Evolution of soil moisture mapping from statistical models to integrated mechanistic and geoscience-aware approaches

High-resolution soil moisture mapping plays a crucial role in advancing the understanding of hydrological processes, optimizing water resource management, and enhancing drought monitoring. Various soil moisture mapping approaches have been developed based on diverse observation and modeling strategies. However, existing reviews primarily focus on remote sensing-based downscaling methods, with limited attention to integrated modeling strategies that combine data-driven and knowledge-driven approaches. Moreover, a systematic induction and comparative analysis of them within the context of advances in observation techniques and modeling paradigms remain lacking. This study provides a comprehensive review of the evolution of high-resolution soil moisture mapping by establishing a systematic classification framework. This framework highlights the transition of soil moisture mapping approaches from mechanistic or statistical modeling to integrated approaches that combine both. The proposed classification categorizes them into three types: (1) mechanistic approaches, (2) statistical approaches, and (3) mechanistic-statistical integrated approaches. Furthermore, this study explores key future directions for high-resolution soil moisture mapping, including optimizing cross-scale relationships and scale effects, balancing physical constraints and computational efficiency, and advancing toward intelligent and geoscience-aware mapping. The development of cognitive geoscience-based intelligent mapping methods and the establishment of a “statistical–mechanistic–intelligent” scaling paradigm are expected to provide more accurate, efficient, and intelligent solutions for soil moisture mapping. • We reviewed the evolution of soil moisture mapping approaches. • A proposed framework evaluates and compares their advantages and limitations. • Future research should follow the “statistics-mechanism-intelligence” paradigm.