Mauvaise Terre Creek Watershed TMDL Report

              Mauvaise Terre Creek
Watershed
TMDL Report
Illinois
Environmental
Protection Agency
Bureau of Water
P. O. Box 19276
Springfield, IL 62794-9276 August 2007
IEPA/BOW/07-008
Printed on Recycled Paper
TABLE OF CONTENTS
STAGE 1 REPORT: Mauvaise Terre Creek, Mauvaise Terre Lake,
North Fork Mauvaise Terre Creek
First Quarterly Progress Report – Watershed Characterization
Second Quarterly Progress Report – Model Recommendation
Third Quarterly Progress Report – Monitoring Recommendation
Fourth Quarterly Progress Report – Public Participation
STAGE 2 REPORT: North Fork Mauvaise Terre Creek
Introduction
Field Investigation Overview
Water Sample Collection and Field Measurements
Discharge Measurements
Sediment Oxygen Demand and Continuous DO Monitoring
Quality Assurance Review
Attachment
Attachment A. Quality Assurance Project Plan (QAPP)
STAGE 3 FINAL APPROVED TMDL: Mauvaise Terre Creek,
Mauvaise Terre Lake
Introduction
Problem Identification
Required TMDL Elements
Watershed Characterization
Description of Applicable Standards and Numeric Targets
Development of Water Quality Models
TMDL Development
Public Participation and Involvement
Adaptive Implementation Process
References
Attachments
Model Files
Responsiveness Summary
IMPLEMENTATION PLAN, Mauvaise Terre Creek, Mauvaise Terre Lake
Summary
Introduction
Watershed Description
TMDL Summary
Implementation Approach
Implementation Alternatives
Identifying Priority Areas for Controls
Reasonable Assurance
Monitoring and Adaptive Management
References
STAGE 3 FINAL APPROVED TMDL, North Fork Mauvaise Terre Creek
Introduction
Problem Identification
Required TMDL Elements
Watershed Characterization
Description of Applicable Standards and Numeric Targets
Development of Water Quality Models
TMDL Development
Public Participation and Involvement
Adaptive Implementation Process
References
Attachments
Manganese Load Duration Curve
Implementation Plan
Responsiveness Summary
Final Stage 1 Progress Report
Prepared for Illinois Environmental Protection Agency
April 2005
Mauvaise Terre Creek Watershed
Mauvaise Terre Creek (ILDD04)
Mauvaise Terre Lake (SDL), North Fork Mauvaise Terre Creek
(DDC), Mauvaise Terre Creek (DD04)
Limno-Tech, Inc.
www.limno.com
This page is blank to facilitate double sided printing.
First Quarterly Progress Report
Prepared for Illinois Environmental Protection Agency
August 2004
Mauvaise Terre Creek Watershed:
Mauvaise Terre Creek (ILDD04)
Mauvaise Terre Lake (SDL), North Fork Mauvaise Terre Creek
(DDC), Mauvaise Terre Creek (DD04)
Limno-Tech, Inc.
www.limno.com
This page is blank to facilitate double sided printing.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page i
TABLE OF CONTENTS
EXECUTIVE SUMMARY ................................................................................................ 1
Background..................................................................................................................... 1
Methods .......................................................................................................................... 1
Results............................................................................................................................ 2
INTRODUCTION .............................................................................................................. 3
TMDL Process................................................................................................................ 3
Illinois Assessment and Listing Procedures ................................................................... 3
List of Identified Watershed Impairments ...................................................................... 4
WATERSHED CHARACTERIZATION .......................................................................... 5
Methods .......................................................................................................................... 5
Mauvaise Terre Creek Watershed Characterization ....................................................... 6
DATABASE DEVELOPMENT AND ANALYSIS ........................................................ 17
Data Sources and Methods ........................................................................................... 17
CONFIRMATION OF CAUSES AND SOURCES OF IMPAIRMENT ........................ 21
Mauvaise Terre Lake (SDL)......................................................................................... 21
North Fork Mauvaise Terre Creek (DDC).................................................................... 23
Mauvaise Terre Creek (DD 04) .................................................................................... 24
CONCLUSIONS............................................................................................................... 25
NEXT STEPS ................................................................................................................... 25
REFERENCES ................................................................................................................. 26
APPENDIX A. DATA SOURCES AND LOCAL CONTACTS..................................... 27
APPENDIX B: PHOTOGRAPHS.................................................................................... 31
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page ii
LIST OF TABLES
Table 1. Impaired waterbodies in the project watershed .................................................... 5
Table 2. Major Soil Associations in the Watershed.......................................................... 10
Table 3. Land Cover Distribution, Mauvaise Terre Lake Watershed............................... 14
Table 4. Land Cover Distribution, North Fork Mauvaise Terre Creek Watershed .......... 14
Table 5. Land Cover Distribution, Entire Mauvaise Terre Creek Watershed .................. 15
Table 6. Percent of Morgan County fields, by crop, with indicated tillage system......... 15
Table 7. Percent of Scott County fields, by crop, with indicated tillage system ............. 15
Table 8. Water quality data summary for Mauvaise Terre Lake (SDL)........................... 18
Table 9. Water quality data summary for North Fork Mauvaise Terre Creek (DDC)...... 19
Table 10. Water quality data summary for Mauvaise Terre Creek (DD04) ..................... 19
LIST OF FIGURES
Figure 1. Point source dischargers, impaired waterbody segments, and other watershed
characteristics.............................................................................................................. 8
Figure 2. Major soil associations in the Mauvaise Terre Watershed................................ 11
Figure 3. Current land cover in the project watershed...................................................... 16
Figure 4. Sampling stations in the project watershed ....................................................... 20
Figure 5. Total phosphorus vs. total suspended solids in Mauvaise Terre Lake .............. 21
Figure 6. Total phosphorus profiles in Mauvaise Terre Lake (near the dam) .................. 22
Figure 7. Fecal coliform and total suspended solids concentrations in Mauvaise Terre
Creek......................................................................................................................... 24
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 1
EXECUTIVE SUMMARY
This is the first in a series of quarterly status reports documenting work completed on the
Mauvaise Terre Creek project watershed. The objective of this report is to provide a
summary of Stage 1 work that will ultimately be used to support Total Maximum Daily
Load (TMDL) development in the project watershed.
Background
Section 303(d) of the 1972 Clean Water Act requires States to define impaired waters and
identify them on a list, which is referred to as the 303(d) list. The State of Illinois
recently issued the draft 2004 303(d) list (IEPA, 2004a), which is available on the web at:
http://www.epa.state.il.us/water/tmdl/303d-list.html. The Clean Water Act requires that a
Total Maximum Daily Load (TMDL) be completed for each pollutant listed for an
impaired waterbody. TMDLs are prepared by the States and submitted to the U.S. EPA.
In developing the TMDL, a determination is made of the greatest amount of a given
pollutant that a waterbody can receive without exceeding water quality standards and
designated uses, considering all known and potential sources. The TMDL also takes into
account a margin of safety, which reflects scientific uncertainty, as well as the effects of
seasonal variation.
As part of the TMDL process, the Illinois Environmental Protection Agency (IEPA) and
several consultant teams have compiled and reviewed data and information to determine
the sufficiency of available data to support TMDL development. As part of this review,
the data were used to confirm the impairments identified on the 303(d) list and to further
identify potential sources causing these impairments. The results of this review are
presented in this first quarterly status report.
Next, the Illinois EPA, with assistance from consultants, will recommend an approach for
the TMDL, including an assessment of whether additional data are needed to develop a
defensible TMDL.
Finally, Illinois EPA and consultants will conduct the TMDLs and will work with
stakeholders to implement the necessary controls to improve water quality in the
impaired waterbodies and meet water quality standards. It should be noted that the
controls for nonpoint sources (e.g., agriculture) would be strictly voluntary.
Methods
The effort completed in the first quarter included: 1) a site visit and collection of
information to complete a detailed watershed characterization; 2) development of a water
quality database and data analyses; and 3) synthesis of the watershed characterization
information and the data analysis results to confirm the sufficiency of the data to support
both the listing decision and the sources of impairment that are included on the draft 2004
303(d) list of impaired waterbodies.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 2
This evaluation focuses on the following waterbodies and associated sources of
impairment:
• Mauvaise Terre Lake: phosphorus, nitrate, manganese
• North Fork Mauvaise Terre Creek: low dissolved oxygen, manganese
• Mauvaise Terre Creek (below Town Brook): fecal coliform bacteria
Results
The available data, though in some cases very limited, support the listed impairments of
the three waterbodies in the Mauvaise Terre watershed. Potential sources of phosphorus
and nitrate to Mauvaise Terre Lake include agricultural sources, existing sediments,
recreation activities, and possibly failing private sewage disposal systems. The primary
source of manganese to both Mauvaise Terre Lake and North Fork Mauvaise Terre Creek
may be background sources due to naturally high concentrations in area soils; in-place
lake sediments may also contribute. The primary potential source of low dissolved
oxygen in North Fork Mauvaise Terre Creek is agricultural runoff. Potential sources of
fecal coliform bacteria to Mauvaise Terre Creek include livestock operations, agricultural
runoff, and sewage disposal, including municipal sewage, CSO discharges, and private
disposal systems.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 3
INTRODUCTION
This Stage 1 report describes initial activities related to the development of TMDLs for
impaired waterbodies in the Mauvaise Terre Creek watershed. Stage 1 efforts included
watershed characterization activities and data analyses, to confirm the causes and sources
of impairments in the watershed. This section provides some background information on
the TMDL process, and Illinois assessment and listing procedures. The specific
impairments in the Mauvaise Terre Creek watershed are also described.
TMDL Process
Section 303(d) of the 1972 Clean Water Act requires States to define impaired waters and
identify them on a list, which is called the 303(d) list. The State of Illinois recently
issued the draft 2004 303(d) list (IEPA 2004a), which is available on the web at:
http://www.epa.state.il.us/water/tmdl/303d-list.html. Section 303(d) of the Clean Water
Act and EPA's Water Quality Planning and Management Regulations (40 CFR Part 130)
require states to develop Total Maximum Daily Loads (TMDLs) for water bodies that are
not meeting designated uses under technology-based controls. The TMDL process
establishes the allowable loading of pollutants or other quantifiable parameters for a
water body based on the relationship between pollution sources and instream conditions.
This allowable loading represents the maximum quantity of the pollutant that the
waterbody can receive without exceeding water quality standards. The TMDL also takes
into account a margin of safety, which reflects scientific uncertainty, as well as the effects
of seasonal variation. By following the TMDL process, States can establish water
quality-based controls to reduce pollution from both point and nonpoint sources, and
restore and maintain the quality of their water resources (USEPA, 1991).
As part of the TMDL process, the Illinois Environmental Protection Agency (IEPA) and
several consultant teams have compiled and reviewed data and information to determine
the sufficiency of available data to support TMDL development. As part of this review,
the data were used to confirm the impairments identified on the 303(d) list and to further
identify potential sources causing these impairments. The results of this review are
presented in this first quarterly status report.
Next, the Illinois EPA, with assistance from consultants, will recommend an approach for
the TMDL, including an assessment of whether additional data are needed to develop a
defensible TMDL.
Finally, Illinois EPA and consultants will conduct the TMDLs and will work with
stakeholders to implement the necessary controls to improve water quality in the
impaired waterbodies and meet water quality standards. It should be noted that the
controls for nonpoint sources (e.g., agriculture) would be strictly voluntary.
Illinois Assessment and Listing Procedures
Water quality assessments in Illinois are based on a combination of chemical (water,
sediment and fish tissue), physical (habitat and flow discharge), and biological
(macroinvertebrate and fish) data. Illinois EPA conducts its assessment of water bodies
using a set of five generic designated use categories: public water supply, aquatic life,
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 4
primary contact (swimming), secondary contact (recreation), and fish consumption
(IEPA, 2004b). For each water body, and for each designated use applicable to the water
body, Illinois EPA’s assessment concludes one of three possible “use-support” levels:
• Fully supporting (the water body attains the designated use);
• Partially supporting (the water body attains the designated use at a reduced level);
or
• Not supporting (the water body does not attain the designated use).
All water bodies assessed as having partial or nonsupport attainment for any designated
use are identified as “impaired.” Waters identified as impaired based on biological
(macroinvertebrate, macrophyte, algal and fish), chemical (water, sediment and fish
tissue), and/or physical (habitat and flow discharge) monitoring data are placed on the
303(d) list. Potential causes and sources of impairment are also identified for impaired
waters.
Following the U.S. EPA regulations at 40 CFR Part 130.7(b)(4), the Illinois Section
303(d) list was prioritized on a watershed basis. Illinois EPA watershed boundaries are
based on the USGS ten-digit hydrologic units, to provide the state with the ability to
address watershed issues at a manageable level and document improvements to a
watershed’s health (IEPA, 2004a).
List of Identified Watershed Impairments
The impaired waterbody segments included in the project watershed are listed in Table 1
below, along with the cause of the listing. These impairments were identified in the draft
2004 303(d) list (IEPA, 2004a). Those impairments that are the focus of this report are
shown in bold font in Table 1. Note that unless otherwise noted, for purposes of this
report, “Mauvaise Terre Creek” refers to the stream section below Town Brook (below
both Mauvaise Terre Lake and North Fork Mauvaise Terre Creek), while “Mauvaise
Terre Creek” refers to waters upstream of Mauvaise Terre Lake. On the draft 2004
303(d) list, Mauvaise Terre Lake (SDL) was listed as being in partial support of the
overall use, aquatic life, and public water supply designated uses, and in nonsupport of
primary contact (swimming) and secondary contact (recreation) designated uses. North
Fork Mauvaise Terre Creek (DDC) was listed as being in partial support of the aquatic
life designated use. Mauvaise Terre Creek (DD04) was identified as being full support of
the following designated uses: aquatic life and fish consumption. Mauvaise Terre Creek
is in nonsupport of the primary contact recreation (swimming) designated use.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 5
Table 1. Impaired waterbodies in the project watershed
Waterbody
segment Waterbody name
Size
(miles/acres)
Year
Listed Listed for1
SDL
Mauvaise Terre
Lake 172 1994
Manganese, Phosphorus,
Nitrate, total suspended
solids, excess algal growth
DDC
N. Fork Mauvaise
Terre Creek 14.03 2004
Manganese, low dissolved
oxygen, total nitrogen, total
suspended solids
DD 04 Mauvaise Terre
Creek 36.55 1998 Fecal coliform
1Bold font indicates cause will be addressed in this report. Other potential causes of impairment listed for
these waterbodies do not have numeric Water Quality Standards and are not subject to TMDL development
at this time.
The remaining sections of this report include:
• Watershed characterization: discussion of methods for information compilation
and a detailed characterization of the watershed
• Database development and data analysis: discussion of data sources and methods
of data analysis
• Confirmation of causes and sources of impairment: assessment of sufficiency of
data to support the listing and identification of potential sources contributing to
the impairment
• Conclusions
WATERSHED CHARACTERIZATION
The purpose of watershed characterization was to obtain information describing the
watershed to support the identification of sources contributing to manganese and total
phosphorus impairments. Watershed characterization activities were focused on gaining
an understanding of key features of the watershed, including geology and soils, climate,
land cover and uses, and urbanization and growth. Active watershed organizations were
also identified. The methods used to characterize the watershed, and the findings are
described below.
Methods
Watershed characterization was conducted by compiling and analyzing data and
information from various sources. Where available, data were obtained in electronic or
Geographic Information System (GIS) format to facilitate mapping and analysis. To
develop a better understanding of land management practices in the watershed, calls were
placed to local agencies to obtain information on crops, pesticide and fertilizer
application practices, tillage practices and best management practices employed. A site
visit was conducted on June 28, 2004.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 6
After the watershed boundaries for the impaired waterbodies (Table 1) in the project
watershed were delineated in GIS using topographic and stream network (hydrography)
information, other relevant information was obtained. Information obtained and
processed for mapping and analysis purposes included:
• current land cover,
• current cropland,
• State and Federal lands,
• soils,
• point source dischargers,
• public water supply intakes,
• roads,
• railroads,
• state, county and municipal
boundaries,
• landfills,
• oil and gas wells,
• coal mines,
• dams,
• data collection locations, and
• location of 303(d) listed lakes
and streams.
To better describe the watershed and obtain information related to active local watershed
groups, data collection efforts, agricultural practices, and septic systems, calls were
placed to county-level officials at the Natural Resources Conservation District (NRCS),
Soil and Water Conservation District, (SWCD), Agricultural Extension Office, and
Health Department. A list of data sources and calls made is included in Appendix A.
Other information compiled for this task related to climate, population growth and
urbanization. These data were obtained from State and Federal sources, including the
National Weather Service, U.S. Census Bureau, and the State of Illinois.
Mauvaise Terre Creek Watershed Characterization
The Mauvaise Terre Creek watershed is located in Morgan and Scott counties in west-central
Illinois. The three waterbodies of concern are Mauvaise Terre Lake (SDL), North
Fork Mauvaise Terre Creek (DDC), and Mauvaise Terre Creek downstream of Town
Brook (DD04). Mauvaise Terre Lake and North Fork Mauvaise Terre Creek lie in
Morgan County, while Mauvaise Terre Creek flows through both Morgan and Scott
Counties.
Mauvaise Terre Lake was constructed by damming the upper part of Mauvaise Terre
Creek (above the North Fork). The lake has a surface area of 172 acres and serves as a
source of drinking water for Jacksonville and several surrounding communities. Most of
the water supply, however, comes from wells located 26 miles from the Jacksonville
(City of Jacksonville, 2004). The combined drainage area of the three impaired
waterbodies is approximately 164 square miles. Mauvaise Terre Lake is approximately
“L” shaped, with an arm extending west from the inlet, and a second arm extending north
to the dam. Mauvaise Terre Lake is connected near the corner of the “L” to a smaller lake
called Morgan Lake.
Figure 1 shows a map of the watershed, and includes some key features such as
waterways, impaired waterbodies, public water intakes and other key features. The map
also shows the locations of point source discharges that have a permit to discharge under
the National Permit Discharge Elimination System (NPDES).
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 7
The following sections provide a broad overview of the characteristics of the Mauvaise
Terre Creek watershed.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 8
Figure 1. Point source dischargers, impaired waterbody segments, and other
watershed characteristics
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 9
Geology and Soils
Information on soils and topography was compiled in order to understand whether the
soils are a potential source of manganese. Figure 2 shows the major soil associations in
the Mauvaise Terre Creek watershed. These are also listed in Table 2.
Of primary concern for this evaluation are the soils in the North Fork Mauvaise Terre
Creek and Mauvaise Terre Lake watersheds, since these waters are listed for manganese.
As discussed below, many of the soils in the Mauvaise Terre watershed contain
manganese and iron oxide concretions or accumulations and are also somewhat acidic.
This could result in manganese and iron moving into solution and being transported in
base flow and/or runoff.
The official soil series descriptions (Soil Survey Staff, 2004) describe the Ipava series as
consisting of “very deep, somewhat poorly drained, moderately slowly permeable soils
formed in loess on uplands”, with slopes ranging from 0 to 5 percent. The Sable series
consists of “very deep, poorly drained, moderately permeable soils formed in loess on
nearly level broad summits of moraines and stream terraces. Slope ranges from 0 to 2
percent.” The Sable series also has “very weakly cemented iron and manganese
concretions throughout” in five of the seven soil horizons (8-47 inches deep) (Soil Survey
Staff, 2004). Upper soil horizons (to 31 inches for Ipava and 23 inches for Sable) in
these two series are described as slightly to moderately acidic. The Tama series consists
of “deep, well and moderately well drained, moderately permeable soils formed in loess
on upland and high stream benches.” Slope ranges from 0 to 20 percent, and these soils
are characterized as strongly acid from zero to 45 inches deep (Soil Survey Staff, 2004).
The Rozetta series consists of “very deep, well drained soils formed in loess on uplands.
Permeability is moderate. Slope ranges from 0 to 25 percent.” This series is described as
moderate to strongly acid (0 to 50 inches deep), with some horizons (21-29 inches deep)
having “masses of iron and manganese accumulation” (Soil Survey Staff, 2004). The
Keomah series consists of “very deep, somewhat poorly drained soils formed in loess on
uplands and high stream terraces. They are moderately slowly to slowly permeable.
Slopes range from 0 to 5 percent”, and most horizons (0 to 47 inches deep) are
characterized as moderately to strongly acid. Four of the nine soil horizons in this series
are also described as having “fine iron and manganese concretions” (Soil Survey Staff,
2004). The Hickory series consists of “very deep, well drained, moderately permeable
soils on dissected till plains. They formed in till that can be capped with up to 20 inches
of loess. Slope ranges from 5 to 70 percent.” The upper horizons (up to 58 inches deep)
are characterized as strongly to very strongly acid, and have “fine rounded black iron-manganese
nodules” at 26-58 inches (Soil Survey Staff, 2004).
The upper part of the watershed (eastern half) is relatively flat, while west of
Jacksonville, the topography is more rolling. The high point in the watershed is located
about 0.5 mile west of Jacksonville, with an elevation of approximately 720 feet above
mean sea level. The watershed drains to the Illinois Creek, with an elevation at the
mouth of approximately 425 ft.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 10
Table 2. Major Soil Associations in the Watershed
Soil Map Units (MUID) Acres Percentage
North Fork Mauvaise Terre Creek Watershed
Ipava-Sable-Tama (Il003) 24,178 76.9%
Rozetta-Keomah-Hickory (Il036) 7,244 23.1%
Mauvaise Terre Lake Watershed
Ipava-Sable-Tama (Il003) 16,513 75.1%
Rozetta-Keomah-Hickory (Il036) 5,479 24.9%
Mauvaise Terre Creek Watershed
Ipava-Sable-Tama (Il003) 65,950 62.8%
Worthen-Littleton-Elburn (Il013) 259 0.2%
Beaucoup-Lawson-Darwin (Il029) 91 0.1%
Rozetta-Fayette-Hickory (Il034) 2,015 1.9%
Rozetta-Keomah-Hickory (Il036) 34,172 32.6%
Plainfield-Bloomfield-Sparta (Il056) 64 0.1%
Wakeland-Birds-Belknap (Il068) 1,818 1.7%
Ipava-Virden-Herrick (Il072) 592 0.6%
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 11
Figure 2. Major soil associations in the Mauvaise Terre Watershed
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 12
Climate
The Mauvaise Terre Creek watershed has a temperate climate and has cold winters and
hot summers. The National Weather Service (NWS) maintains a weather station at
Jacksonville through the Cooperative Observer Program (COOP). Climate data are
archived at the National Climatic Data Center (NCDC) and summaries are available on
the web page of the Illinois State Climatologist Office (Illinois State Water Survey,
2004). The average long-term precipitation (1971-2000) recorded at Jacksonville (Station
114442) is 38.47 inches. The maximum annual precipitation is 60.05 inches (1993) and
the minimum annual precipitation is 21.64 inches (1901). On average, there are 106.5
days with precipitation of at least 0.01 inches and 9.8 days with precipitation greater than
1 inch. Average snowfall is approximately 23.3 inches per year.
Average maximum and minimum temperatures recorded at Jacksonville are 34.4 oF and
15.0 oF, in January and 87.0 oF and 62.7 oF in July (1971-2000 data). The average
temperature recorded in January is 24.7 oF and the average temperature recorded in July
is 74.9 oF.
Land Cover and Use
Runoff from the land surface contributes pollutants to nearby receiving waters. In order
to understand sources contributing to the lake impairments, it was necessary to
characterize land cover in the watershed. Land cover and land uses in the watersheds are
shown in Figure 3, and listed in Tables 3 through 5. The predominant land use in the
watershed is agriculture, shown in yellow on the map. Approximately 65.8% of the
Mauvaise Terre Creek watershed (exclusive of the Mauvaise Terre Lake and North Fork
Mauvaise Terre Creek watersheds) is cropland, while croplands make up 84.0% of the
Mauvaise Terre Lake watershed and 90.8% of the North Fork Mauvaise Terre Creek
watershed. Crops are primarily a corn-soy rotation, with a small amount of wheat
(University of Illinois Agricultural Extension, 2004). Wheat is primarily grown for
livestock operations, either for straw or manure application. Corn represents 55 to 59%
of the total cropland.
According to estimates prepared by the Illinois Department of Agriculture (2002), in
Morgan County approximately 57% of the corn croplands and 5% of the soybean crops
are tilled using conventional tillage methods that leave little or no residue on the surface
(Table 6). Approximately 24% of the corn and 16% of the soybeans are tilled by reduced
tillage methods, which can reduce soil loss in comparison to conventional methods by
30%. The remaining 20% of corn croplands and 79% of soybean crops are planted either
using mulch-till methods, in which at least 30% residue of the previous year’s crop
remains on the land after planting the new crop, or without any tillage prior to planting, a
process that can reduce soil loss by up to 75% (IDOA, 2002). Mulch-till and no-till are
considered conservations tillage systems that can significantly reduce soil loss. Local
agency staff (NRCS, Agricultural Extension) confirmed that these estimates are
reasonable. The Morgan County Agricultural Extension suggested that the percentages
for conventional till might be a little high, and a lot of farmers are going to “strip till”
methods.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 13
Scott County tillage practices are presented in Table 7. Approximately 18% of the corn
crops and 1% of the soybean crops in the county are tilled using conventional tillage,
36% of the corn and 5% of the soybeans are tilled by reduced tillage methods, and 46%
of the corn crops and 94% of the soybeans are planted using either mulch-till (22% of
corn and 36% of soybeans) or no-till (24% of corn and 58% of soybeans) methods
(IDOA, 2002). All of the small grain croplands are planted using mulch-till or no-till.
Local agency staff (NRCS, Agricultural Extension) confirmed that these estimates are
reasonable.
Management practices within the watershed vary by individual producer, but include
things like buffer strips (Morgan County Agricultural Extension, personal
communication). Many producers are taking advantage of cost-share programs through
the NRCS. Buffer strips and streambank stabilization programs were observed during the
June 28, 2004 site visit. A sign was also noted in the eastern part of the watershed,
touting a nutrient management project. NRCS staff have indicated that in the western
part of the watershed, downstream of the lake, flooding and erosion are a concern to
producers, many of whom are experiencing lower yields due to topsoil loss.
The yellow areas on Figure 3 indicating agricultural land use include livestock
operations. There are livestock operations throughout the watershed. Livestock are
primarily cattle, with some hog lots. There may be a few landowners with goats, sheep,
or horses, but cattle and hogs dominate (NRCS, personal communication). During the
June 28, 2004 site visit, cattle and llamas were observed in the lower Mauvaise Terre
Creek watershed, while horses (and evidence of horses being ridden on local roads) were
observed in the Mauvaise Terre Lake watershed. The Morgan County NRCS is getting
involved in grazing management and prescribed grazing, working on programs to limit
access to creeks and streams, and improve the quality of pastures, which also reduces
runoff. Many beef producers are going to management-intensive grazing, trying to
distribute manure better, and use buffer strips. Morgan County has had several
applications for cost-share funds for prescribed grazing; there have been more
applications than available funds. The producers are interested in doing something about
the problem (Morgan County NRCS, personal communication).
The green areas on Figure 3 show forested lands (ranging from approximately 2.0% of
the North Fork Mauvaise Terre Creek watershed to approximately 12.8% of the lower
Mauvaise Terre Creek watershed), which are both upland and floodplain. Also shown on
the map (in red) are areas of low/medium and high density development. These areas
indicate the locations of the towns and residential communities in the watershed.
Jacksonville is the major urban area; the City lies entirely within the Mauvaise Terre
Creek watershed. The City of South Jacksonville is also within the watershed. Other
towns in the watershed include Exeter and Oxville. A portion of the town of Chapin also
lies in the watershed.
The Morgan County Health Department indicated that the Jacksonville area has sewers,
and perhaps a small area northwest of Jacksonville known as Marnico Village, but the
rest of the watershed is on private disposal systems. Local maps note a sewage disposal
location near Marnico Village; during the site visit on June 28, 2004, a pond was
observed at this location. This pond appeared to be entirely covered by a mat of algae.
The Health Department estimated that in the area near Mauvaise Terre Creek, 85% of
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 14
homes are on private sewage disposal septic systems. They are not aware of surface
discharging systems in the area. The Morgan County Health Department permits and
inspects all septic systems and is unaware of any failing systems in the watershed.
There are several point source discharges in the watershed, including sewage disposal for
the City of Jacksonville, food production facilities (ACH Food Company and Nestle), and
several oil wells near North Fork Mauvaise Terre Creek. Jacksonville also has a
combined sewer system and permitted combined sewer overflows (CSOs).
Interstate 72 passes through the watershed, crossing Mauvaise Terre Creek upstream of
the lake. Other major roads include U.S. Highway 67 and State Route 104. Most of the
other roads outside Jacksonville and South Jacksonville are unpaved rural roads.
Parkland and other recreational uses are in proximity to Mauvaise Terre Lake. There is a
municipal park surrounding Morgan Lake, which is connected to Mauvaise Terre Lake.
Parklands extend along the southwestern corner of Mauvaise Terre Lake. In addition to
picnic areas, playgrounds, and a public swimming pool, the park includes a municipal
golf course. The park drains to Morgan and Mauvaise Terre Lakes. There is also a large
private golf course, the Jacksonville Country Club, along the northeast side of the “L”
formed by the two arms of the lake. The country club is in close proximity to the lake,
and culverts were observed draining to the lake from the golf course.
Table 3. Land Cover Distribution, Mauvaise Terre Lake Watershed
Land Cover Type Area (Acres) Percent of Total
Agriculture1 18,468 84.0%
Urban 1,395 6.3%
Grassland 1,145 5.2%
Forest 398 1.8%
Wetland 342 1.6%
Water 216 1.0%
Barren 15 0.1%
Source: Illinois Department of Agriculture, 1999-2000 data (http://www.agr.state.il.us/gis/)
1 Agriculture is primarily comprised of corn (56%) and soybeans (43%), with lesser amount of winter
wheat and other small grains.
Table 4. Land Cover Distribution, North Fork Mauvaise Terre Creek Watershed
Land Cover Type Area (Acres) Percent of Total
Agriculture1 28,520 90.8%
Grassland 1,744 5.6%
Urban 461 1.5%
Forest 418 1.3%
Wetland 226 0.7%
Water 34 0.1%
Barren 9 0.0%
Source: Illinois Department of Agriculture, 1999-2000 data (http://www.agr.state.il.us/gis/)
1 Agriculture is primarily comprised of corn (59%) and soybeans (40%), with lesser amount of winter
wheat and other small grains.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 15
Table 5. Land Cover Distribution, Entire Mauvaise Terre Creek Watershed
Land Cover Type Area (Acres) Percent of Total
Agriculture1 80,879 77.1%
Grassland 8,158 7.8%
Urban 7,192 6.9%
Forest 5,558 5.3%
Wetland 2,548 2.4%
Water 393 0.4%
Barren 130 0.1%
Source: Illinois Department of Agriculture, 1999-2000 data (http://www.agr.state.il.us/gis/)
1 Agriculture is primarily comprised of corn (57%) and soybeans (42%), with lesser amount of winter
wheat and other small grains.
Table 6. Percent of Morgan County fields, by crop, with indicated tillage system
Tillage system
Conventional
Till1
Reduced-
Till2 Mulch-Till3 No-Till3
Corn 57 24 3 17
Soybean 5 16 38 41
Small grain 0 0 21 79
Source: Illinois Department of Agriculture (2002)
1 Residue level 0 – 15%
2 Residue level 16-30%
3 Residue level > 30%
Table 7. Percent of Scott County fields, by crop, with indicated tillage system
Tillage system
Conventional
Till1
Reduced-
Till2 Mulch-Till3 No-Till3
Corn 18 36 22 24
Soybean 1 5 36 58
Small grain 0 0 30 70
Source: Illinois Department of Agriculture (2002)
1 Residue level 0 – 15%
2 Residue level 16-30%
3 Residue level > 30%
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 16
Figure 3. Current land cover in the project watershed
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 17
Urbanization and Growth
Jacksonville is the major urban area within the watershed; the City lies entirely within the
Mauvaise Terre Creek watershed. The City of South Jacksonville is also within the
watershed. Small towns in the watershed include Exeter and Oxville. A portion of the
town of Chapin also lies in the watershed.
The current population of Morgan County, which contains the Mauvaise Terre Lake and
North Fork Mauvaise Terre Creek watersheds, as well as part of the Mauvaise Terre
Creek watershed, is approximately 36,616 (U.S. Census Bureau, 2000). Illinois
Population Trends (State of Illinois, 1997) predict an increase in population of
approximately 7.5% between 2000 and 2010 for Morgan County. The current population
of Scott County, which includes the lower portion of the Mauvaise Terre Creek
watershed, is approximately 5,537 (U.S. Census Bureau, 2000). Illinois Population
Trends (State of Illinois, 1997) predict an increase in population for Scott County of
approximately 8.7% between 2000 and 2010.
Hydrology
There is one USGS flow gage in the watershed. This gage is on the North Fork Mauvaise
Terre Creek near Jacksonville, IL (USGS gage number 05586000). The drainage area
upstream of this gage is 29.1 square miles. Data available at this location include water
quality data (collected between October 1974 and February 1981), daily flow
measurements (collected between December 1949 and September 1975) and peak flow
measurements. Only peak annual streamflow measurements are currently being reported
at this location.
Watershed Organizations
Local watershed organizations with an interest in watershed management are important
for successful implementation of TMDLs. The Illinois Watershed Management
Clearinghouse indicates that there may be a local watershed group for the Mauvaise Terre
watershed. However, an attempt at calling the contact person listed was unsuccessful.
DATABASE DEVELOPMENT AND ANALYSIS
A water quality database was developed and the data were analyzed to confirm the
sufficiency of the data to support both the listing decision and the sources of impairment
that are included on the draft 2004 303(d) list.
Data Sources and Methods
All readily available existing data to describe water quality in the impaired lakes were
obtained. Sources contacted for data include the Illinois Environmental Protection
Agency (State and Regional offices) and the United States Geologic Survey (USGS). All
available and relevant data were then compiled in electronic format along with sample
location and collection information, in a project database. A list of data sources is
included in Appendix A.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 18
Summaries of readily available water quality data are presented for Mauvaise Terre Lake
in Table 8 below, for North Fork Mauvaise Terre Creek in Table 9, and for Mauvaise
Terre Creek in Table 10. Sampling station locations are shown in Figure 4.
Some data are also available for parameters that may be related to the sources if
impairment, including dissolved phosphorus, chlorophyll a, and total suspended solids.
The water quality data were analyzed to confirm the cause of impairment for each
waterbody and, in combination with the watershed characterization data, an assessment
was made to confirm the sufficiency of the data to support the listing decision and the
sources of impairment that are included on the draft 2004 303(d) list. Analysis methods
included computing summary statistics, evaluating trends and correlations, and using
graphical analysis to discern relationships in the data.
Table 8. Water quality data summary for Mauvaise Terre Lake (SDL)
Sample
location and
parameter Criterion
Period of
record and
number of
data points Mean (mg/l)
Maximum
(mg/l)
Minimum
(mg/l)
Mauvaise Terre Lake, Near Dam Midway Between Spillway (Station SDL-1)
Manganese 150 ug/l
April-Oct
2002
5 samples
183 420 67
Phosphorus 0.05 mg/l 1990-2002
47 samples 0.162 0.344 0.015
Nitrate 10 mg/l 1990-2002
47 samples 3.91 12 <0.01
Mauvaise Terre Lake, 800 yd E. of Ramp N. of Docks (Station SDL-2)
Phosphorus 0.05 mg/l 1992 & 2002
10 samples 0.202 0.284 0.087
Nitrate 10 mg/l 1992 & 2002
10 samples 3.93 10 <0.01
Mauvaise Terre Lake, Mid Lake South of Red Brick House (Station SDL-3)
Phosphorus 0.05 mg/l 1992 & 2002
10 samples 0.248 0.370 0.118
Nitrate 10 mg/l 1992 & 2002
10 samples 4.72 13 <0.01
*note that data are for nitrate + nitrite, but water quality standard and listing are for nitrate
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 19
Table 9. Water quality data summary for North Fork Mauvaise Terre Creek (DDC)
Sample location
and parameter Criterion
Period of
record and
number of
data points Mean (mg/l)
Maximum
(mg/l)
Minimum
(mg/l)
North Fork Mauvaise Terre Creek, 0.5 M NE of Jacksonville (Station DDC11)
Dissolved oxygen 5 mg/l June 2001;
1 sample 7.8 7.8 7.8
Manganese 150 ug/l June 2001;
1 sample 78 78 78
North Fork Mauvaise Terre Creek, 3 Mi E of Jacksonville (Station DDC12)
Dissolved oxygen 5 mg/l July 2001;
2 samples 7.68 13.3 2.05
Manganese 150 ug/l
July-Oct.
2001;
2 samples
1,205 2,300 110
Table 10. Water quality data summary for Mauvaise Terre Creek (DD04)
Sample
location and
parameter
Criterion
(cfu/100 ml)
Period of
record and
number of
data points Mean (mg/l)
Maximum
(mg/l)
Minimum
(mg/l)
Mauvaise Terre Creek, 1.5 mi NE of Merritt (Station DD04)
Fecal coliform
400 cfu/100ml
in < 10% of
samples
Geomean <
200 cfu/100
ml
1990-2004,
97 samples 5,388 240,000 <50
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 20
Figure 4. Sampling stations in the project watershed
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 21
CONFIRMATION OF CAUSES AND SOURCES OF IMPAIRMENT
Water quality data were evaluated, in combination with the watershed characterization
data, to:
1. assess the sufficiency of the data to support the listing decision; and
2. identify suspected or known sources of impairment.
Mauvaise Terre Lake (SDL)
Mauvaise Terre Lake is listed on the 303(d) list as impaired by phosphorus, nitrate, and
manganese. The available data support the listing for phosphorus. Only three of the 67
available samples did not exceed the water quality criterion of 0.05 mg/l. On average,
sample results exceed the criterion by 2.4 to 4 times the criterion. Concentrations are
generally highest at the sampling location nearest the inlet, and lowest at the dam,
suggesting watershed sources may be significant.
There is not a strong relationship between total and dissolved phosphorus, suggesting that
there may be multiple sources of phosphorus. Total phosphorus generally increases with
increasing total suspended solids (Figure 5), suggesting a significant contribution from
runoff or resuspended sediments. Phosphorus data were collected at different depths at
station SDL-1 on two occasions; both of these show higher concentrations lower in the
water column (Figure 6), which may suggest resuspension of in-place sediments as a
source.
Figure 5. Total phosphorus vs. total suspended solids in Mauvaise Terre Lake
Total Phosphorus vs. Total Suspended Solids
y = 231.9x + 10.4
R2 = 0.3
0
50
100
150
200
250
300
0.0 0.2 0.4 0.6 0.8 1.0
Total Phosphorus (mg/L)
Total Suspended Solids
(mg/L)
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 22
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0 0.02 0.04 0.06 0.08 0.1 0.12
Concentration (mg/L)
Depth (ft)
4/11/2002
6/7/2002
Figure 6. Total phosphorus profiles in Mauvaise Terre Lake (near the dam)
The available nitrate data support the listing decision. Overall, nearly 20% of the nitrate-nitrite
samples exceeded the nitrate water quality criterion of 10 mg/l (note that data are
only available for nitrate + nitrite, while the water quality criterion is for nitrate). Among
the most recent samples, collected in 2002, 27% exceeded the criterion. A comparison
with total nitrogen concentrations in the lake indicates that nitrate is the largest
component of total nitrogen. The nitrate-nitrite samples show significant seasonality,
with high concentrations in spring and low concentrations in summer.
For manganese, the available data are limited, but support the listing decision. IEPA
guidelines (IEPA, 2004a) for identifying manganese as a cause in lakes state that the
public water supply use is not supported if, in untreated water, greater than 10% of the
observations exceed the applicable standard, for water samples collected in 1999 or later,
and for which results are readily available. Two of the five samples (40%) collected in
2002 exceeded the public water supply criterion of 150 ug/l. One sample exceeded the
criterion by 70 ug/l, while the other exceeded by 270 ug/l. Data were insufficient to
discern relationships with other parameters.
Potential Sources
The Illinois EPA (IEPA, 2004a) defines potential sources as known or suspected
activities, facilities or conditions that may be contributing to impairment of a designated
use. Illinois EPA (IEPA, 2004a) identified habitat modification, stream bank
modification/ destabilization, recreation and tourism activities, forest/grassland/parkland,
and unknown sources as potential sources of impairment. (Note that these potential
sources were identified for all listed causes of impairment, not only those evaluated in
this report.)
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 23
Based on a review of available information, including telephone calls to local agency
staff, site visits, and evaluation of the available water quality data, the following potential
sources of phosphorus were identified:
• Agricultural sources
• Recreational activities (i.e., golf courses)
• Existing in-lake sediment sources
Private sewage disposal systems may also be a source, although the Morgan County
Health Department was not aware of failing systems in the watershed.
The following potential sources of nitrate were identified:
• Agricultural sources
• Recreational activities (i.e., golf courses)
Agricultural fertilizer is the most likely source. Private sewage disposal systems may
also be a source, although the Morgan County Health Department was not aware of
failing systems in the watershed.
It appears that the primary source of manganese is natural background sources. Many of
the soils in the Mauvaise Terre watershed contain manganese concretions or
accumulations and are also somewhat acidic (Soil Survey Staff, 2004). This could result
in manganese moving into solution and being transported in base flow and/or runoff.
Lake sediments may also be a potential source, releasing manganese to the water column
when dissolved oxygen is low. No point source discharges of manganese were identified.
The observed levels of manganese are likely due to the natural geochemical environment
and most likely reflect natural background conditions. For this reason, the general use
standard may be difficult to attain.
North Fork Mauvaise Terre Creek (DDC)
North Fork Mauvaise Terre Creek is listed on the 303(d) list as impaired for manganese
and dissolved oxygen. Very few data are available, with only three measurements each
for manganese and dissolved oxygen. It is difficult to draw firm conclusions from these
limited data. However, the available data confirm that the listings are appropriate.
For dissolved oxygen, the single measurement at station DDC11 did not violate the water
quality criterion. At station DDC12, one of the two measurements violated the criterion
of 5 mg/l. Insufficient data are available to assess relationships to other parameters.
However, it is worth noting that North Fork Mauvaise Terre Creek is also listed as
impaired by nitrogen and suspended solids. The nitrogen impairment suggests that
excess nutrients may be leading to phytoplankton blooms and subsequent reductions in
D.O.
For manganese, a single sample (out of a total of three) exceeded both the drinking water
criterion (150 ug/l) and the general use criterion of 1,000 ug/l. The other two samples did
not exceed either criterion. While it is difficult to draw conclusions from such a limited
data set, it is noteworthy that the highest manganese concentration also corresponded to
the highest total suspended solids in the data set.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 24
Potential Sources
The Illinois EPA (IEPA, 2004a) defines potential sources as known or suspected
activities, facilities or conditions that may be contributing to impairment of a designated
use. Based on a review of available information, including telephone calls to local
agency staff, site visits, and evaluation of the available water quality data, the following
potential sources of low dissolved oxygen were identified:
• Agricultural sources
Illinois EPA (IEPA, 2004a) identified agriculture and crop-related sources as potential
sources of impairment.
As discussed previously, some soils in the watershed are known to contain manganese. It
appears likely that the primary source of manganese is natural sources. The apparent
correspondence between high manganese and high total suspended solids, while based on
only one sample, lends credence to soils as a source. The Illinois EPA (IEPA, 2004a)
lists “unknown sources” as a suggested source of impairment.
Mauvaise Terre Creek (DD 04)
Mauvaise Terre Creek is listed on the 303(d) list as impaired by fecal coliform bacteria.
The available data support this listing. Data are available for a single sampling location,
station DD04. Of the 97 fecal coliform samples collected at this station, 49 were
collected between May and October. An analysis of the May – October fecal data
revealed that 36 of the 49 fecal samples (73%) were greater than 400 cfu/100 ml.
A comparison of fecal coliform levels to total suspended solids concentrations (Figure 7)
suggests that fecal coliform increases with increasing suspended solids concentration.
Figure 7. Fecal coliform and total suspended solids concentrations in Mauvaise
Terre Creek
y = 2.0874x0.4856
R2 = 0.4441
1
10
100
1000
10000
1 10 100 1000 10000 100000 1000000
Fecal coliform (cfu/100ml)
Total suspended solids (mg/l)
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 25
Potential Sources
The Illinois EPA (IEPA, 2004a) defines potential sources as known or suspected
activities, facilities or conditions that may be contributing to impairment of a designated
use. Through a review of available information, including telephone calls to local agency
staff, site visits, and evaluation of the available water quality data, the following potential
sources of fecal coliform were identified:
• Livestock operations
• Runoff from manure-fertilized cropland
• Municipal sewage disposal
• Jacksonville combined sewer overflows
• Private sewage disposal systems
The Illinois EPA listed “unknown” sources as the source of the impairment. The data
suggest that agricultural runoff in particular is a likely source of the impairment. The
apparent relationship in the data between fecal coliform and total suspended solids
suggests a watershed source (such as runoff) for the fecal coliform. Livestock operations
are present throughout the watershed. During the June 2004 site visit, the smell of
manure was apparent at several locations in the lower watershed, although the exact
source was unclear. There are also several municipal sewage discharges to the creek, as
well as private septic systems, that may be contributing to the impairment.
CONCLUSIONS
The available data, though in some cases very limited, support the listed impairments of
the three waterbodies in the Mauvaise Terre watershed. Potential sources of phosphorus
and nitrate to Mauvaise Terre Lake include agricultural sources, existing sediments,
recreation activities, and possibly failing private sewage disposal systems. The primary
source of manganese to both Mauvaise Terre Lake and North Fork Mauvaise Terre Creek
may be background sources due to naturally high concentrations in area soils, with
possible contributions from in-place sediments. The primary potential source of low
dissolved oxygen in North Fork Mauvaise Terre Creek is agricultural runoff. Potential
sources of fecal coliform bacteria to Mauvaise Terre Creek include livestock operations,
agricultural runoff, and sewage disposal, including municipal sewage, CSO discharges,
and private disposal systems.
NEXT STEPS
In the upcoming quarter, methods, procedures and models that will be used to develop
TMDLs for the project watershed will be identified and described. This description will
include documentation of any important assumptions underlying the recommended
approach (methods, procedures and models) and a discussion of data needed to support
the development of a credible TMDL.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 26
REFERENCES
City of Jacksonville, 2004. Utilities Department Water Supply Information,
http://www.jacksonvilleil.com/utilities2.htm
Illinois Department of Agriculture, 2002. 2002 Illinois Soil Conservation Transect
Survey Summary.
Illinois Environmental Protection Agency (IEPA), 2004a. Draft Illinois 2004 Section
303(d) List. Bureau of Water, Watershed Management Section. April 2004.
IEPA/BOW/04-005 http://www.epa.state.il.us/water/watershed/reports/303d-report/
303d-2004.pdf
Illinois Environmental Protection Agency (IEPA), 2004b. Final Draft Illinois Water
Quality Report 2004. Bureau of Water, May 2004. IEPA/BOW/04-006.
http://www.epa.state.il.us/water/water-quality/305b/305b-2004.pdf
Illinois State Water Survey, Illinois State Climatologist Office. Illinois Climate Summary
for Jacksonville, Illinois.
http://www.sws.uiuc.edu/atmos/statecli/Summary/114442.htm
Soil Survey Staff, Natural Resources Conservation Service, United States Department of
Agriculture. Official Soil Series Descriptions [Online WWW]. Available URL:
"http://soils.usda.gov/soils/technical/classification/osd/index.html" [Accessed 3
August 2004].
State of Illinois. 1997 edition. Illinois Population Trends 1990-2020.
United States Census Bureau, 2000. Census 2000 Data for the State of Illinois.
http://www.census.gov/census2000/states/il.html
United States Environmental Protection Agency (USEPA). 1991. Guidance for Water
Quality-based Decisions: The TMDL Process. EPA 440/4-91-001, Office of Water,
Washington, DC.
University of Illinois Extension, 2004. Personal communication.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 27
APPENDIX A. DATA SOURCES AND LOCAL CONTACTS
Table A-1. Data sources
Data description Agency Website
Climate summaries Illinois State Water Survey http://www.sws.uiuc.edu/atmos/statecli/in
dex.htm
NPDES permit limits United States Environmental
Protection Agency
http://www.epa.gov/enviro/html/pcs/pcs_q
uery.html
Aerial photography Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/webdo
cs/doqs/graphic.html
Coal mines: active and
abandoned - polygons part 1
Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Coal mines: active and
abandoned - polygons part 2
Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Coal mines: active and
abandoned – points
Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Coal mine permit boundaries Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
County boundaries Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Cropland
United States Department of
Agriculture, National Agricultural
Statistics Service, via Illinois
Department of Agriculture
http://www.agr.state.il.us/gis/pass/nassdat
a/
Dams National Inventory of Dams (NID) http://crunch.tec.army.mil/nid/webpages/ni
d.cfm
Elevation United States Geological Survey http://seamless.usgs.gov/viewer.htm
Federally-owned lands Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Hydrologic cataloging units Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Hydrography United States Geological Survey http://nhd.usgs.gov/
Impaired lakes Illinois Environmental Protection
Agency
http://maps.epa.state.il.us/website/wqinfo/
Impaired streams Illinois Environmental Protection
Agency
http://maps.epa.state.il.us/website/wqinfo/
Land cover Illinois Department of Agriculture http://www.agr.state.il.us/gis/
Landfills Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Municipal boundaries U.S. Census Bureau
Municipal boundaries Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
National Pollutant Discharge
Elimination System (NPDES)
permitted sites
United States Environmental
Protection Agency
NPDES discharge data Illinois Environmental Protection
Agency
Nature preserves Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Oil wells United States Geological Survey http://energy.cr.usgs.gov/oilgas/noga/
Railroads Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Roads Illinois Natural Resources http://www.isgs.uiuc.edu/nsdihome/
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 28
Data description Agency Website
Geospatial Data Clearinghouse
Roads – state highways Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Roads – U.S. highways Illinois Natural Resources
Geospatial Data Clearinghouse http://www.isgs.uiuc.edu/nsdihome/
Roads- detailed road network U.S. Census Bureau http://www.census.gov/geo/www/tiger/tige
rua/ua_tgr2k.html
Survey-level soils
United States Department of
Agriculture Natural Resources
Conservation Service
http://www.il.nrcs.usda.gov/technical/soils/
ssurgo.html
State-level soils
United States Department of
Agriculture Natural Resources
Conservation Service
http://www.il.nrcs.usda.gov/technical/soils/
statsgo_inf.html - statsgo8
State boundary Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
State conservation areas Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
State forests Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
State fish and wildlife areas Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
State parks Illinois Natural Resources
Geospatial Data Clearinghouse
http://www.isgs.uiuc.edu/nsdihome/
Topographic map quadrangle
index
Illinois Natural Resources
Geospatial Data Clearinghouse http://www.isgs.uiuc.edu/nsdihome/
Topographic map quadrangles Illinois Natural Resources
Geospatial Data Clearinghouse http://www.isgs.uiuc.edu/nsdihome/
USGS stream gages Illinois State Water Survey
Water quality data Illinois Environmental Protection
Agency
Watersheds Illinois Environmental Protection
Agency
http://maps.epa.state.il.us/website/wqinfo/
Water supply – Public water
supply intakes Illinois State Water Survey
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 29
Table A-2. Local and state contacts
Contact Agency/
Organization
Contact
Means Phone # Subject
Aaron
Dufelmeier
Morgan County
Agricultural Extension
Telephone 217-479-4627 Nutrient and pathogen
sources, management
practices
Quentin
Lucassen
Morgan County Health
Department
Telephone 217-245-5111 Onsite sewage disposal,
potential sources of
contaminants
Matt Bunger Morgan County NRCS Telephone 217-243-1535 ext
3
Nutrient and pathogen
sources, agricultural
practices
Brenda Scott County Agricultural
Extension
Telephone 217-742-9572 Referred us to Morgan
County office
Reg Scott County Farm
Service Agency
Telephone 217-742-9561 ext
2
Referred us to Morgan
County NRCS
Rich Nickels Illinois Department of
Agriculture Telephone 217-782-6297 Requested Cropland
Transect Survey
Sue Ebetsch Illinois State Data Center Telephone 217-782-1381 Requested Population
projection report
Laura Biewick U.S. Geological Survey Telephone 303-236-7773 GIS data for oil & gas
wells
Kathy Brown Illinois State Water
Survey Telephone 217-333-6778 USGS gage locations;
water supply intakes
Sharie Heller SW Illinois GIS resource
Center Telephone 618-566-9493 Discussed CRP maps
Steve Sobaski Illinois Department of
National Resources e-mail ssobaski@dnrmail
.state.il.us
Formal request for
conservation related GIS
files
Don Pitts
United States Department
of Agriculture Natural
Resources Conservation
Service
Telephone 217-353-6642
Potential sources of iron
and manganese in south-central
Illinois surface
waters.
Tony Meneghetti IEPA Telephone
and e-mail
217-782-3362
Anthony.Meneghe
tti@epa.state.il.us
Lake data and SWAPs
Dave Muir IEPA Marion Regional
office
Personal
visit 618-993-7200 Assessment data used in
303(d) and 305(b) reports
Tim Kelly IEPA Springfield Regional
office
Telephone
and e-mail
217-786-6892
Tim.Kelly@epa.st
ate.il.us
NPDES DMR data
Jeff Mitzelfelt IEPA e-mail jeff.mitzelfelt@epa
.state.il.us
Websites for GIS
information
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 30
This page is blank to facilitate double sided printing.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 31
APPENDIX B: PHOTOGRAPHS
Agricultural and park lands draining to Morgan Lake (adjacent to Mauvaise Terre Lake).
Park lands draining to Morgan Lake (adjacent to Mauvaise Terre Lake).
Culverts draining to Morgan Lake (adjacent to Mauvaise Terre Lake).
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 32
Mauvaise Terre Lake: From Vandalia Rd looking north, an industrial facility is on the
west side of the lake, and a golf course on the east side
Golf course on the east side of Mauvaise Terre Lake
The golf course, Jacksonville Country Club, is not directly adjacent to the lake. Country
Club Drive, which follows the north and east shores of the lake, runs between the course
and the lake. However, at least one culvert was observed between a pond at the Country
Club and the lake
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 33
Drainages from the west side of Country Club Drive to the eastern arm of the lake
Surface foam and filamentous algae in Mauvaise Terre Lake
Mauvaise Terre Lake at Vandalia Rd.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 34
Mauvaise Terre Creek at Rte 104
Mauvaise Terre Creek at Rte 104
Mauvaise Terre Creek below the lake, at Johnson St.
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 35
West of Jacksonville along Rte 67/104 near Mauvaise Terre Creek
Mauvaise Terre Creek at Mt. Zion Rd, north of Hwy 104, near Marnico Village
Mauvaise Terre Creek at Mt. Zion Rd., just south of Apple Rd
Quarterly Progress Report August 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 36
Mauvaise Terre Creek, along Markham Rd.
Mauvaise Terre Creek, along Willow Branch Rd
Mauvaise Terre Creek along Willow Branch Rd. (stream stabilization program)
Second Quarterly Progress Report
Prepared for Illinois Environmental Protection Agency
October 2004
Mauvaise Terre Creek Watershed
Mauvaise Terre Creek (ILDD04)
Mauvaise Terre Lake (SDL), North Fork Mauvaise Terre Creek
(DDC), Mauvaise Terre Creek (DD04)
Limno-Tech, Inc.
www.limno.com
This page is blank to facilitate double sided printing.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page i
Table of Contents
EXECUTIVE SUMMARY ...............................................................................................1
Results...........................................................................................................................2
INTRODUCTION/PURPOSE..........................................................................................3
IDENTIFICATION OF POTENTIALLY APPLICABLE MODELS AND
PROCEDURES TO BE USED IN TMDL DEVELOPMENT ......................................4
Watershed Methodologies and Modeling Frameworks .................................................4
Water Quality Methodologies and Modeling Frameworks .........................................10
MODEL SELECTION....................................................................................................15
General Guidelines.......................................................................................................15
Model Selection for the Mauvaise Terre Creek Watershed.........................................16
DATA NEEDS FOR THE METHODOLOGIES TO BE USED.................................25
REFERENCES.................................................................................................................25
List of Tables
Table 1. Summary of Potentially Applicable Models for Estimating Watershed Loads... 5
Table 2. Summary of Potentially Applicable Models for Estimating Water Quality....... 11
Table 3. Water Quality Data Summary for Mauvaise Terre Creek (DD04)..................... 18
Table 4. Water Quality Data Summary for North Fork Mauvaise Terre Creek (DDC) . 18
Table 5. Water Quality Data Summary for Mauvaise Terre Lake (SDL) ........................ 19
Table 6. Recommended Modeling Approaches for Mauvaise Terre Creek (DD04)........ 20
Table 7. Recommended Modeling Approaches for North Fork Mauvaise Terre Creek
(DDC) ....................................................................................................................... 23
Table 8. Recommended Modeling Approaches for Mauvaise Terre Lake (SDL)........... 23
List of Figures
Figure 1. Calculation of a Flow Duration Curve (from Freedman et al., 2003) .............. 21
Figure 2. Calculation of a Load Duration Curve (from Freedman et al., 2003) .............. 21
Figure 3. Load Duration Curve with Observed Loads (from Freedman et al., 2003) ..... 22
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page ii
This page is blank to facilitate double sided printing.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 1
EXECUTIVE SUMMARY
This is the second in a series of quarterly status reports documenting work completed on
the Mauvaise Terre Creek project watershed. The objective of this report is to provide a
summary of Stage 1 work that will ultimately be used to support Total Maximum Daily
Load (TMDL) development in the project watershed.
Background
Section 303(d) of the 1972 Clean Water Act requires States to define impaired waters and
identify them on a list, which is referred to as the 303(d) list. The State of Illinois
recently issued the draft 2004 303(d) list (IEPA, 2004), which is available on the web at:
http://www.epa.state.il.us/water/tmdl/303d-list.html. The Clean Water Act requires that a
Total Maximum Daily Load (TMDL) be completed for each pollutant listed for an
impaired water body. TMDLs are prepared by the States and submitted to the U.S. EPA.
In developing the TMDL, a determination is made of the greatest amount of a given
pollutant that a water body can receive without exceeding water quality standards and
designated uses, considering all known and potential sources. The TMDL also takes into
account a margin of safety, which reflects scientific uncertainty, as well as the effects of
seasonal variation.
As part of the TMDL process, the Illinois Environmental Protection Agency (IEPA) and
several consultant teams have compiled and reviewed data and information to determine
the sufficiency of available data to support TMDL development. As part of this review,
the data were used to confirm the impairments identified on the 303(d) list and to further
identify potential sources causing these impairments. The results of this review were
presented in the first quarterly status report.
The intent of this second quarterly status report is to:
• Identify and briefly describe the methodologies/procedures/models to be used in
the development of TMDLs
• Document important assumptions underlying the recommended methodologies
• Identify the data needs for the methodologies to be used in TMDL development,
including an assessment of whether additional data are needed to develop credible
TMDLs
In future phases of this project, Illinois EPA and consultants will develop the TMDLs and
will work with stakeholders to implement the necessary controls to improve water quality
in the impaired water bodies and meet water quality standards. It should be noted that the
controls for nonpoint sources (e.g., agriculture) would be strictly voluntary.
Methods
The effort completed in the second quarter included: 1) summarizing potentially
applicable model frameworks for TMDL development, 2) Recommending specific model
frameworks for application to the three impaired waterbodies in the Mauvaise Terre
Creek watershed, and 3) Making a determination whether sufficient data exist to allow
development of a credible TMDL. Selection of specific model frameworks was based
upon consideration of three separate factors, consistent with the guidance of DePinto et al
(2004):
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 2
• Site-specific characteristics: The characteristics define the nature of the
watershed and water bodies. For Mauvaise Terre Creek below Town Brook, the
relevant site-specific characteristics include a watershed with predominantly
agricultural land use, and a creek impaired by fecal coliform. For Mauvaise Terre
Lake, the relevant site-specific characteristics include a watershed with
predominantly agricultural land use and a lake impaired by manganese, total
phosphorus and nitrate. For North Fork Mauvaise Terre Creek, the relevant site-specific
characteristics include a watershed with predominantly agricultural land
use and a creek impaired by manganese and low dissolved oxygen.
• Management objectives: These objectives consist of the specific questions to be
addressed by the model. For this application, the management objective is to
define a credible TMDL.
• Available resources: This corresponds to the amount and time and data available
to support TMDL development. Water quality data currently exist for Mauvaise
Terre Creek, North Fork Mauvaise Terre Creek and Mauvaise Terre Lake. One
aspect of this work is to define whether or not the existing data are sufficient to
allow development of a credible TMDL.
Results
Several modeling frameworks potentially applicable for developing TMDLs were
identified, spanning a range of detail from simple to complex. Selection of a specific
modeling framework is complicated by the fact that the definition of a “credible” TMDL
depends upon the level of detail to be contained in the implementation plan. If the goal of
the TMDL implementation plan is to define the primary sources of impairment and
quickly identify the general level of reduction required, relatively simple models can be
used to develop a credible TMDL. If the goal of the TMDL implementation plan is to
explicitly define the specific levels of controls required, more detailed models (and
additional data) are required to develop a credible TMDL. Specific recommendations are
provided which correspond to the level of detail provided in other Illinois TMDL
implementation plans conducted to date.
The recommended approach for Mauvaise Terre Creek consists of developing a load-duration
curve to address fecal coliform impairments. This will allow for determination
of the degree of impairment under different flow conditions and the respective
importance of dry weather and wet weather fecal coliform sources. Results from the load-duration
curve can also be used to identify the approximate level of source control needed
under each set of flow conditions.
The recommended approach for North Fork Mauvaise Terre Creek consists of using the
water quality model QUAL2E to address dissolved oxygen problems. Manganese
impairments will be addressed via spreadsheet calculations. Watershed loads for this
segment will be defined using an empirical approach. QUAL2E was selected for
dissolved oxygen modeling because it is the most commonly used water quality model
for addressing low flow conditions. Because problems appear to be restricted to low flow
conditions, watershed loads are not expected to be significant contributors to the
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 3
impairment. For this reason, an empirical approach was selected for determining
watershed loads.
The recommended approach for Mauvaise Terre Lake consists of using the GWLF and
BATHTUB models to address total phosphorus, manganese and nitrate problems in
Mauvaise Terre Lake. Specifically, GWLF will be applied to calculate phosphorus and
nitrate loads to the reservoir from different land uses, over a time scale consistent with
their nutrient residence times in Mauvaise Terre Lake. BATHTUB will then be used to
predict the relationship between nutrient (phosphorus and nitrate) load and resulting in-lake
phosphorus and dissolved oxygen concentrations, and resulting potential for
manganese release from sediments. This relationship will be used to define the dominant
sources of nutrients to the lake, and the extent to which they must be controlled to attain
water quality standards. The BATHTUB model was selected because it does not have
extensive data requirements (and can therefore be applied with existing data), yet still
provides the capability for calibration to observed Mauvaise Terre Lake data. GWLF was
selected as the watershed model because it can provide loading information on the time-scale
required by BATHTUB, with moderate data requirements that can be satisfied by
existing data.
Alternative model frameworks are also provided that will support the development of
differing levels of TMDL implementation plans. Some of these frameworks will require
no additional data collection; however, other frameworks have significantly greater data
requirements, and their use would require additional data collection.
INTRODUCTION/PURPOSE
This Stage 1 report describes intermediate activities related to the development of
TMDLs for impaired water bodies in the Mauvaise Terre Creek watershed. Earlier Stage
1 efforts included watershed characterization activities and data analyses, to confirm the
causes and sources of impairments in the watershed.
The remaining sections of this report include:
• Identification of potentially applicable methodologies to be used in TMDL
development: This section describes the range of potentially applicable
watershed loading and water quality methodologies that could be used to conduct
the TMDL, and identifies their strengths and weaknesses.
• Model selection process: This section describes how management objectives,
available resources and site-specific conditions in the Mauvaise Terre Creek
watershed affect the recommendation of specific methodologies.
• Selection of specific methodologies and future data requirements: This
section provides specific recommendation of methodologies for the Mauvaise
Terre Creek watershed, along with the data needed to support application of the
methodologies.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 4
IDENTIFICATION OF POTENTIALLY APPLICABLE MODELS AND
PROCEDURES TO BE USED IN TMDL DEVELOPMENT
Development of TMDLs requires: 1) a method to estimate the amount of pollutant load
being delivered to the water body of interest from all contributing sources, and 2) a
method to convert these pollutant loads into an in-stream (or in-lake) concentration for
comparison to water quality targets. Both of these steps can be accomplished using a
wide range of methodologies, ranging from simple calculations to complex computer
models. This section describes the methodologies that are potentially applicable for the
three 303(d) listed waterbodies in the Mauvaise Terre Creek watershed, and is divided
into separate discussions of watershed methodologies and receiving water quality model
frameworks.
Watershed Methodologies and Modeling Frameworks
Numerous methodologies exist to characterize watershed loads for TMDL development.
These include:
• Empirical Approaches
• Unit Area Loads/Export Coefficients
• Universal Soil Loss Equation
• Watershed Characterization System (WCS) Sediment Tool
• Generalized Watershed Loading Functions (GWLF) Model
• Agricultural Nonpoint Source Pollution Model (AGNPS)
• Hydrologic Simulation Program - Fortran (HSPF)
• Better Assessment Science Integrating point and Nonpoint Sources (BASINS)/
Nonpoint Source Model (NPSM)
• Storm Water Management Model (SWMM)
• Soil & Water Assessment Tool (SWAT)
This section describes each of the model frameworks and their suitability for
characterizing watershed loads for TMDL development. Table 1 summarizes some
important characteristics of each of the models relative to TMDL application.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 5
Table 1. Summary of Potentially Applicable Models for
Estimating Watershed Loads
Model
Data
Needs
Output
Timescale
Potential
Accuracy Calibration
Applicability for
TMDL
Empirical
Approach High Any High N/A
Good for defining
existing total load;
less applicable for
defining individual
contributions or future
loads
Unit Area
Loads Low Annual
average Low None
Acceptable when
limited resources
prevent development
of more detailed
model
USLE Low Annual
average Low
Requires data
describing
annual average
load
Acceptable when
limited resources
prevent development
of more detailed
model
WCS
Sediment
Tool
Low Annual
average Low
Requires data
describing
annual average
load
Acceptable when
limited resources
prevent development
of more detailed
model
GWLF Moderate Monthly
average Moderate
Requires data
describing flow
and
concentration
Good for mixed use
watersheds;
compromise between
simple and more
complex models
SWMM Moderate Continuous Moderate
Requires data
describing flow
and
concentration
Primarily suited for
urban watersheds
AGNPS High Continuous High
Requires data
describing flow
and
concentration
Primarily suited for
rural watersheds;
highly applicable if
sufficient resources
are available
HSPF High Continuous High
Requires data
describing flow
and
concentration
Good for mixed use
watersheds; highly
applicable if sufficient
resources are
available
SWAT High Continuous High
Requires data
describing flow
and
concentration
Primarily suited for
rural watersheds;
highly applicable if
sufficient resources
are available
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 6
Empirical Approaches
Empirical approaches estimate pollutant loading rates based upon site-specific
measurements, without the use of a model describing specific cause-effect relationships.
Time series information is required on both stream flow and pollutant concentration.
The advantage to empirical approaches is that direct measurement of pollutant loading
will generally be far more accurate than any model-based estimate. The approach,
however, has several disadvantages. The empirical approach provides information
specific to the storms that are monitored, but does not provide direct information on
conditions for events that were not monitored. Statistical methods (e.g., Preston et al.,
1989) can be used to integrate discrete measurements of suspended solids concentrations
with continuous flow records to provide estimates of solids loads over a range of
conditions.
The primary limitation of empirical techniques is their inability to separate individual
contributions from multiple sources. This problem can be addressed by collecting
samples from tributaries serving single land uses, but most tributary monitoring stations
reflect multiple land uses. The EUTROMOD and BATHTUB water quality models
described below contain routines that apply the empirical approach to estimating
watershed loads.
Unit Area Loads/Export Coefficients
Unit area loads (also called export coefficients) are routinely used to develop estimates of
pollutant loads in a watershed. An export coefficient is a value expressing pollutant
generation per unit area and unit time for a specific land use (Novotny and Olem, 1994).
The use of unit areal loading or export coefficients has been used extensively in
estimating loading contributions from different land uses (Beaulac 1980, Reckhow et al.
1980, Reckhow and Simpson 1980, Uttormark et al. 1974). The concept is
straightforward; different land use areas contribute different loads to receiving waters.
By summing the amount of pollutant exported per unit area of land use in the watershed,
the total pollutant load to the receiving system can be calculated.
These export coefficients are usually based on average annual loads. The approach
permits estimates of current or existing loading, as well as reductions in pollutant export
for each land use required to achieve a target TMDL pollutant load. The accuracy of the
estimates is dependent on good land use data, and appropriate pollutant export
coefficients for the region. EUTROMOD is a spreadsheet-based modeling procedure for
estimating phosphorus loading and associated lake trophic state variables, which can
estimates phosphorus loads derived from watershed land uses or inflow data using
approaches developed by Reckhow et al. (1980) and Reckhow and Simpson (1980). The
FLUX module of the BATHTUB software program estimates nutrient loads or fluxes to a
lake/reservoir and provides five different algorithms for estimating these nutrient loads
based on the correlation of concentration and flow. In addition, the potential errors in
loading estimates are quantified.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 7
Universal Soil Loss Equation
The Universal Soil Loss Equation (USLE), and variations of the USLE, are the most
widely used methods for predicting soil loss. When applied properly, the USLE can be
used as a means to estimate loads of sediment and sediment-associated pollutants for
TMDLs. The USLE is empirical, meaning that it was developed from statistical
regression analyses of a large database of runoff and soil loss data from numerous
watersheds. It does not describe specific erosion processes. The USLE was designed to
predict long-term average annual soil erosion for combinations of crop systems and
management practices with specified soil types, rainfall patterns, and topography.
Required model inputs to the USLE consist of:
• Rainfall erosivity index factor
• Soil-erodibility factor
• Slope length factor reflecting local topography
• Cropping-management factor
• Conservation practice factor
Most of the required inputs for application of the USLE are tabulated by county Natural
Resources Conservation Service (NRCS) offices.
There are also variants to the USLE: the Revised USLE (RUSLE) and the Modified
USLE (MUSLE). The RUSLE is a computerized update of the USLE incorporating new
data and making some improvements. The basic USLE equation is retained, but the
technology for evaluating the factor values has been altered and new data introduced to
evaluate the terms for specific conditions. The MUSLE is a modification of USLE, with
the rainfall energy factor of the USLE replaced with a runoff energy factor. MUSLE
allows for estimation of soil erosion on an event-specific basis.
While the USLE was originally designed to consider soil/sediment loading only, it is also
commonly used to define loads from pollutants that are tightly bound to soils. In these
situations, the USLE is used to define the sediment load, with the result multiplied by a
pollutant concentration factor (mass of pollutant per mass of soil) to define pollutant load.
The USLE is among the simplest of the available models for estimating sediment and
sediment-associated loads. It requires the least amount of input data for its application
and consequently does not ensure a high level of accuracy. It is well suited for screening-level
calculations, but is less suited for detailed applications. This is because it is an
empirical model that does not explicitly represent site-specific physical processes.
Furthermore, the annual average time scale of the USLE is poorly suited for model
calibration purposes, as field data are rarely available to define erosion on an annual
average basis. In addition, the USLE considers erosion only, and does not explicitly
consider the amount of sediment that is delivered to stream locations of interest. It is best
used in situations where data are available to define annual loading rates, which allows
for site-specific determination of the fraction of eroded sediment that is delivered to the
surface water.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 8
Watershed Characterization System (WCS) Sediment Tool
The Watershed Characterization System (WCS) Sediment Tool was developed by EPA
Region 4. The Watershed Characterization System is an ArcView-based application used
to display and analyze GIS data including land use, soil type, ground slope, road
networks, point source discharges, and watershed characteristics. WCS has an extension
called the Sediment Tool that is specifically designed for sediment TMDLs. For each grid
cell within the watershed, the WCS Sediment Tool calculates potential erosion using the
USLE based on the specific cell characteristics. The model then calculates the potential
sediment delivery to the stream grid network. Sediment delivery can be calculated using
one of the four available sediment delivery equations: a distance-based equation, a
distance slope-based equation, an area-based equation, or a WEPP-based regression
equation.
The applicability of WCS for estimating sediment loads for TMDLs is similar to that of
the USLE in terms of data requirements and model results; i.e., it is relatively simple to
apply but has the potential to be inaccurate. It provides three primary enhancements over
the USLE: 1) Model inputs are automatically incorporated into the model through GIS
coverages; 2) Topographic factors are calculated in the model based on digital elevation
data; and 3) The model calculates the fraction of eroded sediment that is delivered to the
surface water. It is only applicable to sediment TMDLs whose target represents long-term
loading conditions. Because its predictions represent average annual conditions, it is not
suitable for predicting loads associated with specific storm events. Like the USLE, it is
does not lend itself to model calibration unless data are available to define annual loading
rates.
Generalized Watershed Loading Functions Model (GWLF)
The Generalized Watershed Loading Functions Model (GWLF) simulates runoff and
sediment loadings from mixed-use watersheds. It is a continuous simulation model (i.e.,
predicts how concentrations change over time) that uses daily time steps for weather data
and water balance calculations. Sediment results are provided on a monthly basis. GWLF
requires the user to divide the watershed into any number of distinct groups, each of
which is labeled as rural or urban. The model does not spatially distribute the source
areas, but simply aggregates the loads from each area into a watershed total; in other
words, there is no spatial routing. Erosion and sediment yield for rural areas are estimated
using monthly erosion calculations based on the USLE (with monthly rainfall-runoff
coefficients). A sediment delivery ratio based on watershed size and a transport capacity
based on average daily runoff are then applied to the calculated erosion to determine how
much of the sediment eroded from each source area is delivered to the watershed outlet.
Erosion from urban areas is considered negligible.
GWLF provides more detailed temporal results than the USLE, but also requires more
input data. Specifically, daily climate data are required as well as data on processes
related to the hydrologic cycle (e.g., evapotranspiration rates, groundwater recession
constants). By performing a water balance, it has the ability to predict concentrations at a
watershed outlet as opposed to just loads. It lacks the ability to calculate the sediment
delivery ratio that is present in the WCS sediment tool. Because the model performs on a
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 9
continuous simulation basis, it is more amenable to site-specific calibration than USLE or
the WCS sediment tool.
Agricultural Nonpoint Source Pollution Model (AGNPS)
The Agricultural Nonpoint Source Pollution Model (AGNPS) is a joint USDA-Agricultural
Research Service and -Natural Resources Conservation Service system of
computer models developed to predict nonpoint source pollutant loadings within
agricultural watersheds. The sheet and rill erosion model internal to AGNPS is based
upon RUSLE, with additional routines added to allow for continuous simulation and
more detailed consideration of sediment delivery.
AGNPS was originally developed for use in agricultural watersheds, but has been
adapted to allow consideration of construction sources.
AGNPS provides more spatial detail than GWLF and is therefore more rigorous in
calculating the delivery of eroded sediment to the receiving water. This additional
computational ability carries with it the cost of requiring more detailed information
describing the topography of the watershed, as well as requiring more time to set up and
apply the model.
Hydrologic Simulation Program – Fortran (HSPF)
The Hydrologic Simulation Program – Fortran (HSPF) uses continuous rainfall and other
meteorologic records to compute stream flow hydrographs and pollutographs. HSPF is
well suited for mixed-use (i.e., containing both urban and rural land uses) watersheds, as
it contains separate sediment routines for pervious and impervious surfaces. HSPF is an
integrated watershed/stream/reservoir model, and simulates sediment routing and
deposition for different classes of particle size. HSPF was integrated with a geographical
information system (GIS) environment with the development of Better Assessment
Science Integrating point and Nonpoint Sources (BASINS). Although BASINS was
designed as a multipurpose analysis tool to promote the integration of point and nonpoint
sources in watershed and water quality-based applications, it also includes a suite of
water quality models. One such model is Nonpoint Source Model (NPSM). NPSM is a
simplified version of HSPF that is linked with a graphical user interface within the GIS
environment of BASINS. HSPC is another variant of the HSPF model, consisting of the
equations used by HSPF recoded into the C++ programming language.
HSPF provides a more detailed description of urban areas than AGNPS and contains
direct linkage to a receiving water model. This additional computational ability carries
with it the cost of requiring more detailed model inputs, as well as requiring more time to
set up and apply the model. BASINS software can automatically incorporate existing
environmental databases (e.g., land use, water quality data) into HSPF, although it is
important to verify the accuracy of these sources before using them in the model.
Storm Water Management Model (SWMM)
The Storm Water Management Model (SWMM) is a comprehensive computer model for
analysis of quantity and quality problems associated with urban runoff. SWMM is
designed to be able to describe both single events and continuous simulation over longer
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 10
periods of time. SWMM is commonly used to simulate urban hydraulics, although its
sediment transport capabilities are not as robust as some of the other models described
here.
Soil & Water Assessment Tool (SWAT)
The Soil & Water Assessment Tool (SWAT) is a basin-scale, continuous-time model
designed for agricultural watersheds. It operates on a daily time step. Sediment yield is
calculated with the Modified Universal Soil Loss Equation. It contains a sediment routing
model that considers deposition and channel erosion for various sediment particle sizes.
SWAT is also contained as part of EPA’s BASINS software.
SWAT is a continuous time model, i.e., a long-term yield model. The model is not
designed to simulate detailed, single-event flood routing. SWAT was originally
developed strictly for application to agricultural watersheds, but it has been modified to
include consideration of urban areas.
Water Quality Methodologies and Modeling Frameworks
Numerous methodologies exist to characterize the relationship between watershed loads
and water quality for TMDL development. These include:
• Spreadsheet Approaches
• EUTROMOD
• BATHTUB
• WASP5
• CE-QUAL-RIV1
• CE-QUAL-W2
• EFDC
This section describes each of the methodologies and their suitability for defining water
quality for TMDL development. Table 2 summarizes some important characteristics of
each of the models relative to TMDL application.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 11
Table 2. Summary of Potentially Applicable Models for Estimating Water Quality
Model Time scale
Water body
type
Spatial
scale Data Needs
Pollutants
Simulated
Applicability for
TMDL
Spreadsheet
approaches
Steady
State
Creek or
lake 0- or 1-D Low
DO,
nutrients,
algae,
metals
Good for
screening-level
assessments
EUTROMOD Steady
State Lake 0-D Low
DO,
nutrients,
Algae
Good for
screening-level
assessments
BATHTUB Steady
State Lake 1-D Moderate
DO,
nutrients,
algae
Good for
screening-level
assessments; can
provide more
refined
assessments if
supporting data
exist
QUAL2E Steady
State Creek 1-D Moderate
DO,
nutrients,
algae,
bacteria
Good for low-flow
assessments of
conventional
pollutants in rivers
WASP5 Dynamic Creek or
lake 1-D to 3-D High
DO,
nutrients,
metals,
organics
Excellent water
quality capability;
simple hydraulics
CE-QUAL-RIV1
Dynamic Creek 1-D High
DO,
nutrients,
algae
Good for
conventional
pollutants in
hydraulically
complex rivers
HSPF Dynamic Creek or
lake 1-D High
DO,
nutrients,
metals,
organics,
bacteria
Wide range of
water quality
capabilities,
directly linked to
watershed model
CE-QUAL-W2
Dynamic Lake 2-D
vertical High
DO,
nutrients,
algae, some
metals
Good for
conventional
pollutants in
stratified lakes or
impoundments
EFDC Dynamic Creek or
lake 3-D High
DO,
nutrients,
metals,
organics,
bacteria
Potentially
applicable to all
sites, if sufficient
data exist
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 12
Spreadsheet Approaches
A wide range of simple methods are available to describe the relationship between
pollutant loads and receiving water quality, for a variety of situations including rivers and
lakes. These methods are documented in Mills et al. (1985). These approaches do not
require specific computer software, and are designed to be implemented on a hand
calculator or computer spreadsheet. These approaches have the benefit of relatively low
data requirements, as well as being easy to apply. Because of their simplistic nature, these
approaches are best considered as screening procedures incapable of producing highly
accurate results. They do provide good initial estimates of the primary cause-effect
relationships.
EUTROMOD
EUTROMOD is a spreadsheet-based modeling procedure for estimating phosphorus
loading and associated lake trophic state variables, distributed by the North American
Lake Management Society (Reckhow 1990). The modeling system first estimates
phosphorus loads derived from watershed land uses or inflow data using approaches
developed by Reckhow et al. (1980) and Reckhow and Simpson (1980). The model
accounts for both point and nonpoint source loads. Statistical algorithms are based on
regression analyses performed on cross-sectional lake data. These algorithms predict in-lake
phosphorus, nitrogen, hypolimnetic dissolved oxygen, chlorophyll, and
trihalomethane precursor concentrations, and transparency (Secchi depth). The model
also estimates the likelihood of blue-green bacteria dominance in the lake. Lake
morphometry and hydrologic characteristics are incorporated in these algorithms.
EUTROMOD also has algorithms for estimating uncertainty associated with the trophic
state variables and hydrologic variability and estimating the confidence interval about the
most likely values for the various trophic state indicators.
BATHTUB
BATHTUB is a software program for estimating nutrient loading to lakes and reservoirs,
summarizing information on in-lake water quality data, and predicting the lake/reservoir
response to nutrient loading (Walker 1986). It was developed, and is distributed, by the
U.S. Army Corps of Engineers. BATHTUB consists of three modules: FLUX, PROFILE,
and BATHTUB (Walker 1986). The FLUX module estimates nutrient loads or fluxes to
the lake/reservoir and provides five different algorithms for estimating these nutrient
loads based on the correlation of concentration and flow. In addition, the potential errors
in loading estimates are quantified. PROFILE is an analysis module that permits the user
to display lake water quality data. PROFILE algorithms can be used to estimate
hypolimnetic oxygen depletion rates, area-weighted or mixed layer average constitutent
concentrations, and similar trophic state indicators. BATHTUB is the module that
predicts lake/reservoir responses to nutrient fluxes. Because reservoir ecosystems
typically have different characteristics than many natural lakes, BATHTUB was
developed to specifically account for some of these differences, including the effects of
non-algal turbidity on transparency and algae responses to phosphorus.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 13
BATHTUB contains a number of regression equations that have been calibrated using a
wide range of lake and reservoir data sets. It can treat the lake or reservoir as a
continuously stirred, mixed reactor, or it can predict longitudinal gradients in trophic state
variables in a reservoir or narrow lake. These trophic state variables include in-lake total
and ortho-phosphorus, organic nitrogen, hypolimnetic dissolved oxygen, metalimnetic
dissolved oxygen, and chlorophyll concentrations, and Secchi depth (transparency).
Uncertainty estimates are provided with predicted trophic state variables. There are
several options for estimating uncertainty based on the distribution of the input and in-lake
data. Both tabular and graphical displays are available from the program.
QUAL2E
QUAL2E is a one-dimensional water quality model that assumes steady-state flow, but
allows simulation of diurnal variations in dissolved oxygen and temperature. It is
supported by the U.S. EPA Center for Exposure Assessment Modeling (CEAM) in
Athens, Georgia. The model simulates the following state variables: temperature,
dissolved oxygen, biochemical oxygen demand, ammonia, nitrate, organic nitrogen,
inorganic phosphorus, organic phosphorus, algae, and conservative and non-conservative
substances. QUAL2E also includes components that allow implementation of
uncertainty analyses using sensitivity analysis, first-order error analysis, or Monte Carlo
simulation. QUAL2E has been used for wasteload allocation purposes throughout the
United States. QUAL2E is also linked into EPA’s BASINS modeling system.
The primary advantages of using QUAL2E include its widespread use and acceptance,
and ability to simulate all of the conventional pollutants of concern. Its disadvantage is
that it is restricted to one-dimensional, steady-state analyses.
WASP5
WASP5 is EPA’s general-purpose surface water quality modeling system. It is supported
by the U.S. EPA Center for Exposure Assessment Modeling (CEAM) in Athens, Georgia.
The model can be applied in one, two, or three dimensions and is designed for linkage
with the hydrodynamic model DYNHYD5. WASP5 has also been successfully linked
with other one, two, and three-dimensional hydrodynamic models such as RIVMOD,
RMA-2V and EFDC. WASP5 can also accept user-specified advective and dispersive
flows. WASP5 provides separate submodels for conventional and toxic pollutants. The
EUTRO5 submodel describes up to eight state variables in the water column and bed
sediments: dissolved oxygen, biochemical oxygen demand, ammonia, nitrate, organic
nitrogen, orthophosphate, organic phosphorus, and phytoplankton. The TOXI5 submodel
simulates the transformation of up to three different chemicals and three different solids
classes.
The primary advantage of using WASP5 is that it provides the flexibility to describe
almost any water quality constituent of concern, along with its widespread use and
acceptance. Its primary disadvantage is that it is designed to read hydrodynamic results
only from the one-dimensional RIVMOD-H and DYNHYD5 models. Coupling of
WASP5 with multi-dimensional hydrodynamic model results will require extensive site-specific
linkage efforts.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 14
CE-QUAL-RIV1
CE-QUAL-RIV1 is a linked hydrodynamic-water quality model, supported by the U.S.
Army Corps of Engineers Waterways Experiment Station (WES) in Vicksburg,
Mississippi. Water quality state variables consist of temperature, dissolved oxygen,
carbonaceous biochemical oxygen demand, ammonia, nitrate, organic nitrogen,
orthophosphate, coliform bacteria, dissolved iron, and dissolved manganese. The effects
of algae and macrophytes can also be included as external forcing functions specified by
the user.
The primary advantage of CE-QUAL-RIV1 is its direct link to an efficient hydrodynamic
model. This makes it especially suitable to describe river systems affected by dams or
experiencing extremely rapid changes in flow. Its primary disadvantage is that it
simulates conventional pollutants only, and contains limited eutrophication kinetics. In
addition, the effort and data required to support the CE-QUAL-RIV1 hydrodynamic
routines may not be necessary in naturally flowing rivers.
HSPF
HSPF (Hydrological Simulation Program - FORTRAN) is a one-dimensional modeling
system for simulation of watershed hydrology, point and non-point source loadings, and
receiving water quality for both conventional pollutants and toxicants (Bicknell et al,
1993). It is supported by the U.S. EPA Center for Exposure Assessment Modeling
(CEAM) in Athens, Georgia. The water quality component of HSPF allows dynamic
simulation of both conventional pollutants (i.e. dissolved oxygen, nutrients, and
phytoplankton) and toxics. The toxics routines combine organic chemical process
kinetics with sediment balance algorithms to predict dissolved and sorbed chemical
concentrations in the upper sediment bed and overlying water column. HSPF is also
linked into EPA’s BASINS modeling system.
The primary advantage of HSPF is that it exists as part of a linked watershed/receiving
water modeling package. Nonpoint source loading and hydrodynamic results are
automatically linked to the HSPF water quality submodel, such that no external linkages
need be developed.
CE-QUAL-W2
CE-QUAL-W2 is a linked hydrodynamic-water quality model, supported by the U.S.
Army Corps of Engineers Waterways Experiment Station (WES) in Vicksburg,
Mississippi. CE-QUAL-W2 simulates variations in water quality in the longitudinal and
lateral directions, and was developed to address water quality issues in long, narrow
reservoirs. Water quality state variables consist of temperature, algae, dissolved oxygen,
carbonaceous biochemical oxygen demand, ammonia, nitrate, organic nitrogen,
orthophosphate, coliform bacteria, and dissolved iron.
The primary advantage of CE-QUAL-W2 is the ability to simulate the onset and
breakdown of vertical temperature stratification and resulting water quality impacts. It
will be the most appropriate model for those cases where these vertical variations are an
important water quality consideration. In un-stratified systems, the effort and data
required to support the CE-QUAL-W2 hydrodynamic routines may not be necessary.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 15
EFDC
EFDC (Environmental Fluid Dynamics Code) is a three-dimensional hydrodynamic and
water quality model supported by the U. S. EPA Ecosystems Research Division. EFDC
simulates variations in water quality in the longitudinal, lateral and vertical directions,
and was developed to address water quality issues in rivers, lakes, reservoirs, wetland
systems, estuaries, and the coastal ocean. EFDC transports salinity, heat, cohesive or
noncohesive sediments, and toxic contaminants that can be described by equilibrium
partitioning between the aqueous and solid phases. Unique features of EFDC are its
ability to simulate wetting and drying cycles, it includes a near field mixing zone model
that is fully coupled with a far field transport of salinity, temperature, sediment,
contaminant, and eutrophication variables. It also contains hydraulic structure
representation, vegetative resistance, and Lagrangian particle tracking. EFDC accepts
radiation stress fields from wave refraction-diffraction models, thus allowing the
simulation of longshore currents and sediment transport.
The primary advantage of EFDC is the ability to combine three-dimensional
hydrodynamic simulation with a wide range of water quality modeling capabilities in a
single model. The primary disadvantages are that data needs and computational
requirements can be extremely high.
MODEL SELECTION
A wide range of watershed and water quality modeling tools is available and potentially
applicable to develop TMDLs for waterbodies in the Mauvaise Terre Creek watershed.
This chapter presents the general guidelines used in model selection process, and then
applies these guidelines to make specific recommendations. In summary, two alternative
approaches can be considered for Mauvaise Terre Creek, three for Mauvaise Terre Lake,
and one approach is recommended for North Fork Mauvaise Terre Creek. The final
selection of approach is dependent upon the level of implementation to be immediately
conducted for the TMDLs. The recommendation provided here for Mauvaise Terre Creek
and Mauvaise Terre Lake assumes a level of implementation that is consistent with other
recent Illinois TMDLs.
General Guidelines
A wide range of watershed and water quality modeling tools is available and potentially
applicable to develop TMDLs. This section provides the guidelines to be followed for the
model selection process, based upon work summarized in (DePinto et al, 2004). Three
factors will be considered when selecting an appropriate model for TMDL development:
• Management objectives: Management objectives define the specific purpose of
the model, including the pollutant of concern, the water quality objective, the
space and time scales of interest, and required level or precision/accuracy.
• Available resources: The resources available to support the modeling effort
include data, time, and level of effort of modeling effort
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 16
• Site-specific characteristics: Site-specific characteristics include the land use
activity in the watershed, type of water body (e.g. lake vs. river), important
transport and transformation processes, and environmental conditions.
Model selection must be balanced between competing demands. Management objectives
typically call for a high degree of model reliability, although available resources are
generally insufficient to provide the degree of reliability desired. Decisions are often
required regarding whether to proceed with a higher-than-desired level of uncertainty, or
to postpone modeling until additional resources can be obtained. There are no simple
answers to these questions, and the decisions are often made using best professional
judgment.
The required level of reliability for this modeling effort is one able to “support
development of a credible TMDL”. The amount of reliability required to develop a
credible TMDL depends, however, on the degree of implementation to be included in the
TMDL. TMDL implementation plans that require complete and immediate
implementation of strict controls will require much more model reliability than an
implementation plan based upon adaptive management which allows incremental
controls to be implemented and includes follow-up monitoring of system response to
dictate the need for additional control efforts.
The approach to be taken here regarding model selection is to provide recommendations
which correspond to the level of detail provided in other Illinois TMDL implementation
plans conducted to date. Alternative methodologies are also provided that will support the
development of differing levels of TMDL implementation plans. For each approach, the
degree of implementation that can be supported to produce a credible TMDL will be
provided. Specific recommendations are provided which correspond to the level of detail
provided in other Illinois TMDL implementation plans conducted to date.
Model Selection for the Mauvaise Terre Creek Watershed
Tables 1 and 2 summarized the characteristics of the various watershed and water quality
methodologies with potential applicability to TMDL development. This section reviews
the relevant site-specific characteristics of the systems, summarizes the data available,
and provides recommended approaches. Data needs, assumptions, and level of TMDL
implementation support are provided for each of the recommended approaches.
Site Characteristics
Watershed characterization for the Mauvaise Terre Creek watershed was provided in the
first quarterly status report (LTI, 2004). In summary, the Mauvaise Terre Creek
watershed is located in Morgan and Scott counties in west-central Illinois. The three
waterbodies of concern are Mauvaise Terre Lake (SDL), North Fork Mauvaise Terre
Creek (DDC), and Mauvaise Terre Creek downstream of Town Brook (DD04). Mauvaise
Terre Lake and North Fork Mauvaise Terre Creek lie in Morgan County, while Mauvaise
Terre Creek flows through both Morgan and Scott Counties.
Mauvaise Terre Lake was constructed by damming the upper part of Mauvaise Terre
Creek. The lake has a surface area of 172 acres and serves as a source of drinking water
for Jacksonville and several surrounding communities. Most of the water supply,
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 17
however, comes from wells located 26 miles from Jacksonville (City of Jacksonville,
2004). The combined drainage area of the three impaired waterbodies is approximately
164 square miles. Mauvaise Terre Lake is approximately “L” shaped, with an arm
extending west from the inlet, and a second arm extending north to the dam. Mauvaise
Terre Lake is connected near the corner of the “L” to a smaller lake called Morgan Lake.
Land use in each of the three watersheds is predominantly agricultural. Crops are
primarily a corn-soy rotation, with a small amount of wheat. Urban areas comprise
approximately 7% of the Mauvaise Terre Creek watershed, 6% of the Mauvaise Terre
Lake watershed and 1.5% of the North Fork Mauvaise Terre Creek watershed.
Jacksonville is the major urban area; the City lies entirely within the Mauvaise Terre
Creek watershed. The City of South Jacksonville is also within the watershed. Other
towns in the watershed include Exeter and Oxville. A portion of the town of Chapin also
lies in the watershed. The Morgan County Health Department indicated that the
Jacksonville area has sewers, and perhaps a small area northwest of Jacksonville known
as Marnico Village, but the rest of the watershed is on private disposal systems. The
Morgan County Health Department permits and inspects all septic systems and is
unaware of any failing systems in the watershed. There are several point source
discharges in the watershed, including sewage disposal for the City of Jacksonville, food
production facilities (ACH Food Company and Nestle), and several oil wells near North
Fork Mauvaise Terre Creek. Jacksonville also has a combined sewer system and
permitted combined sewer overflows (CSOs).
The listing of Mauvaise Terre Creek on the Illinois 303(d) list for impairment due to fecal
coliform has been confirmed based on a review of the data. The listing of Mauvaise
Terre Lake for manganese, total phosphorus and nitrate and North Fork Mauvaise Terre
Creek for manganese and low dissolved oxygen have similarly been confirmed.
Potential sources of phosphorus and nitrate to Mauvaise Terre Lake include agricultural
sources, existing sediments, recreation activities, treated combined sewer discharges, and
possibly failing private sewage disposal systems. The primary source of manganese to
both Mauvaise Terre Lake and North Fork Mauvaise Terre Creek may be background
sources due to naturally high concentrations in area soils. In-place sediments may also
contribute to elevated water column concentrations in the lake. The primary potential
source of low dissolved oxygen in North Fork Mauvaise Terre Creek is agricultural
runoff. Potential sources of fecal coliform bacteria to Mauvaise Terre Creek include
livestock operations, agricultural runoff, and sewage disposal, including municipal
sewage, CSO discharges, and private disposal systems.
Data Available
Tables 3, 4 and 5 provide a summary of available water quality data from the first
quarterly status report (LTI, 2004). This amount of data is sufficient to confirm the
presence of water quality impairment, but not sufficient to support development of a
rigorous watershed or water quality model. Specific items lacking in this data set include
tributary loading data for all pollutants of concern, data describing the distribution of total
phosphorus, nitrate, manganese and fecal coliform throughout the watershed, and
continuous flow data. A USGS gage is located in a nearby watershed on Spring Creek
near Springfield (05577500), but a more accurate estimate of flows for the three
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 18
waterbodies would be obtained from a gage located within the Mauvaise Terre
watershed.
Table 3. Water Quality Data Summary for Mauvaise Terre Creek (DD04)
Sample
location and
parameter
Criterion
(cfu/100 ml)
Period of
record and
number of
data points Mean (mg/l)
Maximum
(mg/l)
Minimum
(mg/l)
Mauvaise Terre Creek, 1.5 mi NE of Merritt (Station DD04)
Fecal coliform
400 cfu/100ml
in < 10% of
samples
Geomean <
200 cfu/100
ml
1990-2004,
97 samples 5,388 240,000 <50
Table 4. Water Quality Data Summary for
North Fork Mauvaise Terre Creek (DDC)
Sample location
and parameter Criterion
Period of
record and
number of
data points Mean (mg/l)
Maximum
(mg/l)
Minimum
(mg/l)
North Fork Mauvaise Terre Creek, 0.5 M NE of Jacksonville (Station DDC11)
Dissolved oxygen 5 mg/l June 2001;
1 sample 7.8 7.8 7.8
Manganese 150 ug/l June 2001;
1 sample 78 78 78
North Fork Mauvaise Terre Creek, 3 Mi E of Jacksonville (Station DDC12)
Dissolved oxygen 5 mg/l July 2001;
2 samples 7.68 13.3 2.05
Manganese 150 ug/l
July-Oct.
2001;
2 samples
1,205 2,300 110
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 19
Table 5. Water Quality Data Summary for Mauvaise Terre Lake (SDL)
Sample
location and
parameter Criterion
Period of
record and
number of
data points Mean (mg/l)
Maximum
(mg/l)
Minimum
(mg/l)
Mauvaise Terre Lake, Near Dam Midway Between Spillway (Station SDL-1)
Manganese 150 ug/l
April-Oct
2002
5 samples
183 420 67
Phosphorus 0.05 mg/l 1990-2002
47 samples 0.162 0.344 0.015
Nitrate 10 mg/l 1990-2002
47 samples 3.91 12 <0.01
Mauvaise Terre Lake, 800 yd E. of Ramp N. of Docks (Station SDL-2)
Phosphorus 0.05 mg/l 1992 & 2002
10 samples 0.202 0.284 0.087
Nitrate 10 mg/l 1992 & 2002
10 samples 3.93 10 <0.01
Mauvaise Terre Lake, Mid Lake South of Red Brick House (Station SDL-3)
Phosphorus 0.05 mg/l 1992 & 2002
10 samples 0.248 0.370 0.118
Nitrate 10 mg/l 1992 & 2002
10 samples 4.72 13 <0.01
*note that data are for nitrate + nitrite, but water quality standard and listing are for nitrate
Recommended Approaches
This section provides recommendations for specific modeling approaches to be applied
for the Mauvaise Terre Creek watershed TMDLs. Two alternative sets of approaches are
provided for Mauvaise Terre Creek and three are provided for Mauvaise Terre Lake.
One approach is recommended for the North Fork Mauvaise Terre Creek. The
recommended approaches are presented in Tables 6, 7 and 8, with each approach having
unique data needs and resulting degree of detail.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 20
Table 6. Recommended Modeling Approaches for Mauvaise Terre Creek (DD04)
Modeling
Approach
Pollutants
considered
Watershed
Model
Water
Quality
Model
Additional
data needs
Level of TMDL
implementation
supported
Recommended
Fecal
coliform
Load
duration
curve
None
Identify whether
sources occur
during dry or wet
weather; and
identify
approximate level
of control needed
Alternative
Fecal
coliform HSPF HSPF
Tributary flow
and coliform
concentrations
at multiple
locations
Define specific
sources of bacteria
and detailed
control strategies
The recommended approach for Mauvaise Terre Creek consists of developing a load-duration
curve to address fecal coliform impairments. A load-duration curve is a
graphical representation of observed pollutant load compared to maximum allowable load
over the entire range of flow conditions. Such a graph can be developed by 1)
developing a flow duration curve by ranking the daily flow data from lowest to highest,
calculating the percent of days these flows were exceeded, and graphing the results as
shown in Figure 1; 2) translating the flow duration curve into a load duration curve by
multiplying the flows by the water quality standard as shown in Figure 2; and 3) plotting
observed pollutant loads (measured concentrations times stream flow) on the same graph
as shown in Figure 3.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 21
Figure 1. Calculation of a Flow Duration Curve (from Freedman et al., 2003)
Figure 2. Calculation of a Load Duration Curve (from Freedman et al., 2003)
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 22
Figure 3. Load Duration Curve with Observed Loads (from Freedman et al., 2003)
The load duration curve provides information to:
• Help identify the issues surrounding the problem and differentiate between point
and nonpoint source problems, as discussed immediately below;
• Address frequency of deviations (how many samples lie above the curve vs. those
that plot below), and duration (potentially how long the deviation is present)
questions; and
• Aid in establishing the level of implementation needed, by showing the magnitude
by which existing loads exceed standards for different flow conditions.
The location of loads that plot above the load duration curve is meaningful. Loads which
plot above the curve in the area of the plot defined as being exceeded 85-99 percent of
the time are considered indicative of point source influences on the water quality. Those
loads plotting above the curve over the range of 10-70 percent exceedence likely reflect
nonpoint source load contributions. NPS loads are pollution associated with runoff or
snowmelt from numerous, dispersed sources over an extended area. Some combination of
the two source categories lies in the transition zone of 70-85 percent exceedence. Those
loads plotting above the curve at exceedences less than 10 percent or more than 99
percent reflect extreme hydrologic conditions of flood or drought (Freedman et al, 2003).
The load duration curve approach will identify broad categories of coliform sources and
the extent of control required from these sources to attain water quality standards.
The alternative approach for Mauvaise Terre Creek consists of applying the HSPF model
to define watershed loads for all fecal coliform sources and using the water quality
component of this model to simulate in-stream concentrations and water quality response.
This approach, coupled with intensive monitoring, would define specific sources of
bacteria and identify detailed control strategies necessary to attain water quality
standards.
Second Quarterly Progress Report October 2004
Mauvaise Terre Creek Watershed
Limno-Tech, Inc. Page 23
Table 7. Recommended Modeling Approaches for North Fork Mauvaise Terre
Creek (DDC)
Modeling
Approach
Pollutants
considered
Watershed
Model
Water
Quality
Model
Additional
data needs
Level of TMDL
implementation
supported
Recommended
Dissolved
Oxygen Empirical
approach
QUAL2E Low flow
stream surveys
Identify primary
sources to be
controlled, and
approximate level
of control needed
Manganese
Empirical
approach
Spreadsheet
approach
Low flow
stream surveys
Identify manmade
versus natural
sources
The recommended approach for North Fork Mauvaise Terre Creek consists of using the
water quality model QUAL2E to address dissolved oxygen problems. Manganese
impairments will be addressed via spreadsheet calculations. Watershed loads for this
segment will be defined using an empirical approach. QUAL2E was selected for
dissolved oxygen modeling because it is the most commonly used water quality model
for addressing dissolved oxygen for low flow conditions. Because problems appear to be
restricted to low flow conditions, watershed loads are not expected to be significant
contributors to the impairment. For this reason, an empirical approach was selected for
determining watershed loads. The recommended approach (in conduction with additional
monitoring described below) will identify the primary sources of dissolved oxygen to be
controlled, as well as the level of control needed
Table 8. Recommended Modeling Approaches for Mauvaise Terre Lake (SDL)
Modeling
Approach
Pollutants
considered
Watershed
Model
Water
Quality
Model
Additional
data needs
Level of TMDL
implementation
supported
Recommended
Manganese,
Total
Phosphorus,
Nitrate
GWLF BATHTUB
None Identify primary
sources to be
controlled; and
approximate level
of control needed
Alternative 1
Manganese,
Total
Phosphorus,
Nitrate
None BATHTUB None
Identify
approximate level
of control needed
Alternative 2
Manganese,
Total
Phosphorus,
Nitrate
SWAT CE-QUAL-W2
Tribut