Soil Moisture Retrieval in Irrigated Winter Wheat Fields Using Sentinel Data and Machine Learning Algorithms

Soil moisture is crucial for water resource management, irrigation planning and agricultural productivity. This study evaluated the performance of five machine learning algorithms: boosting, random forest, support vector machine, K‐nearest neighbour and decision tree, for estimating soil moisture using radar backscattering coefficients from Sentinel‐1 imagery. Measurements were collected at 48 georeferenced points in the Qorveh–Dehgolan Plain during different growth stages of winter wheat on 2 April, 14 April, 8 May and 13 June 2022, coinciding with Sentinel‐1 overpasses. Algorithm performance was assessed for both individual dates and combined datasets using mean absolute error (MAE) and Pearson's correlation coefficient ( r ). The results showed that boosting and random forest outperformed the other algorithms, achieving correlation values up to 0.89 and MAPE below 13%, whereas accuracy decreased with increasing vegetation cover and crop growth stages. Overall, integrating machine learning algorithms with satellite imagery provides an effective approach for soil moisture estimation and water resource management.