Soil moisture prediction using remote sensing and machine learning algorithms: A review on progress, challenges and opportunities

Authors: LAMICHHANE, MANOJ - South Dakota State University; MEHAN, SUSHANT - South Dakota State University; Mankin, Kyle

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/8/2025
Publication Date: 7/11/2025
Citation: Lamichhane, M., Mehan, S., Mankin, K.R. 2025. Soil moisture prediction using remote sensing and machine learning algorithms: A review on progress, challenges and opportunities. Remote Sensing. 17(17). Article e2397. https://doi.org/10.3390/rs17142397.
DOI: https://doi.org/10.3390/rs17142397

Interpretive Summary: Soil moisture drives crop production in semi-arid areas like eastern Colorado. But soil moisture varies dramatically across the landscape and over time. And accurate data are expensive and time-consuming to collect, and difficult to come by. In this paper, we explore and assess many recent advances in using satellite remote-sensing and machine-learning data processing to estimate soil moisture. We summarize strengths and weaknesses of the various methods, highlight promising methods, and recommend directions for future study, including improving spatial resolution.

Technical Abstract: Accurate soil moisture estimation is crucial for a better understanding of agricultural water management and hydrological and ecological processes, especially in the era of extreme weather conditions, including droughts and floods. Although in-situ soil moisture measurement provides accurate data, it could be expensive and time-consuming and not capture the ground heterogeneity. In recent decades, advancements in remote sensing have provided great opportunities to estimate soil moisture at spatial scales. There have been various techniques, from empirical methods to complex machine learning algorithms, to estimate soil moisture using satellite outputs. Empirical and physically based models require extensive ground parameters to predict soil moisture with high accuracy. However, in recent times, machine learning has gained popularity in capturing the complex nonlinearity between the predictive variables (including reflectance, brightness temperature, and backscatter coefficients along with topographic, soil, and weather variables) and the response variable (soil moisture). While the previous studies focused on review of soil moisture prediction based on empirical and physical-based models, this paper offers a comprehensive review of the progress, challenges, and opportunities provided by remote sensing and machine learning models for the estimation of soil moisture and future outlook. In this study, we analyzed the results from 121 publications published from 2010 to 2023. This comprehensive review not only highlights the most effective algorithms and data sources for estimation of soil moisture but also offers valuable insights for future research to enhance prediction accuracy by overcoming the current limitations.