Ayari et al. (2026) Comparing 1-km Sentinel-1 surface soil moisture with coarser-resolution satellite data for agricultural drought monitoring in Mediterranean regions

Identification

• Journal: Agricultural Water Management
• Year: 2026
• Date: 2026-04-11
• Authors: Emna Ayari, Mehrez Zribi, Nemesio J. Rodríguez-Fernández, Clement Albergel, Nadia Ouaadi, Pietro Stradiotti, Nicolas Baghdadi
• DOI: 10.1016/j.agwat.2026.110341

Research Groups

• CESBIO, CNES/CNRS/INRAE/IRD/UT3-Paul Sabatier, Université de Toulouse, Toulouse, France
• Department of Crop Production Ecology, Swedish University of Agricultural Sciences, Umeå, Sweden
• European Space Agency Climate Office, ECSAT, Harwell Campus, Oxfordshire, Didcot, United Kingdom
• CRSA, Mohammed VI Polytechnic University, Ben Guerir, Morocco
• Climate and Environment Remote Sensing Group, Department of Geodesy and Geoinformation, Technical University of Vienna, Austria
• CIRAD, CNRS, INRAE, TETIS, University of Montpellier, AgroParisTech, Montpellier, France

Short Summary

This study evaluates the potential of a 1-kilometer surface soil moisture (SSM) product (HRSM) derived from Sentinel-1 and Sentinel-2 data for agricultural drought monitoring in Mediterranean regions, comparing its performance with coarser-resolution satellite SSM products (SMAP, ESA CCI) and root zone soil moisture (RZSM). The HRSM product shows good coherence with coarser products, uniquely identifying drought in agricultural areas, but its lower revisit frequency can miss short-duration rainfall events compared to daily or sub-daily products.

Objective

• To evaluate the potential of a 1-kilometer surface soil moisture (SSM) product (HRSM) derived from Sentinel-1 and Sentinel-2 data for detecting periods of agricultural drought in Mediterranean regions.
• To intercompare the HRSM product with coarser spatial resolution SSM products (SMAP and ESA CCI) and ESA CCI root zone soil moisture (RZSM) products.
• To calculate and analyze a newly proposed drought index based on monthly averages of SSM and RZSM to identify and confirm agricultural drought events.

Study Configuration

• Spatial Scale:
  • Two Mediterranean study sites: Occitanie region in France and a major part of Tunisia (north of the desert region), including various bioclimatic regions (humid, subhumid, semi-arid, arid).
  • HRSM product: 1 km spatial resolution, covering agricultural fields and grasslands.
  • SMAP SSM product: ~43 km native resolution, provided as 36 km EASE-Grid samples.
  • ESA CCI SSM and RZSM products: 0.25° regular grid.
  • ESA CCI RZSM depth layers: 0–10 cm, 10–40 cm, 40–100 cm, and 0–100 cm.
  • CHIRPS precipitation product: 0.05° grid resolution.
• Temporal Scale:
  • HRSM: January 2017 to September 2022 (Tunisia) and January 2017 to November 2022 (Occitanie), with a revisit frequency of 6 days (then 12 days from December 2021).
  • SMAP SSM: January 2017 to December 2022, with a revisit frequency of 2-3 days.
  • ESA CCI SSM and RZSM: January 2017 to December 2021, with daily time stamps.
  • CHIRPS precipitation: Monthly data from 2017 to 2022.
  • Drought index calculated using monthly averages of soil moisture.

Methodology and Data

• Models used:
  • Sentinel Soil Moisture Mapping (S²MP) algorithm for HRSM generation.
  • Water Cloud Model (WCM) inverted using an Artificial Neural Network (ANN) for SSM retrieval from Sentinel-1/2 data (for HRSM).
  • Integral Equation Model (IEM) for bare soil contribution in WCM (for HRSM).
  • Tau-Omega radiative transfer equation approximation for SMAP SSM retrieval.
  • TU Wien water retrieval package (WARP) for active SSM products in ESA CCI.
  • Land Parameter Retrieval Model (LPRM) for passive SSM products in ESA CCI.
  • Exponential filter (EF) for deriving ESA CCI RZSM from SSM.
  • Cumulative Distribution Function (CDF) method for scaling ESA CCI products to GLDAS-Noah climatology.
• Data sources:
  • Sentinel-1 (C-band Synthetic Aperture Radar) and Sentinel-2 (optical) data for HRSM.
  • Soil Moisture Active Passive (SMAP) L3 SSM operational product (L-band radiometer).
  • European Space Agency Climate Change Initiative (ESA CCI) combined SSM (v07.1) and RZSM (v08.1) products, merging data from multiple radiometers (TMI, Windsat, SSM/I, AMSRE, AMSR2, SMOS, FYs, GPM, SMAP) and scatterometers (AMI-WS, ASCATs).
  • Climate Hazards Group InfraRed Precipitation with Station (CHIRPS) for precipitation estimates.
  • 100 m resolution land cover map (CGLS-LC100) for masking non-agricultural/non-grassland areas in HRSM.
  • In-situ SSM measurements (for HRSM validation and RZSM exponential filter calibration via ISMN).
  • Global Land Data Assimilation System (GLDAS)-Noah (as a scaling reference for ESA CCI).

Main Results

• The HRSM product showed overall good coherence with SMAP and ESA CCI SSM time series over both Occitanie and Tunisia. HRSM and SMAP were generally more correlated than HRSM and CCI SSM.
• Over humid and subhumid regions in Tunisia, the best agreement was observed between the CCI SSM and HRSM time series.
• Drought events identified by the SSM-based drought index (from HRSM, SMAP, CCI) were consistently observed with the ESA CCI RZSM-based drought index (0-1 m depth).
• HRSM uniquely identifies drought events exclusively in grasslands and agricultural fields due to its native high spatial resolution and masking of other land covers.
• SMAP and ESA CCI SSM products, with their higher revisit frequencies (2-3 days and daily, respectively, compared to HRSM's 6-12 days), were more effective at capturing rapid SSM dynamics in response to rainfall and thus identifying all drought events.
• Discrepancies in drought identification were noted, particularly in semi-arid and arid regions, often attributed to differences in temporal resolution and the ability to capture ephemeral rainfall events. For example, HRSM sometimes indicated wet conditions when SMAP/CCI indicated dry, due to capturing specific rainfall events on its less frequent acquisition dates.
• The agreement between ESA CCI RZSM L1 (0-10 cm) and CCI SSM (0-5 cm) was good, with deeper RZSM layers showing smoother dynamics due to attenuated surface processes.

Contributions

• First comprehensive evaluation of the 1-kilometer HRSM product (derived from Sentinel-1 and Sentinel-2 synergy) for agricultural drought monitoring in Mediterranean regions.
• Detailed intercomparison of HRSM with widely used coarser-resolution satellite SSM products (SMAP, ESA CCI) and ESA CCI RZSM, highlighting their respective strengths and limitations for drought detection.
• Demonstrates the unique advantage of HRSM in providing drought information specifically for agricultural fields and grasslands by leveraging its high native spatial resolution for land cover masking.
• Provides insights into the trade-offs between spatial and temporal resolution of satellite soil moisture products for effective agricultural drought monitoring in heterogeneous Mediterranean landscapes.
• Validates the consistency of SSM-based drought indices with RZSM-based indices, reinforcing the utility of surface soil moisture for agricultural drought assessment.

Funding

• The European Space Agency Climate Change Initiative for Soil Moisture (grant 4000126684/19/I-NB).

Citation

@article{Ayari2026Comparing,
  author = {Ayari, Emna and Zribi, Mehrez and Rodríguez-Fernández, Nemesio J. and Albergel, Clement and Ouaadi, Nadia and Stradiotti, Pietro and Baghdadi, Nicolas},
  title = {Comparing 1-km Sentinel-1 surface soil moisture with coarser-resolution satellite data for agricultural drought monitoring in Mediterranean regions},
  journal = {Agricultural Water Management},
  year = {2026},
  doi = {10.1016/j.agwat.2026.110341},
  url = {https://doi.org/10.1016/j.agwat.2026.110341}
}