Safir,G.R., M. Colunga-Garcia, S.H. Gage, and Y. Johnson.
Presented at the All Scientist Meeting (2002-10-04 )
We used a Modeling Applications System Integrative Framework to simulate corn yield in the Upper Midwest Region from 1972 to 1990. The model used in this study, which was developed by Muchow and Sinclair, simulates daily maize growth and productivity based on daily weather inputs. Although this model is site specific it has regional application potential and can be adapted for its use in cold climates. We included in the model a variable to account for water availability an algorithm for cold temperature correction. Model outputs were compared with actual yields based on USDA-NASS databases. The application of the temperature correction increased the simulated corn yields with the coldest spring years having the greatest increases. The inclusion of the water availability information resulted in much smaller changes. Large simulated increases in corn yields (30-40%) occurred under water saturation conditions. In 1988 (drought year) the simulation model predicted corn yields more accurately than in 1990. One reason for the higher correlation between model predictions and actual yields in 1988 may be that under conditions of high yield and high water availability, stresses other than lack of water such as disease or insect will play a larger role in reducing simulated corn yields.
Back to meeting | Show |