Integrating APEX Output for Cultivated Cropland with SWAT Simulation for Regional Modeling

The purpose of the Conservation Effects Assessment Project (CEAP) cropland national assessment is to quantify the environmental benefits of conservation programs at the regional and national levels, which include both onsite and instream water quality benefits. Modeling is an effective tool for environmental assessment at the regional and national scale due to the complexities in nature at this scale. Two simulation models, the Agricultural Policy Environmental eXtender (APEX) and the Soil and Water Assessment Tool (SWAT), were used for the CEAP cropland national assessment. A subset of National Resources Inventory (NRI) sample points was selected to serve as "representative fields" for the CEAP cropland survey to determine conservation practices currently in use. The NRI-CEAP points for cultivated cropland were simulated using APEX. The APEX results were aggregated and combined with modeling results from SWAT for uncultivated land uses. The combined modeling results at 8-digit watershed outlets were routed downstream in SWAT for estimating the offsite effects of conservation practices. The regional modeling involved three major steps: APEX setup for simulating conservation practices, calibration of water yield at the 8-digit watershed level (1961-1990), and development of sediment delivery ratios (SDR) for transporting sediment from cultivated cropland and uncultivated land to the 8-digit watershed outlet before combining model outputs for further routing downstream. The objective of this article is to address the use of APEX to model CEAP cultivated cropland, APEX simulation of conservation practices, and SDR development and to test the combined modeling of flow and sediment yield for the upper Mississippi River basin. Simulated annual and monthly flows at key gauging stations along the river basin and annual sediment yields at Valley City and Alton/Grafton, Illinois, were compared with observed values. Test results are promising for using the combined modeling systems for large-scale studies and for performing scenario analyses to evaluate conservation practices. The limitations of this national modeling include the uncertainties associated with data (channel dimensions, conditions of structural practices) and the inability to account for streambank and channel erosion at the 8-digit watershed level. Therefore, while using absolute predictions for individual scenarios, consideration should be given to various uncertainties. Both data updating and the development of an enhanced SWAT model with streambank/channel erosion components and particle size distribution are underway for reducing the level of uncertainty in future national-scale modeling efforts.