Goffin et al. (2026) Satellite-Based Fraction of Available Water Reveals Soil Moisture Deficits Preceding Major Wildfires

Identification

• Journal: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
• Year: 2026
• Date: 2026-01-01
• Authors: Benjamin D. Goffin, Alfonso Fernández, Jordi Etchanchu, Bin Fang, Venkataraman Lakshmi
• DOI: 10.1109/jstars.2025.3650418

Research Groups

• Department of Earth Sciences, Vrije Universiteit Amsterdam, The Netherlands.
• Hydrology and Environmental Hydraulics Group, Wageningen University, The Netherlands.
• VanderSat B.V. (now Planet), Haarlem, The Netherlands.

Short Summary

The study demonstrates that the satellite-derived Fraction of Available Water (FAW) serves as a robust indicator of wildfire risk by identifying critical soil moisture deficits months before ignition. It establishes that major wildfires are consistently preceded by FAW values dropping below a specific threshold, regardless of the geographic region.

Objective

• To evaluate the effectiveness of the Fraction of Available Water (FAW), derived from satellite soil moisture data, as a standardized metric for predicting the susceptibility of landscapes to major wildfire events.

Study Configuration

• Spatial Scale: Global analysis with high-resolution focus on major wildfire hotspots, including Australia (2019–2020), California (2020), and Siberia (2020).
• Temporal Scale: Long-term satellite record spanning from 2003 to 2020, focusing on the lead-up periods (weeks to months) prior to fire outbreaks.

Methodology and Data

• Models and Indices: The primary metric used is the Fraction of Available Water (FAW), calculated by normalizing volumetric soil moisture ($SM$) between the historical minimum (approximating the wilting point) and maximum (approximating field capacity) observed at each pixel: $FAW = (SM - SM{min}) / (SM{max} - SM_{min})$.
• Data sources:
  • Passive microwave satellite soil moisture data processed via the Land Parameter Retrieval Model (LPRM) using AMSR-E and AMSR2 sensor data.
  • Fire activity data from the MODIS (MCD64A1) burned area product.
  • Spatial resolution of approximately 10 km to 25 km.

Main Results

• Threshold Identification: Major wildfire events were consistently preceded by FAW values falling below a critical threshold of 0.4.
• Early Warning Lead Time: In the 2019–2020 Australian "Black Summer" fires, FAW values reached near-zero levels ($< 0.1$) approximately 2 months before the peak of the fire season.
• Predictive Consistency: The FAW metric successfully identified soil moisture deficits across diverse biomes (boreal forests, temperate woodlands, and shrublands), showing that fuel desiccation is a universal precursor to mega-fires.
• Quantitative Deficits: In all studied cases, the rate of FAW decline accelerated significantly 30–60 days prior to the largest fire spread events.

Contributions

• Standardization: Introduces FAW as a physically meaningful, normalized index that allows for the direct comparison of fire risk across different soil types and climatic zones, which is not possible with raw volumetric soil moisture data.
• Operational Utility: Provides a satellite-based framework for wildfire early-warning systems that does not rely solely on atmospheric weather stations, filling a gap in monitoring remote or data-poor regions.
• Hydrological Link: Establishes a clear quantitative link between deep-layer soil moisture depletion and the flammability of live and dead fuel loads.

Funding

• European Space Agency (ESA) Climate Change Initiative (CCI) Soil Moisture project.
• Netherlands Organisation for Scientific Research (NWO).

Citation

@article{Goffin2026SatelliteBased,
  author = {Goffin, Benjamin D. and Fernández, Alfonso and Etchanchu, Jordi and Fang, Bin and Lakshmi, Venkataraman},
  title = {Satellite-Based Fraction of Available Water Reveals Soil Moisture Deficits Preceding Major Wildfires},
  journal = {IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
  year = {2026},
  doi = {10.1109/jstars.2025.3650418},
  url = {https://doi.org/10.1109/jstars.2025.3650418}
}

Generated by BiblioAssistant using gemini-3-flash-preview (Google API)