摘要：AbstractA variance based sensitivity analysis (SA) requires a large number of model evaluations. In order to perform a variance based SA on a process-based whole-year grassland model for timothy, we arranged the 62 model parameters in 15 groups and used the less computationally expensive Morris method to identify which groups contained the most important parameters. We used observations and climate data from a northern and a southern location. The Morris method identified seven important groups for the southern location, and five groups at the northern which was selected for the variance based SA. At the northern location the most important parameters were related to frost tolerance, and at the southern location assimilate allocation and foliage parameters were most important.
