MOSAIC: Crop yield prediction - compiling several years' soil and remote sensing information

Predicting spatial crop yield based on underlying field processes remains an enigma in agricultural management. The aim of this study was to evaluate the usefulness of a normalized difference vegetation index (NDVI) derived from remote sensing for spatial crop yield prediction. The prediction was achieved using an autoregressive state model based on normalized grain yield and NDVI data. Comparison of several years' results should indicate how stable derived sets of state-space coefficients were, and if they could be applied for predictions. Transition coefficients were estimated for four consecutive years (1997-2000) with different crops growing on the same field site. Except for the last year, coefficients were relatively stable in time. A common model was used for all years, and prediction accuracy of yields was approximately 1 t ha(-1). This result indicates the promising applicability of NDVI observations in combination with state-space models for crop yield prediction.