Parameter Estimation for Crop Models

摘要: The adjustment of the parameters in mechanistic crop models to field data, using an automatic procedure, is essential ensure efficient and objective use measured data. However, it general numerically impossible, any case undoubtedly unwise, adjust all model There currently no widely accepted solution this problem. This paper proposes a new approach parameter adjustment, applies corn growth development. One begins by defining criterion goodness-of-fit, which should be adapted goal modeling exercise, corresponding prediction error. For latter we propose cross validation version goodness-of-fit criterion. In Step 1 algorithm, one orders according how much each improves model. second step, number actually adjusted chosen minimize error has advantage explicitly quality as As by-product, leads adjusting relatively few (in our example, 3 out 26 potentially adjustable parameters), considerably reduces numerical problems. procedure quite straightforward apply, although does require substantial computing time.