Submitted to: International Journal of Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: May 16, 2007
Publication Date: March 18, 2008
Citation: Fang, H., Liang, S., Hoogenboom, G., Teasdale, J.R., Cavigelli, M.A. 2008. Corn-yield estimation through assimilation of remotely sensed data into the csm-ceres-maize model. International Journal of Remote Sensing. 29(10):3011-3032.
Interpretive Summary: Crop yields are influenced by climate, genetics, crop management, and the physical and chemical properties of soils, and may vary considerably both spatially and temporally. Accurate,objective, reliable and timely predictions of crop yields over large areas are critical for national food security through policy making on import/export plans and prices. Remote sensing data have been widely used for crop yield estimation. This method is eventually a statistical model and cannot predict the time-dependent processes of growth and field formation that are critical for real-time yield forecasting. The most promising method is the combination of
remote sensing and crop growth modeling. In this study, we have developed a procedure using the data assimilation method to estimate crop yield at the regional scale from remote sensing MODIS
(Medium resolution imaging spectroradiometer) data. The crop models that encompass the Decision Support System for Agrotechnology Transfer (DSSAT) and a copy radiative transfer model are coupled, and the MODIS EVI data at 250m resolution and LAI product at 1 km resolution were assimilated into the coupled model to adjust the key parameters in the crop growth model. The
results compared favorably with the USDA/NASS agricultural statistics data at the county level and are very encouraging. This approach will be useful to scientists and policy makers interested in estimating crop yields at a regional scale.
One of the applications of crop simulation models is to estimate crop yield during the current growing season. Several studies have tried to integrate crop simulation models with remotely sensed data through data assimilation methods. This approach has the advantage of allowing reinitialization of model parameters with remotely sensed observations to improve model performance. In this study, a commonly used crop growth model, CERES in DSSAT (Decision Support System for Agrotechnology Transfer) was integrated with the MODIS (Moderate Resolution Imaging Spectroradiometer) leaf area index (LAI) data products for estimating corn yield in the state of Indiana. This procedure, inversion of crop simulation model (ICSM), facilitates several different user input modes and outputs a series of agronomic and biophysical parameters, including crop yield. The estimated corn yield in 2000 compared reasonably well with USDA NASS statistics for most counties. Planting, emergence and maturation dates, and N fertilizer application rates were also estimated. Further studies will include investigating the model uncertainties and using other MODIS products, such as the enhanced vegetation index (EVI).