Watershed based plan for the Illinois River Watershed

              WATERSHED BASED PLAN
FOR THE
ILLINOIS RIVER WATERSHED
Prepared By:
Oklahoma Conservation Commission
Water Quality Division
2800 N. Lincoln Blvd., Suite 160
Oklahoma City, OK 73105
(405) 522-4500
- 1 -
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 2 -
ILLINOIS RIVER WATERSHED BASED PLAN
Table of Contents
LIST OF TABLES 3
LIST OF FIGURES 6
PREFACE 7
INTRODUCTION 11
WATERSHED CHARACTERIZATION 13
HISTORICAL DATA 18
CAUSES and SOURCES 61
LOAD REDUCTIONS 78
MANAGEMENT MEASURES 83
CRITERIA 92
PUBLIC OUTREACH 95
TECHNICAL AND FINANCIAL ASSISTANCE 104
IMPLEMENTATION SCHEDULE AND INTERIM MILESTONES 109
MONITORING PLAN 113
REFERENCES 125
APPENDIX A: Implementation Plan for the 2007 Illinois River 319
Cost-share Program 130
APPENDIX B: Comments from NPS Working Group 149
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 3 -
List of Tables
Table 1. Land cover in the Illinois River basin 16
Table 2. Selected parameters from the Census of Agriculture 16
Table 3. Streamflow statistics based on USGS data, 2000-2004 18
Table 4. Physico-chemical data from the Illinois River, 1974 19
Table 5. Sources of nutrient loading to Lake Tenkiller, 1974-1975 20
Table 6. OSDH water quality data 21
Table 7. Nutrient and flow data from the Illinois River watershed, 1981-1982 23
Table 8. Nitrogen and phosphorus loadings in the Illinois River watershed, 1981-1982 24
Table 9. Illinois River basin phosphorus data up to 1986 26
Table 10. Illinois River basin nitrogen data up to 1986 26
Table 11. Arkansas SCS stream ranking in the Illinois River watershed 28
Table 12. Oklahoma SCS stream ranking in the Illinois River watershed 28
Table 13. OCC stream ranking in the Illinois River watershed 29
Table 14. Water quality data from small streams in the Illinois River basin, 1990-1992 32
Table 15. Significant water quality trends from 1980-1992 34
Table 16. Comparison of water quality data from 1980-1981 with 1991-1992 35
Table 17. Four year averages for each OCC sampling location along the Illinois River,
1992-1996 37
Table 18. Nutrient load calculations for the Camp Paddle Trails and Tahlequah
sampling locations along the Illinois River 38
Table 19. Lake Tenkiller nutrient data, 1992-1993 39
Table 20. Estimated nutrient loads by source and type for three flow regimes into Lake
Tenkiller 40
Table 21. Estimates of point source discharge quantities of total phosphorus to the Horseshoe Bend area (1991-1993) 41
Table 22. Estimated distribution of total nitrogen load between background point and
nonpoint sources at Horseshoe Bend 46
Table 23. Estimated distribution of total phosphorus load between background point and
nonpoint sources at Horseshoe Bend 46
Table 24. Relative reduction in mean annual total phosphorus concentration and load
with a 25% reduction in NPS inputs 47
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 4 -
Table 25. Relative reduction in mean annual total nitrogen concentration and load with a
25% reduction in NPS inputs 47
Table 26. Relative reduction in mean annual total phosphorus concentration and load
with a 50% reduction in nonpoint source inputs 48
Table 27. Relative reduction in mean annual total nitrogen concentration and load with a
50% reduction in nonpoint source inputs 48
Table 28. Phosphorus trends in Arkansas portion of the Illinois River watershed 49
Table 29. Annual loads for total phosphorus, soluble reactive phosphorus, total nitrogen,
and dissolved nitrite + nitrate nitrogen at the Illinois River south of Siloam
Springs, AR 50
Table 30. Mean water quality values in or near the Barren Fork basin, 1999-2004 56
Table 31. Flow-weighted nutrient concentrations 58
Table 32. Annual loads for the Illinois River at Highway 59 bridge in AR 58
Table 33. Phosphorus loads and concentrations in the Illinois River, 1997-2004 59
Table 34. Impaired streams in the Illinois River watershed in OK, 2008 62
Table 35. Estimates of point source discharge quantities of total phosphorus to the
Horseshoe Bend area of Lake Tenkiller, 1991-1993 64
Table 36. Estimated annual phosphorus loads from WWTPs in the Illinois River basin
from 1990-2001 64
Table 37. Estimated annual effluent loads from WWTPs in the Illinois River basin from
2007-2020 66
Table 38. 1988 estimates of commercial animals in the Illinois River watershed 69
Table 39. Estimated number and type of birds produced in the OK portion of the Illinois
River basin 69
Table 40. Public sewer data for OK counties in the Illinois River watershed 72
Table 41. Modified land cover distribution in the Illinois River watershed 76
Table 42. Contributions of total phosphorus at subbasin gages used for SWAT 79
Table 43. Predicted phosphorus loads to Lake Tenkiller at various waste application rates
and point source concentrations 81
Table 44. Total phosphorus load reaching Lake Tenkiller for different scenarios based
on SWAT 82
Table 45. Best management practices installed through the 2007 319 project 88
Table 46. Annual total phosphorus loads for 1990-2206, 1990-2006 with new point
sources, and 2020 with predicted land cover changes 92
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 5 -
Table 47. Best management practice implementation projects / efforts identified for implementation 105
Table 48. Identified education and outreach funding efforts / needs 106
Table 49. Identified funding needs for monitoring 107
Table 50. Specific funding needs identified for computer modeling 108
Table 51. Schedule and load reduction goals (Interim and Long-term) 111
Table 52. Schedule for 2007 Illinois River Watershed 319 Riparian Program 111
Table 53. Ambient stream monitoring stations 115
Table 54. OCC monitoring sites in Illinois River watershed 118
Table 55. OCC analytical parameters and sampling frequency 119
Table 56. USGS parameters and sampling frequency for streams 120
Table 57. OWRB stream and lake monitoring sample variables 121
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 6 -
List of Figures
Figure 1. Illinois River watershed 11
Figure 2. Major tributaries and towns in the Illinois River watershed 13
Figure 3. Elevation in the Illinois River watershed 14
Figure 4. Landuse in the Illinois River watershed 16
Figure 5. Sampling sites for OSDH survey 22
Figure 6. USGS monitoring sites, 1980-2002 52
Figure 7. Average annual total phosphorus at USGS sites, 1980-2002 53
Figure 8. Total phosphorus trends at Oklahoma USGS sites, 1980-2002 54
Figure 9. Total phosphorus loads from WWTPs with a significant discharge in the
Illinois River basin 65
Figure 10. Significant urban locations and total phosphorus loads from WWTPs 66
Figure 11. Permitted potential pollution sources in Illinois River basin 68
Figure 12. Estimated average soil test phosphorus for pastures receiving waste and not currently receiving waste 70
Figure 13. Per unit area sediment yield by land cover from upland areas as predicted by
SWAT 75
Figure 14. Total phosphorus load per unit area as predicted by SWAT 76
Figure 15. Sediment yield per unit area as predicted by SWAT 77
Figure 16. Total phosphorus reaching Lake Tenkiller by source based on SWAT 77
Figure 17. Upland total phosphorus load per unit area by land cover and by state 78
Figure 18. Locations of USGS gaging stations used to calibrate SWAT 79
Figure 19. Index map for riparian targeting field book 86
Figure 20. Example of modeling result for riparian targeting 87
Figure 21. Monitoring sites in the Illinois River watershed 119
Figure 22. OWRB monitoring sites on Lake Tenkiller 123
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 7 -
PREFACE
The Illinois River watershed spans the Oklahoma-Arkansas border in the northeastern part of the state and is located in Benton, Washington, and Crawford Counties in Arkansas and Delaware, Adair, Cherokee, and Sequoyah Counties in Oklahoma. The watershed encompasses 1,069,530 total acres (approximately 1,600 square miles), with 54% located in Oklahoma. The Illinois River is designated as a State Scenic River, and, as such, it is recognized as one of Oklahoma’s most valuable water resources for reasons ranging from aesthetic and recreational value to high water quality as a drinking water source. In addition, Lake Tenkiller (Tenkiller Ferry Reservoir), which was formed by impounding the Illinois River in 1953 to provide flood control and hydroelectric power, is recognized as one of the state’s most aesthetic lakes, with water clear enough to provide exceptional recreational opportunities. Lake Tenkiller has also become a public water supply source for area municipalities.
It has been recognized since at least the early 1980's that the Illinois River and Lake Tenkiller were experiencing water quality degradation, primarily perceived as decreased clarity and frequent algae blooms in the lake. As substantial research indicated that these perceptions were based on actual problems, efforts began to focus on the potential sources of the problems. Initial research concluded that the watershed was impacted by excess nutrients and indicated that potential sources included wastewater effluent from both
Illinois River 2007
nonpoint sources such as the substantial poultry industry, nurseries, and various other agricultural sources. Streambank erosion due to loss of riparian zones and cattle access to streams was also impacting the water resources. Much of the research concluded that watersheds with the most intense landuse, primarily those with the greatest concentration of poultry and cattle, were the greatest contributors to the water quality problems.
Lake Tenkiller received a Nutrient Limited Watershed designation in 2006 due to low dissolved oxygen and an established relationship between nutrients and algae. The Clean Lakes Study data from 1992 and 1993 showed a substantial increase in chlorophyll-a over that observed in the 1974 national eutrophication study. Recent OWRB monitoring shows that the increased algae levels persist (OWRB 2005). While the average Trophic State Index (TSI) is less than 62, it is frequently exceeded. The Clean Lakes study called for nutrient reductions to limit the increased levels of algae growth. Tenkiller Lake has also been shown to be impaired by low dissolved oxygen in its hypolimnion such that the Fish and Wildlife Propagation Beneficial Use is not supported. Lake Tenkiller is on Oklahoma’s 2008 303(d) list of impaired waterbodies for total phosphorus, dissolved oxygen, and Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 8 -
chlorophyll-a. In addition, four segments of the Illinois River, as well as Chicken Creek, Town Branch of Tahlequah Creek, Ballard Creek, Caney Creek, Barren Fork Creek, Tyner Creek, Peacheater Creek, Battle Branch, Sager Creek, and two segments of Flint Creek are not supporting designated uses due to nutrients and/or pathogens (see Table 33 for details). This corresponds to 171 miles of impaired Oklahoma streams and 13,470 acres of impaired lake water.
Two lawsuits have resulted from these documented water quality problems. In 1986, the State of Oklahoma sued to stop the City of Fayetteville’s discharge into the Illinois River. The suit reached the U.S. Supreme Court in 1992, where the court ruled that the downstream state’s water quality laws must be met, but the upstream state was given the liberty to determine how best to accomplish this. In 2006, the Oklahoma State Attorney General filed a lawsuit against eleven poultry integrator companies for their role in polluting the Illinois River watershed. This lawsuit is currently underway.
An extensive amount of data has been collected for many years in this watershed assessing physical, chemical, and biological parameters. In addition, considerable efforts have already been made to address the sources of the water quality problems in the basin, and extensive work is planned for the near future. These efforts include reductions in point source loading due to cooperation between the Oklahoma Department of Environmental Quality (ODEQ) and cities of Tahlequah and Stillwell, education programs developed by the Oklahoma Scenic Rivers Commission (OSRC), the Cherokee County Conservation District, and the Oklahoma Conservation Commission (OCC), and various programs to reduce nonpoint source loading from agricultural sources in the watershed. Arkansas point source discharges have been reduced, and several Arkansas programs have been implemented to address pollution in the Illinois River watershed. Many of these studies and programs will be discussed in this document.
Both Arkansas and Oklahoma have worked with the USDA Farm Services Agency to fund Conservation Reserve Enhancement Program (CREP) Riparian Restoration in the watershed. Oklahoma is seeking additional matching funding to expand the size of its CREP program beyond approximately 9,000 acres. The States of Arkansas and Oklahoma continue to work cooperatively to seek solutions to nonpoint source pollution problems in the watershed by funding programs including riparian protection, watershed education, streambank stabilization, and alternative uses or more effective uses of poultry waste such as waste to energy, waste composting, or waste conversion to more appropriately formulated fertilizer formulas which can allow excess phosphorus to be transferred out of the watershed while nitrogen can be reapplied in the watershed at levels that are environmentally sound. Through poultry waste transfer programs, the states have worked cooperatively with the poultry industry to fund approximately $1.6 million worth of poultry waste transfer out of the Illinois and neighboring Eucha/Spavinaw watersheds.
The OWRB’s “1996 Diagnostic and Feasibility Study on Tenkiller Lake” recommended an 80% reduction of total phosphorus to return Lake Tenkiller to more acceptable conditions and halt the further degradation of water quality in the lake. A 40% reduction of the total phosphorus load to Lake Tenkiller, based on 1980-1993 data and the 1996 study, was Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 9 -
agreed upon by the states of Oklahoma and Arkansas as the initial goal for implementation in the watershed. This corresponded to a decrease of 132,855 kg/yr. The U.S. Environmental Protection Agency (USEPA) is currently developing a TMDL for the entire Illinois River watershed, including Lake Tenkiller, through a contract with a national environmental firm. This TMDL is slated for release in January of 2011. Until the release of that TMDL, goals for water quality improvement will be based on the initial reduction goal from the lake study and two SWAT (Soil and Water Assessment Tool) modeling efforts by Storm et al. (2006; 2008).
The 2006 SWAT model results estimated that 330,000 kg total phosphorus per year reached Lake Tenkiller between 1997 and 2001. The model predicted that 35% of the loading was due to point sources, leaving 65% to nonpoint sources. According to this modeling, reducing the application of poultry waste to pastures, improving pastures, and reducing the discharge of the major point sources in the watershed could dramatically improve the soluble phosphorus loading in the watershed, as well as the bacteria level, in a relatively short time frame.
Specifically, it was estimated that exporting waste from the watershed could reduce that loading by 15%, eliminating overgrazed pasture could reduce phosphorus loading by 6%, and converting all pasture to forest land would reduce loading by 55%. The model predicted that 50% of the load was due to nonpoint sources such as pastures with high phosphorus level soils, grazing, row crops/small grains, and other sources. The report goes on to say that a combination of waste export, point source improvements, pasture conversion to hayland and forest, and conversion of cropland to pasture or forest will be required to meet load reduction goals that will ultimately be necessary to attain water quality standards.
The potential of BMPs to improve water quality in this watershed has been demonstrated in a subwatershed, the Peacheater Creek watershed. A paired watershed study was conducted comparing water quality in Peacheater Creek before and after implementation of BMPs with Tyner Creek, where no BMPs were implemented. After implementation of BMPs, which included animal waste management, riparian management and improvement, pasture planting and nutrient management, offsite watering, and construction of heavy use areas for animal feeding and waste storage, total phosphorus loading was approximately 66% less than would have been expected without any BMP implementation. Total nitrogen loading was decreased by 57%, and dissolved oxygen was increased by 3%. In addition, benthic macroinvertebrate communities were significantly improved during the critical summer indexing period, and streambank erosion and nutrient loading from streambank erosion were significantly reduced.
This plan will present in detail the proposed expansion of riparian protection actions which are presently occurring or planned in the Illinois River watershed, as well as attempt to summarize the main historical research on water resources in the basin and what has already been done to remediate problems in the watershed. Although the success of the project depends on cooperation between the states of Oklahoma and Arkansas, this Watershed Based Plan (WBP) will focus only on Oklahoma’s pollution programs. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 10 -
Arkansas is similarly developing a WBP for the Arkansas portion of the watershed. These plans will eventually be combined into one basin-wide management plan, with the ultimate goal to restore beneficial use support to all waterbodies in the watershed through the coordination of efforts, both among agencies and between states. The recommendations established in the TMDL for the watershed will be used to update this plan once the TMDL is released.
This WBP has been developed with a great deal of local support. Due to its high priority as a state resource, the Illinois River has attracted the attention of many citizens and agencies. The foundation for this WBP began as the Illinois River Watershed Comprehensive Basin Management Plan (IRCBMP), a document developed in 1999 as part of a 319 project. The IRCBMP was a compilation of existing studies, reports, management recommendations, etc. as developed by numerous entities that were active in the watershed. It was reviewed extensively by the Oklahoma NPS Working Group and was modified to meet expectations and recommendations of this review. In turn, based on Clean Water Action Plan guidelines, the IRCBMP was then modified into a Watershed Restoration Action Strategy (WRAS) which was developed in_1999. One foundational document for these plans was the Oklahoma Scenic Rivers Illinois River Management Plan developed in 1998. Each of these documents had widespread input from locals in the watershed and from Oklahoma agencies. Comments received from members of the Oklahoma Nonpoint Source Working Group after review of this draft of the Illinois River WBP are included in Appendix B.
Throughout the plan, the spelling of Barren Fork Creek may deviate slightly to include “Baron” or “Barron.” This is an artifact of early studies and errors on old maps, but all three spellings denote the same waterbody. Standardization has been attempted as much as possible, but some figures still have an erroneous spelling. In addition, due to the multitude of studies in this watershed, units of measure may switch from metric to standard throughout the text. Again, standardization has been attempted, but both metric and standard units are present, depending on the source of certain figures, tables, and estimates. The OCC will try to correct these deficiencies in future updates of this WBP.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 11 -
INTRODUCTION
In 1997, a nationwide strategy to protect water quality was initiated which resulted in the development of the Clean Water Action Plan (CWAP). The CWAP established goals and implementation schedules for numerous strategies dealing with point and nonpoint sources. Oklahoma’s Office of Secretary of Environment (OSE) was designated as the state lead agency to implement the provisions of the CWAP in Oklahoma. Under OSE’s leadership, Oklahoma has successfully met the CWAP requirement to establish a Unified Watershed Assessment (UWA) strategy. Oklahoma’s UWA is a written document whose development and implementation relied upon input from the state’s UWA Work Group. Through the UWA process, the Work Group identified 150 “Category I” watersheds in Oklahoma that were recognized as significantly impaired and in need of immediate federal and state funding to target restoration activities. The top ten of these watersheds were scheduled for action to address nonpoint source (NPS) pollution. The Illinois River watershed is one of these high priority watersheds.
Cherokee Co.Delaware Co.Adair Co.Sequoyah Co.ArkansasOklahomaBenton Co.Washington Co.Crawford Co.
Figure 1. The Illinois River Watershed. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 12 -
The Nonpoint Source Program and Grants Guidelines for States and Territories for FY 2004 and Beyond requires a Watershed Based Plan (WBP) to be completed prior to implementation using incremental funds. The guidance defines the 9 key components to be addressed in a watershed-based plan, much of which builds from the strategies outlined in the Watershed Restoration Action Strategy (WRAS). These components include: 1) identification of causes and sources that will need to be controlled to achieve load reductions, 2) estimate of load reductions expected from the management measures described, 3) a description of the management measures that will need to be implemented to achieve load reductions, 4) an estimate of the amounts of technical and financial assistance needed, associated costs, and/or the sources or authorities who will bear responsibility, 5) an information/education component that will be used to enhance public understanding of the project and encourage early participation in the overall program, 6) a schedule for implementing the NPS management measures identified in this plan that is reasonably expeditious, 7) a description of interim, measurable milestones for determining whether control actions are being implemented, 8) a set of criteria that can be used to determine whether loading reductions are being achieved over time and substantial progress is being made or whether the Watershed Plan or Total Maximum Daily Load (TMDL) needs to be revised, and 9) a monitoring component to evaluate the effectiveness of the implementation efforts over time.
In order for the WBP to become an integral part of the entire watershed restoration program, it must be amenable to revision and update. The Illinois River WBP has been developed as a dynamic document that will be revised periodically to incorporate the latest information, address new strategies, and define new partnerships between watershed shareholders. Of particular note, this WBP was developed under an accelerated timeline in order to allow the Oklahoma Conservation Commission the opportunity to compete for Clean Water Act Section 319 funding in the Illinois River watershed. Consequently, this WBP may not fully incorporate all relevant data or modeling. The U.S. Environmental Protection Agency (USEPA) is currently developing a TMDL for the entire Illinois River watershed, including Lake Tenkiller, through a contract with a national environmental firm. This TMDL is slated for release in January of 2011. It is understood that the water quality goals set forth in this WBP will be revised after the release of this TMDL. The WBP will also be updated when the results of major modeling or monitoring studies are completed.
As it evolves, this WBP will become a collaborative effort with Arkansas and will continue to evolve as the partnership evolves. It is anticipated that at least biannual revisions may be necessary and that the responsibility for such revisions will rest primarily with the Oklahoma Conservation Commission (OCC), with support from the Office of the Secretary of the Environment (OSE) and the agencies involved with the NPS Working Group. Federal and state funding allocations for future water quality projects designed to address the Illinois River Watershed problems should not be based solely upon their inclusion in this WBP; rather, the WBP should be considered a focal point for initial planning and strategy development.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 13 -
WATERSHED CHARACTERIZATION (element 1)
The Illinois River watershed (Hydrologic Unit Code 11110103) extends from Northwestern Arkansas to Northeastern Oklahoma and is located in Benton, Washington, and Crawford Counties in Arkansas and Delaware, Adair, Cherokee, and Sequoyah Counties in Oklahoma. The Illinois River drains approximately 1,069,530 total acres in Arkansas and Oklahoma (approximately 54% in Oklahoma). The river is impounded to form Lake Tenkiller (Tenkiller Ferry Reservoir), and it was once impounded at the state line to form Lake Frances. The Lake Frances Dam was compromised in the 1990s, and now only the remains of the lake exist. Major tributaries into the Illinois River and Lake Tenkiller include Osage Creek, Clear Creek, Muddy Fork Creek, and Cincinnati Creek in Arkansas, and Flint Creek, Ballard Creek, Caney Creek, and Barren Fork Creek in Oklahoma (Figure 2).
Figure 2. Major tributaries and towns in the Illinois River watershed.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 14 -
Physical / Natural Features
The watershed lies within the Ozark Highlands and Boston Mountains Ecoregions, with the majority of the Oklahoma portion of the watershed in the Ozark Highland Ecoregion. The Ozark Highlands ecoregion is characterized by oak-hickory forests on well-drained soils of slopes, hills, and plains. Trees are of medium height (20 to 60 feet) with a relatively open canopy which allows a thick understory of slow-growing shrubs and trees. Areas of exposed rock are common. Blackjack oak, post oak, white oak, black hickory, and winged elm are the common overstory trees, and coral berry, huckleberry, and sassafras are representative of the understory. A taller forest community is found in protected ravines and on moist or north-facing slopes where soils are deeper and well drained. These forests are 60 to 90 feet high and consist of an overstory of sugar maples, white oaks, chinquapin oak, and hickory, with an understory of redbud, flowering dogwood, pawpaw, spice bush, sassafras and coral berry. Mosses, ferns, and liverworts are abundant on the moist forest floor. Bottomland hardwood forests of oak, sycamore, cottonwood, and elm exist along floodplains of larger streams (OCC 1998; Woods et al. 2005). Presently, rugged areas are forested and nearly level sites are used for pastureland or hayland. Elevation ranges from 300 to 1,800 ft (Figure 3).
The streams of the Ozark Highlands are typically clear, high gradient, riffle and pool type with coarse gravel, cobble, boulder, and bedrock substrates of limestone, dolomite, and chert. Base flows usually are maintained during the dry season by springs and seeps. Widespread karst features include caves, sinkholes, and springs. These features support a variety of rare species such as Gray and Ozark big-eared bats and the Ozark cavefish. Both habitat diversity and species richness are high, and sensitive fish species are common. Minnows, sunfishes, and darters are plentiful. The banded sculpin and slender madtom occur in small streams, and the southern redbelly dace inhabits headwaters. The shadow bass is nearly limited to the region. Other common fishes include the orangethroat darter, stippled darter, greenside darter, fantail darter, northern hogsucker, white sucker, Ozark minnow, cardinal shiner, and bigeye shiner. The most important game species is the smallmouth bass (ODAFF 2010c).
Figure 3. Elevation in the Illinois River Watershed (Storm et al. 2006)
The Illinois River watershed provides habitat for certain species that are both dependent on high water quality and of special conservation status. For example, the Illinois River supports a significant freshwater mussel community, including populations of the Neosho mucket (Lampsilis rafinesqueana) and rabbitsfoot mussel (Quadrula cylindrica cylindrica). Both of these mussels are candidate species for listing under the Endangered Species Act Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 15 -
(USFWS 2009), and the mucket also is listed by the State of Oklahoma as a state endangered species (OSS 2010). The Illinois River is considered to harbor one of only two remaining viable populations of the mucket, and even these populations are experiencing declines (NMWG 2005).
The southern-most section of the watershed lies in the Boston Mountains ecoregion. This ecoregion “is mountainous, forested, and underlain by Pennsylvanian sandstone, shale, and siltstone. It is one of the Ozark Plateaus; some folding and faulting has occurred but, in general, strata are much less deformed than in the Ouachita Mountains. Maximum elevations are higher, soils have a warmer temperature regime, and carbonate rocks are much less extensive than in the Ozark Highlands...Upland soils are mostly Ultisols that developed under oak-hickory and oak-hickory-pine forests. Today, forests are still widespread; northern red oak, southern red oak, white oak, and hickories usually dominate the uplands, but shortleaf pine grows on drier, south- and west-facing slopes underlain by sandstone“(Woods et al. 2005).
The Boston Mountains ecoregion streams are clear, extremely high gradient, riffle and pool type with gravel, cobble, boulder, and bedrock substrates of sandstone, shales, and limestone. There is little streamflow in the dry season because there are few springs and seeps in the Boston Mountains. The fish fauna of the Boston Mountains are nearly as species rich and diverse as the fauna in the Ozark Highlands ecoregion. Summer flow in many small streams is limited or non-existent but isolated, enduring pools may occur. Elevation ranges from 650 to 2,600 ft (Figure 3).
Major soils within the basin are in the Captina, Clarksville, Enders, Jay, Linker, Mountainberg, Nella, Nixa, Noark, Razort, Steprock, and Waben series (USDA 1992). The majority of the higher reaches of the watershed are Clarksville-Nixa-Noark: deep, loamy cherty soils, moderately to well drained, moderately to rapidly permeable. These soils are derived from cherty limestone. Soils in the vicinity of Lake Tenkiller are Enders-Linker-Mountainberg-Nella: deep, loamy, gravelly, or stony soils derived from acid sandstone, siltstone, and shale. These well drained soils range from very slowly permeable to moderately rapidly permeable.
Average annual precipitation in the Oklahoma portion of the Illinois River watershed is about 50 inches, with May and June being the wettest months. Temperatures average near 59 degrees, with a range from an average daytime high of 91 degrees in July to an average low of 27 degrees in January (www.climate.ocs.ou.edu).
Land Use
Nearly half of the Oklahoma portion of the Illinois River watershed is forested, with most of the remaining land used for hay production or pasture (Table 1; Figure 4). The major agricultural industry in the Oklahoma portion of the watershed is poultry, and a significant number of cattle are also raised. Row crops and small grains comprise a small percentage of landuse (Table 1), with wheat, sorghum, soybeans, and various vegetables being grown in small quantities in the watershed.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 16 -
Table 1. Land cover in the Oklahoma portion of the Illinois
River basin from 2001 LandSat (Storm et al. 2006).
Land Cover
Fraction of Basin
Forest
45.90%
Hay
15.42%
Well Managed Pasture
24.34%
Poorly Managed Pasture
7.98%
Rangeland
0.60%
Roads
0.16%
Urban
2.91%
Water
2.04%
Row Crop/Small Grains
0.64%
Figure 4. Landuse in the Illinois River watershed
(Storm et al. 2006).
The Census of Agriculture data in Table 2 shows that poultry production, both broilers and layers/pullets, remained relatively stable from 1992 to 2002 in Adair, Cherokee, and Delaware Counties in Oklahoma, with the exception of a sharp decline in Adair County between 1992 and 1997. The number of cattle produced in the watershed increased quite significantly over this ten-year period, particularly in Delaware County. Hay production also increased during this period.
Table 2. Selected parameters from the Census of Agriculture, 1992, 1997, 2002.
County, State
Agricultural Product
2002
1997
1992
Adair, OK
Broilers
10,888,560
12,147,732
27,739,248
Cattle & calves
59,033
56,443
51,732
Layers and pullets
517,615
934,267
1,658,694
Hay (acres)
38,312
40,242
32,267Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 17 -
County, State Agricultural Product 2002 1997 1992
Cherokee, OK
Broilers
3,442,615
3,336,028
3,930,352
Cattle & calves
45,573
46,277
37,103
Layers and pullets
cannot be disclosed
101,594
cannot be disclosed
Hay (acres)
38,450
31,390
27,097
Delaware, OK
Broilers
29,785,875
28,493,904
26,359,308
Cattle & calves
74,719
68,997
59,856
Layers and pullets
791,272
913,014
778,974
Hay (acres)
59,484
51,231
45,927
The Illinois River watershed supports a poultry industry with a capacity estimated to produce over 35 million birds annually. Storm et al. (2006) estimated that a total of 231,000 tons (210,000,000 kg) of poultry waste were produced in the Illinois River basin each year from about 475 poultry houses. This amount of waste was calculated to contain approximately 10,400,000 kg nitrogen and 2,930,000 kg phosphorus.
The region’s upland and bottomland forests support a small but active forest products industry. According to the U.S. Forest Service’s Timber Product Output report for 2005 (USDA Forest Service 2008), roundwood timber harvest from Adair, Cherokee, Delaware, and Sequoyah counties totaled 2,298 thousand cubic feet, of which 99.7% was hardwood. This represented a 15% increase over survey data from 2002. Since 2005, the annual timber harvest has likely declined in parallel with the overall economic downturn. The primary forest products directory maintained by Oklahoma Forestry Services currently shows eight wood processing plants in or near the watershed, with 21 additional plants in Benton, Crawford, and Washington counties in Arkansas. Over the next five to ten years, in addition to traditional forest products, the region’s forests will likely attract increased interest for biomass energy and wood pellets (ODAFF 2010b).
Human Population
Approximately 243,000 people live in the Illinois River watershed (2000 US Census). About 170,000 (70%) live in urban areas, with the majority residing in Arkansas. There has been rapid population growth in the watershed, especially in Northwest Arkansas, which reported a 34% increase from 1990 (115,075) to 2000 (174,691). The population of
Oklahoma cities in the Illinois River basin also increased during this time from a total of 15,365 to 20,623 (25%).
The Oklahoma portion of the Illinois River basin contains only small urban areas. The largest of these is Tahlequah, with a population of approximately 16,000 (2005 estimate). Stilwell, the county seat of Adair County, has a population of just over 3,200. According to the 2006 U.S. Census, the population of Adair County increased by 6% from 2000 to 2006 to 22,317, Cherokee County increased by 5.6% to 44,910, and Delaware County increased by 8% to 40,061.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 18 -
Waterbody Conditions
Streamflow in the Oklahoma portion of the Illinois River basin is highly variable, but it generally is highest as the river reaches Tahlequah (USGS database), shortly after which it flows into Lake Tenkiller. Table 3 presents streamflow data collected at five USGS gaging stations during the 2000-2004 time period.
Table 3. Streamflow statistics based on USGS data, 2000-2004 (Tortorelli and Pickup 2006).
Station name
Drainage area (sq. mi.)
Mean annual streamflow
(cfs)
Daily mean streamflow,
2000-2004 (cfs)
2000-2002
2001-2003
2002-2004
Minimum
Maximum
Illinois River
near Watts
635
639
539
552
83
19,200
Flint Creek
near Kansas
110
105
78
94
10
7,820
Illinois River
at Chewey
820
745
616
645
94
26,000
Illinois River
near Tahlequah
959
990
787
829
93
32,800
Barren Fork
at Eldon
307
327
250
270
23
22,300
Lake Tenkiller (Tenkiller Ferry) was completed in 1952 by the U.S. Army Corps of Engineers for flood control and hydropower. At normal pool, Lake Tenkiller has a surface area of 12,906 acres, 130 miles of shoreline, and a volume of 1,054,862,170 cubic yards. The lake drains an area of approximately 1,610 square miles, has a mean depth of 52 feet, and a maximum depth of 138 feet near the dam.
HISTORICAL DATA
Numerous projects have assessed the water quality and biological communities of the Illinois River and its tributaries, starting as early as the 1950s and continuing to the present. These projects have not been coordinated to cover all areas of concern, nor have they been conducted in a consistent manner. In addition, some of the conclusions drawn from these studies may not appear completely valid based on the data presented, but rather present some of the historic viewpoints and even biases that have affected activities in the watershed throughout its history. Despite these limitations, a substantial amount of information exists upon which to characterize water quality in the basin. Many of these early studies were reviewed and summarized in 1991 in a report titled "Evaluation and Assessment of Factors Affecting Water Quality of the Illinois River in Arkansas and Oklahoma” (Meyer and Parker 1991). In 1999, the Oklahoma Conservation Commission released the “Comprehensive Basin Management Plan for the Illinois River Basin in Oklahoma” (OCC 1999a), which summarized the most important water quality studies up to that date.
This section of the WBP will include a chronological synopsis of the research in the Illinois River watershed. This summary includes some studies discussed in the 1999 document as well as older and more recent documents. The intent of this review is not to present all Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 19 -
of the information which has been collected, but rather to give an overview of the larger, more intensive studies. The reader is referred to the original texts if additional or more detailed information is required.
A. A Preliminary Study of the Water Quality of the Illinois River in Arkansas (Kittle et al. 1974)
This study, paid for by the Illinois River Property owners of Arkansas, Inc. and performed by personnel from the University of Arkansas, concluded that the Illinois River was “unpolluted” based on assessment of water quality parameters and biota at eight sites along the river. This data was intended to provide a baseline from which to monitor the changes expected to occur with the proposed construction of two sewage treatment plants at Savoy and Siloam Springs, Arkansas. Sites 5-8 were located below the confluence of Osage Creek, which receives effluent from the towns of Springdale and Rogers, Arkansas. Increases in most parameters were observed at these sites relative to sites upstream of the Osage Creek confluence (Table 4). It was concluded that additional discharges would be detrimental to the future water quality of the Illinois River and lead to a more eutrophic state both in the river and in the downstream lakes.
Table 4. Physico-chemical data from the Illinois River, June 29 and 30, 1974.
Monitoring Station
Dissolved Oxygen (mg/l)
pH
Turbidity (FIU)
Chloride (mg/l)
Ammonia (mg/l)
Nitrate (mg/l)
Filterable
Ortho-phosphate (mg/l)
Total Ortho-phosphate (mg/l)
IR-1
7.6
7.9
12
9.99
0.329
1.76
0.083
0.134
IR·2
9.6
8.3
10
9.99
0.486
1.84
0.085
0.996
IR-3
8.8
8.2
12
9.99
0.244
1.57
0.085
0.138
IR-4
8.5
7.5
10
9.99
0.317
1.63
0.078
0.142
IR-5
7.6
7.9
12
11.50
0.329
2.23
0.271
0.446
IR-6
8.2
8.1
14
11.00
0.289
2.19
0.267
0.460
IR-7
10.9
8.5
11
11.00
0.301
1.99
0.252
0.424
lR-8
10.3
8.4
16
10.50
0.374
1. 96
0.203
0.342
Avg.
8.9
8.1
12
10.49
0.334
1.89
0.166
0.385
B. Report on Tenkiller Ferry Reservoir, Cherokee and Sequoyah Counties, Oklahoma (USEPA 1977b) and Report on Lake Frances, Adair County, Oklahoma (USEPA 1977a)
As part of the National Eutrophication Survey, water quality data was collected and analyzed for Lake Tenkiller and Lake Frances in order to compile information on nutrient sources, concentrations, and impacts. Lake Tenkiller was found to be eutrophic and phosphorus-limited. Nonpoint sources were estimated to contribute 84.5% of the total phosphorus in the lake (Table 5). Point sources in Oklahoma were estimated to contribute 15.5% of the total annual phosphorus loading, with Tahlequah responsible for 8%, Stilwell Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 20 -
for 6.5%, and Westville for 1%. The net annual accumulation of nutrients in Lake Tenkiller was estimated to be 49,745 kg of phosphorus and 526,670 kg of nitrogen.
Table 5. Sources of nutrient loading to Lake Tenkiller based on monthly grab samples, 1974-1975.
Source
Location
kg P / yr
% of total
kg N / yr
% of total
Flow (m3/sec)
NPS
Illinois River
68,875
63.2
1,750,390
67.4
23.68
Barren Fork
8,605
7.9
434,890
16.8
8.45
Minor tributaries
13,685
12.6
321,290
12.4
9.04
Municipal
STPs
Tahlequah
8,725
8
18,015
0.7
Westville
1,135
1
3,400
0.1
Stilwell
7,110
6.5
12,995
0.5
Misc.
Septic
20
<0.1
705
<0.1
Direct Precipitation
895
0.8
55,265
2.1
Total
109,050
2,596,950
Data collected from Lake Frances similarly indicated eutrophication and phosphorus limitation, with extremely high nutrient concentrations as well as high turbidity. The net accumulation of phosphorus was estimated to be 18,240 kg/yr, while the net nitrogen accumulation in the lake was 258,240 kg/yr.
In 1981-1982, a diagnostic and feasibility study for Lake Frances was performed by the USEPA which indicated that the primary cause of the observed eutrophication was phosphorus entering from the Springdale and Rogers wastewater treatment plants (Threlkeld 1983). Nutrients were retained in Lake Frances for only a short period of time before flowing into the Illinois River. This was thought to be a major contributor to the degradation of the water quality in the Illinois River downstream of the lake.
C. Nutrient Contributions to the Illinois River in Arkansas: A Preliminary Investigation (Bowen 1978)
This Master’s thesis examined nitrogen and phosphorus at four locations on the Illinois River in Arkansas as well as at three municipal wastewater plants discharging into the watershed. Sites on the river were sampled six times in 1977 in addition to two storm events, while two samples were obtained from each of the wastewater plants. During low flows, phosphorus loadings from municipal wastewater treatment plants accounted for approximately 90 percent of the total phosphorus within the watershed, but contributions of phosphorus and nitrogen from nonpoint sources during the base flow sampling period were significant. Based on the storm sampling results, contributions of nutrients from nonpoint sources were thought to exceed the contributions from point sources annually. Concentrations of phosphorus in the Illinois River were in exceedance of the levels set forth in 1981 Arkansas Water Quality Standards (0.100 mg/L TP in streams and 0.050 mg/L in lakes), and these levels were such that exceedance of standards would continue even if point source contributions of phosphorus were eliminated within the watershed.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 21 -
D. Water Quality Survey of the Illinois River and Tenkiller Ferry Reservoir (OSDH 1978)
The Oklahoma State Department of Health Water Quality Laboratory conducted an intensive 3 week study of Tenkiller Reservoir and the Illinois River upstream of the reservoir in 1976 to examine point and nonpoint sources of pollution and their impact on the watershed. Water chemistry data collected from 1975-1977 at USGS ambient monitoring stations in the watershed were examined as well. Biological samples were also collected. The primary goal of this project was to provide baseline data to determine necessary regulatory actions to abate deterioration of water quality in the basin. The data obtained was limited by sample size (ranging from 1-26 samples), so the values given in Table 6 are not necessarily representative of average annual values for the sites. As stated in the report, “the design and nature of this study…are such that there may be a proclivity to overextend data or to base assumptions on limited investigations.”
Table 6. Summary of OSDH water quality data. All values are in mg/L. Sample sizes are in parentheses. Site numbers correspond to the map, Figure 5.
Site #
Site Description
Total Nitrogen
Total Phosphorus
TKN
281
Illinois River-just above Lake Frances on Hwy 59 (in Arkansas)
2.3 (1)
0.14 (1)
0.9 (1)
2
Illinois River-below Lake Frances at Watts
2.9 (4)
0.20 (21)
1.3 (22)
283
Illinois River-above confluence of Flint Creek
2.2 (9)
0.13 (9)
1.1 (9)
284
Illinois River-below confluence of Flint Creek
2.1 (4)
0.10 (3)
1.6 (4)
274
Illinois River-at Comb's Bridge
1.4 (4)
0.07 (4)
0.8 (4)
301
Illinois River-east of Tahlequah
2.1 (3)
0.08 (10)
1.1 (10)
288
Illinois River-below confluence of Tahlequah Crk
2.4 (1)
0.12 (1)
1.0 (1)
291
Illinois River-above confluence of Barren Fork
2.2 (4)
0.06 (4)
1.3 (4)
256
Illinois River-below confluence of Barren Fork
2.1 (5)
0.10 (4)
0.7 (5)
200
Flint Creek-near Kansas, OK
3.3 (8)
0.12 (25)
1.2 (26)
270
Flint Creek-above confluence of Illinois River
2.6 (8)
0.11 (7)
1.2 (8)
302
Barren Fork-at Eldon
1.6 (8)
<0.09 (24)
0.8 (26)
289
Barren Fork-above Welling Bridge, above camp
1.8 (4)
<0.09 (4)
1.2 (4)
290
Barren Fork-above Welling Bridge, below camp
1.8 (4)
<0.09 (4)
1.2 (4)
202
Barren Fork-at Welling Bridge
1.4 (7)
<0.09 (8)
0.9 (7)
292
Barren Fork-above confluence of Illinois River
1.7 (4)
<0.09 (4)
1.2 (4)
20
Tahlequah Creek-above STP
2.2 (4)
0.14 (14)
0.9 (14)
201
Tahlequah Creek-below STP
2.8 (5)
1.09 (5)
0.9 (5)
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 22 -
Figure 5. Map of sampling sites for OSDH survey.
Major conclusions from this study are summarized below:
1)
Lake Frances was determined to be in the late stages of eutrophication due to heavy siltation and elevated nutrient levels from the Illinois River in Arkansas. Comparative water quality directly above the headwaters and at points downstream from the Lake Frances Dam suggested high nutrient loading to the Illinois River in Arkansas, which is passing through Lake Frances relatively quickly rather than being filtered out in the impoundment.
2)
Flint Creek was determined to be of inferior water quality, and point source discharge from the city of Siloam Springs sewage treatment facility was surmised to be the major factor creating this condition. Flint Creek was determined to be a major contributor of nutrients to the Illinois River, particularly during high-flow conditions. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 23 -
3)
Recreational activities in the lower Flint Creek drainage and in various segments of the Barren Fork and the Illinois River did not appear to contribute significant nutrient loading, but biological communities appeared to be disturbed at and below areas of high public usage.
4)
Based on a limited sampling regime, the Tahlequah sewage plant effluent appeared to exert little impact on the Illinois River (from less than 1% to 3% of total nutrient loading), although there was a definite increase in nutrients just below the discharge. Stormwater runoff from an urbanized area had higher nutrient loading values than rural runoff in this drainage basin.
5)
The Barren Fork was determined to be of superior water quality with no detrimental impact on the Illinois River.
6)
Non-point sources were determined to contribute approximately 95% of the nutrient loading to the Illinois River drainage basin in Oklahoma; hence, regulatory action was not thought to be necessary.
7)
The water quality of the Illinois River was determined to improve from Lake Frances to below Barren Fork.
E. Illinois River Data Summary 1981-1982 (OSDH 1983)
The Oklahoma State Department of Health and Oklahoma Scenic Rivers Commission monitored the Illinois River at 13 sites from 1981-1982 in order to calculate nutrient loadings. This study found that nitrogen and phosphorus levels increased below the river’s confluence with Town Branch. Tahlequah’s sewage treatment plant was found to discharge good quality effluent but was not able to handle sludge properly and was severely impacted by inflow and infiltration after rainfall. “Follow-up action” was taken to improve the Tahlequah plant. Nutrient levels are given in Table 7, and nutrient loads are presented in Table 8.
Table 7. Nutrient and flow data from the Illinois River watershed in Oklahoma, 1981-1982.
Site Description
Site Name
Mean Flow (MGD)
Mean Total Nitrogen (mg/L)
Mean Total Phosphorus (mg/L)
lllinois River near Watts
1955
121.5
2.00
0.278
lllinois River below hog farms at Watts and Kamp Paddletrails
OSRC 1
123.5
1.98
0.252
lllinois River 100 yards above Flint Creek confluence
OSRC 2
113.4
1.69
0.220
lllinois River at Chewey Bridge
OSRC 3
182.7
1.50
0.172
lllinois River downstream from Chewey
OSRC 4
182.7
1.60
0.196
lllinois River below Echota Public Use Area
OSRC 5
182.7
1.28
0.118
lllinois River above Tahlequah
1965
168.1
1.28
0.105
lllinois River below Town Branch (Tahlequah) confluence
0SRC 6
181.0
2.54
0.505
Sager Creek 100 feet above confluence with Flint Creek
0SRC 9
9.5
3.12
1.080
Flint Creek north of West Siloam Springs
0SRC 8
18.8
1.35
0.017
Flint Creek near Kansas
1960
26.5
1.29
0.103
Barren Fork at Proctor
0SRC 7
64.4
1.26
0.173
Barren Fork near Eldon
1970
71.0
1.37
0.081
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 24 -
Table 8. Nitrogen and phosphorus loadings in the Illinois River watershed in Oklahoma, 1981-1982.
Site
Kg/Yr/Ha Nitrogen
103Kg/Yr Nitrogen
Kg/Yr/Ha Phosphorus
103Kg/Yr Phosphorus
1955
2.4
403
0.3
45
OSRC 1
2.5
413
0.3
44
OSRC 2
1.7
306
0.2
34
OSRC3
2.3
492
0.2
52
OSRC4
2.6
567
0.2
49
OSRC 5
2.0
479
0.2
38
1965
1.9
468
0.2
33
OSRC 6
3.0
756
0.6
158
OSRC 9
27.6
51
7.5
14
OSRC 8
4.2
40
0.2
2
1960
2.0
58
0.2
7
OSRC 7
2.6
194
0.1
9
1970
3.0
238
0.1
11
F. Water Quality Assessment of the Illinois River Basin, Arkansas (Terry et al. 1984)
In 1984, the USGS and the Arkansas Department of Pollution Control and Ecology assessed the water quality of the Illinois River, Muddy Fork, Spring Creek, and Osage Creek in northwest Arkansas above Lake Frances in order to calibrate steady-state stream models (Terry et al. 1984). The models were used to simulate changes in instream water resulting from proposed changes in nutrient loading from wastewater. None of the four streams met 1981 Arkansas state standards for dissolved oxygen (4.0 mg/L), total phosphorus (0.100 mg/L), or fecal coliform bacteria (geometric mean of 200 colonies/100 mL and no more than 10% of samples greater than 400 colonies/100 mL during recreation season). The water temperature in Spring Creek and Osage Creek downstream from the Springdale and Rogers wastewater-treatment plants, respectively, also exceeded Arkansas standards. Analysis of data and modeling results indicated that significant nutrient loads were being contributed to the streams during runoff periods in addition to the load due to treatment plant effluent.
Neither the Illinois River nor Muddy Fork were projected to meet Arkansas dissolved oxygen standards (Arkansas Department Pollution Control and Ecology 1981) with any of the proposed effluents from the proposed Fayetteville and existing Prairie Grove WWTPs. Osage and Spring Creeks were projected to be able to meet standards if effluents were not allowed to exceed certain values (see report for details). The phosphorus concentrations in the Illinois River during the study period ranged from 0.03-0.61 mg/L, while the tributaries had a range of 0.03-0.80 mg/L phosphorus (approximately 45% of samples exceeded 0.10 mg/L). Organic nitrogen in both the Illinois River and its tributaries ranged from 0.00-1.10 mg/L, with stormwater runoff values between 0.71-1.50 mg/L phosphorus.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 25 -
G. An Intensive Survey of the Illinois River (Arkansas and Oklahoma) in August 1985 (Gatstatter and Katko 1986)
An USEPA study of the Illinois River basin in Oklahoma and Arkansas in August 1985 examined background phosphorus concentrations at 24 mainstem and tributary sites. Osage Creek had much higher phosphorus concentrations than the other sites; concentrations in Osage Creek were from 7 to 60 times higher than background concentrations and increased the Illinois River total phosphorus concentrations by 3 to 10 times. This was attributed to the effluent discharged into Osage Creek from the Springdale and Rogers WWTPs. The amount of phosphorus in Osage Creek was substantially affecting the water quality of the Illinois River above Lake Frances, as well as the water quality within the lake itself. In addition, this elevated phosphorus was found to affect water quality in the Illinois River as far as 20 miles downstream of Lake Frances. Muddy Creek, which receives effluent from the Prairie Grove WWTP, was found to have total phosphorus concentrations from zero to five times higher than background conditions, representing a relatively small contribution to the nutrient load in the Illinois River as a whole.
Total phosphorus concentrations in Flint Creek ranged from 4 to 7 times the background amounts, likely due to wastewater effluent from Siloam Springs; however, this was not having a significant effect on the Illinois River at its confluence since the phosphorus concentrations were high at this location (due to the Osage Creek inflow). High background inorganic nitrogen concentrations (>2.5 mg/L) were observed in the upper basin, where no point sources were located. This was thought to be due to land application of animal waste.
H. Evaluation and Assessment of Factors Affecting Water Quality of the Illinois River in Arkansas and Oklahoma (Meyer and Parker 1991)
This report attempted to gather all data collected up to 1986 concerning water quality in the Illinois River Basin into a single document and to interpret the results. One of the major areas of focus was the identification of trends in the data over time and space which are discussed in the following sections.
Total Phosphorus{tc \l3 "1. Total Phosphorus}
Spatial trends - statistically significant decrease in concentration from the Arkansas border to Tahlequah.
- statistically significant increase in concentration below Osage Creek.
Temporal trends - statistically significant increases at nine of seventeen sites.
Mean values were in excess of the recommended level of 0.05 mg/L at all sites with some being exceptionally high. The data summary for phosphorus is included in Table 9. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 26 -
Table 9. Illinois River Basin phosphorus data up to 1986. All sites are located on the Illinois River unless otherwise stated.
Station ID
Site Description
Site #
n (months)
Total Phosphorus as P (mg/L)
Mean
Median
SD
USGS 07195000
Osage Cr. nr. Elm Springs
1
134
1.082
0.755
0.927
SR 0.5
Lake Frances, SW end
2
14
0.313
0.295
0.100
USGS 07195500
Hwy 54, N of Watts
3
170
0.293
0.198
0.313
SR 1
Below Watts
4
64
0.265
0.233
0.151
SR 2
Above Flint Cr. confluence
5
66
0.225
0.192
0.176
USGS 07195860
Sager Cr., W of state line
6
117
1.496
0.820
1.021
USGS 07196000
Flint Cr. at Hwy 33
7
127
0.188
0.172
0.090
SR 3
W of Chewey
8
66
0.211
0.184
0.098
SR 4
Round Hollow State Park
9
66
0.201
0.170
0.081
SR 4.5
Comb’s Bridge, W of Ellersville
10
14
0.200
0.187
0.090
SR 5
2 mi. above USGS 07196500
11
66
0.181
0.133
0.295
USGS 07196500
Hwy 62, NE of Tahlequah
12
127
0.130
0.100
0.133
SR 6
Just below Tahlequah STP
13
62
0.845
0.387
0.936
SR 6.3
Above Barren Fork confluence
14
11
0.154
0.118
0.074
USGS 07197000
Barren Fork at Hwy 51
15
126
0.079
0.044
0.102
Nitrite/Nitrate {tc \l3 "2. Nitrite/Nitrate}
Spatial trends - statistically significant decrease in concentration from the Arkansas border to Tahlequah.
- increase in concentration below Osage Creek.
Temporal trends - statistically significant increases at most sites.
Mean values were high at all sites and exceeded recommended values of 1.0 mg/L. The data for summary is included in Table 10.
Table 10. Illinois River Basin nitrogen data up to 1986.
Station ID
Site Description
Site #
n (months)
Total Nitrogen as N (mg/L)
Mean
Median
SD
USGS 07195000
Osage Cr. nr. Elm Springs
1
108
4.081
4.000
1.262
SR 0.5
Lake Frances, SW end
2
14
1.843
1.625
0.749
USGS 07195500
Hwy 54, N of Watts
3
110
1.510
1.200
0.873
SR 1
Below Watts
4
64
1.819
1.800
0.966
SR 2
Above Flint Cr. confluence
5
66
1.673
1.400
1.491
USGS 07195860
Sager Cr., W of state line
6
80
2.888
2.250
1.031
USGS 07196000
Flint Cr. at Hwy 33
7
98
1.291
1.100
0.679
SR 3
W of Chewey
8
66
1.480
1.475
0.778
SR 4
Round Hollow State Park
9
66
1.459
1.300
0.797
SR 4.5
Comb’s Bridge, W of Ellersville
10
14
1.357
0.417
0.647
SR 5
2 mi. above USGS 07196500
11
66
1.293
1.200
0.953
USGS 07196500
Hwy 62, NE of Tahlequah
12
96
1.052
0.800
0.718
SR 6
Just below Tahlequah STP
13
62
2.245
1.600
1.619
SR 6.3
Above Barren Fork confluence
14
10
1.266
1.200
0.550
USGS 07197000
Barren Fork at Hwy 51
15
98
0.914
0.700
0.628 Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 27 -
Nitrogen/Phosphorus Ratios {tc \l3 "3. Nitrogen/Phosphorus Ratios}
The ratio of nitrogen to phosphorus found during baseflow conditions is important in understanding the ability of the water to support algal growth and for management purposes, as the addition of a limiting nutrient would accelerate algal growth. There is some range of opinion concerning the N:P ratio at which one or the other element becomes the factor responsible for limiting algal growth. The majority of research indicates that at N:P ratios of less than 10-16, nitrogen is the limiting nutrient, while phosphorus becomes limiting at higher ratios.
Nitrogen/phosphorus ratios are much lower from the river main stem and main tributaries than for the smaller tributaries. It can be seen by comparing the data from the two data sets that nitrogen values are relative similar, while phosphorus values are much higher at the main stem sites. This suggests that point sources of phosphorus are playing a major role in maintaining high river values.
Nutrient Sources {tc \l3 "4. Nutrient Sources}
Considerable attention was paid to the identification of nutrient sources, especially in regard to phosphorus loading. It was estimated that phosphorus loading from point versus nonpoint sources was approximately equal during low flow conditions but that nonpoint sources exceeded point sources during normal or high flows.
In terms of annual loading of phosphorus it was estimated that the loading at the upper end of Lake Tenkiller was 21% from point sources and 79% from nonpoint sources. Total point source loading of phosphorus was estimated to account for 12% of the Oklahoma total.
Effects on Lake Tenkiller {tc \l3 "5. Effects on Lake Tenkiller}
The primary conclusion that was drawn from the data was that phosphorus loading exceeds the level that would cause Lake Tenkiller to become eutrophic, as predicted by Vollenweider's model.
I. Illinois River Cooperative River Basin Resource Base Report (USDA 1992){tc \l2 "B. ILLINOIS RIVER COOPERATIVE RIVER BASIN RESOURCE BASE REPORT}
The objectives of this report were to better define water quality problems of the Illinois River basin, to prioritize watersheds needing project action to improve water quality, and to develop separate water quality project plans on high priority watersheds in Arkansas and Oklahoma. This report covers a wide variety of subjects, including natural resources, human resources, problems, concerns, ongoing activities, and recommendations. The main outputs of the report include three systems for designating priority watersheds developed by three different agencies: Arkansas Soil Conservation Service (SCS), Oklahoma SCS, and the Oklahoma Conservation Commission (OCC). These results are seen in Tables 11, 12, and 13. The Arkansas SCS system was developed using agricultural nonpoint potential source data, land use, municipal water supply locations, benthic data, and chemical data. The Oklahoma SCS system was developed using Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 28 -
agricultural nonpoint potential source data, land use, and watershed size. The OCC system was developed using agricultural nonpoint potential source data and water sampling data. The highest priority watersheds for both states are generally low order streams or headwater streams. Many of the highest priority subwatersheds in Oklahoma were tributaries of the Barren Fork Creek.
Table 11. Arkansas SCS stream ranking in the Illinois River watershed.
Rank
Watershed
County
Score
Rank
Watershed
County
Score
1
Clear Creek
Washington
3202
20
Cincinnati Creek
Washington
NG
2
Upper Osage
Benton
3197
21
Lower Moores Creek
Washington
NG
3
Little Osage
Benton
3186
22
Goose Creek
Washington
NG
4
Blair Creek
Washington
2684
23
Fly Creek
Washington
NG
5
Barren Fork Creek
Washington
2400
24
Kinion Creek
Washington
NG
6
Spring Creek
Benton
2281
25
Brush Creek
Washington
NG
7
Upper Moores Creek
Washington
2279
26
Muddy Fork of Ill. River
Washington
NG
8
Ballard Creek
Washington
2163
27
Sager Creek
Benton
NG
9
Flint Creek
Benton
2134
28
Lick Branch
Benton
NG
10
Upper Illinois River
Washington
2094
29
Robinson Creek
Benton
NG
11
Lower Osage Creek
Benton
2082
30
Gallatin Creek
Benton
NG
12
Ruby Creek
Washington
2037
31
Evansville Creek
Washington
NG
13
Gum Springs Creek
Benton
NG
32
Lake Wedington
Washington
NG
14
Fish Creek
Washington
NG
33
Puppy Creek
Benton
NG
15
Little Flint Creek
Benton
NG
34
Cross Creek
Benton
NG
16
Wildcat Creek
Washington
NG
35
Frances Creek
Benton
NG
17
Galey Creek
Benton
NG
36
Chambers Creek
Benton
NG
18
Hamstring Creek
Washington
NG
37
Pedro Creek
Benton
NG
19
Wedington Creek
Washington
NG
NG: not given in report
Table 12. Oklahoma SCS stream ranking in the Illinois River watershed.
Rank
Watershed
County
Rank
Watershed
County
1
Tyner Creek
Adair
31
Pumpkin Hollow
Adair
2
Peacheater Creek
Adair
32
Mulberry Hollow
Cherokee
3
Ballard Creek
Adair
33
Dry Creek and Bolin Hollow
Adair, Cherokee, Sequoyah
4
Green Creek
Adair
34
Cedar Hollow & Tully Hollow
Cherokee
5
Tahlequah & Kill H., Rock Branch
Adair
35
Field Hollow
Cherokee, Adair
6
Battle Branch Creek
Delaware
36
Dripping Springs
Adair, Delaware
7
Shell Creek
Adair
37
Smith Hollow
Adair
8
Evansville Creek
Adair
38
Goat Mountain
Adair
9
Mollyfield, Peavine Hollow
Cherokee
39
Walltrip Branch
Adair, Cherokee
10
Scraper Hollow
Adair
40
Tailholt Creek
Adair, Cherokee
11
Peavine Branch
Adair
41
Mining Camp Hollow North
Cherokee
12
England Hollow
Adair
42
Linder Bend & Saw Mill Hollow
Sequoyah
13
Tate Parrish
Adair
43
Luna Branch
Adair
14
Bidding Creek
Adair
44
Pettit Branch
Cherokee, Sequoyah
15
South Briggs
Cherokee
45
Pine Hollow
Sequoyah
16
West Branch
Adair
46
Park Hill Branch
Cherokee
17
Sager Creek
Delaware
47
South Proctor Branch
Adair Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 29 -
18
Hazelnut Hollow
Delaware
48
Snake & Cato Creek
Sequoyah
19
Blackfox, Winset Hollow
Adair, Cherokee, Delaware
49
Elk Creek
Cherokee, Sequoyah
20
Bluespring Branch
Cherokee
50
Terrapin Creek
Sequoyah
21
Fagan Creek
Delaware
51
Mining Camp Hollow South
Cherokee
22
Crazy Creek
Delaware
52
Burnt Cabin Creek
Sequoyah
23
Negro Jake Hollow
Adair, Cherokee
53
Sizemore Creek
Cherokee, Sequoyah
24
Fall Branch
Adair
54
Proctor Mountain Creek
Adair, Cherokee
25
North Briggs Hollow
Cherokee
55
Ross Branch & Tahlequah Cr.
Cherokee
26
Calunchety Hollow
Delaware
56
Kirk Springs & Sawmill Hollow
Adair, Cherokee
27
Falls Branch
Cherokee
57
Dripping Springs Hollow
Cherokee
28
Steeley Hollow
Cherokee
58
Dennison Creek
Adair
29
Beaver Creek
Adair, Delaware
59
Welling Creek
Cherokee
30
Five Mile Hollow
Delaware
60
Telemay & Dog Hollow
Cherokee
Table 13. OCC stream ranking in the Illinois River watershed.
Prioritization Based on Phosphorus
Prioritization Based on Nitrogen
HU*
Name
Rank
HU*
Name
Rank
509
Tyner (Lower & Upper)
1
512
Peacheater
1
330
Kill, Rock & Tahlequah
337
Ballard
337
Ballard (Lower)
610
Fagan
609
Sager
604
Battle Branch
518
Shell
518
Shell
604
Battle Branch
514
England
514
England
315
Mollyfield
325
Fall Branch (East)
606
Hazelnut
333
Tate Parrish
2
521
West
2
610
Fagan
609
Sager
521
West
515
Green
504
Field
509
Tyner (Lower & Upper)
321
Fall Branch
333
Tate Parrish
310
Cedar & Tully
330
Kill, Rock, & Tahlequah
513
Scraper
607
Crazy
323
Black Fox & Winset
603
Calunchety
519
Peavine (E&W)
3
513
Scraper
3
607
Crazy
519
Peavine (E & W)
331
Dripping Springs Br.
404
Bidding
315
Mollyfield
334
Beaver
309
Pumpkin
331
Dripping Springs Br.
603
Calunchety
520
Evansville (L&U)
512
Peacheater
325
Fall Branch (E)
606
Hazelnut
602
Five Mile
408
Goat
4
402
Negro Jake
4
219
Bolin & Dry
408
Goat
507
Walltrip Branch
227
Parkhill
334
Beaver
409
Mulberry
520
Evansville (L&U)
323
Black Fox & Winset Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 30 -
Prioritization Based on Phosphorus Prioritization Based on Nitrogen
HU* Name Rank HU* Name Rank
227
Parkhill
312
Steeley
403
Tailholt
326
Luna
404
Bidding
507
Walltrip Branch
302
Ross & Town Branch
5
407
Smith
5
515
Green
309
Pumpkin
510
South Proctor (E&W)
510
South Proctor (E&W)
204
Linder Bend
403
Tailholt
401
Negro Jake
321
Fall Branch
213
Terrapin
310
Cedar & Tully
225
Mining Camp South
502
Mining Camp North
215
Sizemore
302
Ross & Town Branch
218
Elk
6
216
Petit
6
207
Burnt Cabin
212
Pine
326
Luna
504
Field
407
Smith
219
Bolin & Dry
312
Steeley
605
Bluespring Branch
602
Five Mile
506
South Briggs Hollow
216
Petit
509
Proctor Mountain
212
Pine
307
North Briggs Hollow
409
Mulberry
7
225
Mining Camp South
7
502
Mining Camp North
215
Sizemore
506
South Briggs Hollow
209
Cato & Snake
605
Bluespring Branch
204
Linder Bend
309
Kirk Spr./Sawmill
511
Dennison
209
Cato & Snake
319
Kirk Spr./Sawmill
307
North Briggs Hollow
218
Elk
314
Dog & Telemay
213
Terrapin
Missing Data
Missing Data
226
Dripping Spr. Hollow
207
Burnt Cabin
508
Proctor Mountain
314
Dog & Telemay
511
Dennison
226
Dripping Spr. Hollow
503
Welling Creek
503
Welling Creek
HU* Hydrologic Unit Number
The report also included recommendations for improving environmental quality of the basin. Water quality plans were completed for Upper Osage, Little Osage, and Clear Creeks in Arkansas in 1992 and for Shell and Ballard Creeks in Oklahoma in 1991. These plans suggested voluntary adoption of conservation practices by producers, with technical assistance provided by the SCS, and cost share incentives provided by the ASCS, with a strong education and information program to correct and prevent agricultural source nonpoint source pollution. Additional recommendations made in the report based on a review of studies included: Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 31 -
1. Continued support of governor’s animal waste task force in Arkansas as a means to coordinate agency programs and projects and identify inadequacies, overlap, and/or conflict in animal waste regulations or guidelines.
2. A complete review of existing regulation, legislation, and agency policies concerning animal waste in Oklahoma to determine deficiencies.
3. A comprehensive study of groundwater quality coordinated with nonpoint source programs where possible, and continued support of ongoing groundwater monitoring.
4. Continued streamlining and development of new practices to protect water quality.
5. Further development and support of technology to compost and market poultry waste as a soil improvement.
6. Continued development of water quality farm plans, particularly in priority watersheds in response to local concerns and needs.
7. Development of an intensive educational program to educate the public, landowners, and operators about the extent of the nonpoint source pollution problem, the potential of their operation to contribute to the problem, and sources of available assistance.
8. Encouragement of innovative development and implementation of measures to protect, improve, or enhance water quality in the basin by:
• evaluation of existing programs, laws, and policies to determine potential contributions to water quality improvement and necessary modifications and expansions.
• identification of need and development of new programs.
• establishment of an effective monitoring program.
• establishment of a governor’s advisory group in Oklahoma to support water quality issues and provide a forum for economic growth while minimizing impacts on the environment.
9. Development of phosphorus discharge limits based on the cumulative phosphorus capacities in Lake Tenkiller and the Illinois River, to be included in all point source discharge permits.
J. Water Quality in the Subwatersheds of the Illinois River Basin (OCC 1992)
{tc \l2 "D. WATER QUALITY IN THE SUBWATERSHEDS OF THE ILLINOIS RIVER BASIN}
Sixty-two small streams in the Illinois River watershed were monitored by the OCC during 1990-1992 to determine the extent of nonpoint source (NPS) pollution occurring from land uses in small watersheds and to rank the watersheds as part of the BMP implementation process. Streams were monitored on a quarterly basis under baseflow conditions and twice per year during runoff events. The data from these collections are summarized in Table 14. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 32 -
Table 14. Water quality data from small streams in the Illinois River basin, 1990-1992.
From column 3 it can be seen that the average N:P ratio is much greater than 16. In only 4 of 64 streams was the N:P ratio less than 16, and only one was less than 10. From these data it was inferred that, as a basin-wide phenomenon, phosphorus availability is much more important in determining levels of algal growth than nitrogen; therefore, the discussion of nutrient levels focused on phosphorus. It was also inferred from this ratio and the high average nitrogen value that adequate nitrogen existed in these streams to support luxuriant algal growth.
Total Nitrogen
Baseflow
(mg/L)
Total Phosphorus
Baseflow
(mg/L)
N:P Ratio
Baseflow
(%)
Total Nitrogen
Runoff event
(mg/L)
Total Phosphorus
Runoff event
(mg/L)
Nitrogen
(runoff/baseflow)
(%)
Phosphorus
(runoff/baseflow)
(%)
Minimum
0.18
0.001
8.51
0.24
0.004
0.41
0.31
Maximum
6.40
0.752
660
6.63
0.731
3.39
32.00
Mean
1.48
0.041
79
1.74
0.058
1.23
1.93*
* = maximum value omitted (value = 2.41 with outlier)
Phosphorus values were distributed as follows:
Range (mg/L) # of stream segments
<0.005 - <0.020 31
0.020 - <0.050 20
>0.050 13
From these data it was concluded that phosphorus was adequate to support rich algal growth in many streams of the Illinois River Basin, although it was inadequate in concentration relative to the amount of nitrogen present. This conclusion may seem somewhat contradictory as it suggests that phosphorus is both plentiful yet limiting. This type of contradictory evidence supports an assertion that algal productivity is closely tied to the abundance of some other nutrient or factors such as light or substrate. The identity of this nutrient or factor could not be determined from study results.
The mean total nitrogen for all stream segments tested was 1.48 mg/L with the values being distributed as follows:
Range (mg/L) # of stream segments
0.18 - 0.89 23
0.90 - 2.00 21
>2.00 20
These data indicated that approximately two-thirds of the streams in the basin had nitrogen values which could result in eutrophic conditions. With twenty streams having values greater than 2.00 mg/L, it was apparent that nitrogen levels were high enough to be a cause of concern for stream quality as well as downstream loading. These data also supported the conclusion that nitrogen was not a limiting factor for algal growth.
The data was also examined in terms of the relative concentration of nutrients under baseflow versus runoff conditions. As can be seen in the last two columns of Table 8, both Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 33 -
nitrogen and phosphorus were elevated in runoff conditions. In some cases this was extreme while in other streams, water appears to have been diluted. However, on average, nitrogen concentration increased approximately 23% while phosphorus increased 93%. Given the increased discharge during runoff events and the fact that the values gathered probably do not represent maximum event concentrations, it was concluded that runoff of nutrients was an important contributor to stream and subsequently river water quality.
K. Illinois River Basin—Treatment Prioritization Final Report (Sabbah et al. 1995){tc \l2 "E. ILLINOIS RIVER BASIN-- TREATMENT PRIORITIZATION FINAL REPORT}
The OCC contracted with Oklahoma State University to use more sophisticated methods such as geographical information systems analysis to coordinate different types of data and prioritize subwatersheds in the Illinois River Basin (Sabbah et al. 1995). This report was an attempt to more closely relate land use and water quality information. The effort used the SIMPLE (Spatially Integrated Models for Phosphorus Loading and Erosion) modeling system developed by OSU to estimate watershed-level sediment and phosphorus loading to surface water bodies.
A section of the report dealt with identification and rank of potential phosphorus and sediment sources in the Peacheater Creek and Battle Branch Creek watersheds. Data layers were assembled including a digital elevation model, soil data, and current land use information assembled by the Oklahoma Cooperative Extension Service. Historical rainfall records (1950-1989) were used to run 40 one-year simulations. Long-term averages of runoff, sediment, and phosphorus loadings were estimated for each field and used to predict fields with high environmental risk potentials.
Average annual sediment loading from fields in the Battle Branch Watershed ranged from 0.00 - 0.88 Mg/ha. Predicted sediment loading was highest along the stream channel and from pasture, crop land, and hay meadows as opposed to woodlands. Average annual total phosphorus loading to the stream ranged from 0.00 kg/ha - 9.34 kg/ha. Highest loadings came from fields with high soil test phosphorus levels and from cropped fields, pastures, and hay meadows. Highest loadings were also seen in the headwaters of the watershed, as opposed to lower in the watershed, suggesting BMP implementation should focus on headwater areas and then move downstream.
Average annual sediment loading from fields to Peacheater Creek ranged from 0.00 - 0.96 kg/ha. Again, predicted sediment loading was highest along stream channels and from hay meadows and crop land. Average annual total phosphorus loading to the stream in Peacheater Creek ranged from 0.01 - 34.88 kg/ha. Highest loadings came from hay and pasture land and were associated with high soil phosphorus levels. These high soil P levels were believed to result from application of poultry waste and perhaps from pasturing cattle. Again, areas providing the highest phosphorus loading were concentrated in the headwaters. This suggested BMP implementation should focus in headwaters before Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 34 -
downstream areas.
Two critical ideas are supported by this report. The first is that much of the soil erosion in these watersheds happens along stream courses and is probably associated with stream bank erosion. The second is that much of the phosphorus comes from the headwaters of the watershed, thus remediation efforts should concentrate in this area.
L. Oklahoma Scenic Rivers Commission—River Trend Study (Lynch 1992) {tc \l2 "C. OKLAHOMA SCENIC RIVERS COMMISSION - RIVER TREND STUDY}
The data from samples collected by the Oklahoma Scenic Rivers Commission was analyzed to determine existing and historic water quality conditions, as well as any trends which might be present. An excellent historic data base exists for several sites where monthly samples were collected since December 1980. This report covered the analysis of approximately 120 samples collected between 12/1980 and 10/1992 from each of the following sites: Kamp Paddle Trails, Fiddlers Bend, Chewey Bridge, Round Hollow, Echota Bend, Illinois River below the Tahlequah Creek confluence, Flint Creek, and Sager Creek.
Other sites were sampled less frequently due to changes in sample site location and other factors; therefore, less data existed from these sites, and that which exists may be temporally disrupted or may cover a limited duration. Despite these limitations, some of this data was very useful in interpreting stream conditions. This included the following sites: Peavine Hollow, No Head Hollow, Barren Fork Creek, Hwy 59 bridge (Arkansas), Hwy 16 bridge (Arkansas), Illinois River above Osage Creek (Arkansas), and Illinois River above Flint Creek.
Trend analysis was used to determine long-term changes in water quality using the Seasonal Kendall Tau test. Taken as a whole, the data from the long-term sites showed few trends, and those trends which existed were of a low magnitude. This indicated that there was little change in the quality of water at these sites over the almost twelve year sampling period. However, there was a high degree of variance in the data such that the values fluctuated widely from month to month. Some of this fluctuation was due to changes in river volume; therefore, if values could have been looked at in terms of loading, the data would probably have been more uniform. The wide degree of data variance probably masked some trends. Trends which were found to be statistically significant (95% confidence level) are listed in Table 15.
Table 15. Significant water quality trends from 1980-1992.
Site
Trend
Parameter
Site
Trend
Parameter
Kamp Paddle Trails
positive
turbidity
Round Hollow
negative
COD
Fiddlers Bend
negative
COD
Echota Bend
negative
COD
Fiddlers Bend
negative
phosphorus
Echota Bend
positive
turbidity
Chewey Bridge
negative
COD
IR blw. Tahlequah Cr.
negative
COD
Chewey Bridge
positive
phosphorus
IR blw. Tahlequah Cr.
positive
turbidity Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 35 -
Chewey Bridge
positive
turbidity
The best overall conclusion that could be drawn from this data was that chemical oxygen demand (COD) appeared to be dropping at several sites, but turbidity seemed to be increasing. Given the amount of variance in the data, these analyses were largely unsatisfactory; therefore, long-term changes were looked using time sequence data to compare average values during early years to that of later years. In this case, data averages for the first two years were compared to those of the last two years of sample collection as listed in Table 16. On the whole, averages from the two time periods were not very different, which corroborates the findings that there was not much of a trend over the years of the study. Again, there was considerable variation within the two-year periods; therefore, mean values may have been weighted by unusual events, and differences in means may not be statistically significant.
Table 16. Comparison of water quality data from 1980-1981 with data from 1991-1992.
Site
Date
COD (mg/L)
Total Nitrogen (mg/L)
Total Phosphorus (mg/L)
TSS (mg/L)
Turbidity (NTU)
Kamp Paddle Trails
80/81
10.6
2.02
0.253
17.6
11.1
91/92
6.6
2.49
0.236
20.1
12.3
Fiddlers Bend
80/81
7.1
1.78
0.223
9.5
4.1
91/92
3.7
2.22
0.170
6.4
3.9
Chewey Bridge
80/81
6.3
1.62
0.195
7.2
4.4
91/92
4.5
1.98
0.170
4.3
5.0
Round Hollow
80/81
6.6
1.71
0.196
6.3
3.2
91/92
4.0
2.02
0.166
5.2
3.1
Echota Bend
80/81
6.8
1.40
0.090
5.4
2.8
91/92
4.1
1.93
0.115
5.9
2.8
IR blw. Tahlequah
80/81
8.7
2.45
0.475
11.9
4.7
91/92
7.6
4.37
0.825
4.5
2.5
Barren Fork Creek
80/81
4.6
1.59
0.152
2.2
1.2
91/92
4.4
1.85
0.315
2.7
1.5
Flint Creek
80/81
4.5
1.54
0.041
3.1
2.7
91/92
3.7
2.14
0.111
4.5
1.5
Sager Creek
80/81
6.9
3.13
1.008
2.4
1.1
91/92
11.3
5.76
0.724
1.8
1.9
Total nitrogen increased at all sites between the two periods. Although these increases were not generally of a large magnitude, the fact that they occurred at all sites led to the conclusion that nitrogen loading had increased in the Illinois River. There was no consistent increase or decrease in total phosphorus values among the sites, but these values were all very high.
Of all the data, the increases in Flint Creek and Barren Fork Creek were the most significant. The values from the samples collected the first year at Flint Creek were uniformly low and often below the detection limit of 0.005 mg/L. These values began to Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 36 -
rise during 1982, but the two-year average was still quite low compared to other sites. The 1991-1992 values from this site were much higher and indicated a real change in phosphorus concentrations over the study period. A similar situation occurred in Barren Fork Creek, where seventeen of the first twenty-four samples collected contained phosphorus concentrations below the detection limit. The 1991-1992 values were greatly increased, indicating a definite change in water quality in this river.
The concentration of TSS did not change much over the study period. The values were similar down the course of the river with the exception of Kamp Paddle Trails, which was much higher than other sites, probably due to the dislodging of sediments from Lake Frances. From the data in this study, it could not be concluded that any observable changes had occurred between 1980 and 1992.
Results of the data analysis indicated that a significant portion of the nutrients in the river were coming from across the Arkansas border; however, significant contributions were occurring within Oklahoma, too. From the data it was obvious that sewage treatment plant discharges posed a major threat to river quality, although it was difficult to assess the magnitude of this contribution relative to that from non-point sources based on these data. Contributions of nutrients within Oklahoma between Fidler’s Bend and Tahlequah were surmised to be almost entirely nonpoint source in nature.
The contribution of nutrients and sediment from Lake Frances was of concern, also. Given the deteriorating structural conditions of the dam in 1992, it was possible that almost all of the accumulated lake sediment would eventually be discharged into the river as it meandered across the lake bed unless corrective measures were taken. Given the levels of nutrients in the river, it was not surprising that Lake Tenkiller was experiencing nutrient problems as demonstrated by accelerated eutrophication. The lake is expected to continue to degrade at a rapid rate until these nutrient levels are significantly reduced.
M. Report on Water Quality for the Illinois River (Canty 1996)
As an expansion of the river trend study summarized above, the OSRC and OCC continued monitoring at 14 sites in the Illinois River watershed monthly from 1992-1996. Observed nutrient concentrations were excessive at all 10 Illinois River sites as well as at the sites on three tributaries. Nitrate and ortho-phosphorus were the predominant contaminants, and total phosphorus levels greatly exceeded the suggested USEPA concentration of 0.05 mg/L. Total phosphorus values typically ranged from 0.16 – 0.25 mg/L, with Sager Creek, located approximately three miles from the Siloam Springs WWTP, having the highest average of 0.62 mg/L. Total nitrogen values were also very high, ranging from 1.37-2.69 mg/L, approximately ten times higher than USEPA “unpolluted” values. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 37 -
Table 17. Four year averages for each OCC sampling location along the Illinois River, selected parameters (1992-1996).
SITE
SITE (abbrev.)
Total Nitrogen
TKN
N03
Total Phosphorus
Ortho-Phosphorus
TSS
Turbidity
COD
IR upstream of Osage Creek
IRUO
2.26
0.47
1.81
0.14
0.08
20.32
9.47
6.89
IR at Highway 16
HWY
2.66
0.47
2.19
0.25
0.20
60.05
6.57
8.58
IR at Kamp Paddle Trails
CMP
2.69
0.51
2.19
0.22
0.18
45.97
34.08
5.58
IR upstream of Flint Creek
IRUF
2.38
0.39
1.96
0.17
0.11
6.64
5.27
3.51
Flint at Fagan Creek
FAG
2.51
0.52
2.02
0.12
0.07
6.62
3.40
3.84
Sager Creek
SAG
5.90
0.55
5.56
0.62
0.53
5.09
3.64
5.83
Flint Creek upstream of IR
FLT
2.36
0.36
1.99
0.12
0.10
8.20
3.11
3.22
IR downstream of Flint Creek
IRDF
2.43
0.43
2.00
0.20
0.14
26.84
12.55
4.91
IR at Round Hollow
RND
2.35
0.46
1.68
0.18
0.14
24.84
4.60
5.96
IR at No Head Hollow
NH
2.17
0.42
1.75
0.18
0.12
33.17
7.38
5.91
IR at Echota Bend
ECH
2.06
0.44
1.63
0.16
0.12
24.51
4.50
4.94
IR at Tahlequah
TAL
1.95
0.37
1.63
0.17
0.10
27.33
9.98
5.61
IR upstream of Barren Fork
lRUB
1.85
0.36
1.52
0.17
0.13
25.52
5.82
5.16
Barren Fork Creek
BFK
1.37
0.30
1.12
0.12
0.06
9.67
2.96
3.62
Review of the four year average nutrient data (Table 17) indicated an increase in turbidity at four sites (Kamp Paddle Trails, No Head Hollow, Flint Creek near Fagan Creek, and Echota Bend); however, only one site was of significant concern. Kamp Paddle Trails had an increasing trend of 0.59 NTU/year which was thought to be due to eroding lake bed sediments from Lake Frances. Conversely, turbidity at Sager Creek decreased. Four sites experienced a small but significant decrease in the amount of total phosphorous (Kamp Paddle Trails, the Illinois River upstream of Flint Creek, Round Hollow, and Sager Creek). Trend analysis at Sager Creek indicated a more significant decrease of 0.044 mg/L per year in phosphorous over the fifteen year time period, which was assumed to be due to the sewage treatment plant upgrade implemented by the city of Siloam Springs. Significant, positive trends in total nitrogen were observed at all seven sampling sites evaluated, with the increase in concentration varying from 0.036 to 0.232 mg/L per year. The reason for the increase in nitrogen was probably due to increased agriculture, recreation, and urban development in the watershed; however, no responsible source of pollution could be identified by this study.
Review of the four year average nutrient data indicated an increase in nutrient pollution (nitrate, total nitrogen, orthophosphate, total phosphorous, and COD) between the Illinois River upstream of Osage Creek and the Highway 16 sampling locations. The increase in pollution was thought to be due to nutrient loading coming from Osage Creek, which contains wastewater effluent from the cities of Springdale and Rogers along with nonpoint source pollution from the surrounding watershed. Since the flow volume of Osage Creek is considerably greater than the Illinois River prior to the confluence, it was assumed that the elevated nutrient concentrations observed at the Highway 16 site were due primarily to watershed activities in the Osage Creek area, specifically those activities related to the tremendous urban development in the watershed. Nutrient concentrations increased as follows: nitrate increased 1.2 times, orthophosphate more than doubled (0.08 mg/L to 0.20 mg/L), and total phosphorous increased from 0.14 mg/L to 0.25 mg/L. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 38 -
Sager Creek, which receives sewage treatment effluent from the city of Siloam Springs, also had elevated concentrations of nutrients, notably higher than any other site for nitrate, total nitrogen, orthophosphate, and total phosphate. Sager Creek had historically shown exceptionally high levels of nutrients. Despite these findings, the effects of the wastewater treatment plant on Sager Creek appeared minimal based on fish and macroinvertebrate assessments as well as periphyton monitoring.
In order to assess the nutrient load due to Oklahoma contributions, the point where the river enters Oklahoma (Kamp Paddle Trails site) and the Tahlequah sampling points were compared. Since the river increases in volume by roughly 1.5 times between these two locations, it can be misleading to make a direct comparison of nutrient concentration without correcting for flow. As a means for direct comparison, the nutrient loadings in kilograms per year were calculated for the total phosphorous and total nitrogen parameters at these two sites. Average discharge volumes from the USGS gauging stations at Watts and Tahlequah from 1990-1994 were used for flow estimates. Average nutrient concentrations from the 1992-1996 time period were used to represent average river concentrations. Loadings were calculated by multiplying average concentration (mg/L) by average discharge (cfs) to produce an annual load (kg/yr) (Table 18).
Water quality data from the USGS and ODEQ were compared with the OCC data to verify accuracy of the loading estimates (Table 18). Moderate differences were expected due to temporal and spatial sampling variation, but there was no statistical difference between the data sets with the exception of the Tahlequah total phosphorous data. The USGS Tahlequah site had a significantly lower total phosphorous value than the OCC and the ODEQ data. Low sample size was thought to be the cause of the lower USGS value, and the OCC and ODEQ estimates were considered more accurate.
Table 18. Nutrient load calculations for the Camp Paddle Trails and the Tahlequah sampling locations along the Illinois River.
Agency and Site
Avg Discharge (cfs)
Total Phosphorus (mg/L)
Total Nitrogen (mg/L)
Total Phos. Loading (kg/yr)
Total Nitrogen Loading (kg/yr)
OCC Kamp Paddle Trails
863.6
0.22
2.68
169,500
2,065,000
OCC Tahlequah
1313.6
0.17
1.97
199,200
2,309,000
USGS Watts
863.6
0.208
2.55
161,800
1,964,000
USGS Tahlequah
1313.6
0.085
1.77
99,600
2,065,000
DEQ Watts
0.296
2.37
DEQ Tahlequah
0.150
1.74
Analysis of the loading data indicated that approximately 169,500 kg/year of total phosphorous and 2,065,000 kg/year of total nitrogen were entering the state of Oklahoma from Arkansas. Comparing these figures with the Tahlequah values suggested that there was a significant increase in total phosphorous and total nitrogen within the state of Oklahoma (29,700 and 244,000 kg/year, respectively). This suggests that watershed influences within Oklahoma are contributing significantly to the nutrient load in Lake Tenkiller. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 39 -
N. Clean Lakes Phase I Diagnostic and Feasibility Study of Lake Tenkiller (Jobe 1996){tc \l2 "F. CLEAN LAKES PHASE I DIAGNOSTIC AND FEASIBILITY STUDY OF LAKE TENKILLER}
The OWRB contracted with Oklahoma State University Water Quality Research Laboratory to conduct an USEPA Phase I Clean Lakes Study on Lake Tenkiller to diagnose the problems and recommend solutions. OSU WQRL studied the lake intensively between April 1992 and October 1993. Samples were collected at eight stations in and below the lake. Water quality in the Illinois River and its tributaries was also analyzed for purposes of the study.
The study determined that water quality in Lake Tenkiller was showing signs of degradation. Symptoms included periodic algae blooms, excessive algal growth, and extensive hypolimnetic anoxia throughout stratified periods. The lake was classified as eutrophic based on nitrogen, phosphorus, and chlorophyll a concentrations, which were excessive when compared to published criteria (Table 19).
Tble 19. Lake Tenkillernutrient data, 1992-1993.
a
PARAMETER
STATION
MEAN
MEDIAN
SD
n
o-PHOSPHATE
(mg/l)
1
0.11
0.09
0.05
16
2
0.05
0.04
0.03
18
3
0.04
0.03
0.03
18
4
0.04
0.03
0.03
18
5
0.03
0.02
0.03
18
6
0.02
0.01
0.02
18
7
0.02
0.01
0.02
18
TOTAL
PHOSPHORUS
(mg/l)
1
0.14
0.12
0.07
16
2
0.08
0.08
0.03
18
3
0.08
0.08
0.04
18
4
0.08
0.07
0.04
18
5
0.05
0.05
0.03
18
6
0.04
0.02
0.04
18
7
0.03
0.02
0.04
18
NITRATE
(mg/l)
1
1.27
1.18
0.56
16
2
0.53
0.46
0.44
17
3
0.49
0.36
0.45
18
4
0.46
0.34
0.42
18
5
0.38
0.21
0.38
18
6
0.44
0.30
0.40
18 Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 40 -
PARAMETER
STATION MEAN MEDIAN SD
n
7
0.47
0.30
0.36
18
TOTAL
NITROGEN
(mg/l)
1
2.25
2.18
1.00
16
2
1.45
1.16
0.75
17
3
1.40
1.23
0.77
17
4
1.34
1.17
0.66
17
5
1.06
0.79
0.60
17
6
0.97
0.74
0.59
17
7
1.01
0.74
0.64
17
The study estimated the total nutrient loading to the lake and partitioned that estimate by source. These estimates (Table 20) represent loading to the lake from both Oklahoma and Arkansas. The loads were predominantly derived from nonpoint sources during high flows, although point sources contribute significant amounts of nutrients during low flows. These nutrient loads, especially the nonpoint fractions, had increased significantly since 1974 but had stabilized since 1985-1986. The load estimates in Table 20 have been adjusted downward based on calculations to account for instream nutrient decay. Estimated nutrient loading from point sources before application of this decay correction is shown in Table 21, and detailed explanation of the load calculations is located in the study.
Table 20. Estimated nutrient loads, by source and type, for three flow regimes into Lake Tenkiller.
Source
Estimated Average Load at Horseshoe Bend
kg/yr (%)
Estimated Low Flow Contribution at Horseshoe Bend
kg/yr (%)
Estimated Medium Flow Contribution at Horseshoe Bend
kg/yr (%)
Estimated High Flow Contribution at Horseshoe Bend
kg/yr (%)
N
P
N
P
N
P
N
P
Background
550000 (23.9)
25000
(11.0)
35200
(22.8)
1600
(9.7)
208450
(23.9)
5225
(10.9)
306350
(24.0)
18175
(11.2)
Point Source
61605 (2.7)
12547
(5.5)
35793
(23.2)
7290
(44.1)
19406
(2.2)
3952
(8.2)
6407
(0.5)
1305
(0.8)
Nonpoint Source
1688980 (73.4)
190078
(83.5)
83345
(54.0)
7628
(46.2)
643869
(73.9)
38968
(80.9)
961795
(75.5)
143482
(88.0)
Total
2300585
227625
154338
(6.71)
16518
(7.26)
871725
(37.89)
48145
(21.15)
1274552
(55.40)
162962
(71.59)
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 41 -
Table 21. Estimates of point source discharge quantities of total phosphorus to the
Horseshoe Bend Area of Lake Tenkiller (1991 to 1993 data).
The excessive nutrient loads have increased algal growth and thus compromised water clarity throughout the lake and its tributaries. Nutrient limitation analysis indicated that the lake was phosphorus limited in the lower end (near the dam), variably limited (phosphorus, nitrogen, and light) in the midreaches, and probably light limited in the headwaters. Based on these results, it was concluded that source control of phosphorus loading was the optimum management alternative. Accumulation of toxics in the lake water and sediments and resident fish did not appear to be a problem.
After considering the feasibility and effectiveness of control measures, the report recommended a 30 - 40% reduction in headwater phosphorus loads be implemented as a short-term goal and a 70 - 80 % reduction as a long-term goal. Since both of these goals still indicated a significant risk of hypolimnetic anoxia, it was further recommended that re-aeration devices be installed in the tailrace to protect the downstream trout fishery.
The report recommended the following programs be initiated to attempt to reduce phosphorus contamination within the basin:
1. Voluntary switch to non-phosphate detergents by all lakeside residents and the cities of Tahlequah and Watts, OK and Rogers and Springdale, AK.
2. Implementation of best management practices upstream from Lake Tenkiller to minimize contributions of phosphorus in surface water runoff from agricultural fertilizer and waste and poultry waste applications.
3. Continue to work with point source dischargers, to the extent possible within the watershed, to minimize discharges of nutrients, including phosphorus
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 42 -
4. Establish a citizens’ monitoring group for basic water quality analysis and evaluation, thus affording a more robust assessment of management effectiveness.
O. Determining the Nutrient Status of the Upper Illinois River Basin Using a Lotic Ecosystem Trophic State Index (Matlock et al. 1996){tc \l2 "G. DETERMINING THE NUTRIENT STATUS OF THE UPPER ILLINOIS RIVER BASIN USING A LOTIC ECOSYSTEM TROPHIC STATE INDEX}
The Clean Lakes Study determined that Lake Tenkiller was phosphorus limited at the lower end, variably limited by nitrogen, phosphorus, and light availability in the mid-reaches, and light limited at the upper end. However, it was unknown whether the Illinois River was limited by the same factors. One goal of this study was to determine which nutrients most often limit primary productivity in tributaries to the Illinois River.
The watersheds of three tributaries to the Illinois River were chosen based on availability of historical water quality data, similar land use, and similar size. These were Peacheater Creek, Tyner Creek, and Battle Creek. Although Battle Creek watershed was smaller than Peacheater and Tyner Creek watersheds, all had predominantly pasture and range land use (63 to 68 percent) and substantial forest cover (32 to 36 percent). The main difference in land uses among the three watersheds was the degree of anthropogenic activity.
The study used in situ nutrient limitation assays to estimate limiting nutrients in the three creeks. Six nutrient enrichment treatments were tested: 1) Nitrate - 5 ppm, 2) Phosphate - 5 ppm, 3) Nitrate and phosphate - 5 ppm, 4) Micronutrients - from Weber et al. (1989) at 200 times concentration, 5) Total nutrients, consisting of treatments 3 and 4, combined, and 6) Control- deionized water. Periphytometers were colonized in a run 0.3 m deep above a riffle for 14 days. Growth surfaces were protected from grazers with an aluminum screen. Assays were conducted in April and October 1995.
Comparisons of the treatment means suggested that Battle Creek was phosphorus limited in the spring 1995 but limited by something other than nutrients during the fall, possibly light availability, which would be affected by turbidity. Peacheater Creek appeared to be co-limited by nitrogen and phosphorus during both spring and fall sampling. Tyner Creek appeared to be limited by some factor other than nutrients during the spring and co-limited during the fall. Conclusions of the report focused on the variable status of growth limiting factors in tributaries of the Illinois River. The variability of growth limiting factors in these streams suggests they are primarily impacted by nonpoint source pollution. Nonpoint sources vary temporally as well as they do in substance and nature of pollution. A stream impacted by point sources would be expected to have a more consistent growth limiting factor between seasons than was seen in these results. The findings of this report support conclusions of previous studies that nutrients and sediment are problematic in the Illinois River Basin and that nonpoint sources as well as point sources are contributing to the water quality problems.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 43 -
P. Analysis of Bank Erosion on the Illinois River in Northeast Oklahoma (Harmel 1997){tc \l2 "H. ANALYSIS OF BANK EROSION ON THE ILLINOIS RIVER IN NORTHEAST OKLAHOMA}
One source of increased turbidity in the Illinois River, its tributaries, and Lake Tenkiller, as well as increased bedload in the Illinois River and its tributaries, was believed to be streambank erosion. However, the magnitude of the contribution of streambank erosion had not been investigated until OSU and the OCC completed a survey of bank erosion on the Illinois River in 1996-1997. This project involved completion of several milestones:
1. Initial bank characterization, selection of banks for detailed study, and detailed characterization of selected banks were performed and reported in the Bank and Reach Characterization Report.
2. Long-term bank erosion was measured from aerial photographs and reported in the Aerial Photograph Erosion Analysis Report.
3. Short-term bank erosion was measured in the field at selected sites along the length of the river.
Initial Bank Characterization{tc \l3 "1. Initial Bank Characterization}
In July 1996, 193 bank segments along the length of the Illinois River from below Lake Frances dam to Horseshoe Bend on the upper portion of Lake Tenkiller were characterized. Data was generally collected only on eroding banks; however, several stable banks were characterized to provide a comparison. Data collected included length, height, angle, river position, location, material, vegetation type and percent cover, root depth and density, maximum water depth, bankfull depth, and percent flow in the near bank region under bankfull flow conditions. Banks were then grouped according to physical and vegetative conditions and hydrologic influence. At least one bank from each group (36 sites) was selected for detailed characterization. Selected sites were characterized with Rosgen Level III stream reach condition evaluation. Twenty-three of the 36 sites were characterized as C4c-channels, 11 as C4, and 2 as F4. C4c and C4 channels are gravel dominated, slightly entrenched, gentle gradient, riffle/pool channels with high width/depth ratios. These channels, characterized by depositional features, are very susceptible to shifts in stability caused by flow changes and sediment delivery from the watershed. F4 channels have similar characteristics but are entrenched. Channel bars were common, and bank erosion rates were likely high due to mass-wasting of the steep banks.
Aerial Photograph Erosion Analysis{tc \l3 "2. Aerial Photograph Erosion Analysis}
USDA-SCS 1:7920 scale aerial photographs taken in 1958, 1979, and 1991 were analyzed to estimate long-term bank erosion. Analysis yielded information on the 193 initially characterized sites in addition to 28 other significant erosional / depositional areas (generally greater than 0.5 acres lost by erosion or gained by deposition). Measurements included maximum lateral erosion, lateral erosion and/or deposition, land surface area, and length. For the period between 1958 and 1979, maximum lateral erosion averaged 67 ft, lateral erosion averaged 37 ft or 1.7 ft/yr, and lateral deposition averaged 47 ft or 2.2 Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 44 -
ft/yr. A total of 64 acres of land was eroded, and 78 acres was deposited. The length of eroding areas averaged 1014 ft, and the length of depositional areas averaged 999 ft. For the period from 1979 to 1991, maximum lateral erosion averaged 74 ft, lateral erosion averaged 41 ft or 3.6 ft/yr, and lateral deposition averaged 5 ft or 0.4 ft/yr. A total of 195 acres of land surface area was eroded and 13 acres was deposited. The length of eroding areas averaged 1131 ft. and the length of depositional areas averaged 665 ft.
The river width, measured at each 0.5 river mile from bank tracings indicated that the river was widening, with increased width in the downstream direction. Average river width for 1979 and 1991 was 175 ft and 206 ft, respectively. River width in the first 21 mile section averaged 147 ft in 1958, 158 ft in 1979, and 185 ft in 1991. For miles 21 to 42, average width increased from 169 ft in 1979 to 195 ft in 1991. Average width on the lower third of the river increased from 199 ft in 1979 to 239 ft in 1991. Overall, the Illinois River became an average of 18% wider between 1979 and 1991.
The impact of riparian vegetation was measured using long-term erosion data. Relationships tested included maximum lateral erosion rate for forested, grassed, and mixed sites, maximum lateral erosion rate for forested, grassed, and mixed sites given the site eroded between 1958 and 1991, and percent of grassed, forested, and mixed bank length that eroded or received deposition. Between 1979 and 1991, mean erosion was greater on grassed and mixed land than on forested land but the change was not statistically significant. From 1958 to 1979, mean values were significantly different between forested, grassed, and mixed sites. Although mean values were generally lowest on forested areas, data indicated that major erosion could occur on forested as well as grassed and mixed sites and minor erosion could occur on grassed and mixed vegetation sites as well as forested sites.
The lengths of erosional and depositional areas were compared to vegetation data to determine the percent of forested, grassed, and mixed vegetation area length that eroded or received deposition. In both time periods, grassed areas had the greatest percent length of erosion and deposition and forested areas had the least. Over the two comparison periods, grassed areas were almost twice as likely to experience detectable erosion compared to mixed vegetation areas and 3.5 times more than forested areas.
Field Measurement of Bank Erosion{tc \l3 "3. Field Measurement of Bank Erosion}
Short-term streambank erosion was measured with bank pins and cross-section surveys from September 1996 to July 1997. Erosion was measured after major flow events (exceeding 9000 cfs at the Tahlequah gage station) in September 1996, twice in November 1996, and in February 1997. Erosion was measured for 33 and 29 sites (out of 36 sites) after the second and fourth major flow events, respectively. After the first and third events, only 11 and 18 sites were measured due to lost pins.
Cumulative erosion after the four major flow events averaged 4.5 ft and ranged from -0.03 to 26.5 ft. Erosion was also measured once after two at or near bankfull events that occurred in spring and summer 1997. Erosion from these two events from averaged 0.40 ft and ranged from 0.00 to 2.35 ft. This study was conducted during a wet year when Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 45 -
streamflow volume and frequency of significant flow events exceeded normal conditions. The average flow was 1123 cfs from August 1, 1996 to July 31, 1997, representing a 20% increase from normal conditions and a 3.0 year return period. Flow events also occurred with greater or equal to a 2 year return period during the course of this sampling. Data from the surveys indicated that several sites experienced moderate to major aggradation. Other sites experienced degradation, although to a lesser degree than the aggrading sites experienced aggradation.
The impact of riparian vegetation was evaluated on short-term erosion data. Cumulative erosion for 27 sites after four major flow events was compared to riparian vegetation data. Differences in bank erosion between forested, grassed, and mixed sites suggested mean erosion from grassed and mixed sites exceeded that of forested sites. However, large variability among the vegetation types caused none of the differences to be statistically significant. Substantial erosion occurred on some forested sites while little erosion occurred on some grassed sites.
Conclusion{tc \l3 "Conclusion}
One of the major sources of sediment in the Illinois River basin was likely streambank erosion. Much of the watershed was grassland or forested (92%). Although clearing of forested areas for pasture was increasing, this area still represented only a small portion of the watershed. Estimated inputs of sediment from bank erosion (3.5 million tons of material between 1979 and 1991) indicated this to be a significant, perhaps the major source, contributing to bedload in the river and sedimentation of Lake Tenkiller.
Long-term erosion analysis indicated that natural riparian forested vegetation was important in reducing and preventing bank erosion on the Illinois River. Grassed banks were 3.5 times more likely to erode than forested banks and almost twice as likely at mixed vegetation banks.
In addition, the river was changing to a wider, shallower, perhaps braided river. Data showed that in addition to extensive bank erosion, the river had widened from an average of 175 ft in 1979 to 206 ft in 1991. Both the width to depth ratio and the sinuosity in many reaches of the river approached or fulfilled the Rosgen criteria for a braided channel. Many channel reaches showed signs of aggradation, which can follow a cycle of high sediment input (either from upland or bank erosion), increased in-channel deposition, and increased bank erosion.
Q. An Investigation of the Sources and Transport of NPS Nutrients in the Illinois River Basin in Oklahoma and Arkansas (Gade 1998)
The focus of this study was to estimate the quantity of nutrients delivered to Lake Tenkiller at the Horseshoe Bend area, as well as to identify the sources of those nutrients. Autosamplers were installed at two locations in Oklahoma, one on Barren Fork Creek and the other on the Illinois River near Tahlequah. In 1993, two high flow events were analyzed at the first site, and three high flow events were assessed at the other site. Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 46 -
Results indicated that nitrogen concentrations decreased with initial increase in discharge due to dilution. Phosphorus concentrations gradually declined after an initial peak due to
runoff. High flows delivered the greatest mass of nitrogen and phosphorus (54% and 61%). Osage Creek was found to be the main point source contributor to the Illinois River.
Historical water quality data from 1980-1993 was examined for eight USGS sites in the basin in order to determine whether changes had occurred over time. A significant increase in both the concentrations and loads of both nitrogen and phosphorus was seen at most sites during this time period. Gade estimated (from a QUAL2EU model) that 2-3% of the total nitrogen entering Lake Tenkiller was from point sources and that about 73% was from nonpoint sources (NPS) (Table 21). About 84% of the total phosphorus load entering the lake was from NPS and 6% was from point sources (Table 22). During low or base flow, a larger percentage of the phosphorus load was from point sources (about 44.1%), but this was dramatically reduced when the flow increased.
Table 22. Estimated distribution of total nitrogen load between background, point, and nonpoint sources at the Horseshoe Bend area of Lake Tenkiller.
Source
Estimated Average Total Nitrogen Load at Horseshoe Bend (kg/yr)
Estimated Low Flow Contribution at Horseshoe Bend (kg N /yr)
Estimated Medium Flow Contribution at Horseshoe Bend (kg N /yr)
Estimated High Flow Contribution at Horseshoe Bend (kg N /yr)
Background
550,000 (23.9%)
35,200 (22.8%)
208,000 (23.9%)
306,000 (24.0%)
Point Source
61,600 (2.7%)
35,800 (23.2%)
19,400 (2.2%)
6,400 (0.5%)
Nonpoint Source
1,690,000 (73.4%)
83,400 (54.0%)
644,000 (73.9%)
962,000 (75.5%)
Total
2,300,600
154,400
871,400
1,274,400
(6.7% of Total)
(37.9% of Total)
(55.4% of Total)
Table 23. Estimated distribution of total phosphorus load between background, point, and nonpoint sources at the Horseshoe Bend area of Lake Tenkiller.
Source
Estimated Average Total Phosphorus Load at Horseshoe Bend (kg/yr)
Estimated Low Flow Contribution at Horseshoe Bend (kg P /yr)
Estimated Medium Flow Contribution at Horseshoe Bend (kg P /yr)
Estimated High Flow Contribution at Horseshoe Bend (kg P /yr)
Background
25,000 (11.0%)
1,600 (9.7%)
5,230 (10.9%)
18,200 (11.2%)
Point Source
12,500 (5.5%)
7,290 (44.1%)
3,950 (8.2%)
1,300 (0.8%)
Nonpoint Source
190,000 (83.5%)
7,630 (46.2%)
39,000 (80.9%)
143,000 (88.0%)
Total
227,500
16,520
48,180
162,500
(7.3% of Total)
(21.2% of Total)
(71.4% of Total)
SIMPLE models showed that areas with high soil phosphorus (due to long-term waste application) were the greatest contributors of NPS phosphorus. Specifically, Osage, Barren Fork, Flint, Benton, and Clear Creeks were the subbasins that delivered the greatest quantities of nutrients. Gade noted the dramatic increase in number of poultry houses since 1980 in the Illinois River basin and observed that the subbasins with the Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 47 -
greatest densities of poultry houses delivered the greatest amount of nutrients.
QUAL2EU models were used to estimate the effects of 25% and 50% NPS nutrient reduction on loading and concentration in the basin and Lake Tenkiller (Tables 23-26). Loading of both phosphorus and nitrogen to Lake Tenkiller was found to be excessive even in the absence of NPS contributions (100% reduction).
Table 24. Relative reduction in mean annual total phosphorus concentration and load with a 25% reduction in nonpoint source inputs.
USGS GagingStation Identification
Simulated Mean Annual Total Phosphorus Conc. (mg/L)
Simulated Mean Annual Total Phosphorus Conc. With 25% NPS Reduction (mg/L)
Change (%)
Simulated Mean Annual Total Phosphorus Load (kg/yr)
Simulated Mean Annual Total Phosphorus Load With 25% NPS Reduction (kg/yr)
Change (%)
07194800
0.37
0.29
-22
39,400
30,200
-23
07195400
0.48
0.40
-17
197,000
161,000
-18
07195500
0.32
0.26
-19
189,000
154,000
-19
07196500
0.24
0.20
-17
223,000
183,000
-18
Horseshoe Bend
0.23
0.19
-17
291.000
241,000
-17
07195000
0.39
0.36
-17
88,700
80,800
-9
07196000
0.29
0.24
-17
37,400
31,000
-17
07196900
0.24
0.19
-21
11,900
9,310
-22
07197000
0.16
0.12
-25
51,200
40,900
-20
Table 25. Relative reduction in mean annual total nitrogen concentration and load with a 25% reduction in nonpoint source inputs.
USGS GagingStation Identification
Simulated Mean Annual Total Nitrogen Conc. (mg/L)
Simulated Mean Annual Total Nitrogen Conc. With 25% NPS Reduction (mg/L)
Change (%)
Simulated Mean Annual Total Nitrogen Load (kg/yr)
Simulated Mean Annual Total Nitrogen Load With 25% NPS Reduction (kg/yr)
Change (%)
7194800
2.9
2.2
-24
303,000
234,000
-23
7195400
3.8
2.9
-24
1,540,000
1,180,000
-23
7195500
2.6
2
-23
1,510,000
1,180,000
-22
7196500
2.1
1.7
-19
1,960,000
1,550,000
-21
Horseshoe Bend
1.9
1.5
-21
2,480,000
1,970,000
-21
7195000
2.4
1.9
-21
543,000
433,000
-20
7196000
2.4
1.9
-21
308,000
243,000
-21
7196900
2.1
1.7
-19
102,000
81,000
-21
7197000
1.5
1.2
-20
495,000
404,000
-18
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 48 -
Table 26. Relative reduction in mean annual total phosphorus concentration and load with a 50% reduction in nonpoint source inputs.
USGS GagingStation Identification
Simulated Mean Annual Total Phosphorus Conc. (mg/L)
Simulated Mean Annual Total Phosphorus Conc. With 50% NPS Reduction (mg/L)
Change (%)
Simulated Mean Annual Total Phosphorus Load (kg/yr)
Simulated Mean Annual Total Phosphorus Load With 50% NPS Reduction (kg/yr)
Change (%)
07194800
0.37
0.20
-46
39,400
20,900
-47
07195400
0.48
0.30
-38
197,000
124,000
-37
07195500
0.32
0.21
-34
189,000
120,000
-37
07196500
0.24
0.15
-38
223,000
142,000
-36
Horseshoe Bend
0.23
0.15
-35
291,000
189,000
-35
07195000
0.39
0.32
-18
88,700
72,600
-18
07196000
0.29
0.19
-34
37,400
24,500
-34
07196900
0.24
0.14
-42
11,900
6,730
-43
07197000
0.16
0.09
-44
51,200
30,400
-41
Table 27. Relative reduction in mean annual total nitrogen concentration and load with a 50% reduction in nonpoint source inputs.
USGS GagingStation Identification
Simulated Mean Annual Total Nitrogen Conc. (mg/L)
Simulated Mean Annual Total Nitrogen Conc. With 50% NPS Reduction (mg/L)
Change (%)
Simulated Mean Annual Total Nitrogen Load (kg/yr)
Simulated Mean Annual Total Nitrogen Load With 50% NPS Reduction (kg/yr)
Change (%)
07194800
2.9
1.6
-45
303,000
165,000
-46
07195400
3.8
2.0
-47
1,540,000
822,000
-47
07195500
2.6
1.5
-42
1,510,000
854,000
-43
07196500
2.1
1.2
-43
1,960,000
1,130,000
-42
Horseshoe Bend
1.9
1.1
-42
2,480,000
1,460,000
-41
07195000
2.4
1.4
-42
543,000
321,000
-41
07196000
2.4
1.4
-42
308,000
177,000
-43
07196900
2.1
1.2
-43
102,000
59,700
-41
07197000
1.5
0.9
-40
495,000
312,000
-37
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 49 -
R. Recent Total Phosphorous Loads in the Illinois River Watershed in Arkansas Compared to Loads in 1980-1993 (Maner 1998)
This investigation assessed decreasing phosphorus loads in the Arkansas portion of the Illinois River watershed. An overall phosphorus load reduction of 20.1% was observed during the 1991-1995 period as compared to the 1980-1993 period. Further improvement was noted in the 1993-1997 period, with a 22.9% load reduction from the 1980-1993 period and a total phosphorus concentration of 0.210 mg/L versus 0.311 mg/L (Table 27).
Table 28. Phosphorus trends in the Arkansas portion of the Illinois River watershed.
Phosphorus concentration (mg/L)
Phosphous load (kg/yr)
1980-1993
1991-1995
1993-1997
1980-1993
1991-1995
1993-1997
% reduction
Total IR watershed
0.311
0.21
221,425
176,948
146,665
22.9
Sager Cr.
1.102
0.844
20,668
14,488
29.9
Barren Fork
0.151
0.104
7,692
5,434
23.4
Flint Cr.
0.077
0.055
3,483
3,047
12.5
The point versus nonpoint sources loads were estimated to be 17% and 68% of the total phosphorus load based on observed in-stream decay rates of the known point source load. Considering “end-of-pipe” values from point sources with no decay of phosphorous, the point source contribution comprised 45% of the total phosphorus load, the nonpoint source load was 40% of the total load, and background sources accounted for 15% of the load. It was assumed that the actual load contributions are somewhere between the two estimates.
S. Phosphorus and Nitrogen Concentrations and Loads at Illinois River, South of Siloam Springs, AR 1997-1999 (Green and Haggard 2001)
In this USGS report, an analysis of phosphorus and nitrogen collected bimonthly from 1997-1999 and during storm events is discussed. The results indicated that both point and nonpoint sources were affecting water quality. Annual flow-weighted concentrations and yields were determined using regression load estimates based on data collected from 1997-1999.
Flow-weighted nutrient concentrations and nutrient yields were 10-100 times greater than national averages for undeveloped basins. Most of the phosphorus load was contributed during surface runoff, while nitrogen showed a different trend (Table 28): about 15% of total phosphorus from base flow and 85% from runoff; 72% of soluble reactive phosphorus was from runoff; about 46% total nitrogen from base flow and 54% from runoff; 42% nitrate-nitrite nitrogen from runoff and 58% from base flow.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 50 -
Table 29. Annual loads for total phosphorus, soluble reactive phosphorus, total nitrogen, and dissolved nitrite plus nitrate nitrogen at Illinois River south of Siloam Springs, AR. All values in kilograms per year.
1997
1998
1999
Total phosphorus
All data
257,000
217,000
260,000
Baseflow (BF)
38,000
33,700
39,200
Surface runoff (SRO)
201,000
248,300
194,000
Sum of BF plus SRO data
239,000
282,000
233,200
Soluble reactive phosphorus
All data
150,000
130,000
160,000
Baseflow (BF)
34,300
30,700
35,000
Surface runoff (SRO)
100,000
115,300
104,000
Sum of BF plus SRO data
134,000
146,000
139,000
Total nitrogen
All data
2,000,000
1,700,000
2,100,000
Baseflow (BF)
1,100,000
750,000
1,200,000
Surface runoff (SRO)
1,100,000
1,300,000
1,200,000
Sum of BF plus SRO data
2,200,000
2,050,000
2,400,000
Dissolved nitrite plus nitrate nitrogen
All data
1,310,000
1,160,000
1,440,000
Baseflow (BF)
967,000
682,000
1,070,000
Surface runoff (SRO)
593,000
652,000
659,000
Sum of BF plus SRO data
1,560,000
1,334,000
1,729,000
T. Phosphorus Sources in an Ozark Catchment, USA: Have We Forgotten Phosphorus from Discrete Sources? (Haggard et al. 2003)
Water samples were obtained from 30 sites in the Illinois River basin, and USGS data from 1997-2001 was used in order 1) to determine the average annual phosphorus load in the Illinois River near the Oklahoma-Arkansas state line, 2) to assess the relative contributions of point and nonpoint sources of phosphorus, and 3) to identify major phosphorus sources at base flow. Results from this study indicated that total phosphorus levels in the Illinois River were 7 to 9 times the criterion of 0.037 mg/L, with runoff values ranging between 11 and 12 times the criterion and baseflow values 5 to 6 times the criterion. Total phosphorus increased significantly during surface runoff events, with an average annual phosphorus load during base flow of approximately 34,000 kg and 174,000 kg during surface runoff conditions. This corresponds to 84% of the average total load being transported during surface runoff conditions.
Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 51 -
Of the average total load from 1997 through 2001, almost 45% was thought to be from municipal WWTPs in the basin. Springdale WWTP contributed almost 83% of the total average annual phosphorus load from point sources. About 35% of the phosphorus observed during runoff conditions was surmised to be from resuspension of instream sediment. These findings suggest that discrete sources of phosphorus and sediment-bound phosphorus must be considered in facilitating reductions of instream phosphorus concentrations.
U. Water Quality and Biological Assessment of Selected Segments in the Illinois River Basin and Kings River Basin, Arkansas (Parsons 2004)
This report presented water quality and aquatic biological data for several streams in the Illinois River basin in Arkansas in order to provide data that could be used to evaluate support of aquatic life criteria. The primary concern of this project was the impact of excessive nutrient concentrations on instream biological communities. Water quality data was collected and analyzed three times at each of eight sites in Arkansas. Sites included locations above and below the WWTPs of Rogers, Springdale, Prairie Grove, and Berryville. In addition, two biological and habitat assessments were performed at each location.
The results indicated that low dissolved oxygen and exceedances of the Arkansas 24-hour dissolved oxygen fluctuation standard subjected aquatic life to stress. Nutrient levels and total dissolved solids were consistently higher at sites downstream of wastewater treatment plants (WWTP) as opposed to sites upstream of the plants. Fourteen percent of the TDS samples exceeded Arkansas standards. Total phosphorus surpassed the 0.1 mg/L guideline for TP in Arkansas’ Water Quality Standards in 58% of samples, most notably at every site located immediately downstream of a WWTP. Total nitrogen values ranged from 0.987 to 8.498 mg/L, with the highest values detected downstream of the Springdale WWTP (from 4.672 to 8.498 mg/L). This study found that nutrient loading at the sites selected was due to WWTP discharge but noted that these findings could have been influenced by the nature of the low flow condition sampling.
Studies cited prior to this found that instream sediments acted as a phosphorus sink at sites immediately downstream of WWTPs, releasing high levels of phosphorus to the streams. Another Arkansas study compared total phosphorus data from previous studies with recent collections. The results indicated that total phosphorus concentrations in storm flow had decreased while those of base flow remained stable, suggesting that best management practices in the watershed were reducing the amount of total phosphorus reaching the Illinois River (Parsons 2004).
The two sites on the Illinois River immediately upstream of Oklahoma yielded results indicating habitats supportive of aquatic life, despite high phosphorus levels and an overabundance of periphyton. However, the lack of many sensitive macroinvertebrate species was noted as a concern. Sedimentation and alteration of the hydrologic regime were proposed reasons for the reduced numbers of pollution intolerant species. Urban Illinois River Watershed
Watershed Based Plan
Accepted January 2011
- 52 -
and agricultural sediment loads contributed phosphorus to the stream while decreasing valuable habitat for aquatic organisms. In the headwaters, sediment seemed to be the pollutant of greatest concern, as opposed to lower in the watershe