Using a Process Based Model to Simulate the Effects of Drainage and Land Use Change on Hydrology, and Sediment and Nutrient Transport in the Midwestern United States

Abstract:

Historically the Midwestern United State was a region dominated by prairie grasses and wetlands. To make use of the rich soils underlying these fertile environments, farmers converted the land to agriculture and currently the Midwest is a region of intensive agricultural production, with agriculture being a predominant land use. The Midwest is a region of gentle slopes, tight soils, and high water tables, and in order to make the lands suitable for agriculture, farmers have installed extensive networks of ditches to drain off excess surface water and subsurface tiles to lower the water table and remove excess soil water in the root zone that can stress common row crops, such as corn and soybeans. The combination of tiles, ditches, and intensive agricultural land practices radically alters the landscape and hydrology. As part of the Minnesota River Basin Integrated Study we are simulating nested watersheds in a sub-basin of the Minnesota River Basin, Seven Mile Creek, using the physics-based watershed model GSSHA (Gridded Surface Subsurface Hydrologic Analysis) to simulate water, sediment, and nutrients. Representative of the larger basin, more than 80% of the land in the watershed is dedicated to agricultural practices. From a process perspective, the hydrology is complicated, with snow accumulation and melt, frozen soil, and tile drains all being important processes within the watershed. In this study we attempt to explicitly simulate these processes, including the tile drains, which are simulated as a network of subsurface pipes that collect water from the local water table. Within the watershed, tiles discharge to both the ditch/stream network as well as overland locations, where the tile discharge appears to initiate gullies and exacerbate overland erosion. Testing of the methods on smaller basins demonstrates the ability of the model to simulate measured tile flow. At the larger scale, the model demonstrates ability to simulate flow and sediments. Sparse nutrient data limit the assessment of nutrient simulations. The models are being used to asses an array of potential future land use scenarios, including predevelopment and increased agricultural use. Results from these simulations will be presented. Preliminary results indicate that tile drains increase discharge and erosion in the watershed.