Application of DSSAT Model to Simulate Corn Yield under Long-Term Tillage and Residue Practices

Abstract

Long-term (1991-2013) field experiments were conducted to study the effects of different tillage and residue management practices on grain corn production in eastern Canada. Three different tillage practices, namely conventional tillage (CT), reduced tillage (RT), and no tillage (NT), along with two residue management practices, namely with residue (R) and without residue (NR), were considered. Field measurements of grain yield, biomass, soil organic carbon (SOC), soil moisture, and mineral nitrogen were used to evaluate the performance of the DSSAT model and to understand the impacts of long-term tillage and residue management practices. The observed corn yield, biomass, LAI, and nitrate-nitrogen (NO3-N) were not found to be significantly different under the various tillage and residue practices. The observed soil moisture showed a significant difference in the upper 10 cm soil layer. The SOC pool at 0-20 cm depth was reduced by 9.5% for CT and increased by 17.3% and 7.6% for RT and NT treatments, respectively. The DSSAT model was able to simulate corn and biomass yield, LAI, soil moisture, and SOC with RMSE ranging from 2% to 31%, indicating reasonable model performance. The model did not provide accurate results for NO3-N and soil moisture simulations, with RMSE ranging from 34% to 78%. Further improvements in the model are needed to better simulate soil moisture and N dynamics under different tillage and residue practices.