Calibration of Crop Growth Simulation Model for Irrigated Cropland Using Remote Sensing Derived Leaf Area Index

Optimizing the irrigation scheduling is an important decision task in irrigated agricultural areas. The performance of different irrigation scheduling strategies may be simulated and modeled using a crop growth simulation model. To realistically simulate crop growth and the impact of alternative irrigating schedules, the crop growth model needs to be calibrated for specific crop, soil, and climate zone. In this study, the World Food Studies (WOFOST) model is selected as the crop growth simulation model. One proven straightforward approach to calibrate WOFOST model is to use remote-sensing-derived leaf area index (LAI) as a cross-validation indicator since LAI is one of the outputs. Optimization approach can be used to find the optimal initial parameters on adjustable parameters of WOFOST. Locally calibrated models are then used to simulate the crop growth at approximate location for evaluating alternative irrigation scheduling. Two LAI products are used: one is MODIS MCD15A3H 4-day composite and another is for selected area from Landsat using a simple model. They represent two different spatial resolutions. The experiment area is in Nebraska. Crops are corn and soybean. Results are evaluated with selected ground truth and at aggregated level against statistic report. The crop yield estimate shows improved accuracy using the calibration.