Satellite-Based Estimation of Soil Moisture Content in Croplands: A Case Study in Golestan Province, North of Iran

Soil moisture content (SMC) plays a critical role in soil science via its influences on agriculture, water resources management, and climate conditions. There is broad interest finding relationships between groundwater recharge, characteristics, plant properties for the quantification of SMC. The objective this study was to assess potential optical satellite imagery estimating SMC over cropland areas. For purpose, we collected 394 samples as targets Gonbad-e Kavus Golestan province north Iran, where variety crop types are cultivated. As input data, first computed several spectral indices from Sentinel 2 (S2) Landsat 8 (L8) images, such Normalized Difference Water Index (NDWI), Modified (MNDWI), Salinity (NDSI), then analyzed their with surveyed using four machine learning regression algorithms: random forests (RFs), XGBoost, extra tree decision (EDT), support vector (SVM). Results revealed high rather similar correlation measured values both S2 L8 data. EDT algorithm yielded highest accuracy, an R2 = 0.82, MAE 3.74, RMSE 1.08 0.88, 2.42, images. also that MNDWI, NDWI, NDSI responded most sensitively estimation.