 30. The limited contact recreation criteria are directly applied as water quality
6
targets for all TMDLs except segment 4 of Rapid Creek where the more stringent immersion recreation criteria were applied. DENR expects that meeting the numeric E. coli criteria will lead to conditions necessary to support relevant narrative criteria. Appendix A of the submittal discusses how the geometric mean criterion and the single sample maximum criterion are equally protective of the recreation uses, therefore the varying precedents set in the original fecal coliform TMDLs for selection of TMDL targets was repeated for these E. coli TMDLs. Assessment: EPA concludes that DENR adequately described the applicable water quality standards and numeric water quality targets for these TMDLs. 3. Loading Capacity - Linking Water Quality and Pollutant Sources
The TMDL submittal must include the loading capacity for each waterbody and pollutant of concern. EPA regulations define loading capacity as the greatest amount of a pollutant that a water can receive without violating water quality standards (40 C.F.R. §130.2(f)). The TMDL submittal must:
• describe the method used to establish the cause-and-effect relationship between the numeric target and the identified pollutant sources. In many instances, this method will be a water quality model;
• contain documentation supporting the TMDL analysis, including the basis for any assumptions; a discussion of strengths and weaknesses in the analytical process; and results from any water quality modeling; and
• include a description and summary of the water quality data used for the TMDL analysis. EPA needs this information to review the loading capacity determination, and load and wasteload allocations, which are required by regulation (40 C.F.R. §130.2). The full water quality dataset should be made available as an appendix to the TMDL or as a separate electronic file. Other datasets used (e.g., land use, flow), if not included within the TMDL submittal, should be referenced by source and year. The TMDL analysis should make use of all readily available data for the waterbody unless the TMDL writer determines that the data are not relevant or appropriate. The pollutant loadings may be expressed as either mass-per-time, toxicity or other appropriate measure (40 C.F.R. §130.2(i)). Most TMDLs should be expressed as daily loads (USEPA. 2006a). If the TMDL is expressed in terms other than a daily load (e.g., annual load), the submittal should explain why it is appropriate to express the TMDL in the unit of measurement chosen. The TMDL submittal must describe the critical conditions and related physical conditions in the waterbody as part of the analysis of loading capacity (40 C.F.R. §130.7(c)(1)). The critical condition can be thought of as the “worst case” scenario of environmental conditions (e.g., stream flow, temperature, loads) in the waterbody in which the loading expressed in the TMDL for the pollutant of concern will continue to meet water quality standards. TMDLs should define the applicable critical conditions and describe the approach used to estimate both point and nonpoint source loads under such critical conditions.
All four original fecal coliform TMDLs relied on the load duration curve approach to define the fecal coliform loading capacity of each stream. Consequently, these E. coli TMDLs, which are based on the fecal confirm TMDL analyzes, used the same approach to establish E. coli loading capacities. A load duration curve is a graphic representation of pollutant loads across various flows. The approach helps correlate water quality conditions to stream flow and provides insight into the variability of source
7
contributions. EPA has published guidance on the use of duration curves for TMDL development (USEPA, 2007) and the practice is well established. Using this approach, DENR set the TMDL equivalent to the loading capacity and expressed the TMDL in colony forming units (CFU) per day at a number of different flow zones as listed in Tables 8, 10, 12, 14 and 16. DENR calculated E. coli TMDLs by multiplying the existing fecal coliform TMDLs by the ratio between the applicable bacteria standards. For example, the East Fork Vermillion River’s fecal coliform TMDL was derived using the target of 1,000 CFU/100mL. The applicable E. coli criterion is now 630 CFU/100mL. Thus, the River’s E. coli TMDL was established by multiplying the original fecal coliform TMDL by 0.63 (i.e., 630/1,000). This approach is equivalent to establishing the E. coli TMDL using the E. coli criterion as the TMDL target and a consistent same stream flow value. The E. coli allocations were subsequently based on the percent contribution established by the original fecal coliform TMDLs. For example, the East Fork Vermillion River’s high flow zone fecal coliform LA was 90% of the TMDL (i.e., 5.55x1013 CFU/day x 0.9 = 5.00x1013 CFU/day) therefore the high flow E. coli LA was established in the same manner (i.e. 3.50x1013 CFU/day x 0.9 = 3.15x1013 CFU/day). In this submittal, DENR verified that the bacterial source assessment and linkage analysis was still accurate and that because conditions had not changed from the previous TMDL submittals, it was acceptable to rely on the fecal coliform loading capacities and allocation schemes for new E. coli TMDLs. Full water quality datasets are included in the original fecal coliform TMDLs. See Appendix B for Beaver Creek (DENR, 2004), Table 5 for Brule Creek (DENR, 2011), and Appendix A for East Fork Vermillion River (DENR, 2012) and Rapid Creek (DENR, 2010). Assessment: EPA concludes that loading capacities were calculated using an acceptable approach, used a water quality targets consistent with water quality criteria, and have been appropriately set at levels necessary to attain and maintain the applicable water quality standards. The pollutant loads have been expressed as daily loads. The critical conditions were described and factored into the calculations and were based on a reasonable approach to establish the relationship between the target and pollutant sources. 4. Load Allocation
The TMDL submittal must include load allocations (LAs). EPA regulations define LAs as the portion of a receiving water's loading capacity that is attributed either to one of its existing or future nonpoint sources of pollution and to natural background sources. Load allocations may range from reasonably accurate estimates to gross allotments (40 C.F.R. §130.2(g)). Where possible, separate LAs should be provided for natural background and for nonpoint sources. In the rare instance that a TMDL concludes that there are no nonpoint sources or natural background for a pollutant, the load allocation must be expressed as zero and the TMDL should include a discussion of the reasoning behind this decision.
The E. coli LAs were based on the percent contribution established by the original fecal coliform TMDLs. For example, the East Fork Vermillion River’s high flow zone fecal coliform LA was 90% of the TMDL (i.e., 5.55x1013 CFU/day x 0.9 = 5.00x1013 CFU/day) therefore the high flow E. coli LA was established in the same manner (i.e. 3.50x1013 CFU/day x 0.9 = 3.15x1013 CFU/day). DENR originally established a single LA for each stream as the allowable load remaining after the WLA and explicit
8
MOS were accounted for (i.e., LA = TMDL – WLA – MOS). For Brule Creek and the East Fork Vermillion River where the WLA equals zero, the calculation can be simplified as LA = TMDL – MOS. Tables 8, 10, 12, 14, and 16 present the LAs across various flow zones. These composite LAs represent all nonpoint source contributions, both human and natural, as one allocation, however, individual nonpoint source categories were characterized in greater depth in the original fecal coliform TMDLs. Assessment: EPA concludes that the LAs provided in the TMDL submittal are reasonable and will result in attainment of the water quality standards. 5. Wasteload Allocations
The TMDL submittal must include wasteload allocations (WLAs). EPA regulations define WLAs as the portion of a receiving water's loading capacity that is allocated to existing and future point sources (40 C.F.R. §130.2(h)). If no point sources are present or if the TMDL recommends a zero WLA for point sources, the WLA must be expressed as zero. If the TMDL recommends a zero WLA after considering all pollutant sources, there must be a discussion of the reasoning behind this decision, since a zero WLA implies an allocation only to nonpoint sources and natural background will result in attainment of the applicable water quality standards, and all point sources have no measurable contribution. The individual WLAs may take the form of uniform percentage reductions or individual mass based limitations for dischargers where it can be shown that this solution meets WQSs and does not result in localized impairments. In some cases, WLAs may cover more than one discharger (e.g., if the source is contained within a general permit).
The Beaver Creek fecal coliform TMDL established a WLA for the City of Valley Springs Wastewater Treatment Facility (NPDES Permit #SD0020923) based on the facility’s average design flow multiplied by the single sample maximum fecal coliform criterion (2,000 CFU/100mL) and a unit conversion factor (DENR, 2004). As explained on page 7, the E. coli WLA was derived from the same calculation with the single sample maximum E. coli criterion (1,178 CFU/100mL) subsituted for the fecal coliform criterion. The E. coli WLA is presented in Table 12 as a continuous load across all three flow zones, however, the facility has only discharged once in the last twenty years. Therefore, the actual E. coli load contributed to Beaver Creek is considered minimal. The Brule Creek fecal coliform TMDL did not establish a WLA (i.e., WLA = 0) (DENR, 2011). As explained on page 7, this submittal characterized and considered establishing a WLA for L.G. Everist, Inc. (NPDES Permit #SD0027928). Ultimately, DENR decided E. coli is not a pollutant of concern in the gravel pit’s intermittent discharge and no E. coli WLA was established (i.e., WLA = 0). The Rapid Creek fecal coliform TMDL established a WLA for the Rapid City Wastewater Reclamation Facility (NPDES Permit #SD0023574) based on the facility’s average design flow multiplied by fecal coliform criteria and a unit conversion factor (DENR, 2010). Fecal coliform WLAs were calculated for both the single sample maximum and geometric mean criterion. As explained on page 7, E. coli WLAs were derived from the same calculations with the E. coli criteria subsituted for the fecal coliform criteria. The E. coli WLAs are presented in Tables 14 and 16 as continuous loads across all five flow zones. This submittal characterized and considered establishing WLAs for two additional point sources: the Rapid City Regional Airport Wastewater Facility (NPDES Permit #SD0028638) and the Rapid Valley Sanitary District (NPDES Permit #SDG86007). Ultimately, DENR decided to not establish E.
9
coli WLAs for these point sources because the Regional Airport Wastewater Facility is a “non-discharging” facility that is only allowed to discharge during an emergency and the Rapid Valley Sanitary District is a rural drinking water system where E. coli is not considered a pollutant of concern in the facility’s discharge. The East Fork Vermillion River fecal coliform TMDL did not establish a WLA (i.e., WLA = 0) (DENR, 2012). As explained on page 7, this submittal characterized and considered establishing WLAs for two CAFOs: the Christopher Katzer Swine Facility (NPDES Permit #SDG100150) and Orland Hutterian Brethren, Inc. (NPDES Permit #SDG109100). Ultimately, DENR determined E. coli WLAs were unnecessary as long as these facilities comply with the existing general CAFO permit requirements to ensure their discharges are unlikely, minimal and indirect loading events. After reviewing current information on point sources in the TMDL watersheds, DENR determined the WLA component of the existing fecal coliform TMDLs did not merit an expansion or reduction. Therefore, because the ratios among all allocations remained the same, including the WLAs, LAs and MOSs, DENR could summarize that E. coli WLAs were established at the same percent contributions established by the original fecal coliform TMDLs. For example, Rapid Creek’s high flow zone fecal coliform WLA was 3.5% of the TMDL (i.e., 2.27x1011 CFU/day ÷ 6.47x1012 CFU/day = 0.035) therefore the high flow E. coli WLA was established at the same proportion (i.e. 3.80x1012 CFU/day x 0.035 = 1.33x1011 CFU/day). The E. coli WLAs are not expected to add new requirements or implementation expectations to permits where bacteria effluent limits already exist. Assessment: EPA concludes that the WLAs provided in the TMDL submittal are reasonable, will result in the attainment of the water quality standards and will not cause localized impairments. The TMDLs account for all point sources contributing loads to impaired segments, upstream segments and tributaries in the watersheds. Where DENR recommends a WLA of zero, EPA concludes the decision was justified and reasonable. 6. Margin of Safety
The TMDL submittal must include a margin of safety (MOS) to account for any lack of knowledge concerning the relationship between load allocations, wasteload allocations and water quality (CWA §303(d)(1)(C), 40 C.F.R. §130.7(c)(1)). The MOS may be implicit or explicit. If the MOS is implicit, the conservative assumptions in the analysis that account for the MOS must be described. If the MOS is explicit, the loading set aside for the MOS must be identified.
The TMDLs contained in this submittal rely on the explicit MOS approach. As described on page 8, all E. coli allocations, including the MOS, were derived using the same relative contribution established by the original fecal coliform TMDL allocations. For Beaver Creek and the East Fork Vermillion River, 10% of the TMDL was assigned to the explicit MOS. For Brule Creek, the MOS was calculated as the difference between the loading capacity at the mid-point and minimum of each of flow zone. For Rapid Creek, the MOS was calculated as the difference between the loading capacity at the 25th and 75th percentile of each of flow zone. The explicit MOS for each E. coli TMDL varies by flow zone and is included in Tables 8, 10, 12, 14, and 16.
10
Assessment: EPA concludes that the TMDL submittal incorporates an adequate explicit margin of safety. 7. Seasonal Variation
The TMDL submittal must be established with consideration of seasonal variations. The method chosen for including seasonal variations in the TMDL must be described (CWA §303(d)(1)(C), 40 C.F.R. §130.7(c)(1)).
The load duration curve method used to establish these TMDLs incorporates variations in stream flow, which in turn, is influenced by other climatic and human factors that change throughout the year. To account for these variations, DENR developed TMDLs at different flow zones as listed in Tables 8, 10, 12, 14, and 16. Additional insight into seasonal variations of bacteria concentrations, stream flow, and pollutant sources are provided in the original fecal coliform TMDLs. Refer to page 188 for the discussion pertaining to Beaver Creek (DENR, 2004), Section 9.0 for East Fork Vermillion River (DENR, 2012) and Section 6.2 for Brule Creek (DENR, 2011) and Rapid Creek (DENR, 2010). Assessment: EPA concludes that seasonal variations were adequately described and considered to ensure the TMDL allocations will be protective of the applicable water quality standards throughout any given year. 8. Reasonable Assurances
When a TMDL is developed for waters impaired by both point and nonpoint sources, EPA guidance (USEPA. 1991) and court decisions say that the TMDL must provide reasonable assurances that nonpoint source control measures will achieve expected load reductions in order for the TMDL to be approvable. This information is necessary for EPA to determine that the TMDL, including the load and wasteload allocations, has been established at a level necessary to implement the applicable water quality standards (CWA §303(d)(1)(C), 40 C.F.R. §130.7(c)(1)). EPA guidance (USEPA. 1997) also directs Regions to work with States to achieve TMDL load allocations in waters impaired only by nonpoint sources. However, EPA cannot disapprove a TMDL for nonpoint source-only impaired waters, which do not have a demonstration of reasonable assurance that LAs will be achieved, because such a showing is not required by current regulations.
Nonregulatory, voluntary-based reasonable assurances are provided for the LAs in the original fecal coliform TMDLs where DENR generally discussed the State’s adaptive management approach to the TMDL process, monitoring strategies that will be used to gage TMDL effectiveness in the future, and core aspects of TMDL implementation strategies. These assurances commonly include the recommendation of specific activities to focus implementation, the identification of watershed partners with shared interests in water quality, and the identification of several potential funding sources. Assessment: EPA considered the reasonable assurances contained in the TMDL submittal and concludes that they are adequate to meet the load reductions.
11
9. Monitoring Plan
The TMDL submittal should include a monitoring plan for all: • Phased TMDLs; and • TMDLs with both WLA(s) and LA(s) where reasonable assurances are provided.
Under certain circumstances, a phased TMDL should be developed when there is significant uncertainty associated with the selection of appropriate numeric targets, estimates of source loadings, assimilative capacity, allocations or when limited existing data are relied upon to develop a TMDL. EPA guidance (USEPA. 2006b) recommends that a phased TMDL submittal, or a separate document (e.g., implementation plan), include a monitoring plan, an explanation of how the supplemental data will be used to address any uncertainties that may exist when the phased TMDL is prepared and a scheduled timeframe for revision of the TMDL. For TMDLs that need to provide reasonable assurances, the monitoring plan should describe the additional data to be collected to determine if the load reductions included in the TMDL are occurring and leading to attainment of water quality standards. EPA guidance (USEPA. 1991) recommends post-implementation monitoring for all TMDLs to determine the success of the implementation efforts. Monitoring plans are not a required part of the TMDL and are not approved by EPA but may be necessary to support the decision rationale for approval of the TMDL.
DENR recognizes that during and after implementation of best management practices, monitoring will be necessary to measure attainment of water quality standards as prescribed in the TMDL submittal. This will generally be accomplished through DENR’s ambient water quality monitoring program at the same stations where data was collected to develop these TMDLs. DENR also maintains the ability to modify the TMDLs and allocations as new data becomes available using an adaptive management approach in accordance with the TMDL revision process previously recommended by EPA. The original fecal coliform TMDLs include additional stream-specific future monitoring recommendations. Refer to page 211 for the discussion pertaining to Beaver Creek (DENR, 2004), Section 11.0 for East Fork Vermillion River (DENR, 2012), Section 8.0 for Brule Creek (DENR, 2011) and Section 10.0 for Rapid Creek (DENR, 2010). Assessment: Monitoring plans are not a required element of EPA’s TMDL review and decision-making process. EPA is taking no action on the monitoring strategy included in the TMDL submittal. 10. Implementation
EPA policy (USEPA. 1997) encourages Regions to work in partnership with States/Tribes to achieve nonpoint source load allocations established for 303(d)-listed waters impaired by nonpoint sources. Regions may assist States/Tribes in developing implementation plans that include reasonable assurances that nonpoint source LAs established in TMDLs for waters impaired solely or primarily by nonpoint sources will in fact be achieved. The policy recognizes that other relevant watershed management processes may be used in the TMDL process. EPA is not required to and does not approve TMDL implementation plans. EPA encourages States/Tribes to include restoration recommendations (e.g., framework) in all TMDLs for stakeholder and public use to guide future implementation planning. This could include identification of a range of potential management measures and practices that might be feasible for addressing the main loading
12
sources in the watershed (see USEPA. 2008b, Chapter 10). Implementation plans are not a required part of the TMDL and are not approved by EPA but may be necessary to support the decision rationale for approval of the TMDL.
In general, based on the makeup of contributing pollutant sources within the watersheds and the fact that nonpoint source contributions require the greatest reductions, DENR encourages that future implementation activities focus on:
• Reducing direct livestock access to streams by increasing alternative watering sources • Enhancing riparian vegetation to stabilize streambanks and filter runoff • Installing filter strips along abutting cropland and pastureland • Properly managing animal waste management systems at confinement facilities • Continuing Rapid City’s septic system inspection and repair program • Assessing the impact of CWA Section 319 projects and revising plans in cooperation with
stakeholders The original fecal coliform TMDLs include additional implementation recommendations such as descriptions of potential funding sources and previously completed restoration projects. Refer to page 213 for the discussion pertaining to Beaver Creek (DENR, 2004), Section 13.0 for East Fork Vermillion River (DENR, 2012), Section 9.0 for Brule Creek (DENR, 2011) and Section 12.0 for Rapid Creek (DENR, 2010). Assessment: Although not a required element of the TMDL approval, DENR discussed how information derived from the TMDL analysis process can be used to support implementation of the TMDL. EPA is taking no action on the implementation portion of the TMDL submittal. 11. Public Participation
EPA policy is that there must be full and meaningful public participation in the TMDL development process. Each State/Tribe must, therefore, provide for public participation consistent with its own continuing planning process and public participation requirements (40 C.F.R. §25.3 and §130.7(c)(1)(ii)). The final TMDL submittal must describe the State/Tribe’s public participation process, including a summary of significant comments and the State/Tribe’s responses to those comments (40 C.F.R. §25.3 and §25.8). Inadequate public participation could be a basis for disapproving a TMDL; however, where EPA determines that a State/Tribe has not provided adequate public participation, EPA may defer its approval action until adequate public participation has been provided for, either by the State/Tribe or by EPA.
The submittal explains the public engagement process DENR followed during development of the TMDL on page 11. A draft TMDL report was released for public comment from September 4th, 2020 to October 8th, 2020. No public comments were submitted. The opportunity for public review and comment was posted on DENR’s website and announced in several local newspapers within watersheds of the four impaired stream segments. Assessment: EPA has reviewed DENR’s public participation process and concludes that DENR involved the public during the development of the TMDL and provided adequate opportunities for the public to comment on the draft report.
13
12. Submittal Letter
The final TMDL submittal must be accompanied by a submittal letter that explicitly states that the submittal is a final TMDL submitted under Section 303(d) of the Clean Water Act for EPA review and approval. This clearly establishes the State’s/Tribe’s intent to submit, and EPA’s duty to review, the TMDL under the statute (40 C.F.R. §130.7(d)(1)). The final submittal letter should contain such identifying information as the waterbody name, location, assessment unit number and the pollutant(s) of concern.
A transmittal letter with the appropriate information was included with the final TMDL report submission from DENR, dated October 13th, 2020 and signed by Paul Lorenzen, Environmental Scientist Manager 1, Water Protection Program. Assessment: EPA concludes that the State’s submittal package clearly and unambiguously requested EPA to act on the TMDLs in accordance with the Clean Water Act and the submittal contained all necessary supporting information.
14
References DENR. 2004. Phase I Watershed Assessment Final Report and TMDL, Central Big Sioux River, Brookings, Lake, Moody, and Minnehaha Counties, South Dakota. Division of Financial and Technical Assistance, South Dakota Department of Environment and Natural Resources, Pierre, South Dakota. DENR. 2010. Fecal Coliform, Escherichia coli Bacteria Total Maximum Daily Loads (TMDLs) for Lower Rapid Creek, Pennington County, South Dakota. Water Resources Assistance Program, South Dakota Department of Environment and Natural Resources, Pierre, South Dakota. DENR. 2011. Fecal Coliform Bacteria Total Maximum Daily Load Evaluation for Brule Creek, Union County, South Dakota. Water Resources Assistance Program, South Dakota Department of Environment and Natural Resources, Pierre, South Dakota. DENR. 2012. Pathogen Total Maximum Daily Load (TMDL) for One Segment of the East Fork of the Vermillion River South Dakota. Water Resources Assistance Program, South Dakota Department of Environment and Natural Resources, Pierre, South Dakota. DENR. 2020. The 2020 South Dakota Integrated Report for Surface Water Quality Assessment. South Dakota Department of Environment and Natural Resources, Pierre, South Dakota. USEPA. 1991. Guidance for water quality-based decisions: The TMDL process. EPA 440-4-91-001. Office of Water, Assessment and Watershed Protection Division and Office of Wetlands, Oceans, and Watersheds, U.S. Environmental Protection Agency, Washington, DC. USEPA. 1997. New policies for establishing and implementing Total Maximum Daily Loads (TMDLs). Office of Water, U.S. Environmental Protection Agency, Washington, DC. USEPA. 2000. Bacterial Indicator Tool User’s Guide. EPA-823-B-01-003. Office of Water, U.S. Environmental Protection Agency, Washington, DC. USEPA. 2001. Protocol for Developing Pathogen TMDLs. EPA 841-R-00-002. Office of Water, U.S. Environmental Protection Agency, Washington, DC. USEPA. 2006a. Establishing TMDL "Daily" Loads in Light of the Decision by the U.S. Court of Appeals for the D.C. Circuit. Office of Water, Office of Wetlands, Oceans, and Watersheds, U.S. Environmental Protection Agency, Washington, DC. USEPA. 2006b. Clarification Regarding “Phased” Total Maximum Daily Loads. Office of Water, Office of Wetlands, Oceans, and Watersheds, U.S. Environmental Protection Agency, Washington, DC. USEPA. 2007. An Approach for Using Load Duration Curves in the Development of TMDLs. EPA-841-B-07-006. Office of Water, Office of Wetlands, Oceans and Watersheds, U.S. Environmental Protection Agency, Washington, DC. USEPA. 2008. Handbook for Developing Watershed Plans to Restore and Protect our Waters. EPA-841-B-08-002. Office of Water, Environmental Protection Agency, Washington, DC. USEPA. 2010. National Pollutant Discharge Elimination System (NPDES) Permit Writers’ Manual, Chapter 6, Water Quality-Based Effluent Limitations. EPA-833-K-10-001. Office of Water, Office of Wastewater Management, Water Permits Division, Washington, DC.
15
USEPA. 2012. Recreational Water Quality Criteria. OW-820-F-12-058. EPA Office of Water, Office of Science and Technology, Washington, DC. USEPA. 2014. Water Quality Standards Handbook: Chapter 1: General Provisions. EPA-820-B-14-008. EPA Office of Water, Office of Science and Technology, Washington, DC. USEPA. 2017. Water Quality Standards Handbook: Chapter 3: Water Quality Criteria. EPA-823-B-17-001. EPA Office of Water, Office of Science and Technology, Washington, DC.
Recommended