Upland soil and streambank erosion, and sediment deposition are also critical water quantity and quality issues in Illinois. Erosion causes loss of fertile soil, streambank erosion causes loss of valuable lands, and both contribute large quantities of sediment in water flowing through streams and rivers that cause turbidity in sensitive biological resource areas and fill streambeds and banks, lakes, and reservoirs. Lake Decatur, Lake Springfield, and Peoria Lake are a few of the examples. Eroded soil and sediment also carry chemicals that pollute water bodies and stream/reservoir beds.
Runoff water from farm fields, collected in creeks or streams through tile drains or small ditches, contains significantly high concentrations of sediment and agricultural chemicals that pollute receiving water bodies during early stages of planting (late spring or early summer). Agricultural chemicals include chemicals applied through fertilizer, herbicides, and pesticides, and chemicals produced naturally most importantly atmospheric deposition, fixation, and mineralization. Understanding and dealing with these complex hydrologic, soil erosion, and sediment and contaminant transport processes and the associated problems have been quite a challenge for scientists and engineers. Mathematical models are becoming invaluable tools to analyze these complex processes and to evaluate land use and best management practices (BMPs) in reducing the damaging effects of flooding, soil erosion, sedimentation, and contamination on drinking water supplies and other valuable water resources. Existing and commonly used models are limited because they are not physically based and cannot simulate the dynamic behaviors of the water and its constituents' movements.
In a 1999 report, New Strategies for America's Watersheds, published by the National Research Council, the Committee on Watershed Management analyzed the current status of watershed modeling for decision making. The Committee concluded that the available models and methods are outdated, and "a major modeling effort is needed to develop and implement state-of-the-art models for watershed evaluation." Existing physically based models are computational and data intensive, and too cumbersome for practical use in large watersheds. Therefore, new physically based and efficient models must be developed to simulate the spatially and temporarily varying physical and chemical processes in watersheds ranging in size from farm fields to river basins.
An effort is underway at the Illinois State Water Survey (ISWS) in which a dynamic watershed simulation model (DWSM) is being developed using physically based governing equations to simulate propagation of flood waves, entrainment and transport of sediment, and all agricultural chemicals commonly used in agricultural and rural watersheds. The model has three major components: hydrology, soil erosion and sediment transport, and nutrient and pesticide transport. These model components, adopted from earlier work of the lead author, have efficient routing schemes based on approximate analytical solutions of the physically based governing equations, and preserving the dynamic behaviors of the water, sediment, and accompanying chemical movements. These and other model formulations and procedures are described in this report. The nutrient and pesticide component is limited to transport of these chemicals with surface runoff and sediment considering adsorption and desorption and based on initial chemical concentrations dissolved with pore water and adsorbed with soil particles on the ground surface and in the near surface soil matrix. Reactions and transformations of the chemicals are not simulated.
During this two-year study, the first two components of the DWSM were tested on subwatersheds of the 925-square-mile Upper Sangamon River basin in east-central Illinois that drains into Lake Decatur, using data previously collected by the ISWS and intensive storm data collected during this study. The ISWS established an extensive monitoring network in this watershed, from which it has been collecting streamflow and nitrogen data since 1993. Monitoring was conducted in response to a legal commitment by the city of Decatur to the Illinois Environmental Protection Agency (IEPA) to reduce concentrations of nitrate-N in the lake to levels below the drinking water standard by the year 2001.
The hydrology component of the DWSM was tested on the entire Upper Sangamon River basin using data collected earlier by the ISWS. The model performed well in predicting peak flows and time to peak flows in the five monitored subwatersheds ranging in size from 38 to 112 square miles. However, the model underpredicted the recession and base flow portions of the hydrographs, as well as the runoff volumes from the larger subwatersheds due to the lack of tile drain and base flow simulation capabilities.
Detailed flow and concentrations of suspended sediment, nitrate-N, phosphate-phosphorous (phosphate-P), atrazine, and metolachlor were collected during 1998 spring storm events at the Big Ditch station, draining a 38-square-mile subwatershed of the Lake Decatur watershed. During 1999, the same types of data, except metolachlor, were collected at Big Ditch and two other stations on the main stem of the Sangamon River: Fisher and Mahomet draining, respectively, 240 and 360 square miles of the Upper Sangamon River or Lake Decatur watershed. Rainfall data were collected from newly established raingages: one at Big Ditch during 1998 and five others throughout the Upper Sangamon River watershed above Mahomet during 1999. Rainfall data from the six stations varied noticeably from station to station, especially between the three eastern and three western stations.
In the above monitored data, all the constituents closely followed the flow hydrograph except nitrate-N. The nitrate-N concentration varied inversely with water discharge, decreasing drastically during rising and peak flows and increasing with the recession and base flow portions of the hydrographs. Such variations may be due to runoff pathways. For example, runoff through subsurface soil and the tile drain contains more nitrate-N than runoff over the ground surface (surface runoff). Therefore, peak flows contributed primarily by surface runoff contain less nitrate-N concentrations than water in the recession and base flows contributed primarily by the tile drain and subsurface flows...