Identification of problem grids within a wheat and corn field by the implementation of a process-oriented precision farming crop growth model

Abstract

The objective of this study was to use the process-oriented precision farming crop growth model APOLLO to identify problem grids for crop production within a field. The model was calibrated for a wheat and a corn field. During the calibration process soil parameters were adjusted iteratively for each grid within the fields, using a simulated annealing algorithm. This procedure gave good yield estimates for most of the grids within the fields. However, simulated yields in some grids of the field remained as outliers and thus did not fit the 1:1- line of the model calibration. The data points were categorized based on the standard deviation between measured and simulated yields to identify outliers. Thus the model calibration provided information about soil parameters causing yield variability in some grids and identified grids in the field with unknown reasons for yield variability. Based on these results crop models can be used to identify grids in the field where the variation in yield is not caused by soil properties or other processes incorporated into the crop model.