Using DAYCENT to quantify impacts of land use conversion to nitrogen-managed switchgrass in the southern U.S.
Use of a simulation model to predict long-term yield, greenhouse gas (GHG) emissions, and water quality impacts can be valuable for assessing land use conversion to bioenergy crops. The objective of this study is to assess the usability of DAYCENT for measuring environmental impacts due to land conversions from cotton and CRP lands (as unmanaged grasses) to switchgrass in the Southern U.S. We use published yield data to calibrate the crop growth parameters and test the calibrated model on independent data sets. We then apply the model to predict other relevant C and N parameters. In the case of cotton, the model simulates long-term mean cotton lint yield within 25% of observed yields across the South and within 4% of yields in the case study area of Darlington County, SC. DAYCENT also matches observed mature switchgrass yields within 25% of the mean in the range of expected fertilization rates across the region and within 6% in the case study area. Long-term simulations predict a decrease in GHG emissions (1.0–3.8 MtCO2-e/ha-yr) and a reduction of nitrate runoff (up to 95%) for conversions from cotton to switchgrass at N application rates of 0–135 kgN/ha. Conversely, conversion from unmanaged grasses to switchgrass resulted in annual increases of net GHG emissions (0.2–1.4 MtCO2-e/ha-yr) for switchgrass at no and low (45 kgN/ha) fertilization rates. Sequestration occurs due to increased soil organic C when higher levels of N are applied. At all levels of fertilization, a reduction of nitrate (50–70%) occurs when converting from unmanaged, unharvested grasses. The amount of nitrate leaching is only slightly sensitive to the fertilization rate applied to the perennial switchgrass. DAYCENT sufficiently models the “carbon debt” from land use conversion from CRP grasslands to managed switchgrass and highlights the importance of fertilization rate. Both C and N parameter results fall within published observed ranges. Thus, the long-term (10–15-year) accuracy of the model for both cotton and switchgrass offers promise as a tool for analyzing land use conversions in terms of N-managed yields and subsequent environmental impacts and benefits.