CropSuiteLite: An Optimized Implementation of the CropSuite Crop Suitability Framework

Aguilar Ariza, Andres; Steward, Peter Richard; Youngberg, Brayden; Rosenstock, Todd Stuart; Zabel, Florian; Knuttel, Matthias

CropSuiteLite: An Optimized Implementation of the CropSuite Crop Suitability Framework

Autores

Aguilar Ariza, Andres

Steward, Peter Richard

Youngberg, Brayden

Rosenstock, Todd Stuart

Zabel, Florian

Knuttel, Matthias

Descripción

CropSuiteLite is a lightweight, optimized Python implementation of the crop suitability modeling framework presented in the CropSuite paper by Zabel et al. (2025) https://gmd.copernicus.org/articles/18/1067/2025/. It is designed to assess how suitable a geographic area is for growing different crops by analyzing climate and soil conditions. This version focuses on performance, scalability, and flexibility, incorporating modern data science libraries and parallel processing to handle large datasets and complex scenarios efficiently. Disclaimer: This repository is an independent implementation based on the concepts, workflow, and structure described in Zabel et al. (2025), “CropSuite: An open modular framework for site-specific crop modeling at global scale.” The official CropSuite framework is available from the CropSuite Zenodo repository https://zenodo.org/records/16759895. New implementations that were included into CropSuiteLite: Parallel Processing: Utilizes parallel processing (concurrent.futures) to accelerate computationally intensive steps such as climate data downscaling and daily climate suitability analysis. Scalable with Xarray and Dask: Offers an alternative processing backend using xarray and dask, enabling out-of-core computation that can handle datasets larger than available RAM. Only available for non-winter crops. Memory management tiling: Offers a manual option to divide large geographic areas into tiles for processing. Original version divides the tiles depending on the RAM size. This was done to make it feasible to run high-resolution analysis on standard hardware. Flexible configuration: Uses YAML files for run configuration, allowing for easy management of different scenarios and climate inputs. Customizable time step: Allows the user to adjust the time step used in climate analysis. In the original version, climate suitability is computed daily; in this implementation, the time step can be customized (e.g. dekads). Detailed user documentation can be found here: https://adaptationatlas.github.io/CropSuiteLite/