United States Environmental Protection Agency Office of Water Regulations and Standards Washington DC 20460 June 1986 Status of Agricultural IMPS Projects —1985 ------- NWQEF» 1985 ANNUAL REF*ORT STATUS of AGRICULTURAL NF»S PROJECTS BY North Carolina Agricultural Extension Service National Water Quality Evaluation Project . Personnel Richard P. Maas Jean Spooner Catherine A. Jamieson Steven A. Dressing Michael D. Smolen - Principal Investigator Frank J. Humenik - Project Director USDA Cooperative Agreement: 12-05-300-472 EPA Interagency Agreement: AD-12-f-0-037-0 Biological & Agricultural Engineering Department North Carolina State University Raleigh, North Carolina 27695 EPA PROJECT OFFICER USDA PROJECT OFFICER James W. Meek Fred N. Swader Nonpoint Source Branch Extension Service Criteria and Standards Division Natural Resources Washington, DC Washington, DC June 1986 ------- This publication was developed by the National Water Quality Evaluation Project, a special project of the North Carolina Agricultural Extension Ser- vice, sponsored by the United States Department of Agriculture Cooperative Agreement 12-05-300-472 and the United States Environmental Protection Agency Interagency Agreement AD-12-f-0-037-0. The contents and views expressed in this document are those of the authors and do not necessarily reflect the policies or positions of the North Carolina Agricultural Extension Service, the United States Department of Agriculture or the United States Environmental Protection Agency. The mention of trade names for products or software does not constitute their endorsement by the North Carolina Agricultural Extension Service, the United States Department of Agriculture or the United States Environmental Protection Agency. ------- FOREWORD . This manuscript is Part I of the 1985 Annual Report by the National Water Quality Evaluation Project to the Project Advisory Committee. Part II, Tech- nical Analysis of Four Agricultural Water Quality Projects, is available as an appendix to this report. Projects analyzed in the appendix include: the Saline Valley RCWP in Michigan, the Tillamook Bay RCWP in Oregon, and the LaPlatte River Watershed Project in Vermont. 111 ------- 8IBWASY The increasing concern about agricultural nonpoint sources (Agricultural NFS) has spawned numerous federal and state Agricultural NFS pollution control programs intended to proiaote or to demonstrate technologies for control of Agricultural NFS pollution of major water resources. The following unresolved issues are addressed in this report: 1) determining what types of water resources are most responsive to NFS control measures. 2) effective means of gaining landowner participation in NFS control efforts. 3) determining the effectiveness of watershed level implementation of Best Management Practices (BMPs). 4) methods for identifying and targeting funds to water quality criti- cal areas. 5) designing cost-effective water quality monitoring systems to docu- ment the effectiveness of Agricultural NFS control efforts. The purpose of the present report is to examine the progress achieved to date by completed and on-going AgNPS control projects in resolving these issues. Twenty Agricultural NFS control projects were selected for more intensive study. Of the twenty projects, four have been reviewed in-depth and are included in an appendix to this report. The National Water Quality Evaluation Project (NWQEP) performed original analysis of water quality data on four of these projects to contribute significant information to the state- of-the-art on Agricultural NFS control. The major findings presented in the report are summarized below: 1. Arid land irrigation canals exhibit the quickest water quality response to BMP implementation. 2. Four years was sufficient time to document statistically signifi- cant sediment concentration reductions in irrigation canals in Idaho where 36% of the land area was treated by BMPs. 3. Streams are the next most responsive water resource type because of their short hydraulic residence time. However, stream water quality is extremely variable, making documentation of improvements difficult. Fecal coliform concentrations appear to be more respon- sive to treatment than sediment. 4. Water resource use impairments based on quantitative standards such as swimming and drinking water supply are easier to address through AgNPS control than qualitative impairments such as eutrophication. LV ------- 5. 100% BMP treatment of a 2,000 acre irrigation block in Washington reduced sediment loss by 80% and phosphorus loss by 50% within one year. 6. Sediment basins along with subsurface drainage and automated water cutback systems in furrow-irrigated fields yielded an 80% reduction in sediment and 50% reduction in phosphorus. 7. The required timeframe for observing water quality results from AgNPS control depends on monitoring design, meterologic variabili- ty, watershed size, water resource type, and pollutant type. Analysis of data from Oregon, Vermont, and Illinois indicates that at least 5 years are required to document improvements in water quality in humid regions. 8. Model results for the Vermont RCWP project suggest that management of all aspects of manure handling can reduce surface water inputs of manure P by 80-90% from northeastern U.S. dairy operations. 9. Monitoring results from the New York MIP project indicate that 50- 90% reduction of manure P can be obtained through barnyard manage- ment practices alone. 10. Eliminating the practice of winter manure spreading may reduce loss of total P (slightly) and ortho-P (substantially). 11. While structural BMPs can be effective in reducing sediment and nutrient losses, they often are not the most cost-effective methods. Our preliminary evaluation shows that conservation tillage and vege- tative cover practices are substantially more cost-effective than sediment basins, diversions, terraces and sediment control struc- tures . 12. Improved fertilizer management appears to be the most cost-effective BMP for reducing nutrient losses in most projects we have examined. This BMP should be equally effective for protection of surface water and groundwater. 13. Bringing approximately 60% of the manure in the Tillamook Bay watersheds under best management has resulted in a 40-50% decrease in log-mean fecal coliform concentrations in the Bay. This result has been documented by water quality monitoring and analysis in this report. , i 14. From water quality data we have determined that 80% BMP implementa- tion on animal production operations in a 700-acre critical area in a Utah watershed has produced 43% reduction in total P, 55% reduc- tion in ortho-P, 59% reduction in TKN and 90% decrease in fecal coli form concentrat ions. 15. Forty percent sediment concentration reduction in irrigation canals has been achieved in the Idaho RCWP project from treatment of 36% of the identified water quality critical area. v ------- 16. The variability in concentration data from water quality monitoring over a four-year time frame is generally so high that a "true" change of 50-60% mean annual concentration is necessary to document statistically significant change. If the monitoring period extends beyond four years or the data analysis accounts for meteorologic variability, sensitivity can be improved substantially. 17. Smaller Agricultural NFS projects have generally obtained higher participation rates than larger projects. 18. Most projects which have a high level of farmer participation have extensively employed radio and newspaper media, one-on-one-contacts with farmers, and relatively high cost-sharing rates. VI ------- TABLE OF CONTENTS Page EXECUTIVE SUMMARY iv TABLE OF CONTENTS vii ABBREVIATIONS OF PROJECTS ix ABBREVIATIONS USED WITHIN THE REPORT x LIST OF TABLES xii LIST OF FIGURES xii CHAPTER 1: INTRODUCTION 1 CHAPTER 2: MAJOR QUESTIONS RELATED TO AGRICULTURAL NFS CONTROL 4 [ 1. Water Resource Treatment Feasibility 4 2. BMP Effectiveness 9 3. Critical Area Selection and Implementation 14 4. Institutional/Organizational Considerations 18 5. Water Quality Monitoring 20 CHAPTER 3: BRIEF SUMMARIES OF AGRICULTURAL NFS PROJECTS REVIEWED 22 RURAL CLEAN WATER PROJECTS (RCWP) 22 Rock Creek, Idaho 22 Prairie Rose Lake, Iowa 24 Highland Silver Lake, Illinois 26 Bonne Idee, Louisiana 28 Double Pipe Creek, Maryland 30 Saline Valley, Michigan 32 Tillamook Bay, Oregon 34 Conestoga Headwater, Pennsylvania 36 Oakwood Lakes-Poinsett South Dakota 38 Snake Creek, Utah 40 Nansemond-Chuckatuck, Virginia 42 St. Albans Bay, Vermont 44 MODEL IMPLEMENTATION PROJECTS (MIP) 46 W. Branch Delaware River, New York 46 Broadway Lake, South Carolina 48 Yakima, Washington 50 ------- OTHER NONPOINT SOURCE PROJECTS 52 Lake Le-Aqua-Na, Illinois 52 Skinner Lake, Indiana 54 Big Stone Lake, South Dakota/Minnesota 57 LaPlatte River Watershed, Vermont 59 Columbia Basin Block 86, Washington 61 REFERENCES 65 Vlll ------- ABBREVIATIONS OF PROJECT NAMES AL- RCWP Lake Tholoco, Alabama DE- RCWP Appoquinimink, Delaware FL- RCWP Taylor Creek, Florida IA- RCWP Prairie Rose Lake, Iowa ID- RCWP Rock Creek, Idaho IL- RCWP Highland Silver Lake, Illinois TN- RCWP Reelfoot Lake, Tennessee LA- RCWP Bonne Idee, Louisiana MA- RCWP Westport River, Massachusetts MD- RCWP Double Pipe Creek, Maryland MI- RCWP Saline Valley, Michigan MN- RCWP Garvin Brook, Minnesota NE- RCWP Long Pine Creek, Nebraska OR- RCWP Tillamook Bay, Oregon PA- RCWP Conestoga Headwater, Pennsylvania SD- RCWP Oakwood Lakes—Poinsett, South Dakota UT- RCWP Snake Creek, Utah VA- RCWP Nansemond-Chuckatuck, Virginia VT- RCWP St. Albans Bay, Vermont WI- RCWP Lower Manitowoc, Wisconsin NY- MIP W. Branch Delaware River, New York SC- MIP Broadway Lake, South Carolina WA- MIP Yakima, Washington IL- LAN Lake Le-Aqua-Na, Illinois IN- BC Black Creek, Indiana IN- SL Skinner Lake, Indiana SD/MN- BSL Big Stone Lake, South Dakota/Minnesota VT-LP LaPlatte River Watershed, Vermont WA-B86 Columbia Basin Block 86, Washington RCWP = Rural Clean Water Program MIP = Model Implementation Program IX ------- ABBREVIATIONS USED WITHIN THE REPORT ACP AGNPS AgNPS ANSWERS ARS ASCS CDF Chi a CLP CM&E CREAMS DO DP ERS FC CIS IPM MIP MLRA MPN NWQEP NPS NPSCA NTU OP PLUARG RCWP SCS STP TKN TN TP TSS USBR USDA USEPA USGS VSS PL-566 108a 208 Agricultural Conservation Program Agricultural Nonpoint Source Pollution Model Agricultural Nonpoint Source (generic) Areal Nonpoint Source Watershed Environment Response Simulation (Model) Agricultural Research Service, USDA Agricultural Stabilization Conservation Service, USDA Cumulative Distribution Frequency Chlorophyll a Clean Lakes Program Comprehensive Monitoring and Evaluation Chemical, Runoff, and Erosion from Agricultural Management Systems (Model) Dissolved Oxygen Dissolved Phosphorous Economic Research Service, USDA Fecal Coliform Geographic Information System Integrated Pest Management Model Implementation Program Major Land Resource Areas Most Probable Number/100 ml National Water Quality Evaluation Project Nonpoint Source Nonpoint Source Critical Area Nessler Turbidity Unit Ortho phosphate Pollution of the Great Lakes from Land Use Activities, Organization Rural Clean Water Program Soil Conservation Service, USDA Sewage Treatment Plant Total Kjeldahl Nitrogen Total Nitrogen Total Phosphorus Total Suspended Solids United States Bureau of Reclamation United States Department of Agriculture United States Environmental Protection Agency United States Geologic Survey Volatile Suspended Solids Public Law 566 Section 108a of the 1972 Clean Water Act Section 208 of the 1972 Clean Water Act x ------- RCWP BMPs by Number [Reference: ASCS 1-RCWP, "National BMPs and Guidelines," Revision 1, Exhibit 10, par. 37, August, 1984] BMP 1 Permanent Vegetative Cover BMP 2 Animal Waste Management System BMP 3 Stripcropping Systems BMP 4 Terrace System BMP 5 Diversion System BMP 6 Grazing Land Protection System BMP 7 Waterway System BMP 8 Cropland Protection System BMP 9 Conservation Tillage Systems BMP10 Stream Protection System BMP11 Permanent Vegetative Cover on Critical Areas BMP12 Sediment Retention, Erosion or Water Control Structures BMP13 Improving an Irrigation and or Water Management System BMP14 Tree Planting BMP15 Fertilizer Management BMP16 Pesticide Management BMPXX Other Developed by Local Coordinating Committee XI ------- - Three types of BMPs were implemented to deal with the animal waste problems in OR-HCWP and NY-MIP: (1) waste storage structures, (2) sub- surface pasture drainage, and (3) milkhouse curbing and guttering. It appears that curbing and guttering of milkhouses is the most cost-ef- fective of these practices, reducing a large proportion (20-40*) of the waste input to streams at low cost compared to other animal waste BMP components. 2c. To what extent do groundwater BMPs conflict with surface water BMPs? Of all common agriculture-related pollutants, only soluble nitrogen and soluble pesticides generally present a potential conflict between efforts to reduce surface water and groundwater inputs. Preliminary results from PA and from other field studies suggest that surface runoff-reducing practices have potential to increase groundwater contamination by nitrate or soluble pesti- cides. Simulations with the CREAMS model, in the PA-RCWP, suggested that conservation tillage has no real effect on groundwater, but that terraces may increase nitrogen transport to groundwater. The IA-RCWP might contribute some perspective on potential conflicts due to terracing if the watershed treatment retains sediment but fails to reduce nutrient flow to the lake. 2d. What degree of sediment reduction can be achieved by BMP implementation at the watershed level? The answer to this question will develop out of all projects' results for various size areas, climates, topographies, soil types, and crops. ID-RCWP has shown significant reductions in irrigation canal sediment concentrations in the subbasins where high levels of sediment control BMPs were installed. Our analyses show that these reductions are in the range of approximately 40-60 percent (NWQEP, 1985a). Additional land treatment data are needed, however, to tie the observed reduction to BMP application unequivocally. Sediment basins and improved irrigation systems implemented on a 2000-acre area reduced sediment loading (WA-B86) 80 percent. Likewise, conversion from furrow to sprinkler irrigation systems resulted in sediment reductions at the edge of fields and often resulted in total elimination of return flows (WA-MIP). Unfortunately, overall reduction in sediment loadings from the watershed was not estimated in the WA-MIP project, and no watershed level sediment reduction has yet been documented on Rock Creek (ID-RCWP). Several other projects are expected to determine potential sediment re- ductions from conservation tillage (MD-RCWP, MI-RCWP, IL-LAN) and terraces (IA-RCWP and PA-RCWP) on the field-scale to the small watershed level (10-8000 acres). 2e. What degree of nutrient reduction can be achieved by BMPs at a water- shed level? Considerable information on nutrient loading and concentration reduction from land treatment has been developed by the VT-RCWP and PA-RCWP projects. The VT-RCWP has projected, based on results from BMP implementation, water quality data, and modeling, that total P loadings from its most extensively implemented subbasin will decrease by 30 percent over the project time frame. A 57 percent decrease in dissolved P is projected. It appears that these 12 ------- CHAPTER 1 INTRODUCTION Control of agricultural nonpoint source (Agricultural NFS) pollution has been recognized as a necessary element to achieve national water quality goals, and in response, numerous federal and state sponsored Agricultural NFS control projects have been initiated. These projects are providing basic institutional/organizational experience, synthesis of results for such projects provides the opportunity to evaluate technologies for future efforts to control agriculture-related water quality problems. This report brings together much of what has been learned to date from these Agricultural NFS projects. We have identified 18 specific questions to assess individual project contributions towards understanding the techno- logical and institutional aspects of Agricultural NFS control. The result of this approach is an overview which will extend the findings and lessons learned from existing water quality projects to the managers of new water quality projects. The Agricultural NFS control programs either currently in progress or completed include the Rural Clean Water Program (RCWP), the Model Implementa- tion Program (MIP), the Agricultural Conservation Program Special Water Quality Projects, the EPA 108a Program, the Clean Lakes Section 314 projects, PL-566 projects, the Wisconsin state projects, and others. Table 1 is a compilation of Agricultural NPS control projects we have identified which include land treatment and water quality monitoring components. We selected 20 of these projects for in-depth consideration of their contribution to present state-of-the-art knowledge. Future reports will analyze other pro- jects from Table 1. For several projects we conducted water quality data analysis beyond that reported previously. (Cassell and Van Calcan, 1983; NWQEP, 1985; Davenport and Lowrey, 1985; Agena et.al, 1985; Neubieser, 1985; Jackson, 1985; Hopkins and Clausen, 1985; Clausen, 1985). An appendix with separate extended chapters devoted to each of these projects is available under separate cover. Finally, it should be noted that in the past 3 years a large number of state-sponsored Agricultural NPS control projects have been authorized and are currently in either the planning or early implementation phase. Most are not included in this report because their contributions will require several years to develop. ------- Table 1 U.S. Agricultural Honpoint Source Projects in HHQEP Data Base. RCUP PROJECTS PROJECT NAME Lake Tholoco, AL Double Pipe Cr, HD Saline Valley, HI St. Albans Bay, VT Tillaiook Bay, OR Cones toga Headwater, PA Long Pine Cr, HE Rock Cr, ID Prairie Rose Lake, IA Taylor Cr, FL Lower Hanitowoc, HI Hanseiond-Chuckatuck, VA Appoquiniiink, DE Highland Silver Lake, IL Reel foot Lake, KY West port River, HA Garvin Brook, HN Oakwoods Lakes- Poinsett, SD Snake Cr, UT Bonne Idee, LA TYPE RCUP+ RCHP+* RCHP+** RCHP+* RCHP+** RCHP+* RCHP+ RCHP+* RCHPi* RCHP+ RCHP+ RCHPt* RCHP+ RCHP+* RCHP+ RCHP+ RCHP+ RCHP+* RCHP+** RCHP+* DATES 1980-91 1980-91 1980-91 1980-91 1980-91 1980-91 1981-91 1981-91 1981-91 1981-91 1981-91 1981-91 1980-90 1980-90 1980-90 1981-91 1981-91 1981-91 1980-90 1980-91 ACRES 51,400 110,000 74,030 33,334 23,540 110,000 58,310 45,000 4,610 110,000 102,000 165,000 30,762 30,946 153,600 37,000 30,720 100,000 700 54,400 PROBLEH TYPE An. waste, seditent Sediient, An. waste Hutrients Nutrients, an. waste An. waste An. waste, nutrients, sediient Sediient, nutrients, an. waste Seditent, an. waste Sediient Hutrients Hutrients, sediient An. waste Turbidity Hutrients, an. waste Sediient An. waste An. waste Hutrients An. waste Pesticide LAHD QUALITY TREAT- HOHI- HEHT X X X X X X X X X X X X X X X X X X X X TORIH6 X X X X X X X X X X X X X X X X X X X X ACP AND HIP PROJECTS Little R, CT L.Q. Drain, ID Blue Cr, IL Dirty Baker's Dz., IHD Saginaw Bay, HI Hall County, HE Chestuee, Cr, TH Mulberry, Cr, TH Indiana Heartland, IH Haple Cr, HE West Br Delaware R, HY ACP+ 1980-83 19,200 ACP ACP+ ACP ACP+ ACP ACP ACP HIP+ HIP+ HIP+* 1976-80 1979-82 1979-83 1979-82 1979-82 1978-80 1978-80 1979-82 1979-82 1978-82 3,704 7,012 23,000 242,636 41,360 85,000 - 103,000 33,088 287,224 Sediient, an. waste X X nutrients Sediient X X Sediient X X Sediient, nutrients X X Hutrients, sediient X X Hutrients, X X An. waste X X An. waste X X Sediient, nutrients X X an. waste Sediient, nutrients X X An. waste, nutrients X X ------- Table 1 U.S. Agricultural Honpoint Source Projects in HWEP Data Base (contined). Broadway Lake, SC Lake Herian, SD Yakiia, HA HIP+* 1978-82 HIP* 1978-82 HIP+* 1978-82 25,183 42,948 26,500 Sediient Sediient, nutrients Sediient X X X OTHER PROJECTS Hatkinsville, 6A Lake Le-Aqua-Ha, IL Black Cr, IH Skinner Lake, IN Chonan R., HC N. Appalachia, NC Union, HC Hayne Lenoir, HC HE Pesticide Iipact I Accessient Prograi Genessee R Basin, HY Defiance County, OH Big Stone Lake, SD I HH LaPlatte R, VT Coluibia Block, HA Henotenee, HI Washington County, HI White Clay Lake, HI ARS 1970-cont CLP.ACP* 1981-86 108A+ 1972-81 CLP* 1977-82 208+ 1979-83 Univ. 1982-84 208 1978-82 208 1978-82 1983-88 - PLUAR6 1975-77 EPA 1981- CLP* 1982-88 * Others PL-566** 1979-90 EPA* 1979-82 PLUAR6 1975-77 108A+ 1976-78 State 4 1975-79 EPA plot studies 2,400 12,038 10,000 13,000 stall 12,700 1,880 4 field studies 1,572,522 2,000 750,000 34,137 2,000 7,940 3,040 Nutrients, sediient, pesticides Nutrients, sediient Sediient, nutrients Nutrients, sediient Nutrients An. Haste An. iiaste An. waste Pesticides, nutrients Nutrients Sediient, nutrients Sediient, nutrients An. Haste Nutrients, sediient Sediient An. Haste X X X X X X X X X X X X X X X X X X X X X * Brief Sundries in this Report. ** Extended Reports and Brief Suuaries in this Report. + Projects Reviewed in other HHQEP Reports. ------- CHAPTER 2 MAJOR QUESTIONS RELATED TO AGRICULTURAL NFS CONTROL Water Resource Treatment Feasibility 1. What types of water resources can be most effectively protected or re- stored through land treatment efforts? Streams, lakes, estuaries, groundwater and irrigation canals are among the impaired water resources being addressed by agricultural NFS control projects. All of these resources are represented in the projects we selected for consideration in this report. Our analyses show with increasing clarity that the water quality of irri- gation canals responds most quickly and effectively to BMP implementation. We have identified statistically significant reductions in canal pollutant load- ing or in concentrations within four years of project initiation in each arid land irrigation project we have studied. (ID-RCWP, UT-RCWP, WA-886, and WA- MIP)* We believe this is because the variability due to meteorological factors is lower, there is greater control over the management of the water resource, and agriculture is the major or sole pollutant source in these projects com- pared to most humid region, non-irrigated projects. In addition, some of the available irrigation BMPs such as conversion of furrow systems to sprinkler or drip systems and installation of sediment basins appear to be highly effective in controlling certain pollutants such as sediment and total phosphorus. The increased analytic control for meteorologic and other sources of variability in these studies has also allowed more efficient monitoring to document the water quality changes. Because of their short hydraulic residence time, streams appear to be the next most treatable water resource. Stream water quality, however, is ex- tremely variable, and actual quantification of improvements continues to prove difficult. Treatability appears to vary as a function of pollutant. Fecal coliform concentrations appear to be more responsive to treatment than other pollutants (OR-RCWP, AL-RCWP). At the other extreme, stream suspended solids concentrations and loadings seem to be much less responsive to land treatment efforts, presumably because of streambank and bedload contributions and the extended time frame during which the stream establishes a new sediment trans- port/deposition steady-state (SC-MIP, AL-RCWP, VT-RCWP). The response of the stream system decreases as size increases, due to the increasing number of sources and the increasing time lag for flushing pollutants from the system. * Project abbreviations in parenthesis indicate source of information. Summaries are presented in the appendix and in other NWQBP reports. ------- The treatability of estuaries varies considerably. IB oost respects the Tillasook Bay, Oregon can be viewed as representing a aost responsive type of estuary because of its small size, very high flushing rate largely forested watershed (SOXs forest cover) , and high concentration of dairy sources directly above the bay. Intensive nanure aanageaient has reduced the concentrations of fecal colifora bacteria in the bay rapidly. In this case changes in contam- ination of the bay were aore easily docuaented than in the tributary streams because changes in bay salinity were used as a Barker to distinguish real changes fmma those attributable to Beteorologic factors. He believe that the Be&tport River estuary (MA-BCE3P) could also show a fecal coliforu response to improved waste Banagenent. At the other extreme is the Nansesond-Chuckatuclk KJW estuary where: 1) the watershed is large amd has multiple land uses, 2) the largest sources of pollutants are quite far upstream and isolated frosa the estuary by irapoundaerats 3) the estuary is large, and 4) there may be a sizable urban WPS. We believe that the Nansesaomd-Chuckatuck system is less® responsive to WPS control than a river system of the msms watershed size and land us®. We have observed that lakes change lesss rapidly them irrigation canals, streams or estuaries in response to land treatment. We assume this is due to longer hydraulic residence tines and recycling of pollutants within the lacustrine system. However, other Banagement tools exist for protecting and restoring iapaired lake uses besides watershed land treatment. These include such practices as weed harvesting, copper sulfate or herbicide treataent, water level control, and dredging. Each of these can have a direct and profound effect both on the lake water quality and on the use impairment. The result is that, although lakes are less responsive to land treatment, there are acre ©xaoples of success in restoring uses impaired by nonpoint sources in lakes than in stresses or estuaries. The aore ssimcosssful strategies have employed both watershed and im-lake The treatability of groundwater resources is difficult to characterize froffl the available data base. The preliminary data from PA-HGWP show that intermediate depth (30-100 ft.) groundwater can be relatively responsive to land raanageaent activity in permeable soils underlain by liaestone. la. What tyjps of AgfflPS-csnis&d isms iqpaircents cmn be cost effectively re- stored through reisedial effort? The most cosmon impaired or potentially iognaired uses addressed by pre- vious and current AgWPS projects are domestic supply, reservoir storage, recreation, fisheries, and aesthetic em joynaent . All of these uses are well represented by the projects given special consideration in this report. Swi»- aing innpairaents nay be the 'easiest to restore because often only a reduction of sueaaertime fecal colifora concentrations below 200/100 al is required (AL- RCWP, VT-ECWP, OR-BCWP). In naay ©ystesas this can be acconplished by treating only the most critical animal production operations. Impairments of domestic water supply nay be the next most amenable to restoration through HPS control. The extent of impairment can be quantitatively defined in terms of established drinking water standards and is often manifested by only a single chemical or physical parameter (PA-RCWP, LA-BCWP, MB-HCWP) . Thus, for example, if maximum nitrate levels can be reduced below 10 ug/1, the impairment has, by defi- nition, been alleviated. At the other extrese, use impairments caused by i 5 ------- eutrophication (e.g. fishery, aesthetic, boating) are often difficult to quantify, and hence, it is difficult to define at what point they have been alleviated. (IN-SL, VT-RCWP, SD/MN-BSL) Generally, there may be a change in the probability that the use will be impaired at any given time as a result of the NFS control effort. Shellfisheries (e.g., OR-RCWP, MA-RCWP) present a very difficult problem because the water must be improved to such a high quality (i.e., median concentration not exceeding 14 npn/100 ml of fecal coliform) to support the use. Ib. What timeframe is required to observe water quality results from a land treatment program? All NFS control projects that progress toward their BMP implementation goals and monitor their water quality effects will ultimately contribute to answering this question. We define an "adequate level" of BMP implementation as the level of implementation that reduces the mean concentration of the pollutant of interest by 30-40* (see Question 5b). To monitor water quality effects, a program should be able to characterize this 30-40% reduction as statistically significant (see Question 5b). Based on our current analyses of projects that appear to meet these criteria, we observe that: 1. Four years is sufficient to observe a significant reduction of sediment concentration in irrigation canals where 40* or more of the land surface is protected by BMPs (ID-RCWP, UT-RCWP, WA-MIP, WA-B86). 2. 100* BMP treatment of sediment sources in a 2000-acre irrigation tract, reduces phosphorus losses significantly within one year (WA- B86). 3. Bringing half of the manure in a northestern watershed under best management does not produce statistically significant stream phospho- rus reductions in a three-year timeframe (VT-RCWP); however, signifi- cant fecal coliform reductions are possible with the same level of management and three year timeframe (OR-RCWP). 4. Erosion control on the majority of a 4000 acre watershed may not produce a statistically significant improvement in lake turbidity within four years (IA-RCWP). However, the anticipated improvements may be masked by natural variability, due to weather and climate. 5. In Alabama, treatment of the majority of agricultural nonpoint sources on 50,,000 acres has produced observable in-lake fecal coli- form reductions but not sediment reduction within four years. ------- Table 2. Anticipated Tineframe for Obtaining and Observing Water Quality Improvement from Watershed Level NFS Control Efforts. Water "Actual Resource Physical/Chemical Type Improvements Pre-Treatment Documented Significant Improvements* Improvements Post-treatment Total Irrigation Canal 0-1 Stream Estuary Lake Ground Water Aquifer 1-5 0-5 2-10 unknown 1 2-3 2-3 2-3 1-2 Years 1-5 1-5 1-5 1-5 1-5 1-2 3-8 2-5 5-13 2-4 5-12 2-6 6-14 unknown unknown ^Assuming BMP implementation of at least 40% effectiveness. The required timefrane for observing water quality results depends on the watershed size, water resource type, hydraulic residence time, the pollutant type, and the starting conditions. The conception of timeframe may be biased by short-term results, however, because major storms or other anomalies in- fluence water quality. In addition, there are two timeframes to consider: the timeframe in which water quality can actually improve to the desired level in a physical, chemical, biological, or aesthetic sense; The other is the time- frame required to document the water quality improvement with monitoring data. The latter timeframe must, by necessity, include a pre-treatment monitoring period, an implementation period and a post-treatment monitoring period. In Table 2 we have charted the time range for each by water resource type. Ic. Do groundwater resources respond rapidly enough to reflect changes in land management within a ten year time frame? The PA-RCWP field studies show that a moderate depth (30-100 ft.) water- table in the Conestoga watershed respond to each major precipitation event, and that almost complete recharge occurs over a one-year period. In this situation, the quality of the groundwater may respond quickly (1-2 years) if there is drastic change in land management. However, in this project, a large soil nutrient reservoir exists which will have to be leached through the system before improvement will be seen. The SD-RCWP plot and field studies should also provide information toward answering this question when those results become available. ------- Id. How much problem definition is needed to select and develop a successful and cost-effective NFS control project? Our approach to problem definition, developed in the 1985 RCWP Cross Project Report (NWQBP, 1985c), outlines three problem definition steps to maximize the probability of a successful (improving water resource) and rela- tively cost-efficient project. 1. Determine the extent to which a use or potential use of the water resource is impaired, and make a best estimate of the economic cost of the use impairment. 2. Estimate the relative magnitude of sources that can be treated through the program versus those that are beyond the program's jurisdiction, such as point sources or background. (These estimates should be updated as more information is obtained). 3. Determine what pollutant(s) is causing the impairment, and estimate how much reduction in that pollutant(s) is needed to achieve the desired effect. (BMP goals should then be set to achieve this amount of reduction). In this analysis we evaluate the apparent effect of problem definition on each project's overall effectiveness. The following are some observations from specific projects which illus- trate the potential effect of problem definition on project success. - In the Black Creek, Indiana 108a project, 30% of their funds were used for streambank erosion control, whereas this source was later estimated to contribute only 5% of sediment load. - Clear problem definition in the total WA-MIP and WA-B86 projects allowed rapid implementation with resulting water quality success. (NWQEP and Harbridge House, 1983; King et.al, 1983) The SC-MIP spent most funds on pasture improvement and sediment ponds below pastures, both of which were later found to have little effect on sediment loadings from the project. (MWQEP and Harbridge House, 1983) - Pre-project source identification indicated that NPS control alone would not solve problems in the VT-RCWP project. Thus, an effort to control point sources was initiated simultaneously rather than waiting for NPS efforts to produce desired results. 8 ------- Effect i^osaoso 2. How effective are the various BPiPs in reducing pollutant inputs to water The water quality effectiveness of a BMP or BMP system is generally site specific. Important factors include proximity to watercourse , surface slope, soil types timing of activities. Magnitude of land disruption, and intensity, frequency, and duration of precipitation. Pieces of the overall picture are coming froa plot and field studies, HCWP projects, and other WPS control projects. These include the following observations: - Managing all aspects of Manure handling properly can reduce surface water inputs of manure phosphorus by 80-90 percent from northern U.S. dairy operations (VT-RCWP). With approximately 60& of manure under best management, a 40-50 percent reduction in Bean fecal coliform concentration has been accomplished (OR-BCWP) . Nutrients (TP, OP, TKW) and fecal coliform concentrations have been reduced (40-65& and 50-90%, respectively) through the implementation of animal waste management sys- tens in a snail irrigated watershed (UT - Barnyard management practices have the potential to reduce TP load- ings 50-90& (MY-MIP) . These practices include diversion of upland flow, use of concrete slab surfaces to prevent surface erosion, and diversion of barn roof water. - Elimination of winter manure spreading reduces the yield of total P (slightly) and ortho P (substantially) (VT-BCWP). Observation with re- spect to nitrogen from the PA-RCWP and VT-1SCWP projects appear to con- flict to some extent. The PA-RCWP model iag studies suggest that, in practice, the benefits of eliminating winter manure spreading are offset by storage factors that make available a large slug of manure nitorgen for transport. (30 percent increase) The VT-RCW field studies, in contrast, showed that TSK export increased 148 percent and amonia-W increased 618 percent as a result of winter manure spreading. The high W losses from the VT-RCWP areas could be due to the fact that winter and early spring events produce the greatest pollutant yields in VT-HCWP when manure is spread during winter. - Evidence of the effectiveness of conservation tillage practices for reducing nutrients and sediment should be available from MD-RCWP and MI- RCWP, and IL-LAW projects within the next 2-3 years. - The IA-RCWP project will provide considerable information about the effectiveness of terraces for the control of turbidity and sedimentation in western Iowa. In PA-RCWP, terraces were installed at the field sites during the past year, and should provide some definitive information on the effects of terracing on transport of pollutants to surface and subsurface water resources. On the basis of pre- implementation water quality data, it is projected that terraces may increase nitrate concen- trations of water transported to surface and groundwater because of the increased exposure time of manure to water. Increases in surface runoff ------- nitrogen loads should be moderated by terraces, however, if they also reduce runoff volume. Suspended sediment and total phosphorus losses in surface runoff should also be reduced by terraces. Sediment basins along with subsurface drainage and treatment of 100% of area with automated irrigation water cutback systems reduced sediment loss 80 percent and reduced phosphorus loss by 50 percent with no significant changes in nitrogen (WA-B86). Our analyses of the ID-RCWP water quality data show conclusively that significant reductions of sediment concentrations have occurred in drainage canals as a result of BMP implementation. However, both BMP-12 (sediment retention basins) and BMP-13 (irrigation management) have been integrated in such a way that it is not possible to distinguish the effects of the two practices. We conclude, therefore, that the combined effect of irrigation and sediment retention BMPs reduces sediment concentrations significantly. Studies of 20 humid region sediment basins show sediment trap efficiencies ranging between 65 and 98 percent depending on incoming sediment concentrations, particle size distribution, retention time, and basin geometry (NWQEP, 1982). Total phosphorus reductions from a subset of those studies have ranged between 10 and 77 percent with a mean of about 50 percent. The Skinner Lake (IN-SL) project with its fine clay soils may approximate "worst case" conditions; only marginal reduction in sediment and nutrients was observed. In general, we would expect humid region sediment basins to be less effective than those in arid irrigation areas because they are less effective under storm-flow conditions. - Conversion from furrow to sprinkler irrigation often totally elimi- nates return flows when proper water management is used. Thus, this BMP can be 100* effective in reducing sediment losses attributable to irri- gation. (WA-MIP) 2a. ffotf do external (uncontrolled) factors such as meteorologic conditions affect the ability of BMPs to protect or restore impaired water re- sources? There are two components to the answer to this question because: (1) the effectiveness of BMPs may be influenced by meteorologic factors, such as the frequency, magnitude, and duration of storm events and (2) the precision of monitoring and data analysis are also affected by these factors. The occur- rence of storm events coincident with field activities (e.g. spreading of fertilizer, manure spreading, or plowing) may change drastically the effec- tiveness of BMPs such as manure management. Therefore, we must distinguish between design characteristics and management options to determine whether BMPs are susceptible to large runoff events and which BMPs are most effective over a wide range of meteorologic conditions. Clearly, there is a probabilis- tic aspect to the answer. Meteorologic variability has only a minor effect in arid, irrigated regions (ID-RCWP, UT-HCWP, WA-MIP, and WA-B86). Even so, our analysis of subbasin data from ID-RCWP indicated that a 40-60% reduction in concentrations is required for a decrease to be statistically signficant. These four projects had statistically significant results within a short (4- year) timeframe. Projects in humid areas require more time to document water quality changes because meteorologic variability affects BMP performance and other water quality factors (IA-RCWP, VA-RCWP, VT-RCWP, VT-LAP, IN-SL, and other projects). 10 ------- The second component refers to experimental and statistical considera- tions. There are aethods to account for part of the variability of natural systems, thereby, improving analytical sensitivity. In our analysis of the estuary of the OR-RCWP, we were able to account for about 30SS of the variabil- ity in the concentration data by considering salinity as a surrogate measure of precipitation and stresua flow in a linear regression aodel. The regression models of stream water quality were improved significantly by adding terras to the model to account for stream discharge (UT-HGW), stream discharge and hydrograph direction (rising or falling) (OH-KCfflP), or stream discharge and season (MI-BCWP). Hiver stage, where controlled by locks, may also be useful as a control variable (LA-KCW), because it appears to be related to concen- trations; regression analysis should verify this relationship. Other tech- niques that are used to reduce the effect of seteorologic variability include pairing watershed responses (VT-BGWP) and pairing upstream and downstream observat ions. (ID-RCWP). 2b. What SfiPs are most cost-effective in addressing water resource impaii— The answer to this question will develop from the results of many NFS projects. A more detailed discussion of this topic will be given at a later time. The following are preliminary observations: While structural BMPs can be effective in reducing sediment and nutrient (especially P) losses, they may not be the most cost-effective. Our preliminary evaluation suggests that ainiBrau tillage and vegetative cover practices are substantially Bore cost-effective (in terms of cost/ton of soil saved/year) than sediment basins, diversions, terraces, and sediment control structures (IN-SL). Structural BMPs, such as sedi- ment basins and irrigation systees, were also found by BBS not to be as cost-effective as practices that reduce erosion on the field, such as conservation tillage (ID-HCWP). (See Status Heport on the CM&K Projects, [WWQEP, 1985] for more details). Results from two projects in Washington state suggest that improved irrigation water management (timing and automatic furrow cutback) is also sore cost-effective than structural alternatives for reducing sediment losses (WA-MIP, WA-B86). - Fertilizer management is a relatively inexpensive practice to re- duce nutrient loss. In terms of cost per pound of phosphorus saved, fertilizer management is estiuated to be raore cost-effective than conser- vation tillage, and both are raore cost-effective than animal waste man- agement (MI-RCWP) when animal waste management involves cost sharing a waste storage structure. Prelisainary results fron field site monitoring (PA-RCWP) suggest that a nutrient management (both fertilizer and animal waste) program, which provides soil testing and precise recommendations that Ejatch nutrient application rates to crop utilization rates, may be the jaost effective BMP for reducing nitrogen losses to groundwater. Preliminary economic analysis for PA-RCWP suggests that the cost of hauling to export cattle and swine manure from the 110,000-acre project area would be unacceptably high. Exporting poultry manure, however, may be Bore cost-effective because it is dry and relatively high in nitrogen content. 11 ------- Three types of BMPs were implemented to deal with the animal waste problems in OR-RCWP and NY-MIP: (1) waste storage structures, (2) sub- surface pasture drainage, and (3) milkhouse curbing and guttering. It appears that curbing and guttering of milkhouses is the most cost-ef- fective of these practices, reducing a large proportion (20-40*) of the waste input to streams at low cost compared to other animal waste BMP components. 2c. To what extent do groundwater BMPs conflict with surface water BMPs? Of all common agriculture-related pollutants, only soluble nitrogen and soluble pesticides generally present a potential conflict between efforts to reduce surface water and groundwater inputs. Preliminary results from PA and from other field studies suggest that surface runoff-reducing practices have potential to increase groundwater contamination by nitrate or soluble pesti- cides. Simulations with the CREAMS model, in the PA-HCWP, suggested that conservation tillage has no real effect on groundwater, but that terraces may increase nitrogen transport to groundwater. The IA-RCWP might contribute some perspective on potential conflicts due to terracing if the watershed treatment retains sediment but fails to reduce nutrient flow to the lake. 2d. What degree of sediment reduction can be achieved by BMP implementation at the watershed level? The answer to this question will develop out of all projects' results for various size areas, climates, topographies, soil types, and crops. ID-RCWP has shown significant reductions in irrigation canal sediment concentrations in the subbasins where high levels of sediment control BMPs were installed. Our analyses show that these reductions are in the range of approximately 40-60 percent (NWQEP, 1985a). Additional land treatment data are needed, however, to tie the observed reduction to BMP application unequivocally. Sediment basins and improved irrigation systems implemented on a 2000-acre area reduced sediment loading (WA-B86) 80 percent. Likewise, conversion from furrow to sprinkler irrigation systems resulted in sediment reductions at the edge of fields and often resulted in total elimination of return flows (WA-MIP). Unfortunately, overall reduction in sediment loadings from the watershed was not estimated in the WA-MIP project, and no watershed level sediment reduction has yet been documented on Rock Creek (ID-RCWP). Several other projects are expected to determine potential sediment re- ductions from conservation tillage (MD-RCWP, MI-RCWP, IL-LAN) and terraces (IA-RCWP and PA-RCWP) on the field-scale to the small watershed level (10-8000 acres). 2e. What degree of nutrient reduction can be achieved by BMPs at a water- shed level? Considerable information on nutrient loading and concentration reduction from land treatment has been developed by the VT-RCWP and PA-RCWP projects. The VT-RCWP has projected, based on results from BMP implementation, water quality data, and modeling, that total P loadings from its most extensively implemented subbasin will decrease by 30 percent over the project timeframe. A 57 percent decrease in dissolved P is projected. It appears that these 12 ------- loading reductions would be even greater except that very high phosphorus levels have built nap in the soils as a result of historical over-application of Manure and coEaercial fertilizer. The total saanure P reduction anticipated froa bringing aanure raider best taanageisent is estimated as 80—90 percent. Our projections from the PA-BCWP field and modeling studies are: 1) that practices such as terracings conservation tillage, and Manure storage will have a relatively Hinor influence on transport of nitrogen to groundwater or to streaai baseflow; and 2) that terraces and conservation tillage will pro- duce SOBS reduction of total phosphorus and total nitrogen transport to sur- face waters. On the other hand, we expect that nutrient EanageBent, matching nitrogen application© to crop needs based on soil and eamure nitrogen tests, will have a significant irapact on nitrogen losses. After an initial flushing period we expect that nitrogen loading reductions to both ground and surface waters will be proportional to the reduction in e«cess JJ applied to soil. Our analysis of OT-RCW, a ©mall project area (700 acres) with five treated aniBal operations identified statistically significant concentration reductions in total P (438), ortho P (55^)s and total Ijeldahl H (5S&). These reductions resulted frosa iaproving aniaal waste Banagesent , and they occurred even though herd sizes increased during the SEES tiussfrsos. An irrigated systea with sediaent basins and water aanageHent practices insalled reduced total P loading 50 percent and reduced dissolved P loading by 20-40 percent (WA-B86). Under the saiae conditions , nitrogen loadings did not decrease. Our analysis of the IB-BCfflP showed significant reductions of total P 'concentrations in three of the subbasins with sediment basins and water management . We found no significant reductions in TIN concentrations. The P reductions were less than the sedisaent reductions s corroborating previous studies which showed that sediment-control BMP& reduce total suspended solids to a greater extent than total P. Other projects eventually will add iraforaation to this topic. MI-RCWP could potentially have results -(on a 1000-8000 acre scale) on nutrient re- ductions froa conservation tillage and anisial waste aanageaent within two years. On a smaller scale (25-150 acres) nutrient reductions free specific BMPs Bay be determined by the MD-BCKP and IL-KCWP. W-IP and VA-HCWP may also show soue nutrient reductions, and results are espected froa studies conducted in the IL-BCWP. 2f . K&at degree of bacterial reduction can be achieved through various levels of With approsieately SO percent of the sanure (OH-BCW) under best aanage- ment, our analyses show 40 to 50 percent reduction in the log-nean fecal coiifona concentrations. Other work in siaall upstreara tributaries containing 1-2 dairies suggests that about SO percent fecal col i fora reduction cam be accoaplished by treating all dairies in the watershed. In addition, we found SO percent reduction in fecal colifora concentra- tions froEi installing animal waste oanageaent on 5 out of 6 critical dairies in the 700-acre UT-KCW project. 13 ------- The ID-RCWP did not implement many animal wast« BMPs. However, they achieved significant reductions in fecal coliform in one subbasin, after the cattle were kept from traversing the stream area. Additional information on possible fecal coliform reductions will come from other projects, especially VT-RCWP and VT-LP, MW-RCWP, and MA-RCWP. 14 ------- 3. Where ia a watershed should BMPs be placed to restore or protect a given water resource? The projects we studied have used a wide variety of criteria to identify critical areas within their boundaries. Among the criteria employed are: 1. distance of the farm from the nearest watercourse, 2. distance from the impaired water resource, 3. erosion rate, 4. nutrient application rates and timing, 5. presence of manure sources, 6. designated high or low priority subbasins, 7. on-site evaluation. It is very difficult to correlate a project's critical area targeting approach with its present or potential water quality results because there are many confounding variables. However, insights into appropriate methods for selecting critical areas can be gained from the approaches and experiences of projects to date. The following discussion highlights some of these approaches and experiences. At one end of the critical area spectrum is the IA-RCWP project where nearly the entire watershed above the lake is targeted for BMP treatment. The watershed is relatively small and designation of all cropland as critical ap- pears to be appropriate. The OR-RCWP provides two valuable insights to critical area selection. First, because the impairment is bacterial contami- nation of shellfish beds, the needed bacterial reduction is very large, and thus, the project concluded that most dairy sources should be classified as critical. Second, OH-HCWP is one of only two projects (ID-HCHP is the other) we have found that have used the "designated subbasin" criteria explicitly. Because one of the tributaries to Tillamook Bay enters the bay very near its outlet to the Pacific Ocean, the project determined that material from this tributary could not reach the shellfish beds. Thus, they eliminated dairy operations on this tributary from cost-sharing consideration. SD/MN-BSL used two models (AGNPS [Young, et al., 1985] and the Minnesota Feedlot Model [Young, et al., 1982]) to identify critical areas in a large watershed (750,000 acres). A modeling approach appears to be appropriate for large watersheds where farm by farm evaluation may be impractical. 15 ------- Ideally, the moat efficient way to select critical areas would b© ©SB the basis of water quality monitoring data froa within the project area. Four current projects (ID-RCWP, IL-KCWP, VT-RCWP, UT-RCWP) are in a position to refine critical areas on this basis. Several projects (MI-HCWP, MD-RCWP, WI-RCWP, VA-RCWP) have used "distance to nearest watercourse" as their priiaary or sole criterion for critical area selection. This is a good first cut but could be improved by considering other factors. The LA-HCWP project is noteworthy in that it designated areas directly adjacent to the impaired water body as "extra-critical" with cost- sharing raised to 90%. Quantitative farm level rating forms have been used to prioritize cost- share applications in several projects (e.g. OR-RCWP, VT-RCWP, WA-MIP). This has proven to be a very useful tool provided that the form is weighted to the appropriate water quality concerns. The PA-RCWP and SD-RCWP have developed targeting criteria specific for groundwater impairments. These focus on soil permeability, nutrient appli- cation rates, and depth to groundwater. 3a. What fraction of watershed critical areas or sources must be treated with BMPs to restore or protect a given water resource? An answer to this question will be forthcoming from the agricultural NFS control projects presently underway. These projects encompass the entire spectrum from almost no BMPs to essentially 100* treatment. As water quality responses are documented, it should be possible to predict water quality for some impariraents based on the fraction of watershed area treated, assuming a valid critical area selection procedure has been followed. Below are some of the more concrete results presently available. 1. In Idaho (ID-RCWP) approximately 40% reduction in irrigation canal sediment concentration has been associated with the treatment of 36% of the identified critical area. 2. The OR-RCWP project has observed 40-50% reduction in log-mean fecal coliform concentration in Tillamook Bay that corresponds with their bringing approximately 60% of the manure produced in the watershed under best management. Further improvements are anticipated from this project as the treatment level approaches 90%. 3. Extensive modeling and monitoring efforts in VT-RCWP indicate that animal waste management is approximately 80% effective in reducing the loss of total phosphorus to area streams. Continued water quality monitoring should document how closely this translates to the watershed level effects. 4. Treatment of 100% of the land in a Columbia Basin irrigation block (WA-B86)reduced sediment yield 80% and total phosphorus yield 50% The BMPs employed had no effect on nitrogen losses from the system. 16 ------- 5. In the UT-RCWP project approximately 80* of the project area dairy farms had waste management BMPs implemented. Our analysis of the water quality data indicates a 43% total P reduction, 55* ortho P reduction, 59* concentration TEN reduction and 90* fecal coliform reduction. t 6. In the NY-MIP 91 of 275 barnyards were treated (154 were of high priority) with only marginally significant reductions (10-15*) in dissolved phosphorus loadings estimated by modeling. 7. In the IL-LAN project approximately 60* of the watershed was treated with conservation tillage. The water quality monitoring program was insufficient to clearly document any water quality change; however, visual improvements in the lake have been observed. 17 ------- Inatitutonal/Organizational CoaaideratioBS 4a. What are the aoat effective means of obtaining fanner participation? A diversity of approaches to obtaining farmer participation in water quality projects is apparent in the projects reviewed for this report. These include: one-on-one contact of project personnel with fanners; high cost share rates on desirable practices; extensive media coverage along with public meetings and field days; supplying services such as pest scouting and soil sampling; targeting effort to specific key individuals in the community; and negative reinforcement such as regulations and economic disadvantages for those who do not comply with project objectives. Project size, also, appears to be a major factor in the implementation rate achieved by projects, perhaps because a drastic increase in work load occurs with increasing project size, and more effective personal interactions and group dynamics occur in smaller projects. Thus, smaller projects appear to be more successful than large projects in obtaining farmer participation. As a corollary, one-on-one attention by project personnel appears to be on© of the most effective means of obtaining farmer participation. A high degree of personal contact was visible in the VT-HCWP, VT-1P, OR-RCWP, MI-RCWP, WA- MIP, and WA-B86 projects. VT-HCWP, in particular, has a person designated specifically to market the program, and WA-B86 involved direct contact between university engineering faculty and each farm operator in the project area. Bach of these projects is notable for its high implementation rate. High cost share rates for a slate of practices that the farmers prefer was apparently an effective means of obtaining participation in the LA-RCWP, UT-RCWP, ID-RCWP, SC-MIP, WAHMIP, and IN-SL projects. The IN-SL project had poor participation rates until it increased its cost share rates from 45 percent to 85 percent. The LA-RCWP obtained the participation of the most critical farms by offering them preferential cost share rates as high as 90%. The WA-B86 achieved 100% participation with severely restricted cost-sharing rates, however, by providing a great deal of personal attention. The UT-RCWP helped its farmers to improve their animal waste management facilities suf- ficiently that participants were able to increase herd sizes substantially, and ID-RCWP assisted its participating farmers to improve their irrigation distribution systems, to provide increased water use efficiency, labor savings, and erosion control benefits. The high rate of nutrient and pesticide management in the SD-RCWP also was, at least in part, due to a simultaneous Extension Service effort that provided pest scouting to those who signed contracts for pest management in the RCWP project. Most successful water quality projects have used radio and newspaper media effectively to develop awareness of their projects. The MI-RCWP, VT- RCWP, and IL-RCWP projects supplemented their media efforts with public meet- ings and highly publicized field days and tours that were helpful in promoting BMPs and in stimulating participation. 18 ------- The PA-RCWP, WA-MIP, IN-BC, and MD-ROJP projects took very specific tar- geting approaches to address their participation objectives. The IN-BC pro- ject was highly successful in obtaining the participation of its large Amish ccwBunity by targeting an intense effort to obtain the approval of the comsau- nity's religious leaders. The PA-RCWP, which also has large Amish and Menno- nite population©5 followed the XW-BC example with ouch less success. The WA- MIP project targeted its efforts toward the trend-setting farmers in the project area, and the MB-SCW project targeted its effort to operators of large farms with the rationale that they could, thereby, achieve extensive iE$>lementatiora with fewer contracts. Targeting was apparently successful in the WA-KHP, but in the MB-KCKJP project, results ar© somewhat equivocal. The high participation rate in the OH-I5CW project is largely attribut- able to the negative consequences that Bay be invoked for dairy farmers that do not cooply with project objectives. All of the fanners in the watershed ©ell their milk to a cooperative which support© the objectives of the project. Any dairy operation that does not oeet specified manure handling criteria i© penalized in the price paid for its HiIk. Participation in the project as- sures that a fan® will sieet the coop's criteria and provides cost sharing assistance. In addition, the OB-KCWP has regulatory statutes which it can and does invoke where necessary. In summary8 small project size and close contact between project person- nel and project area farmers appear to be effective aspects of projects that are successful in achieving participation objectives. High cost share rates or direct assistance for desirable practice© also have been shown to be effec- tive incentives. The threat of negative consequences for those who do not coaply with project objectives has been deaonstrated effective in at least one project, but targeting to key community leaders or large farms has not been effective in all of the projects studied. 4b. How Does Multiplicity of Objectives, e.g. simultaneous groundwater and surface water objectives affect project performance? Two groundwater projects should contribute substantially to answering this question. The experience in SDHSCW0 to date, indicates that severe problesas can arise in multiple-objective projects unless considerable effort is devoted to determining how to coabine objective® properly. The lack of clarity and definition of surface water and groundwater objectives has slowed the development of the SD-iCW project. Infcreation from the PA-RCWP project shows that there are some BMP©, such as fertilizer management„ that can reduce both surface and groundwater iHpainments. These results further suggest that increasing infiltration through terracing or no-till practices without nutrient Banagesaent nay have negative isjpact© on groundwater quality. 19 ------- Sggtgr Quality Monitoring 5a. What are the groundwater levels of pesticides that can be expected in areas with intensive agriculture? There is a paucity of data to address this question. However, the emer- ging picture shows that the occurrence of pesticides in groundwater is a function of application rate, soil type, pesticide type, and aquifer recharge rate. Extensive use of soluble pesticides (e.g. aldicarb) on sandy soils has produced documented groundwater contamination at an increasing number of sites nationwide. Anilide and triazine herbicides in groundwater are the most commonly observed pesticide contaminants, in this concentration range of 1-40 ug/1 (Maas et al, 1984). Herbicide concentrations in the PA-HCWP monitoring wells showed significant increases following field application in the spring. However, the concentrations observed were consistently less than 1 ug/1, not considered sufficient to cause a water use impairment. A major unresolved issue is the effect that increasing usage of runoff reducing practices such as conservation tillage will have on groundwater pesticide concentrations, espec- ially if these practices require an increase in herbicide use. 5b. How much change in nonpoint source pollution must occur to be detectable through water quality monitoring. NWQEP has focused a great deal of attention on this question in the past year, and we anticipate a continuing effort. The answer is fundamental to development of successful NFS control projects. For example, if the goal of a particular project is to produce water quality improvements through BMP imple- mentation, knowing the minimum detectable water quality change is crucial for setting realistic BMP implementation goals (and consequently, for funding, critical area designation, timefrerae, technical assistance, etc.). Our research into the issue has taken two separate, but related approaches. One is to observe what level of pollutant reduction is statis- tically significant in projects with different timeframes and/or with dif- ferent types of water quality monitoring programs. In the other approach, we examine the natural variability of water quality data from BMP implementation projects in a variety of water resource types without regard to the documented improvement in water quality. With a statistical analysis of data vari- ability, we have estimated the duration of monitoring and how many water samples must be taken to document a given level of change. A detailed prelim- inary presentation of this analysis can be found in the Technical Supplement to the 1985 RCWP CM&E Report (NWQEP, 1985b.) 20 ------- With the data we have examined so far, we have found that the sensitivity of a nonpoint source water quality monitoring system increases with time (i.e., increasing number of independent samples). The more of the total variability which can be explained by sources other than BMP implementation, the greater is the potential sensitivity of the monitoring system. Prelimi- nary work suggests that a 50-60% decrease in mean pollutant concentrations in precipitation-driven systems may be the minimum change that is statistically significant without correcting for meteorologic variability. If meteorology- related variables can be accounted for, then 30-40% decrease may be sufficient. Examining the water quality data from specific projects has provided the following information related to this question: 1. In the OR-RCWP project 40-50% reduction in log-mean fecal coliform concentration was required for significance. Fecal coliform data are extremely variable, but we were able to achieve this level of sensitivity by using bay salinity measurements to adjust for mete- orological effects. 2. In the ID-RCWP project we found that 35% reduction of raw mean sediment, concentration (corrected for upstream concentration) over four years was statistically significant. This low value was at- tributed to the fact that meteorological effects on the system are small. 3. Very similar results were found for the UT-RCWP where, approximately 40% reduction in mean sediment concentration was required. Meteoro- logical effects within the UT-RCWP irrigated system, too, are relatively small. Also, in both ID-RCWP and UT-RCWP we used an "upstream-downstream" analytical design which adjusted for incoming sediment concentrations in the irrigation system. 4. The MI-RCWP has an excellent water quality database consisting of 5 years of weekly samples with corresponding streanflow measurements. The data show a significant seasonal effect. Our preliminary analy- sis suggests that a linear regression model which adjusts the concentration data for both streamflow and season may allow a 30% change in phosphorus concentration to be considered significant. 21 ------- Brief Summaries of Agricultural NonPoint Source Projects RURAL CLEAN WATER PROJECTS (RCWF) Rock Creek RCWP Twin Falls County, Idaho MLRA: B-ll I. Project's contributions toward understanding the effectiveness of NPS control efforts: Information on the effectiveness of BMPs in an irrigated system will be gained from this project. After four years of water quality monitoring, signi- ficant sediment concentration reductions have been found in six subbasins. Additional documentation of the relationship between land treatment and water quality will be helpful to establish a cause-and-effect relationship of BMPs to water quality improvements. (For more information see RCWP Status Report on the CM&E Projects, 1985, pp. 35-64.) II. Project Characteristics 1. Project type: RCWP-comprehensive monitoring and evaluation, project area = 45,000 acres. 2. Water resource type: Irrigation canals and streams. 3. Use impairment: Recreation, fishing, and aesthetics. 4. Timeframe: 1981-1991. 5. Water quality at start of project: 1980 flow-weighted mean concentrations at the mouth of Rock Creek: TSS = 158 mg/1 (irrigation season only) TP = 0.123 mg/1 (irrigation season only) TN 3.3 mg/1 (water year) FC = 1182 mpn (geometric mean) 22 ------- 6. Meteorologic factors: Annual rainfall =8.5 inches; USLE "R" factor = 20 7. Water quality monitoring program: Grab samples and instantaneous flow are taken biweekly to weekly at Rock Creek and subbasin stations during the irrigation period; monthly monitoring is per- formed during the non-irrigation season. Samples are analyzed for total phosphorus, dissolved orthophosphate, suspended solids, fecal coliform, Kjeldahl nitrogen, and inorganic nitrogen. 8. BMPs: Focus is on sediment retention structures and irrigation management systems with some permanent vegetative cover on critical areas (RCWP-BMPs 12, 13, and 11). Several other practices were approved, but few are implemented (i.e. RCWP-BMPs 2, 9, 15, and 16). 9. Critical areas: The subbasins were prioritized, however, the imple- mentation of practices has not followed the designated order closely. 10. Incentives: 80% cost-sharing, $50,000 maximum. 11. Economic information: BRS performed analysis. Preliminary esti- mates of total benefits are projected to be much less than the total costs (benefit/cost = 0.2). However, it was projected that the benefits could exceed costs if the project were to emphasize lower cost practices, such as conservation tillage and erosion control, in place of sediment basins and irrigation systems. III. Lessons Learned Arid, irrigated argicultural areas, like the Rock Creek subbasins, appear to respond faster to land treatment than do non-irrigated, humid areas. This is probably due to a relatively low variability in the hydrology and water quality of the irrigated system. Further comparisons with other projects will help to test this hypothesis. Although less variability is present in these data than in other projects in humid regions, analyses show a 40-60% decrease in concentrations is necessary to achieve statistical signifi- cance. More variability is likely to exist in the Rock Creek and Snake River systems that receive the effluent from the irrigation tract, because they are influenced more by meteorologic factors. 23 ------- Prairie Rose Lake, RCWP Shelby County, Iowa MLRA: M-107 I. Project's contributions toward understanding the effectiveness of NFS control efforts: This project is a prime test of whether treating most of the critical erosion areas in a watershed can reduce turbidity and sedimentation in a lake. II. Project Characteristics: 1. Project Type: RCWP budget: $801/849. $446,000 allotted for cost sharing incentives on BMPs. Total watershed area is 4,490 acres of which 80% is cropland. 2. Water Resource Type: Recreational lake, 218 acres, max depth 24ft. The lake is surrounded by state-owned park land. The lake is used for fishing and primary contact recreation with some drinking water withdrawn for use in the park. 3. Use Impairment: The fishery and swimming are degraded by sediment and turbidity. Bathymetric evidence indicates that sedimentation is filling the lake faster than expected. 4. Timeframe: 1980-1991. 5. Water Quality at start of project: In 12 years preceding the proj- ect, there was 10* loss of boating area, 19% loss of storage vol- ume, and major destruction of fish habitat. High concentrations of pesticides were observed in storrawater at the start of project (Dieldrin > 0.0019ug/l and toxaphene > 0.013ug/l; TP averaged 0.23 to 0.5 mg/1 and Chi a averaged 17 to 34ug/l in lake stations during the summer). 6. Meteorologic and hydrologic variability: Climate is subhumid, with precipitation uniformly distributed through the year. Average precipitation: 29 in, USLE R-factor 150 to 175. 7. Monitoring: Bi-weekly monitoring of three lake stations through the summer recreational season (May through September). Additional samples are collected following storms greater than 1 inch of rain- fall during period June through August. 24 ------- 8. Land Treatment: Primary emphasis in this project is on installation of terraces with underground outlet structures and Nutrient and Pesticide Management. Several sediment control structures are also slated for construction. Conservation tillage and fertilizer and pesticide management are generally required along with the terrace contracts. 9. Critical Areas: All cropland in this project is considered critical. 10. Incentives: The project offers 75* cost sharing on all BMPs except Nutrient and Pesticide management, which are not cost shared. An Extension program was conducted early in the project to promote nutrient and pesticide management by offering soil sampling and pest scouting services. 11. Economic Information: Only limited information is available from the RCWP project reports. III. Lessons learned: This project offers the clear opportunity to document whether or not thorough land treatment can accomplish water quality goals. In addition, it will offer insight into the benefits and disadvantages of terraces; i.e. the terraces are expected to reduce soil loss effectively, but terraces may in- crease nutrient transport through subsurface routes. This project is a good example of successful implementation and high participation that can be achieved by a small project with clear objectives. 25 ------- Highland Silver Lake, RCWP Madison County, Illinois MLRA: M-114 I. Project's contributions toward understanding the effectiveness of NFS control efforts: At the present time, there is no certainty the water quality impairment of the lake will be reversed. The field study aspect of the project, if continued, may help to determine if BMPs can effectively reduce the erosion of fine particles of sediment from natric soils. (For more infor- mation see HCWP Status Report on the CM&E Projects, 1985, pp. 65-78.) II. Project Characteristics: 1. Project type: RCWP, comprehensive monitoring and evaluation, project area = 30,640 acres. 2. Water resource type: Stream and lake 3. Use impairment: Recreational, water supply, fish and wildlife. (i 4. Timeframe: 1980-1990 5. Water quality at beginning of project: Average water quality from lake site nearest water intake (May 1981 - April 1983) TSS Turbidity TP TN Chi a mean 27.8 mg/1 54.4 NTU 0.18 mg/1 2.0 mg/1 6.26 ug/1 n 18 17 18 18 17 6. Meteorologic factors: Annual rainfall = 40.5 inches; USLE 'R' factor ~ 200. 26 ------- 7. Water Quality Monitoring Program: 9 lake sites sampled monthly 1 lake outflow site sampled biweekly 3 tributary sites sampled biweekly, discontinued during Oct. 1984 8 field sites sampled during runoff events, not reported and discontinued during Oct. 1984 3 biological tributary sites sampled twice a year 1 channel and streambed survey 1 lake sedimentation survey 8. BMPs: Practices have been selected that will increase ground cover, decrease the velocity of surface runoff, and improve the management of livestock waste (i.e. RCWP BMPs 1,2,4,5,7,8,9,10,11,12,14, and 15). Implementation is behind schedule. 9. Critical areas: Criteria are soil type and slope, which appear to be appropriate. Water quality data should be used to re-evaluate critical areas and where other important sources of pollutants are located. 10. Incentives: 75% cost-sharing with $50,000 maximum. 11. Economic Information: BRS performed analysis; a preliminary esti- mate of the total benefits did not exceed the total costs (benefit/cost = 0.2). The onsite longterm productivity benefits from erosion control are low due to the deep and fertile soils with low slope, and the cost of increasing filtration of drinking water and potential to increase recreational use are less than the cost of the program. III. Lessons Learned: The Highland Silver Lake project had much advance planning. Critical areas were defined and a sound monitoring program was developed. However, BMP implementation level is low and there is a possibility that the current BMPs may not be able to alleviate the lake's water quality problem. 27 ------- Storehouse Parish, LA MLRA: 0-134 I. Project's contributions toward understanding the effectiveness of NFS control efforts. No useful information about BMP water quality relationships has yet come from this project. A major concern of the project is the reduction of pesti- cide residues in water, sediment and biota. Present indications are that little information on BMP effectiveness for reducing pesticide residues will be obtained due to poor coordination of monitoring and land treatment and inappropriate water quality monitoring design. Some information on turbidity reduction from irrigation improvements may be forthcoming. A recent drastic reduction in project size and a corresponding increase in farmer participation rate lends some optimism that this project may yet make a useful contribution to the field of knowledge on NPS control. II. Project Characteristics 1. Project type: RCWP - 66,000 acres (reduced from 220,000 acres). 2. Water resource type: River (bayou). 3. Use impairment: The only documented impairment is the occurrence of excessive organochlorine insecticide residues in fish tissue. 4. Timeframe: 1080-1991. 5. Water quality at start of project: River bottom sediment samples contained 5-400 ppb of various organochlorines during 1982. 6. Meteorologic factors: Mean annual precipitation = 48 inches. USLE 'R' factor = 400. 7. Water quality monitoring program: a. 5 ambient stations (monthly) 53 parameters, b. 2 automated stormwater tributary sites (quarterly storm), c. 3 fish sampling sites (biannual). The monitoring system, while extensive, does not appear to be coordinated with BMP activities nor does it seem to be designed to answer any specific questions. 28 ------- 8. BMPs: All RCWP BMPs have been approved except BMP-3, BMP-6 and BMP- 14. The majority of funds have been spent on BMP-13 (irrigation land leveling, land smoothing, and irrigation water conveyance). As of 9/30/84, 57% of the critical acreage (39* of total project area) was under contract. The 1985 goal was to bring an additional 8300 acres under contract. About 60* of this was actually completed as of 9/30/84. 9. Critical areas: Critical areas are defined as cotton lands and cropland within 3/4 nile of Bayou Bonne Idee. There is no indica- tion of how well contracting has adhered to these criteria. 10. Incentives: Cost-sharing rates .vary from 50-75* depending on BMP. In addition the project has increased the cost-share to 90* for those farms directly bordering the Bayou Bonne Idee. • 11. Economic information: RCWP and farmer BMP expenditures. III. Lesson Learned: It is uncertain at this time whether the project will contribute significant information to the field of NFS control. There is some potential because of the high farmer participation and extensive water quality monitor- ing, in spite of the lack of coordination between the project elements. A major lesson learned is that the original 220,000 acre project area was much too large to achieve observable water quality results with the level of fund- ing available through RCWP. 29 ------- Double Pipe Creek HCWP Carroll County, Maryland MLRA: S-148 I. Project's contributions toward understanding the effectiveness of NPS control efforts: The project will make at least some contribution to knowledge of BMP- water quality relationships. Reporting of BMP implementation accomplishments has been unclear, but it appears that insufficient implementation has occurred to be observable from water quality monitoring. Project area may also be too large to obtain observable improvements with available cost-share funds. The three farm sites (17, 80, and 175 acres) may show water quality effectiveness of specific BMPs within 2—3 years, if the project can get BMPs completed. II. Project Characteristics 1. Project type: RCWP, 110,000 acres. 2. Water resource type: Rivers. 3. Use impairment: Domestic water supply and fishery degraded by sedi- ment and bacterial contamination. Project area contributes nutrients disproportionately to the Chesapeake Bay. 4. Timeframe: 1980-1991. 5. Water quality at beginning of project: Maximum fecal coliform concentrations of 40,000/lOOml during runoff events. Turbidity commonly in excess of 40 NTUs during and after runoff events. 6. Meteorologic factors - Annual rainfall 45 inches, USLE 'R' factor = 200. 7. Water quality monitoring program: Storm and baseflow monitoring was conducted for two years using flow-proportional composite samplers at one mainstream and three farm sites. The post-BMP monitoring phase has been put on hold because of difficulties in getting BMPs applied. The post-BMP monitoring scheme should be sufficient to detect fecal coliform, sediment and turbidity changes in the range of 35-40%. 8. BMPs: Primarily animal waste management, conservation tillage and grassed waterways. 30 ------- 9. Critical Areas: Appropriate critical area criteria have been de- veloped (distance from major streams, size of operation, present conservation status). There is little information on how rigorous- ly these are applied. Accurate targeting to critical areas is very important for this project because the goal is to treat only 15% of the watershed area. 10. Incentives: 75* cost-sharing, $50,000.00 maximum. 11. Economic information: Nothing quantitative is presented. Domestic water treatment costs are reported to be excessive because of high sediment and fecal coliform concentrations. III. Lessons Learned: 1. Project may be a good test of whether 40% pollutant concentration reduction can be achieved by treating identified critical areas that comprise only 10-15% of the watershed. 2. Project personnel very consciously directed recruitment efforts to the large producers in the watershed. Final participation level will indicate whether this was a good strategy. 3. Project area is probably too large for available cost-share funds. 31 ------- Saline Valley, RCWP Washtenaw County, Michigan MLRA: M-lll and L-99 I. Project's contributions toward understanding the effectiveness of NFS control efforts. At the present tine very little information on the actual water quality effectiveness of BMPs has yet developed from the project. However, BMP imple- mentation is approaching levels which can produce water quality improvements, and the water quality monitoring program is well designed to quantify project impact. On this basis we believe that this project may document water quality effectiveness of nutrient control BMPs at a subbasin (1000-8000 acres) level within the next two years. II. Project Characteristics 1. Project type: RCWP - 77,000 acres. 2. Water resource type: Streams and river draining to western basin of Lake Erie. 3. Use impairment: Excessive per area nutrient loading to Lake Erie. 4. Timeframe: 1980-1991. 5. Water Quality at start of project: Highest per acre phosphorus loading to Lake Erie. Majority of TP loading from project area derives from point sources. 6. Meteorologic factors: Mean annual precipitation - 32 inches. USLB 'R' factor = 125. 7. Water quality monitoring program: Weekly grab samples accompanied by stream flow measurements at nine sites since 1980. This monitoring program is appropriate to determine whether mean nutrient concentrations are changing significantly over time. 8. BMPs: The project is focusing on obtaining nutrient loading reduc- tions from animal waste management, conservation tillage, and fer- tilizer management. 14,465 out of 42,428 critical cropland acres were under contract as of 9/30/84 in addition to 19 of 27 critical dairy operations. 32 ------- 9. Critical Areas: Criteria - All cropland and animal operations with- in 1/4 mile of watercourses. 10. Incentives: RCWP cost-sharing up to 75* and $50,000. 11. Economic Information: RCWP allocation for BMPs is $1,880,000.00. This translates to a government expenditure of approximately $25/acre. If farmer contributions and only cropland are consid- ered, the BMP investment is about $50/acre. III. Lessons learned from project: 1. Original 200,000 acre project area was too large to achieve ade- quate, BMP coverage with amount of funding available through RCWP. 2. Monitoring smaller subbasins within the overall project area can associate water quality results with BMP implementation more effi- ciently than monitoring the overall project area. 33 ------- Tillamook Bay RCWP Tillamook County, Oregon MLRA: A-l I. Project's contributions toward understanding the effectiveness of NFS control effort: This project is making an important contribution concerning the ef- fectiveness of animal waste management for improving water quality at a water- shed level. The water quality results to date show that a 40-50% reduction in log-mean fecal coliform concentrations have been achieved by bringing approxi- mately 60% of the animal waste produced in the project area under best manage- ment. A more thorough knowledge of the marginal water quality benefits of increased manure management should be gained from this project as the total treatment approaches 90* of needs over the next two years. II. Project Characteristics 1. Project type: HCWP - 23,540 acres. 2. Water resource type: Estuary and river tributaries. 3. Use impairment: Bacterial contamination of shellfish beds. 4. Timefrarae: 1981-1991. 5. Water quality at beginning of project: Fecal coliform concentra- tions in Tillamook Bay were in excess of public health standards a majority of the time. 6. Meteorologic factors: Mean annual precipitation= 90 to 140 inches depending on elevation. USLE 'R' factor > 50. 7. Water quality monitoring program: 14 bay sites have been grab sam- pled for fecal coliform and salinity since 1960. Several intensive storm samplings have also been conducted in the bay. Grab sampling with flow measurement has been conducted for various periods of time on several tributary sites. This has included both intensive storm and baseflow sampling. 8. BMPs: All activity has been related to improving the management of animal waste. This has included manure storage facilities, im- proved milkhouse conditions (curbing and guttering), and sub-surface drainage on pasture receiving manure applications. Approximately 80% of dairies in the project area were contracted and 60% of implemen- tation work completed by 9/85. 34 ------- 9. Critical area: The project made an extensive effort to identify critical dairy operations. Most of this was done by on-site inspec- tion using an on-farm rating procedure. 10. Incentives: Seventy-five percent RCWP cost-sharing has been avail- able to a maximal of $50,000. The cost of some systems has been very high resulting in a high cost to the dairy fanner. The state of Oregon allows a 60* tax credit for conservation investments over 10 years. The Tillamook Creamery Association reduces the price it pays the dairy farmer for milk if substandard aesthetic or sanitary conditions are observed at the dairy. State water quality regula- tions are also available to compel farmers to comply with project objectives. 11. Economic information: Total BMP cost = $4,000,000. Cost/acre = $170.00. The project is attempting to quantify on-farm and water quality benefits of the project. III. Lessons Learned: 1. Animal waste management can affect water quality improvements in terms of reduced mean fecal coliform concentrations when imple- mented on a 23,000 acre project. 2. Some measurable indicator of the meteorological state is generally needed in a monitoring program to attribute water quality changes to BMP implementation. 3. A pre-BMP water quality data base of at least 2 years greatly facilitates documenting water quality effects of BMPs. 4. A high level of farmer participation can be achieved when agricul- tural and water quality agency personnel work together closely designing and publicizing the program. 5. The combination of financial incentive and environmental regula- tion is effective in achieving high rates of participation. 35 ------- Conestoga Headwater RCWP Lancaster Couoty, Pennsylvania MLRA: S-148 I. Project's contributions toward understanding the effectiveness of NFS control efforts. Project results come exclusively from one 25-acre, intensive field site. A full description of project contributions is provided in the 1985 NWQEP RCWP-CM&E Report. (NWQEP, 1985) Results are summarized below: A. In permeable soils with excess manure, terraces may increase nitrate transport to groundwater and may increase dissolved nutrient concen- trations in surface runoff. Terraces may also reduce sediment and nutrient loadings to surface water by reducing the volume of run- off. B. In this project raanurial nutrients greatly exceed crop needs. Thus, water quality benefits from animal waste storage (e.g. improved timing of applications) are offset because nitrogen that could have been volatilized is conserved in storage. C. Nutrient management BMPs (soil and manure testing, proper matching of application rates, and timing to plant needs) can reduce both ground and surface water nitrogen losses. II. Project Characteristics 1. Project Type: RCWP - Comprehensive Monitoring and Evaluation of 110,000 acres. 2. Water Resource Types: Streams, groundwater. 3. Use Impairment: Groundwater is impaired for domestic water supply by excessive nitrate concentrations. 4. Timeframe: 1981-1991. 5. Water quality at initiation of project: Nitrate levels exceeded lOrag/1 in a majority of wells. Maximum concentrations were over 100mg/l. 6. Meteorologic factors: Average annual precipitation = 42 in. USLB 'R' factor = 175. 36 ------- 7. Water Quality Monitoring Program: The program is comprehensive and monitors both ground and surface waters. The most \intensive monitoring is done on two 25 acre field sites where complete moni- toring of ground and surface water losses is performed. A system of wells and stream gages is also present to detect changes at the project and subbasin levels. 8. BMPs: The BMPs emphasized include animal waste management, ter- races, fertilizer management, grassed waterways, and conservation tillage. Farmer participation to date has been low with only 4* of the project area under BMP contract as of 9/85. 9. Critical Areas: Qroundwater critical areas are indentified as farmland overlying carbonate soils. Directing BMP cost-sharing to critical areas has not been effective because of the lack of farmer part icipat ion. 10. Incentives: Project has RCWP cost-sharing that ranges 40 - 75% de- pending on BMP. The usual on-farm economic incentive for nutrient management appears to be minimized in this project because there is a large excess of manure. Therefore, manure is a 'waste product' rather than a 'resource'. 11. Economic information: The Economic Research Service (ERS) has performed an on-farm economic analysis of this project. This analysis is summarized in the 1985 NWQEP RCWP-CM&E Report. III. Lessons Learned from the Project 1. Most economic incentive for nutrient management is lost when menu- rial nutrients produced in the project area greatly exceed crop and pasture requirements. 2. A 50% cost-share for animal waste management is insufficient to generate very much farmer participation when on-farm manure nu- trients exceed crop needs. 3. Low participation rates undermine the effectiveness of a critical area targeting plan. 37 ------- I. Project's contributions toward understanding the effectiveness of HPS control effort©: This project has a high probability of showing whether conservation tillage is beneficial or detrimental to groundwater quality in highly peraeable soils. II. Project characteristic©: 1. Project Type: HCWP Comprehensive Monitoring and Evaluation. Budget $3,848,157;, $1.24 million allocated for cost sharing incentives on BMP implesaentatioin. Total project area is 106,000 acres in four counties. 2. Water Resource Type: Tfee project is concerned with protection of near surface aquifers and recreational lakes. 3. Use IsapairiBent: Nitrate contamination of aquifer drinking water sup- plies and eutrophication of the recreational lakes. 4. Tisaefrasie: 1981-1991. 5. Water quality at start of project: Mitrate in 27% of 861 private wells sampled exceeded 10 sag/l K; TP in lakes 0.12-0.15 mg/1 P; TP=0.5 mg/1 in tributaries; total -M 3-9 mg/1. 6. Sfeteorologic and faydrologic variability: Average precipitation: 22 in/yr, largely in snowfall. H-factor for USLB is 100. 7. ffcnitoring: Growndwater is monitored 'by groups of wells associated with field sites with known levels of BMP implementation and known geologic formation. 8. Land Treatment: F*k>st project effort is directed toward conservation tillage, fertilizer oanageraent and pesticide taanagement. Minor ef- forts are directed to sediment control with terraces and to animal waste management systems. 9. Critical Areas: The project has developed a scheme for prioritizing farms based on proximity to the lakes for sediment control, and aquifer depth and soil characteristics for groundwater protection. At this tioe there is no indication whether these criteria will be successful. 38 ------- 10. Incentives: The project offers a cost sharing incentive for imple- mentation of conservation tillage and a dollar incentive for imple- mentation of fertilizer management. Pesticide management is en- couraged by a complementary IPM program sponsored by Cooperative Extension. Fertilizer management is still viewed by area farmers more trouble and expense than it is worth. 11. Economic Information: Economic analysis of this project suggests that protection of the groundwater supply may not be worth the cost of the program, but protection of the recreational benefits of Oakwood and Poinsett lakes might justify the cost. III. Lessons learned: Intensive Extension efforts to promote IPM are a strong incentive for implementation of pesticide management as a BMP. A similar effort to assist farmers in soil sampling might also have positive effects. 39 ------- Snake Creek, RCWP Wasatch County, Utah MLRA: E-47 I. Project's contributions Control efforts. toward understanding the effectiveness of NFS The UT-HCWP project has added to our understanding of the effectiveness of BMPs in arid, irrigated areas and the effectiveness of animal waste management system. The project has nearly 100% implementation of a small area. Significant reductions in phosphorus (40-65%), nitrogen (45-60%), and bacteria (50-90%) concentrations were found after animal waste BMPs were implemented. These results were documented with five years of water quality data (two years pre-implementation, one year during implementation, and two years post-implementation), which is a much shorter period than is generally required to document effectiveness for projects in humid, non-irrigated regions. II. Project Characteristics: 1. Project Type: RCWP, project area is approximately 700 acres, near the mouth of a 24,700-acre watershed. 2. Water Resource Type: Streams and a reservoir. 3. Use Impairment: Phosphorus limited eutrophication of Deer Creek Reservoir domestic water supply, recreation, and aesthetic enjoyment. 4. Timeframe: 1980-1990. 5. Water quality at beginning of project (1980-1981 concentrations): Station 14 ISnake Creekl Station 6 iditchi geometric max. mean TP (mg/1) TKN (mg/1) FC (count/100ml) 0.02 0.05 30 0.71 1.00 7500 0.09 0.46 282 0.04 .10 19 geometric max. mean .56 4.6 12,800 0.15 0.77 407 6. Meteorologic factors: USLE 'R' factor ~ 30. 40 ------- 7. Water quality monitoring: Consists of monthly (weekly during spring runoff) grab samples with simultaneous flow measurements; State Health Department analyzed samples for fecal coliform. Several different labs analyzed the samples for nutrients at various times. Different labs reportedly used the same analytical techniques. 8. BMPs: Mainly animal waste management (BMP #2) with storage facili- ties. 9. Critical Area: All of the dairy operations in this small project area were thought to be critical. 10. Incentives: cost-share at 75%, with $50,000 maximum. 11. Economic Information: Limited information available. III. Lessons Learned: This project not only was successful in reducing nutrient and bacterial concentrations, but also was exemplary in its region. Other dairies in the Heber Valley area now are considering installing similar practices after seeing the success of the Snake Creek RCWP. The small area of this project made it ideal for (nearly) complete implementation and ease of tracking Water quality data identified two critical areas: one small reach of the Snake Creek and the Huffaker Ditch. Water quality data indicate that it may not have been necessary to install practices outside of these two critical areas. 41 ------- Nansemond-Chuckatuck, RCWP Suffolk and Isle of Wight Counties Virginia MLRA: T-153A I. Project's contributions toward understanding the effectiveness of NFS control efforts: Little contribution will be obtained from this project unless it concen- trates its implementation efforts to address the phosphorus, nitrogen and bacterial sources in the critical area. II. Project Characteristics: 1. Project Type: HCWP, budget: $2,076,931, 1.5 million for cost sharing incentives on BMPs. Total project area is 161,365 acres including only 44,000 acres of cropland. The project is administered in two counties. 2. Water Resource Type: The project includes two estuaries and seven water supply reservoirs. The combined surface area of reservoirs is 4,850 acres. 3. Use Impairment: Shellfish production in the estuaries is severely impaired, due to bacteria and BOD from nonpoint sources. The water supply reservoirs are not impaired but are threatened by eut rophicat ion. 4. Timeframe: 1981-1991. 5. Water Quality at start of project: Estuary - 3,000 acres of shellfish area has been condemned, chlorophyll a concentrations often exceeds 40 ug/1, and DO is frequently depleted. Reservoirs- phosphate concentrations ranged 0.05 to 0.20 in fall and winter samples, higher values accompanied by high fecal coliform bacteria were observed in some tributary streams. 6. Meteorologic and hydrologic variability: Mild climate, 48 in annual rainfall with periodic summer droughts. R-factor for USLE is 300. 7. Monitoring: Monthly sampling of water supply reservoirs and estu- aries. A thorough longitudinal study of water quality through the length of the estuaries was conducted in 1983. This was supple- mented by monthly sampling at specified stations during 1983 and 1984. 42 ------- 8. Land Treatment: Project effort is directed toward implementation of animal waste management systems, conservation tillage, ferti- lizer and pesticide management, cropland protection systems, and sediment control structures. Implementation prior to October 1984 had not been substantial. 9. Critical Areas: The project designates critical areas on the basis of proximity to the reservoirs, the reservoir tributaries, and the estuaries. The designated critical area accounts for 29,000 acres of farmland. 10. Incentives: Most implementations were cost shared at 75% except for crop cover, waste transportation vehicles, and injectors which were cost shared at 50*. No cost sharing is offered for ferti- lizer management or pesticide management. 11. Economic Information: Limited information available. III. Lessons learned from the project: A large project such as this one has considerable difficulty focusing its activities sufficiently to achieve a measurable impact on the quality of its impaired water resource. 43 ------- St. Albans Bay - RCWP Franklin County, Vermont MLRA: R-142 I. Project's contributions toward understanding the effectiveness of NFS control efforts: This project has made substantial contributions to understanding the water quality effects of animal waste management BMPs and will make greater contributions in the future. These contributions are discussed in detail in the 1985 NWQEP RCWP-CM&E Report. (NWQBP, 1985a) II. Project Characteristics 1. Project type: RCWP-Conprehensive Monitoring and Evaluation, 33,000 acres. 2. Water Resources: Bay of Lake Champlain and small tributaries to bay. 3. Use Impairments: Degraded water quality impairs the bay for swimming, boating, fishing, and aesthetic enjoyment, and has re- duced property values relative to other lake-side property on Lake Champlain. P loading is 24% NFS and 76% municipal point source. 4. Timeframe: 1980-1991. 5. Pre-project water quality: Dissolved P concentration in the inner bay were consistently above levels known to cause eutrophication. Fecal col i form counts at the state park beach were often above 200/100 ml. 6. Meteorologic factors: Mean Annual precipitation = 35 inches USLE 'R' factor =100. 7. Water quality monitoring program: The RCWP Comprehensive Monitoring and Evaluation Project conducts a very detailed WQM program on the bay and tributaries. At least 2 years of pre-BMP monitoring was conducted at most sites against which post-BMP data can be com- pared. 8. BMPs: Animal waste management is the primary BMP being implemented on dairy farms in the project. 12,762 out of 15,257 critical areas in the project area were under RCWP or ACP contract as of 9/84 with about 75% of this amount completed. 44 ------- 9. Critical areas: criteria include amount of manure, distance from watercourse, present manure storage and spreading practices, and manure spreading rates. The project appears to have adhered to these criteria in prioritizing cost-share requests. 10. Incentives: Nearly all RCWP practices carry a 75* cost-share.. 11. Economic information: Economic Research Service (ERS) has conducted an extensive on-site and off-site analysis of project benefits and cost. This analysis is summarized in the 1985 NWQEP RCWP-C.M&E Report. (NWQWP, 1985a) The analysis shows the project to have a high benefit to cost ratio. III. Lessons Learned from Project This project demonstrates that a high level of dairy farmer participation in NFS control can be obtained when: 1. Farmers and the local community are educated about the water quality problt 2. Investments in waste management provide payback in terms of de- creased commercial fertilizer usage and labor. 3. A high initial cost-share rate is provided. The project also demonstrates that very detailed monitoring of land management activities may be needed to relate water quality to land use changes. 45 ------- MODEL IMPLBBNTATION PROJECTS (MIP) West Branch of the Delaware River, MIP Delaware County, New York MLRA: R-140 I. Project's contributions toward understanding the water quality effective- ness of NPS control efforts: Estimates of the effectiveness of barnyard management of dairy farms (potential 50-90* reduction of'TP) were'ga'ined from this project. Manure spreading schedules to minimize phosphorus1 "losses were also developed using soil and manure testing and modeling techniques. Average TP load reduction from these new schedules was estimated to be 35% for the seven farms analyzed by model simulation. Project contributions were beneficial to the beginning phases of several RCWP projects, particularly VT-RCWP. II. Project Characteristics tr ••*'.. 1. Project type: Model Implementation Project, Project area = 287,000 acres. - ' •' '••'•''','• 2. Water resource type: Streams and reservoir; 3. Use impairment: Domestic water supply, eutrophication of reservoir, downstream flow quality. 4. Timeframe: 1978-1982. 5. Water quality at beginning of project: Eutrophic conditions pre- vailed in Cannonsville reservoir. 6. Meteorologic factors: annual rainfall = 40 inches, USLE ' R ' factor 100. 7. Water quality monitoring program: Event and baseflow monitoring of one station at the mouth of the river was performed for 30 months. These water quality data were used in the modeling aspect of the project. They were not sufficient by themselves to document any potential trends. Two field sites (barnyards) were monitored, one site for a year and the other for approximately five months, to document the effectiveness of barnyard practices. 8. BMPs: Emphasis was on animal waste and barnyard management prac- tices, with some erosion control, streambank protection, and other practices. 46 ------- 9. Critical areas: Criteria for critical areas were: a. areas with highest concentrations of apparent problems, b. distance to water course, and c. landowner's willingness to participate. The 275 farms in the project area were prioritized according to these criteria. Out of 154 high priority barnyards, 91 were treated by 1982. 10. Incentives: Cost-sharing ranged from 50% (for permanent vegetative cover) to 90% (for animal and milkhouse waste facilities) for vari- ous practices. III. Lessons Learned: Although many farms were treated with barnyard practices, no net effect on the impaired reservoir has been documented. Modeling of the West Branch Watershed indicated that the majority of TP losses were from direct runoff from manured cropland. Manure spreading schedules to reduce 35% of this portion of the loading were developed, but not widely implemented. Perhaps a total manure management strategy with the implementation of these developed manure spreading schedules would have been more effective. Data from this project have not yet been fully analyzed. Additional analyses could give more insights on the benefits of barnyard management practices and potential water quality gain on a watershed level. 47 ------- Broadway Lake Model Implementation Project Anderson County, SC MLHA: P-136 I. Project's contributions toward understanding the effectiveness of NFS control efforts. Although there are many "lessons learned" (see Section III) from the project, there is little contribution to knowledge about the water quality effectiveness of BMPs. This is due primarily to the project's emphasis on farm ponds below pasture and pasture improvement. Interrupted monitoring of the major water resources also limited any useful results. The Clemson sub- basin monitoring work provided a methodology for calculating the relative sediment contribution of streambank erosion. It showed that streambank erosion could be expected to contribute 20-50% of sediment loading in the southeastern piedmont physiographic province. II. Project Characteristics 1. Project type: Model Implementation Program (MIP) 25,183 acres. 2. Water resource type: 300-acre lake and feeder streams. 3. Use impairments: a) filling of the lake by sediment impairs boating and fishing, b) sedimentation in stream channels may be increasing flood frequency. 4. Timeframe: 1978-1982. 5. Water quality at start of project: Little information is avail- able. A biological monitoring study conducted in 1979 at 4 sites immediately upstream from the lake indicated that the water quality was relatively good overall. Pre-MIP studies at the same sites found high nitrate and sediment concentrations during storm events. 6. Meteorologic factors: Mean annual precipitation = 47.5 inches. USLE 'R' factor = 300. 7. Water quality monitoring program: The monitoring of the lake and lower tributaries was curtailed in the middle of the project time- frame, and the monitoring effort was inadequate to determine water quality effects of the project. A separate monitoring effort by Clemson University in the upper tributaries was designed to deter- mine BMP effects by comparing treated and untreated subbasins. This effort was plagued by inaccurate land use information and by the fact that BMPs were contracted inappropriately in the control ba- sins. 48 ------- 8. BMPs: Sixteen BMPs were approved for the project. The majority of implementation included farm ponds, terraces, conservation tillage and pasture improvement. 9. Critical Areas: Critical areas were identified as cropland and roadside banks. Nearly all program applicants received cost-sharing. 10. Incentives: Cost-sharing rates were 90% for most practices with a maximum of $3,500 per year per farm. 11. Economic information: Only BMP cost figures are available. III. Lessons learned: 1. Treating pastures which have low erosion rates is probably not a cost effective means of reducing watershed sediment loading. 2. Constructing sediment ponds below pastures is, likewise, probably not a cost-effective means of reducing watershed sediment loading. 3. Large projects, such as the MIP, benefit from having a full-time project manager who has both agricultural and water quality know- ledge. 4. Cost-share funds should be made available to watershed NFS control projects on a total project basis rather than year by year. Annual funding can seriously undermine critical area targeting attempts. 49 ------- South Yakioa Model Implementation Project Yakima County, Washington MLRA: B-8 I. Project's contributions toward understanding the effectiveness of NPS control efforts: This project has provided substantial information relating to the water quality improvements from BMP implementation in irrigated agriculture. The project fell far short of its potential contribution because BMP implemen- tation and water quality monitoring ceased at the end of the MIP funding period. There was, consequently, not enough time to achieve or monitor the water quality benefits of the land treatment. Nonetheless the project demon- strated that significant sediment reductions in return flows could be accom- plished with irrigation BMPs. The conversion from furrow to sprinkler or trickle irrigation was particularly effective in reducing both erosion and water usage. Practices which reduce water usage also improve the quality of the return flow. From a water quality perspective, the ideal is to manage irrigation operations to the level at which return flows are eliminated. The monitoring results also demonstrated that high sediment losses are associated with 1) the first irrigations of the season, 2) highly erosive crops and 3) accidents such as broken pipes and direct field-sloughing into drains. II. Project Characteristics 1. Project type: Model Implementation Program (MIP)-26,500 acres. 2. Water resource type: Irrigation canals, river. 3. Use impairment: Excessive sediment from improper irrigation manage- ment was found to impair waters for irrigation uses by filling canals and drains and clogging irrigation equipment. Fisheries are also believed to have been affected, although no documentation is available. 4. Timeframe: 1978-1982. 5. Pre-project water quality: Pre-MIP studies indicated that approxi- mately 80% of sediment loading to the Yakima River was derived from irrigated agricultural sources. The remaining 20% originated from natural upstream sources (snowmelt, streaabank erosion. At the beginning of the project a majority of return flows had suspended solids concentrations greater than 3.5 ml/1 as measured by the Imhoff cone.) 50 ------- 6. Meteorologic factors: Mean annual precipitation is about 10 inches. USLE 'H' factor =20. 7. Water Quality Monitoring Program: Three levels of monitoring were conducted. a. Daily sediment sampling of all MIP-1 individual fields where irrigation was occurring. Collection ditches and drains, 1979- 1981. b. Weekly or biweekly sampling by the conservation district of all supply water and ditches/drains carrying water in or out of pro- ject area. c. An Imhoff cone - 10 percent - systematic sampling of all ir- rigated farms throughout the entire irrigated area of central Washington, 1979-1981. Unfortunately all monitoring ceased at the conclusion of the project. 8. BMPs: Many BMPs were approved for the project. The emphasis was on a) conversion from furrow to sprinkler systems (3,100 acres), b) crop residue use management (5,573 acres), c) water conveyance pipe (1 million feet), d) sediment basins (23), e) subsurface drain system (9740 ft) and f) improved water management (8,100 acres). About 50% of project area benefited from BMP installation. 9. Critical areas: A quantitative on-site evaluation rating form was developed using soil type, crop, slope, irrigation method and return flow system as criteria. The ratings were not strictly adhered to because of the need to generate momentum early in the project and ; because cost-share funds were available only on an annual basis. 10. Incentives: ACP cost sharing up to 90* was available with a maximum of $3,500/yr. However, the practices installed were expensive, so that, in the final analysis, farmers paid about 67% of the total BMP cost. 11. Economic information: No economic information has been reported except BMP costs. III. Lessons learned from Project 1. To generate initial enthusiasm and farmer participation, some com- promise of the critical area targeting is generally be needed. 2. The Imhoff cone is a very effective tool in helping irrigators see their soil loss and thereby promote better management practices. 3. The conversion of furrow to sprinkler irrigation systems is the most effective BMP both for reducing sediment losses and conserving water. 4. A fulltime project manager is a key ingredient to developing a coordinated project. 51 ------- OTHER NONPOINT SOURCE PROJECTS: Lake Le-Aqua-Na Stephenson County, Illinois MLRA: M-108 I. Project's contributions toward understanding the effectiveness of NPS control efforts: This project has an integrated approach to watershed management that in- cludes land use and in-lake treatment. It has a strong probability of a- chieving its goal to cleanup the lake, but it may not fully document the effectiveness of the BMPs implemented in the watershed. It has the potential to demonstrate the effectiveness of conservation tillage. The organization at the local level contributed considerably to the successful implementation of this project. II. Project Characteristics: 1. Project type: Clean Lakes Program along with the Agricultural Conservation Program and the Illinois Dept. of Conservation. Water- shed area = 963 ha (2400 acres). 2. Water resource type: Lake with streams. 3. Use impairment: Recreation, aesthetics, and loss of lake capacity due to sedimentation. 4. Timeframe: Phase I 1981-1983; Phase II 1984-1986. 5. Water quality at beginning of project: 1981 mean lake concentrations TP = 0.323 mg/1 DP = 0.217 mg/1 IN = 1.85 mg/1 Chlorophyll a ranged from 2 to 243 ug/1 with mean = 89.4 ug/1; nuisance algal blooms dominated by blue-green algae. During peak stratification, 51* of lake volume was anoxic. Several winter fish kills have occurred. 6. Neteorologic factors: USLE 'R' factor = 175. 52 ------- 7. Water quality monitoring program: Program is based on a before-and- after implementation sampling scheme. It has 5 stations: one on a creek just above the lake; 3 stations within the lake; and 1 station below the dam. Unfortunately, there are no stations in any of the small subwatersheds of the project area that could be used to docu- ment the effectiveness of specific BMPs treatment. The one station located just above the lake may not be adequate to show significant changes in water quality if hydrologic variability is high during the program period. 8. BMPs: Consist mainly of conservation tillage with some terracing, stripcropping, waterways, sediment basins, and streambank protec- tion. Other non-BMP in-lake treatments include (1) lake de- stratifiers, (2) macrophyte harvesting, (3) chemical algae control (CuSOi), and (4) shoreline stabilization. 9. Critical areas: Criteria for selection of critical areas were (1) distance to water course and (2) erosion rate. 10. Incentives: Cost-share payment for conservation tillage varied with the amount of residue left. Practices other than conservation tillage received 80* cost-share. i III. Lessons Learned: Comparisons (two sample t-test) of 1981 to 1984 water quality data for both the stream and the lake stations showed no significant differences; however, the means of most parameters were lower in 1984 than in 1981. Use of stronger statistical analyses may verify significant decreases over this period. The monitoring scheme may not be adequate to document changes in water quality, especially changes due to BMPs, but visual improvement in the appearance of the lake have been reported. In this respect, the project may be successful with its lake protection/restoration program and may increase the recreational benefits of the area whether or not improvement is verified by chemical monitoring. 53 ------- Skinner Lake, Clean Lakes Program Noble County, Indiana MLRA I. Project's contributions toward understanding the effectiveness of NFS control efforts: Public perception of water quality efforts produced a successful project even though public costs were very high (approx. $1,000 per acre) and measured water quality impact was marginal. II. Project Characteristics: 1. State of Indiana Clean Lakes Project, funded by EPA Great Lakes Pro- gram . $909,000 (50* local funds). Project size: 9977 acres. 2. Water Resource Type: 125-acre lake, mean depth: 14 ft., flushing coefficient varied 2.55 to 5.30 during study. 3. Use impairment: Whole body contact recreation and fishery degraded by eutrophication. 4. Tiaeframe: 1977-1982. 5. Water quality at start of project: TP=0.082 mg/1 in the lake, TN/TP=19 to 220 (considered to be P-limited). 6. Meteorologic and hydrologic variability: Snowmelt produced 21% to 53% of discharge; 35% to 39% of annual streamflow from spring rain- fall. June rainfall accounted for 25% of 1981 annual runoff. An- nual R-factor for USLE is 150. 7. Water Quality Monitoring Program: Bi-weekly in lake and tributary streams 1978 through 1981. Daily sampling was added during spring 1979, spring 1981, and spring 1982. 54 ------- 8. Land Treatment: Practice Contracts Units tons soil cost cost/ton saved/yr Sediment basins 3 Diversions 5 Minimum Tillage 16 Sediment control structures 9 Terraces 17 Vegetative cover 22 Large sediment basin (5acre) each 2700ft 1178ac 22525ft 469ac 590 25 3372 225 2014 3250 $13,363 1,792 12,371 6,886 166,859 13,892 137,011 $23.11* 71.68* 3.67 30.60* 82.85* 4.27 *cost should be amortized over useful life for comparison. 9. Critical Areas: Project defined critical sites as those with severe erosion problems, particularly those on slopes adjacent to the lake or its tributaries. Applications were, prioritized on first come first served basis, then accepted or rejected by SWCD based on: 1) whether it was in the project work plan, 2) seriousness of problem and relationship to agricultural pollution, 3) whether application was for an individual or a group (groups were placed ahead of indi- viduals), 4) urgency of problem, and 5) interest and aptitude of applicant. Locations were rated by proximity to lake or tributaries as A, B, or C. Of 79 participants, 50 were rated as A, 14 as B, and 15 as C. 10. Incentives: Initially, cost sharing incentives were set to range from 25* for repair of tile mains to 50% for construction of sedi- ment basins. These rates were too low to attract many participants, so they were raised to 80 to 85% of total costs. By the end of the project, interest was very high. Participants were not allowed to pick practices that had high cost share rates exclusively when other practices were identified in the conservation plan. 11. Economic Information: None presented in project reports. III. Lessons learned: Control of agricultural nonpoint sources can be very expensive ($909,000 to treat 9,977 acres), particularly if practices such as terraces and large sediment basins are included in the plan. Indications from monitoring within the project period, 1979 to 1982, are that some water quality improvement has occurred, but no follow-up has apparently been done. According to USLE esti- mates, more soil loss was prevented by spending $12,000 on minimum tillage than by spending $166,000 on terraces. Project data show about 10% reduction 55 ------- in sediment, nitrogen, and phosphate delivered to the lake as a result of the large sediment basin. Although this efficiency is very low, any reductions here affect the quality of Skinner Lake directly. 56 ------- Big Stone Lake Restoration Project CLP Minnesota and South Dakota MLRA: M-102A I. Project's contributions toward understanding the effectiveness of NFS control efforts: Although it is too early for this project to show effectiveness of BMPs, it does offer insights into the organizational aspect of large, interstate projects. The cooperative efforts among the nany agencies involved with this project reflect successful communications and planning which are important factors of watershed management. II. Project Characteristics 1. Project type: Clean Lakes Program with some SCS, ASCS, CES, and local funding as well. Watershed area = 750,000 acres. 2. Water resources type: Streams and lake. 3. Use impairment: Recreation and fishing. 4. Timeframe: Phase I 1981-1983; Phase II 1984-1988. 5. Water quality at beginning of project: The lake is hypereutrophic with abundant plant growth, noxious odors, and high sedimentation. 6. Meteorologic factors: USLE 'R' factor = 90. 7. Water quality monitoring program: There are several levels of monitoring: (1) tributary - 9 stations for storm event sampling with base flows sampled quarterly to assess nutrient and sediment concentrations, loadings, and exports, (2) subwatershed monitoring - 3 subwatersheds with baseflow and storm event monitoring, (3) above and below grab samples for assessment of the impact of streambank erosion on water quality, (4) before and after grab sampling of feedlots and, (5) in-lake monitoring - 6 stations, each one having 6 sub- sites to form one composite, sampled monthly for nu- trients, sediments, chlorophyll a, oxygen, and physical parameters. 57 ------- 8. BMPs: Conservation tillage, waterways, and animal waste (feedlot) management. 9. Critical areas were determined via models. Feedlots were rated and prioritized separately from other sources using a feedlot model. The AGNPS model was used to determine critical areas. It was also used to estimate the percent reduction after 2 years of implementation and for the final implementation goals. 10. Incentives: Various agencies provided cost-share funds: feedlot management: 85%, to a maximum of $7500 conservation tillage: $15/acre waterways: 75%, to a maximum of $3500 III. Lessons Learned: This project is beginning to implement land treatment practices and it will be at least a few years before it can demonstrate BMP effectiveness. Its use of models to select critical areas for a large watershed will hopefully lead to successful targeting of resources. The communications and planning developed among the many agencies involved in the pre-project and Phase I periods produced a strong organizational structure for the project. 58 ------- LaPlatte River Watershed, PL83-566 Chittenden County, Vermont MLRA: R-142 I. Project's contributions toward understanding the effectiveness of NFS control efforts: The LaPlatte River Watershed project will contribute knowledge on the effectiveness of manure management (timing and type of spreading on fields) and of barnyard and milkhouse practices. On a watershed scale, the project may reveal the potential water quality benefits from the implementation of animal waste storage facilities. In addition, the study examined the model CREAMS for simulating pollutant losses under northern U.S. climatic conditions found within this watershed. II. Project Characteristics: 1. Project type: PL83-566. The watershed is approximately 34,000 acres. 2. Water resource type: Streams and bay of Lake Champlain. 3. Use impairment: Recreation, aesthetics. 4. Timeframe: 1979-1990. 5. Water quality at beginning of project: (1980 Data) Station 1 (~ 67* of watershed) (ng/D TSS TP TKN min. 1.1 0.113 0.47 max. 64.8 1.406 3.52 median 8.95 0.327 1.02 Station 2 12* of watershed) (mg/1) TSS TP TKN min. 2.9 0.023 0.07 max. 78.3 0.424 3.35 median 15.5 0.90 0.79 6. Meteorologic factors: normal yearly precipitation =33.7 inches; USLE 'R' factor ~ 90. 59 ------- 7. Water quality monitoring program: A comprehensive Monitoring scheme with automatic and continuous monitoring is one of the strong char- acteristics of this project. This scheme includes four stream stations, one point source (STP), two edge of field sites, and other special short-term monitoring projects. More details are presented in the appendix to this report. 8. BMPs: The main focus of this project is on animal waste management systems, with some implementation of erosion and sedimentation con- trol practices, such as conservation cropping systems, strip cropping, contour farming, hayland management, permanent vegeta- tion, diversions, waterways, and streambank protection. Milkhouse waste and barnyard management systems were also included. 9. Critical Areas: Critical areas were not defined for the watershed. Applications for the contracts were ranked subjectively either high or low without supporting evidence as they were received. Some on- site visits did occur. For the most part, there were no criteria for determining critical areas. 10. Incentives: The percentage of cost-share for BMPs in this program varied among the different BMPs as follows: 75% - agricultural waste management (including storage facili- ties) and streambank protection. 60% - waterways, livestock exclusion, and pasture and hayland planting. 50% - diversions and troughs for pasture management. A maximum of $30,000 per item (BMP) was allowed. III. Lessons Learned: It will take several years of water quality monitoring before any signi- ficant trends can be established from the water quality data due to the high variability of meteorologic and hydrologic factors. The model CREAMS was found to be inadequate for predicting runoff, sediment and phosphorus export for the two field sites tested. It was recommended that the model be modified and/or carefully fitted with observed data to yield more accurate estimations of export under northern U.S. climatic conditions. 60 ------- Columbia Basin Block 86 Grant County, Washington MLRA: B-8 I. Project's contributions toward understanding the effectiveness of NFS control efforts. This project has made a tremendous contribution to understanding the water quality effectiveness of irrigated agriculture BMPs. By obtaining nearly 100% land treatment (sediment basins, subsurface drainage, structural modifications of drains, conversion of furrow to center pivot systems) accom- panied by an intensive water quality monitoring program, this project has demonstrated water quality improvements. Sediment loading reductions of ap- proximately 80% and total P reduction of approximately 50% have been rigorous- ly documented. These results are displayed in Figure 1. Sediment control practices were found to have no significant effect on nitrogen loadings. As shown in Table 3, the decrease in sediment yield was a function of controlling the amount of soil loss and trapping sediment in basins. 5000 c O) -o Ol OO 2500 (4472)(2903) 4386 _ 2867 (2463) 2422 Sediment Phosphorus -,3000 (2578) 2556 (1284) 1373 (835) 843 (1152) 1252 (1261) 1404 (506) 501 (976) 1004 2000 — in 3 O c. o _c o. 1000 1977 1978 1979 1980 1981 Figure 1. Net sediment and phosphorus losses from study area, 1977-1981. Note that the values in parentheses were corrected by the USGS to account for errors in compositing samples, (from King et al, 1983) 61 ------- TABLE 3 OVERALL SEASONAL SEDIMENT BASIN RETENTION PERFORMANCE, 1977-1981 (from King et al., 1983). Date 1977 1978 "l979 1980 1981 Number of sediment bas i ns 8 8 15 20 20 Total sediment produced (mt) 5594 4193 3452 3462 4257 Main-drain sediment discharge (mt) 4809 3015 1287 1101 1550 Sediment basin retention (mt) 785 1178 2165 2361 2707 Sed iment retent ion (%) 14 28 63 68 64 II. Project Characteristics 1. Project Type: University, EPA, USGS and USBR joint effort; 2000 acres. 2. Water resource type: Irrigation drains into Columbia River. 3. Use impairment: No impaired uses identified. Project was intended as demonstration of sediment and nutrient reductions achievable through construction of on-fann facilities. 4. Timeframe: 1977-1982. 5. Water quality at start of project: In 1977, 4,386 mt sediment and 2,867 kg P were lost from 2,000 acres project area. 6. Meteorologic factors: Project area receives only 7.3 in/yr p~eci- pitation; USLE *R' factor is less than 20. 7. Water quality monitoring program: A very intensive program was carried out from 1977 through 1981. Samples were taken automat- ically at 2 hour intervals from the main drain and composited into daily samples. These automated samples were correlated with nu- merous replicated concentration in sample form depth-integrated samples taken with a standard U.S.G.S. DH-48 sampler. Samples were also taken at the point of diversion into each farm. In addition approximately 40* of the individual fields were monitored from 1978 through 1981 for runoff volume and sediment using Parshall flumes.This WQM program provided very precise measurements of sedi- ment, P, and N loading both within the project area and discharging from the project area. 62 ------- 8. BMPs: The principal BMPs were sediment basins, buried pipeline, gated pipe, concrete-lined head ditches and center pivot sprinkler systems. The sediment basins were designed so that they would have a sediment holding capacity of at least one-year loading and they were cleaned out 1-2 times per year. The purpose of the other BMPs was to allow irrigation water to be drained away with less erosion of furrows and ditches. In addition, farmers were encouraged to use automatic or semi-automatic water cutback systems to reduce end-of- furrow discharge. Essentially all land in the project area benefited form the BMPs. A pre-project survey determined that all farmers in the study area were ready to participate. 9. Critical areas: All agricultural land in the study area was tar- geted for treatment. It was noted that certain crops (i.e. sugar beets, potatoes) were especially critical sources of sediment. 10. Incentives: A 70% cost-share rate was established for the project with a maximum cost-share of $125.00 per acre benefited. There was also some tax and production benefit incentive especially for larger operations. 11. Economic information: A total of $70,000.00 was spent on BMPs during the study. This amounted to an average of about $35/acre. There is a continuing cost for maintenance of the sediment basins. The BMPs prevented the loss of approximately 12,600 mt of sediment ($5.56/mt) and 5,000 kg of phosphorus. An economic model was developed out of the project data which showed that tax considerations are very important in motivating BMP adoption. III. Lessons learned from project. 1. Sediment losses from a 2,000 acre furrow and sprinkler irrigated area can be reduced by 80% through the combination of subsurface drainage, improved furrow and drain structures and sediment basins. Total P reductions (50%) were significantly less than sediment reductions because of P enrichment on the finer, less easily captured, sediment fraction. These practices had no observable effect on nitrogen loadings. 2. Use of practices to control stream size in individual furrows can reduce sediment in tailwater. 3. Irrigation scheduling has a significant effect on the seasonal sedi- ment loss from a field. Reducing the number of irrigations can reduce sediment losses without affecting productivity in many cases. 4. The Imhoff cone can be used effectively at the farm level to help the irrigator visualize his soil loss and optimize water management for soil retention. 63 ------- 5. Economic modeling of the project showed that a program of variable incentives that depend upon farm size and debt/equity position would be the most efficient expenditure of funds to induce adoption of BMPs. 6. Farmers should be required to pipe center pivot sprinkler overflows to an acceptable, improved drain, especially for new installation of center pivots. 7. Water quality improvements from furrow irrigated agriculture should focus on reducing on-field erosion as well as trapping eroded sedi- ment. 64 ------- REFERENCES Agena, U., M. Wnuk, and C.M. Lawyer. 1985. Prairie Rose Lake Rural Clean Water Program Project. In: Perspectives on Nonpoint Source Pollution. EPA 440/5-85-001. pp. 259-263. Cassell, E.A., and J.V. Calcan. 1983. East and West, Animal Waste Cleanup Pays. In: Using Our Natural Resources. 1983 Yearbook of Agri- culture, pp. 346-353. Clausen, J.C. 1985. The St. Albans Bay Watershed RCWP: A Case Study of Monitoring and Assessment. In: Perspectives on Nonpoint Source Pollution. EPA 440/5-85-001. pp. 21-24. Davenport, T. and J. Lowrey. 1985. Watershed Water Quality Programs: Lessons Learned in Illinois. In: Perspectives on Nonpoint Source Pollution. EPA 440/5-85-001. pp. 256-258. Hopkins, R.B. and J.C. Clausen. 1985. Land Use Monitoring and Assessment for Nonpoint Source Pollution Control. In: Perspectives on Nonpoint Source Pollution. EPA 440/5-85-001. pp. 25-29. Jackson, F.E. 1985. Protecting Tillamook Bay Shellfish With Point/Nonpoint Source Controls. In: Perspectives on Nonpoint Source Pollution. EPA 440/5-85-001. p. 425. King, L.G., B.L. McNeal, F.A. Ziari, and S.C. Matulich. 1983. On-farra Im- provements to Reduce Sediment and Nutrients in Irrigation Return Flow, Project Report Columbia Basin Project, Washington State University, Pull- man, Washington. Maas, P., S.A. Dressing, J. Spooner, M.D. Smolen, F.J. Humenik. 1984. Best Management Practices for Agricultural Nonpoint Source Control. IV. Pes- ticides. National Water Quality Evaluation Project. Biological and Agricultural Dept., North Carolina State University, Raleigh, NC. National Water Quality Evaluation Project. 1985a. Rural Clean Water Program, Status Report on the CM&E Projects. Biological and Agricultural Engineer- ing Dept., North Carolina State University, Raleigh, NC. National Water Quality Evaluation Project. 1985b. Rural Clean Water Program, Status Report on the CM&E Projects. Supplemental Report: Analysis Meth- ods. Biological and Agricultural Engineering Dept., North Carolina State University, Raleigh, NC. 65 ------- National Water Quality Evaluation Project. 1985c. Rural Clean Water Program, Cross Project Evaluation. Biological and Agricultural Engineering Dept., North Carolina State University, Raleigh, NC. National Water Quality Evaluation Project and Harbridge House Inc. 1983. An Evaluation of the Management and Water Quality Aspects of the Model Implementation Program. Biological and Agricultural Engineering Dept., North Carolina State University, Raleigh, NC. Neubieser, M.J. 1985. Rural Clean Water Program: The Experiment Continues. In: Perspectives on Nonpoint Source Pollution. EPA 440/5-85-001. pp. 391-396. Young, R.A., C.A. Onstad, D.D. Bosch, and W.P. Anderson. 1985. Agricultural Nonpoint Source Pollution Model (AGNPS). Minnesota Pollution Control Agency, St. Paul MN and USDA-Agricultural Research Service, Washington, D.C. Young, R. A., M.A. Otterby, and A. Roos. 1982. An Evaluation System to Rate Feedlot Pollution Potential. USDA-Agricultural Research Service, Peoria, IL. 78 pp. 66 ------- |