BestiaPop - A Python package to automatically generate and visualise gridded climate data for crop model applications (oral presentation)

Abstract

Climate data is an essential input for crop models to predict crop growth and development using site-specific (point) or gridded climate data. While point data is usually available in a readily format, gridded data is stored in NetCDF files which are difficult to archived and convert to an input file readable by the Agricultural Production Systems sIMulator (APSIM) or other crop models such as the Decision Support System for Agrotechnology Transfer (DSSAT). We developed BestiaPop, a Python package which allows model users to automatically download gridded climate data in an APSIM and DSSAT format from an Australian (Scientific Information for Land Owners; SILO) and global (NASA- Prediction Of Worldwide Energy Resource; NASA-POWER) climate data source. The package offers the possibility to select a range of grids (0.05° resolution) and years producing files with daily climate data in different formats (CSV, MET, WTH). We (i) compared the Bestiapop performance to download CSV, MET and WTH files using multiprocessing and (ii) tested the performance of the package to generates climate data across areas suitable for potato (Solanum tuberosum L.) in Tasmania, Australia. A total of 1724 climate files across 20 years (1991-2020) were automatically downloaded and the spatio-temporal variability of climate inputs was mapped. The case study reveals that implementing BestiaPop is a useful and efficient tool to automatically download gridded climate data in an APSIM format and could be extended to other crop models and regions across the world.