Formulating cultivation potential index under uncertainty for wheat and canola by incorporating remotely sensed and big data

Abstract

Identifying suitable areas for crop cultivation is crucial for sustainable agriculture and the protection of soil and water resources. This study presents a framework under uncertainty to calculate the Cultivation Potential Index (CPI) for wheat and canola and decreases the inherent subjectivity in the utilized two sets of data layers. The first set includes climatic variables—precipitation and temperature during the growing, flowering, and ripening periods—sourced from CHIRPS and ERA5 satellites. The second set consists of non-climatic variables, such as slope, soil texture, pH, and organic carbon, derived from big data provided by Open Land Map. The framework employs fuzzy C-means (FCM) to calculate the CPI, which is also computed using the FAO standard framework. Results identify classes S1 and S2 representing high-potential areas for wheat and canola cultivation in the Araz basin, northwest Iran, where the accumulative areas swept by these classes occupy 29 and 27 percent of the study area for wheat and canola, respectively. While the CPI based on FCM does not replace the FAO framework, it offers valuable insights by reducing subjectivity and capturing spatial uncertainty. The findings are essential for policymakers and farmers, highlighting areas where crops currently thrive and where changes in cultivation practices could lead to significant improvements.