Advancements in soil moisture estimation through integration of remote sensing and artificial intelligence techniques.
Soil moisture plays a vital role in crop productivity, ecosystem function, climate feedback, and groundwater recharge, yet it remains one of the least directly measured hydrological and hydrogeological variables. Its limitations in spatial coverage, depth sensitivity, and real-time monitoring hinder the soil moisture estimation methods. This review presents a comprehensive overview of recent advancements in soil moisture estimation techniques, covering traditional in-situ, remote sensing, and AI-based methods. A comparative evaluation highlights these methods' operational mechanisms, spatial limitations, and potential synergies with region-specific insights. It also summarizes approaches such as cosmic-ray neutron sensing, Global Navigation Satellite System (GNSS) reflectometry, Synthetic Aperture Radar (SAR) based methods, and multi-sensor data fusion for large-scale surface soil moisture estimation. The present study highlights the key challenges, such as physical interpretability, transparency, and model transferability in AI-based models due to their black-box nature. Incorporation of remote sensing with explainable AI further enhances the performance, trustworthiness, and reliability of the model. In addition, integrating multi-sensor data with in-situ data, machine/deep learning, and physical models enhances the accuracy, spatial and temporal resolution, continuous data availability, and root zone soil moisture estimation. Further research is required in uncertainty handling, evaluating, validating, and improving the transferability of the model across diverse regions. Through citizen science and crowdsourcing, real-time soil moisture monitoring in the data-scarce areas can be achieved. The study concludes by outlining future directions for scalable, interpretable monitoring systems and developing robust systems to support sustainable water resource management and climate-resilient agriculture.
Paper
Full text
Advancements in soil moisture estimation through integration of remote sensing and artificial intelligence techniques.
Semantic Scholar · Environmental Science · 2025
Abstract
Soil moisture plays a vital role in crop productivity, ecosystem function, climate feedback, and groundwater recharge, yet it remains one of the least directly measured hydrological and hydrogeological variables. Its limitations in spatial coverage, depth sensitivity, and real-time monitoring hinder the soil moisture estimation methods. This review presents a comprehensive overview of recent advancements in soil moisture estimation techniques, covering traditional in-situ, remote sensing, and AI-based methods. A comparative evaluation highlights these methods' operational mechanisms, spatial limitations, and potential synergies with region-specific insights. It also summarizes approaches such as cosmic-ray neutron sensing, Global Navigation Satellite System (GNSS) reflectometry, Synthetic Aperture Radar (SAR) based methods, and multi-sensor data fusion for large-scale surface soil moisture estimation. The present study highlights the key challenges, such as physical interpretability, transparency, and model transferability in AI-based models due to their black-box nature. Incorporation of remote sensing with explainable AI further enhances the performance, trustworthiness, and reliability of the model. In addition, integrating multi-sensor data with in-situ data, machine/deep learning, and physical models enhances the accuracy, spatial and temporal resolution, continuous data availability, and root zone soil moisture estimation. Further research is required in uncertainty handling, evaluating, validating, and improving the transferability of the model across diverse regions. Through citizen science and crowdsourcing, real-time soil moisture monitoring in the data-scarce areas can be achieved. The study concludes by outlining future directions for scalable, interpretable monitoring systems and developing robust systems to support sustainable water resource management and climate-resilient agriculture.