Simulating crop rotations to improve estimates of nutrient losses using APSIM in HPC

Abstract

: In New Zealand, policy makers and land stewards are under pressure to change farming practices and to limit their environmental impact. Having information on such impacts is crucial for guiding policy setting and to help making decisions regarding land use change and improving management. We employed the latest advances in biophysical parameters databases, simulation modelling, and computing automation tools to develop simulations of crop rotations for all growing regions in New Zealand. The aim is to provide improved estimates of nutrient losses from cropped farmlands with a wide coverage across the country. We present here the methodology and developments made to streamline the supply of parameters and inputs to the model, the automatic creation of simulations of a variety of cropping systems, and the running of thousands of simulations in a high-performance computing (HPC) environment. These simulations were set up using the Agricultural Production Systems Simulator (APSIM). The developments include: