Li et al. (2026) Field-scale soil moisture retrieval from drone-based L-band radiometry with optical and thermal infrared priors

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-08-04
• Authors: Zixi Li, Yan Li, Rui Tong, Peizhe Cheng, Fuqiang Tian, Zhuang Yao
• DOI: 10.5194/hess-30-4909-2026

Research Groups

• Department of Hydraulic Engineering, Tsinghua University
• State Key Laboratory of Hydroscience and Engineering, Tsinghua University
• Information Center (Hydrology Monitor and Forecast Center), Ministry of Water Resources
• Fujian Key Laboratory of Severe Weather & Key Laboratory of Straits Severe Weather, China Meteorological Administration
• Fujian Meteorological Administration

Short Summary

This study develops an uncertainty-aware Bayesian joint retrieval framework that integrates UAV L-band radiometry with RGB and thermal infrared information to improve field-scale soil moisture estimation under heterogeneous land-surface conditions.

Objective

• To develop a physically informed multi-source prior approach for improving the accuracy of field-scale hydrological soil-moisture observation

Study Configuration

• Spatial Scale: Field-scale (4 × 7 m)
• Temporal Scale: Daytime observations during the growing season in Pengzhou, China

Methodology and Data

• Models used: Dual-polarized τ–ω emission model
• Data sources: UAV-based L-band radiometry, RGB imagery, thermal infrared data, and ground measurements using a portable time-domain reflectometry (TDR) probe

Main Results

• The proposed Bayesian joint retrieval framework achieved improved accuracy in field-scale soil moisture estimation compared to the baseline dual-parameter τ–ω inversion.
• The introduction of fraction vegetation cover (FVC), thermal-raster-based effective-temperature support, and footprint-scale texture descriptors progressively improved the results.

Contributions

• This study provides a physically interpretable pathway for improving field-scale soil moisture estimation under heterogeneous land-surface conditions by integrating multiple information sources.
• The proposed framework can be applied to various agricultural regions with diverse crop types and land-use patterns.

Funding

• Not specified in the provided text.

Citation

@article{Li2026Fieldscale,
  author = {Li, Zixi and Li, Yan and Tong, Rui and Cheng, Peizhe and Tian, Fuqiang and Yao, Zhuang},
  title = {Field-scale soil moisture retrieval from drone-based L-band radiometry with optical and thermal infrared priors},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-4909-2026},
  url = {https://doi.org/10.5194/hess-30-4909-2026}
}