Sensitivity Analysis of Soil Parameters in Crop Model Supported with High-Throughput Computing.

Uncertainty of input parameters in crop models and high costs of their experimental evaluation provide an exciting opportunity for sensitivity analysis, which allows identifying the most significant parameters for different crops. In this research, we perform a sensitivity analysis of soil parameters which play an essential role in plant growth for the MONICA agro-ecosystem model. We utilize Sobol' sensitivity indices to estimate the importance of main soil parameters for several crop cultures (soybeans, sugar beet and spring barley). High-throughput computing allows us to speed up the computations by more than thirty times and increase the number of sampling points significantly. We identify soil indicators that play an essential role in crop yield productivity and show that their influence is the highest in the topsoil layer.