Integrating Satellite Images with Agriculture Modeling for Estimating Field Scale Crop Evapotranspiration and Irrigation Water Demand

This research integrates high-resolution satellite imagery with a simplified process-based agricultural model (CropSyst-W) to estimate crop evapotranspiration (ET), biomass gain, and irrigation demand at both field and pixel scales. The approach addresses limitations of standalone satellite-based ET products, such as METRIC and EEFlux, and traditional process-based models that require intensive input data.

Using Landsat-derived NDVI to estimate canopy cover and crop growth stages, the model was applied to the Columbia Basin Project in Washington state, where approximately 671,000 irrigated acres rely on Columbia River water. Four major crops—alfalfa hay, corn grain, potato, and wheat—were analyzed using climate (gridMET), soil (gSSURGO), and land use data from the Washington State Department of Agriculture.

Results show that CropSyst-W successfully captured spatial and temporal ET variability and reduced overestimation observed in EEFlux late in the growing season. The integrated framework also provides separate estimates of transpiration, soil evaporation, and irrigation demand, offering detailed information to support precision irrigation management.

This work was supported by the USDA National Institute of Food and Agriculture, project #1016467.