 This study aimed to create an SLM-level map of Batafa region in northern Iraq using random forest and extreme gradient boosting models with remote sensing data, supplementary variables, and machine learning algorithms. The XGBoost model exhibited higher accuracy than the RF model, and Band 10, DEM, SAVI, and NDVI were identified as the most important predictors for both models. The methodology has the potential to map a SLM in similar settings and offer significant insights for soil management stakeholders. This article was authored by Helmat S. Kalef, Yasim T. Mustafa and Mohamed A. Fayat.