Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Water Management and Systems Research » Research » Publications at this Location » Publication #417473
Research Project: Improving Resiliency of Semi-Arid Agroecosystems and Watersheds to Change and Disturbance through Data-Driven Research, AI, and Integrated Models

Location: Water Management and Systems Research

Title: Surface soil moisture prediction using multimodal remote sensing data fusion and machine learning algorithms in semi-arid agricultural region

Author
item LAMICHHANE, MANOJ - South Dakota State University
item MEHAN, SUSHANT - South Dakota State University
item Mankin, Kyle

Submitted to: Science of Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/7/2025
Publication Date: 7/8/2025
Citation: Lamichhane, M., Mehan, S., Mankin, K.R. 2025. Surface soil moisture prediction using multimodal remote sensing data fusion and machine learning algorithms in semi-arid agricultural region. Science of Remote Sensing. 12. Article e100255. https://doi.org/10.1016/j.srs.2025.100255.
DOI: https://doi.org/10.1016/j.srs.2025.100255

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. Accurate data are expensive and time-consuming to collect, and difficult to come by. We used satellite data and machine learning methods to estimate surface soil moisture (top 1 ft) at a fine spatial resolution (33 ft square). Our methods explained 72% of the variation in soil moisture throughout a 3-year period in wheat, corn, millet, and fallow fields near Akron, Colorado. This is an important advancement. Better soil moisture data could lead to better understanding of how crop yields vary across the landscape and how to better manage cropping decisions in the future.

Technical Abstract: Precise spatial and temporal soil moisture is one of the limiting variables affecting agricultural decision-making in semi-arid regions. Despite the availability of various soil moisture monitoring products, their utility at the field scale is limited due to their relatively coarse spatial resolutions. Our study aims to predict surface soil moisture (SSM) at a much finer spatial resolution (10 m) using the Synthetic Aperture Radar (SAR) Sentinel-1 C-band and harmonized Landsat Sentinel (HLS) data sets, with the help of in-situ soil moisture measurements (0-30 cm) from crop fields located in the semi-arid locale of Akron, CO, USA. Four machine learning (ML) models: support vector machines (SVM), random forests (RF), gradient boosting machines (GBM), and K-nearest neighbors (KNN), were evaluated based on their performance metrics. Our findings indicated that the GBM excelled, demonstrating superior accuracy when using multi-source data with R2 of 0.72, RMSE of 0.025 cm3/cm3, and RRMSE of 11.86% for unseen verification data. Results also revealed that fitted GBR captured the SSM dynamics in crop fields and different phenological stages of crops. This suggests that the GBR model holds significant promise for field-scale SSM prediction in heterogeneous crop fields and different crop growth stages in semi-arid regions, offering a valuable tool for enhancing agricultural water management and hydrological modeling in the face of global water scarcity challenges.