Home // CSRF 2025, The Second International Conference on Sustainable and Regenerative Farming // View article


Predictive Modeling of Soil Moisture: A Review of Benchmark Datasets, Their Strengths, and Limitations

Authors:
Kamrul Hasan
Arnold Muiruri

Keywords: soil moisture prediction; remote sensing; international soil moisture network; data fusion; machine learning.

Abstract:
Groundwater depletion, primarily driven by un- sustainable irrigation practices in agriculture, has become a pressing global issue. Accurate soil moisture monitoring and prediction are essential for supporting sustainable water resource management. This review contributes to an ongoing research effort aimed at developing a predictive soil moisture modeling framework by integrating signals from sparsely distributed ground-based sensors with satellite-derived datasets, including NASA’s Soil Moisture Active Passive (SMAP) products. As part of this study, a case analysis involving sevaral International Soil Moisture Network (ISMN) stations in the United States is conducted to evaluate the agreement between in-situ and satellite- derived measurements. While both data sources reveal consistent seasonal trends, significant discrepancies in magnitude highlight concerns regarding the reliability of these data as a universal benchmark. The paper provides a comprehensive review of recent advances and persistent challenges in soil moisture prediction, emphasizing the role of ISMN data. The overarching goal is to guide the development of robust, high-resolution tools for precision agriculture and sustainable groundwater management.

Pages: 9 to 12

Copyright: Copyright (c) IARIA, 2025

Publication date: October 26, 2025

Published in: conference

ISBN: 978-1-68558-327-9

Location: Barcelona, Spain

Dates: from October 26, 2025 to October 30, 2025