McNorton et al. (2025) Hydroclimatic Rebound Drives Extreme Fire in California's Non‐Forested Ecosystems

Identification

• Journal: Global Change Biology
• Year: 2025
• Date: 2025-09-01
• Authors: Joe McNorton, Alberto Díaz Moreno, Marco Turco, Jessica Keune, Francesca Di Giuseppe
• DOI: 10.1111/gcb.70481

Research Groups

• UK Centre for Ecology and Hydrology
• Met Office Hadley Centre
• European Centre for Medium-Range Weather Forecasts (ECMWF)
• Tyndall Centre for Climate Change Research, University of East Anglia
• National Institute for Space Research (INPE), Brazil
• Imperial College London
• University of California, Merced
• Wageningen University and Research
• European Commission Joint Research Centre
• BeZero Carbon Ltd.
• Numerous other universities and research institutions globally

Short Summary

This second annual "State of Wildfires" report systematically tracks global and regional fire activity for the March 2024 to February 2025 season, analyzing the causes of prominent extreme wildfire events and projecting their future likelihood under climate change. It finds that global fire-related carbon emissions totaled 2.2 petagrams of carbon (Pg C), 9% above average and the sixth highest since 2003, despite below-average global burned area, primarily driven by extreme fire seasons in South America and Canada, with climate change significantly increasing the likelihood and intensity of these events.

Objective

• To systematically track global and regional fire activity, analyze the causes of prominent extreme wildfire events, and project the likelihood of similar events occurring in future climate scenarios.
• To identify extreme wildfires or seasons (March 2024–February 2025) and place them in context of recent trends.
• To assess the global impacts of extreme wildfire events on population, physical assets, carbon projects, and air quality.
• To diagnose the contributions of weather, fuel dryness, fuel load, ignitions, and suppression to four selected focal events.
• To evaluate the predictability of these focal events by operational systems.
• To attribute focal events to anthropogenic influences, including climate change and socioeconomic factors.
• To project future changes in the probability of these focal events under various climate scenarios.

Study Configuration

• Spatial Scale: Global, disaggregated to continental biomes, ecoregions, countries, and sub-national (level 1 administrative) regions. Focal event analyses at 0.25° or 0.05° resolution.
• Temporal Scale: Current fire season (March 2024–February 2025), historical analysis (since 2002/2003), multi-decadal trends (2003–2019), short- to medium-range forecasts (1–15 days), subseasonal to seasonal forecasts (up to 6 months ahead), and multi-decadal projections (up to 2100).

Methodology and Data

• Models used:
  • Probability of Fire (PoF) model (ECMWF Sparky fire model, XGBoost methodology)
  • Controlar Fogo Local Analise pela Máxima Entropia (ConFLAME) attribution framework (Bayesian inference)
  • HadGEM3-A large ensemble (for Fire Weather Index (FWI) attribution)
  • JULES-ES (Joint UK Land Environment Simulator Earth System model)
  • CanESM5 (Canadian Earth System Model version 5)
  • CLIMADA (for population and physical asset exposure assessment)
  • Copernicus Atmosphere Monitoring Service (CAMS) global model framework (for PM2.5 simulation)
  • FireMIP (Fire Model Intercomparison Project) attribution framework
  • ISIMIP3a/b (Intersectoral Impacts Model Intercomparison Project)
  • Random Forest (RF) algorithm (for burned area anomaly forecasting)
• Data sources:
  • Satellite: NASA MODIS MCD64A1 (burned area, 500 m), MODIS MCD14A1/MYD14A1 (fire radiative power, active fires), VIIRS VNP64A1 (burned area, 750 m), ESA FireCCIS311 (burned area, 300 m), NASA FIRMS (active fires).
  • Observation/Reanalysis: Global Fire Atlas (individual fire characteristics), GFAS (fire carbon emissions, 0.1°), GFED4.1s (fire carbon emissions, 0.25°), ERA5/ERA5-Land (meteorological data, FWI, Standardized Precipitation Evapotranspiration Index (SPEI)), Gridded Population of the World (GPW), LitPop (asset exposure), World Bank (produced capital estimates), EM-DAT (disaster database), Internal Displacement Monitoring Centre (IDMC) (displacement figures), BeZero Carbon Ltd. (carbon project boundaries).

Main Results

• Global Fire Season (2024–2025): Total burned area was 3.7 × 10^6 km^2 (9% below average), ranking 16th since 2002. However, fire carbon emissions reached 2.2 Pg C (9% above average), the sixth highest on record since 2003, indicating more intense fires in carbon-rich ecosystems.
• South America: Experienced an unprecedented fire season with record carbon emissions (263 Tg C, 84% above average) and 120,000 km^2 burned (35% above average). Bolivia, Brazil, and Venezuela saw record-breaking anomalies, characterized by fewer but larger, faster-spreading, and more intense fires.
  • Northeast Amazonia (Focal Event): Record forest burned area (+332% above average). Climate change made extreme fire weather 30–70 times more likely, increasing burned area by approximately 4 times.
  • Pantanal–Chiquitano (Focal Event): Worst fire season on record, with burned area nearly triple the average and carbon emissions 6 times above average. Fine particulate matter (PM2.5) concentrations reached 900 micrograms per cubic meter (µg/m^3), 60 times the WHO daily standard. Climate change made fire weather 4–5 times more likely, increasing burned area by approximately 34 times. Socioeconomic factors also showed a very strong role (amplification factor > 100).
• North America: Second most severe fire year on record, with total carbon emissions of 194 Tg C (112% above average) and burned area of 31,000 km^2 (35% above average).
  • Southern California (Focal Event, January 2025): Catastrophic Palisades and Eaton fires caused 31 fatalities, destroyed 11,750 homes, and resulted in USD 140 billion in total damages. PM2.5 levels peaked at 483 µg/m^3. Climate change increased burned area by approximately 25 times.
• Congo Basin (Focal Event): Highest recorded fire activity (+28% burned area), contributing to a 150% increase in primary forest loss in 2024 versus 2023. Climate change made fire weather 3–8 times more likely, increasing burned area by approximately 3 times.
• Impact Assessments:
  • Population Exposure: Approximately 100 million people globally were exposed to wildfires, with India and the Democratic Republic of the Congo having the highest numbers (~15 million each). Only 0.02% were formally displaced.
  • Physical Asset Exposure: Estimated USD 215 billion in physical assets exposed globally. The Los Angeles fires alone caused USD 140 billion in total economic losses.
  • Carbon Project Exposure: A record 18% of 927 Voluntary Carbon Market (VCM) forestry projects experienced fire in 2024, with 1.6% of project areas burned on average, largely driven by drought.
• Predictability: Weather was the dominant driver of fire activity (40–70% explainability), with fuel availability and dryness increasing in importance for severe fires. Seasonal forecasts successfully captured broad regional patterns of elevated fire danger.
• Future Projections (by 2100 under SSP370): The likelihood of 2024-scale extreme fire events is projected to increase by up to 57% in Northeast Amazonia, 34% in the Pantanal–Chiquitano, and 50% in the Congo Basin. Strong mitigation efforts (SSP126) can limit these increases to below 15% in tropical regions. Southern California's future risk trajectory remains uncertain, with some models suggesting a potential decline due to CO2 fertilization effects on vegetation.

Contributions

• Provides the second annual, systematic, and comprehensive assessment of global and regional wildfire activity, causes, impacts, and future projections.
• Integrates cutting-edge fire observations and modeling with extensive regional expert knowledge to identify and characterize extreme wildfire events.
• Applies advanced methodologies for disentangling the drivers of extreme wildfire events, including the roles of weather, fuel, ignitions, and suppression.
• Quantifies the attribution of recent extreme wildfire events to anthropogenic climate change and socioeconomic factors, using multiple complementary modeling frameworks.
• Introduces new routine regional assessments of wildfire impacts on society, including population exposure, physical asset exposure, carbon project exposure, and air quality degradation.
• Expands forward-looking capabilities by providing seasonal forecasts of burned area and multi-decadal projections of Fire Weather Index at future global warming levels.
• Presents 13 new datasets and model codebases, including updated annual statistics on wildfire extent, outputs from the PoF model, and the codebase for the ConFLAME attribution/projections model.
• Advances the representation of socioeconomic influences on fire by incorporating urban and rural population densities and non-linear response mechanisms in attribution analyses.

Funding

• UK Natural Environment Research Council (NERC) (LTSM2 TerraFIRMA project, NC-International programme grant no. NE/X006247/1, grant no. NE/V01417X/1, ARIES Doctoral Training Partnership grant no. NE/S007334/1)
• UK Department for Science (DSIT) Met Office Hadley Centre Climate Programme and ISPF (UK Met Office Climate Science for Service Partnership (CSSP) Brazil)
• European Commission Joint Research Center (service contract no. 942604)
• Swiss Innovation Agency Innosuisse (grant no. 120.464 IP-SBM)
• State Assignment Project (grant no. FWES-2024-0040)
• FAPESP (grant nos. 2019/25701-8 and 2023/03206-0)
• Westpac Scholars Trust (Westpac Research Fellowship)
• Australian Research Council (Industry Fellowship, grant no. IM240100046)
• UK Engineering and Physical Sciences Research Council (EPSRC; 2696930)
• Brazilian National Council for Scientific and Technological Development (CNPq; #409531/2021-9, #314473/2020-3, #443285/2023-3)
• Spanish Ministry of Science and Innovation (MCIN/AEI/10.13039/501100011033; ONFIRE PID2021-123193OB-I00)
• European Regional Development Fund (ERDF)
• Copernicus Atmosphere Monitoring Service (operated by ECMWF)
• Spanish Ministry of Science, Innovation and Universities (Ramón y Cajal, grant no. RYC2019-027115-I)
• Portuguese Foundation for Science and Technology (FCT; UID/04033/2025 and LA/P/0126/2020)
• Brazilian Institute of Environment and Renewable Natural Resources (IBAMA)/Federal University of Rio de Janeiro (UFRJ; #968711)
• Dragon Capital Chair on Biodiversity Economics
• European Research Council Consolidator Grant (grant no. 101000987)
• PRODIGY project by the BMFTR (grant no. 01LC2324) and Fulbright Amazonia programme

Citation

@article{McNorton2025Hydroclimatic,
  author = {McNorton, Joe and Moreno, Alberto Díaz and Turco, Marco and Keune, Jessica and Giuseppe, Francesca Di},
  title = {Hydroclimatic Rebound Drives Extreme Fire in California's Non‐Forested Ecosystems},
  journal = {Global Change Biology},
  year = {2025},
  doi = {10.1111/gcb.70481},
  url = {https://doi.org/10.1111/gcb.70481}
}