Submitted to: North American Agroforestry Conference
Publication Type: Proceedings
Publication Acceptance Date: 5/18/2013
Publication Date: 6/19/2013
Citation: Senaviratne, A., Udawatta, R.P., Baffaut, C., Anderson, S.H. 2013. APEX simulation: environmental benefits of agroforestry and grass buffers on corn-soybean watersheds. North American Agroforestry Conference. Paper No. 92-18. Interpretive Summary: Scientists and soil and water managers need computer simulation tools such as the Agricultural Policy Environmental Extender (APEX) model to assess sediment and nutrient transport from agricultural fields and the effects of best management practices. In this study, the model was used to simulate the benefits of upland grass and agroforestry buffers in three adjacent no-till corn/soybean field-size watersheds at the Greenley Memorial Research Center, Knox County, Missouri. The model was first used to simulate crop yields, runoff, and sediment in these fields and compare results to 1998-2008 data. Simulated crop yields and runoff were well predicted for the three watersheds but the models could not be calibrated for sediment losses. The buffers reduced average annual runoff by 4 to 5 percent. Increase of buffer widths to 5.5 m and 7.5 m were not effective. The agroforestry buffers located on the steepest part of the landscape were the most effective but this trend was not seen with the grass buffers. The study provides guidance on how to simulate grass and agroforestry buffers with the APEX model. The results contribute to the validation of APEX across the United States and will be useful to scientists and natural resources managers in need of applying and evaluating this tool, and assessing the effectiveness of upland buffers.
Technical Abstract: The Agricultural Policy Environmental Extender (APEX) model has the ability to simulate the effects of vegetative filter strips on runoff and pollutant loadings from agricultural watersheds. The objectives of this study were to calibrate and validate the APEX model for three adjacent watersheds and determine optimum buffer dimensions and placement locations. ArcAPEX and APEX0604 versions were used for the simulations. The simulated corn and soybean yields were within ±13 percent and ±27 percent of the measured yields, respectively. The agroforestry, grass buffer, and control watershed models were calibrated (1998 to 2001) and validated (2002 to 2008) for event-based runoff with r2 and Nash-Sutcliffe Coefficients (NSC) values of 0.7-0.8 and 0.4-0.8, respectively. The models could not be calibrated for sediment losses. The simulated grass and agroforestry buffers reduced average annual runoff by 5.2 percent and 4.3 percent, respectively. Increase of buffer widths to 5.5 m and 7.5 m were not effective. The buffers located on the backslopes were the most effective for the agroforestry watershed but this trend was not seen in the grass buffer watershed. The study provides guidance on how to parameterize APEX to simulate grass and agroforestry buffers. It contributes to the validation of APEX and will be useful to scientists in need of parameterizing the model for watersheds that include upland buffers.