Geographically weighted regression random forest for modeling soil particles

Clay particles play a vital role in determining soil quality, particularly in the fields of agriculture and conservation. However, the complex and non-linear spatial distribution of clay particles is difficult to capture using conventional modeling methods. This study aims to develop a hybrid model, Geographically Weighted Regression Random Forest (GWRRF), which combines the ability of Geographically Weighted Regression (GWR) to capture spatial heterogeneity and the strength of Random Forest (RF) in handling non-linear relationships. The data used in this study were derived from soil texture and local morphologic analysis across 50 observation points in the Kalikonto watershed. The results show that the GWRRF model provides more accurate predictions of clay particle content than the GWR model, with an R² value of 0.735 and a lower RMSE. In contrast, the GWR model achieved an R² value of 0.575. The integration of both methods in GWRRF offers a novel, adaptive, and context-aware approach to understanding the distribution of clay particles, contributing to more precise and sustainable data-driven land management.