A new wildfire regime is emerging in Southern Europe, characterised by larger and more intense fires, and by a fire season that now extends beyond the traditional summer months. In this region, climate projections indicate that fire occurrence and severity will increase faster than the global average due to an increased risk of heatwaves and droughts, as well as the evolution of biodiversity towards more resilient and less fire-prone plant species. These changes in wildfire regimes reveal significant gaps in the tools and technologies needed for implementing comprehensive fire management approaches. The community still faces challenges in predicting which wildfires may escalate into extreme events, and the environmental, climate and health impacts of such events remain poorly understood.
The Southern Europe Biomass BURNing (EUBURN) programme emerged as a concerted response to the need to improve the prevention, monitoring and prediction of wildfire risks in southern Europe. EUBURN integrates a series of multi-year and multi-scale field campaigns, lab studies, satellite remote sensing, and advanced modeling to build the research foundations for understanding the complex interactions between wildfires and the atmosphere. Based on this fundamental research, the EUBURN programme aims to support fire responders, ecosystems and air quality management, while addressing specific climate research requirements by developing new or enhanced operational products, tools and services for monitoring and predicting wildfires and their atmospheric impacts.
The first field campaign SILEX (Smoke from European Wildfire Experiment) of the EUBURN programme took place in southern France from 15 July to 3 August 2025. It had three specific objectives: (i) characterising the interactions between fuel, fire, gases, aerosols, radiation and clouds; (ii) contributing to the development of numerical prediction tools for fire behaviour and atmospheric plume dynamics; and (iii) assessing the uncertainties, biases and limitations of fire and smoke products from ground-based and satellite remote sensing. Ten scientific flights were carried out with the ATR-42 research aircraft equipped with state-of-the-art remote-sensing and in situ instruments to characterise wildfires occurring in southern France, as well as their associated smoke plumes, from the onset of emissions to their regional transport. The main purpose of the presentation is to familiarize the broader scientific community with the EUBURN programme and the SILEX dataset it produced. New findings on fire characteristics, gas and aerosol emissions, physical and chemical aging and cloud condensation nuclei will be emphasized.
How to cite: Denjean, C., Paugam, R., Pelletier, S., Borbon, A., Chiapello, I., Costa, M. J., Senra Rivero, F., Rochoux, M., Salgado, R., Tulet, P., Marino, E., Roman, R., Luna, Y., Tong, G., Ceamanos, X., cambre, A., Di Giuseppe, F., Filippi, J. B., Petetin, H., and Ruffault, J. and the Cyrielle Denjean: A European Initiative on Wildfire Risk and Atmospheric Impacts, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19990, https://doi.org/10.5194/egusphere-egu26-19990, 2026.
Wildfires are powerful forces of nature, shaping ecosystems, degrading air quality, and influencing the climate. Human activities intensify fires through land use change, accidental ignitions, and droughts driven by climate change. However, the complex interactions between climate change, vegetation shifts, and human behavior—and their consequences for wildfires—remain poorly understood. The Dutch innovations in atmospheric satellite sensors SPEXone, EarthCARE and TROPOMI now allow detailed studies of wildfires and their nearby and far-reaching consequences. The recently funded and started INFLAMES-project (Interdisciplinary Network for Fire research from Low Earth Orbit Atmospheric Measurements) aims to combine cutting-edge satellite data with state-of-the-art modeling techniques to unravel how wildfires alter air quality and climate, with a special focus on vegetation’s evolving role—both as a fuel source and a carbon sink in fire-affected regions.
In this presentation, we demonstrate the scientific ambition of the INFLAMES-project. We then show the first scientific results from INFLAMES, including satellite-derived trace gas (NOx, VOCs) and aerosol emission estimates for severe fires in Les Landes, France (August 2022), based on TROPOMI and MODIS observations and evaluated against the GFED emission inventory. We further show the first coincident EarthCARE, PACE and TROPOMI observations of wildfire plume heating-rate profiles over the Pantanal, demonstrating the potential of combined active–passive satellite measurements to directly constrain aerosol radiative effects. Together, these results establish a pathway toward improved quantification of the Aerosol Direct Radiative Effect (ADRE), a major remaining uncertainty in present-day radiative forcing, which will be further addressed using aerosol microphysical constraints from SPEXone on PACE. We conclude by highlighting opportunities for broader community engagement through dedicated workshops and an international summer school.
How to cite: Boersma, K. F., de Graaf, M., Hasekamp, O., Penning de Vries, M., Vrieling, A., Schutgens, N., Koren, G., van Bodegom, P., Kollia, D., Geurts, M., and Chantry, A.: First results from INFLAMES - Interdisciplinary Network for Fire research from Low Earth Orbit Atmospheric Measurements, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10271, https://doi.org/10.5194/egusphere-egu26-10271, 2026.
Nuclear conflict can ignite widespread fires that inject massive quantities of smoke particles into the atmosphere. Using the chemistry–climate model CESM2-WACCM6, we simulate idealized nuclear war scenarios with varying emission magnitudes of black carbon (BC) and primary organic matter (POM) released at 150~300 hPa over a 7-day period. Model results show that absorption of solar radiation by BC and POM leads to stratospheric temperature increases exceeding 50 K. This intense heating enhances the vertical lofting of smoke particles, enabling their transport even into the lower mesosphere and significantly extending their atmospheric residence time to over 4 years, thus leading to long-term environmental and climatic impacts. Even a regional nuclear conflict between India and Pakistan, emitting 5 Tg of BC (IP-5B scenario), results in a global total column ozone reduction exceeding 400 Tg (~12%), comparable in magnitude to that simulated for a large-scale nuclear war between USA and Russia with 16 Tg of BC emissions (UR-16B scenario). The co-emission of POM further amplifying stratospheric ozone depletion, leading to increased ultraviolet (UV) radiation at the surface. This heightened UV exposure poses serious risks to ecosystems and human health.
How to cite: Zhuo, Z., Pausata, F. S. R., Paul, K. J., and Cheung, A. K. H.: Tropical Small-Scale Nuclear War Fire Emissions Cause Greater Ozone Depletion Than Extratropical Large-Scale Conflicts, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10325, https://doi.org/10.5194/egusphere-egu26-10325, 2026.
High latitude wildfire regimes are changing, and boreal regions have seen unprecedented fire activity in recent years. Given the high climate sensitivity of the Arctic and boreal regions, it is important to explore the impacts of these changes. There are also region-specific impacts of biomass burning particular to high latitude regions, such as black carbon (BC) deposition on snow. While many sources of atmospheric pollution are being mitigated, fires are emerging as a growing contributor to poor air quality, both locally to the fire emissions source and across wider regions.
Here we investigate the climate and atmospheric effects of increased biomass burning emissions, including the sensitivity to emission region, focusing on aerosols. We perform idealized emission perturbation experiments in two Earth System Models (CESM2 and NorESM2), where we perturb first all boreal biomass burning emissions and then emissions in smaller regions of interest (boreal North America, East Siberia and West Siberia). These experiments use 2005-2014 as a baseline period, and use the sum of this period as the perturbation, giving an approximately x10 perturbation in the regions of interest, in both fixed SST (30 years) and coupled (200 years) simulations. The strength and location of the aerosol changes studied here (when comparing aerosol optical depth) are comparable to the recent trends in aerosols between 2015-2024 and 2005-2014.
We investigate subsequent effects on modelled atmospheric composition with a focus on the high latitudes, including air quality implications, and climate response, including effective radiative forcing (ERF) and fully-coupled climate response estimates. The preliminary analysis highlights the role of boreal forests in enhancing aerosol optical depth, over the source regions but also extending into the central Arctic, as well as local air pollution levels. Global and Arctic mean ERFs of 0.5 Wm-2 are estimated, with some distinct differences depending on the region of emission, at least for the Arctic average forcing.
How to cite: Lund, M. T., Staniszek, Z., Samset, B. H., Linke, O., and Ekman, A.: Regional to global impacts of boreal biomass burning emissions changes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20299, https://doi.org/10.5194/egusphere-egu26-20299, 2026.
Wildfires are unpredictable combustion events that significantly drive atmospheric emissions and modulate global cloud cover. An extreme example of such an event is the case of the 2023 Canadian wildfires, wherein nearly 5% of Canada’s forested area was burned between May and September 2023 [1]. This event produced the largest wildfire emissions ever recorded in Canada, with plumes extending across the Northern Hemisphere [2]. Aerosol intrusions and associated modifications absorbing and/or scattering can cause variability of solar irradiance [3], while reductions in photovoltaic power anywhere between 13% and 22% can take place because of aerosol optical depth (AOD) increases [4]. Consequently, the plumes resultant from the 2023 Canadian wildfires might have caused significant photovoltaic power losses over North America and Europe.
In this work, the global and local atmospheric impacts of this historic wildfire event are investigated using the EC‑Earth3 Earth system model in the interactive aerosols and atmospheric chemistry configuration (AerChem) [5]. BB emissions from the Copernicus Atmosphere Monitoring Service (CAMS) Global Fire Assimilation System (GFAS) were used through the model to produce two 10-member ensemble simulations, with and without the 2023 Canadian wildfire emissions respectively. The main parameter of interest is the modelled surface downwelling flux anomaly, which enables direct inference of modelled reductions in solar power output.
Model results have shown substantial radiative anomalies during May–September 2023 mainly in North America and Europe, with an average hemispheric shortwave radiation reduction of −4.18 W/m2 leading to PV production deficits. Secondary analyses suggest that surface cooling, which amounted to an average hemispheric temperature anomaly of −0.91 °C and which impacts PV performance, compensated 8–21% of the PV losses, varying by region. The results indicate a total 5-monthly modelled PV generation loss of 6.38 TWh, and the emitted carbon burden equivalent to this reduction in energy production is estimated at 2083 tons of CO2, with a total associated economic deficit of 1.33 billion euros. These findings emphasize the need for integrated transnational strategies in extreme event prediction and wildfire prevention to ensure the continued resilience of renewable energy production.
[1] Roșu, I. A., Mourgela, R. N., Kasoar, M., Boleti, E., Parrington, M., & Voulgarakis, A. (2025). Large-scale impacts of the 2023 Canadian wildfires on the Northern Hemisphere atmosphere. npj Clean Air, 1(1), 22.
[2] Byrne, B., Liu, J., Bowman, K. W., Pascolini-Campbell, M., Chatterjee, A., Pandey, S., ... & Sinha, S. (2024). Carbon emissions from the 2023 Canadian wildfires. Nature, 633(8031), 835-839.
[3] Wendisch, M., & Yang, P. (2012). Theory of atmospheric radiative transfer: a comprehensive introduction. John Wiley & Sons.
[4] Neher I., Buchmann T., Crewell S., Pospichal B. & Meilinger S. (2019). Impact of atmospheric aerosols on solar power. Meteorologische Zeitschrift, 4, 28.
[5] Van Noije, T., Bergman, T., Le Sager, P., O'Donnell, D., Makkonen, R., Gonçalves-Ageitos, M., ... & Yang, S. (2020). EC-Earth3-AerChem, a global climate model with interactive aerosols and atmospheric chemistry participating in CMIP6. Geoscientific Model Development Discussions, 1-46.
How to cite: Rosu, I.-A., Jones, M. W., Grillakis, M., Petrakis, M. P., Kasoar, M., Mourgela, R.-N., and Voulgarakis, A.: Impacts of 2023 Canadian wildfire emissions on solar power over North America and Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9005, https://doi.org/10.5194/egusphere-egu26-9005, 2026.
The rising frequency and intensity of wildfire-driven pyro-cumulonimbus (pyroCb) events constitute an important atmospheric perturbation, injecting massive amounts of smoke into the stratosphere. The Australian Black Summer wildfires of 2019–2020 released about a million tonnes of smoke and gases, causing the most significant stratospheric temperature perturbation since 1991 Pinatubo eruption. This study simulates the evolution of smoke plumes from the Australian wildfires using the UKESM1.1 model. The aerosol and greenhouse gas follow the CMIP6 SSP245 scenario, with 0.62 Tg of total smoke injected into the upper troposphere/lower stratosphere based on estimates from Global Fire Emissions Database (GFED). The simulated aerosol layer expands both vertically and horizontally, with significant lofting in the first month following injection, reaching altitudes of ~30 kms, consistent with CALIPSO observations. The modelled zonal-mean aerosol extinction agrees well with OMPS retrievals, with peak values of around 0.006 km⁻¹. However, the modelled stratospheric AOD is higher (up to ~2 times) than the observations showing the aerosols in the model are more optically efficient. Additional sensitivity tests are ongoing to examine whether a higher initial injection altitude in these simulations might be causing the aerosols to remain in the stratosphere longer and decay more slowly. These findings highlight the need for improved observational constraints and modelling strategies to better quantify the global impacts of wildfire-induced stratospheric smoke.
How to cite: Soni, M., Johnson, B., and Haywood, J.: Wildfire-driven Stratospheric Perturbations:Modelling Insights from the Australian Wildfires, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14942, https://doi.org/10.5194/egusphere-egu26-14942, 2026.
Pyrogenic airborne particle deposition downwind of active wildfires has traditionally been examined primarily as near-term hazards of fire spotting by firebrands or long-range transport of smoke particles (<10 µm). However, wildfires also emit substantial quantities of intermediate-sized airborne particles (10-2000 µm) that carry nutrients and contaminants affecting ecosystems downwind of the fire perimeter. The production, transport, and deposition of these intermediate-sized particles remain understudied. Here we develop a physics-based modeling framework for particle generation at fire lines, lofting by fire-driven convection, transport by prevailing winds, and subsequent ballistic settling. The framework enables characterization of this largely overlooked wildfire deposition footprint. Sensible heat flux from the fire feeds a convective plume capable of lofting particles to heights governed by fire intensity, particle size, shape, and density. Once aloft, particles are carried by ambient winds and ultimately ballistically deposited. The model performance was assessed using a unique dataset of particle deposition measured 3-40 km downwind of the fire front during the 2021 Caldor Fire. Supplemental observations of fire behavior, fuel properties, and meteorological conditions serve as inputs for model evaluation. The framework relies on various assumptions and constraints regarding unknown variables, including the mass fraction of emitted particles (5-7%), particle density (150-300 kg/m³), and drag coefficient formulation (fixed versus size-dependent), whose values were selected based on existing literature and physical plausibility. Over a 16-day sampling period, measured particle deposition ranged from 0.35 to 11.1 g/m². The largest deposition values (9.12-11.10 g/m²) occurred at collection sites closest to the fire (4-8 km), with progressively lower deposition (0.58-2.62 g/m²) observed at distant sites (10-40 km). When extrapolated to the landscape scale, a deposition rate of 10 g/m² over 1 km² corresponds to approximately 10 metric tons of pyrogenic material delivered to ecosystems for two weeks, an amount comparable to inputs from volcanic ashfall events. Within the modeling framework, simulations assuming a particle density of 300 kg/m³ and a pyrogenic emission fraction of 7% most closely matched field observations (RMSE < 1.8 g/m²; modest positive bias 0.8 g/m²; R > 0.90; p > 0.2). This configuration successfully reproduced both the magnitude and spatial gradients of observed pyrogenic mass deposition, demonstrating the framework’s potential to predict and quantify downwind delivery of wildfire-emitted particulate material to ecosystems.
How to cite: Scordo, F., Bavandpour, M., Or, D., Ebrahimian, H., Chandra, S., and Brahney, J.: Quantifying Downwind Deposition of Wildfire-Emitted Particles to Ecosystems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5840, https://doi.org/10.5194/egusphere-egu26-5840, 2026.
Climate change projections point to a sustained rise in emissions from biomass burning (BB), highlighting the need for a comprehensive evaluation of the environmental impacts of BB-derived aerosols (BBA). Of particular importance is the organic aerosol fraction (BBOA), which is chemically reactive and undergoes complex transformations during atmospheric ageing. These processes are especially critical in coastal regions, where strong coupling between atmospheric and marine systems can amplify environmental and ecological risks. In this study, we apply a multidisciplinary framework combining atmospheric chemistry, aerosol characterization, modeling, marine science, and toxicology to investigate the physicochemical properties of BBA, with emphasis on BBOA, and to assess how their atmospheric evolution affects air quality and marine ecosystems.
A comprehensive field campaign was conducted in the central Adriatic region, an area frequently impacted by intense wildfire events yet still poorly characterized in terms of BB influences. During controlled pinewood biomass burning experiments in April 2025, real-time measurements were conducted using state-of-the-art instrumentation, including a Scanning Mobility Particle Sizer (SMPS), an Optical Particle Counter (OPC), gas analyzers, and a CASS system combining an Aethalometer and a Total Carbon Analyzer. In parallel, fine particulate matter (PM2.5), volatile organic compounds (VOCs), and size-resolved aerosols (0.010–32 µm) were collected for comprehensive offline analyses, including the determination of trace metals, major ions, anhydrosugars, polyols, organic carbon, and aerosol oxidative potential.
To link atmospheric processes with marine impacts, laboratory exposure experiments were performed to evaluate the effects of ambient BB aerosols and model black carbon materials on the growth of representative marine phytoplankton species (such as Emiliania huxleyi, Cylindrotheca closterium, Melosira nummuloides, Synechococcus sp.) under controlled conditions (18 °C; 16 h light/8 h dark). These experiments reveal species-specific physiological responses to BB aerosol exposure. Overall, the integrated dataset provides new insights into the properties and evolution of BB aerosols and their cascading impacts on coastal air quality and marine ecosystem health in the Adriatic region, with broader implications for other vulnerable coastal environments.
This work was supported by Croatian Science Foundation project IP-2024-05-6224 ADRIAirBURN.
How to cite: Frka, S., Depolo, A., Arapov, J., Skejić, S., Šantić, D., Cvitešić Kušan, A., Chaux, F., Vicente, E., Alves, C., and Bubola, L.: Linking Air Quality and Marine Ecosystem Responses to Biomass Burning Aerosols in the Adriatic Coastal Zone, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11040, https://doi.org/10.5194/egusphere-egu26-11040, 2026.
The 2019–2020 Australian Black Summer megafires burned over eight million hectares of vegetation and constituted an extreme perturbation to terrestrial carbon cycling, releasing an unprecedented quantity of greenhouse gases to the atmosphere. Quantifying fire-driven emissions remains a key challenge, as emission inventories typically fall into one of two categories; bottom-up approaches (such as with the Global Fire Emission Database, GFED) that rely on burned area, fuel load, and combustion completeness estimates, or top-down approaches, (such as the Global Fire Assimilation System, or GFAS) which scale Fire Radiative Power (FRP) observations to emissions using emission coefficients. Currently, the two most widely used inventories (GFED and GFAS) ultimately rely heavily on uncertain modelled estimates of broad scale biome-specific combustion completeness, which remains a major limitation in constraining carbon fluxes from fires. We apply the Fire Radiative Energy Emission (FREM) approach, a top-down framework that directly links observed Fire Radiative Energy (FRE) to trace gas emissions, thereby reducing reliance on poorly constrained fuel and combustion assumptions. FREM is derived from co-located observations of FRP from the geostationary Himawari satellite and carbon monoxide (CO) from TROPOMI aboard Sentinel-5P. A dataset of 580 cloud-free landscape fires and associated plumes across six major Australian biomes (low woodland savanna, grassland, shrubland, evergreen and deciduous broadleaf forests, and sparse vegetation) was assembled for 2019 to derive biome-specific emission coefficients relating FRE to excess CO. These coefficients, combined with a calculated small-fire correction factor and hourly FRE observations from Himawari, were used to estimate emissions from the Black Summer megafires and to compare FREM-derived fluxes with those from existing inventories (GFAS v1.2, GFED v4.1s, GFED v5.1, and the Fire Energetics and Emissions Research, or FEER). The FREM estimates exhibit coherent spatial and temporal patterns and fall within the spread of emissions reported by these inventories, indicating consistency at regional scales while retaining sensitivity to fire intensity and temporal variability. By utilizing the geostationary FRP observations from Himawari, the FREM approach provides high-temporal-resolution, near-real-time estimates of fire emissions across Australia that are directly linked to observed radiative energy release, and bypasses the need for fuel load and combustion completeness estimations.
How to cite: Maslanka, W., Xu, W., Wooster, M., and He, J.: Quantifying Greenhouse Gas emissions from the Australian Black Summer Megafires using the Fire Radiative Energy Emission (FREM) Approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12220, https://doi.org/10.5194/egusphere-egu26-12220, 2026.
Wildfires emit large quantities of brown carbon (BrC), a class of light-absorbing organic aerosols with poorly constrained climate effects. BrC exhibits highly variable absorptivity, from weakly absorbing chromophores in the near-ultraviolet to strongly absorbing "dark BrC" (d-BrC) extending into the visible spectrum, yet the optical properties, global prevalence, and radiative impact of d-BrC remain poorly understood. Here we show that d-BrC is widespread in wildfire plumes globally, based on integrated analyses of aircraft, ground-based, and satellite observations. We found d-BrC mass absorption efficiencies of 0.5–1.5 m²/g at 500 nm, with absorption often comparable to or exceeding that of black carbon (BC). Implementing these observationally constrained optical properties in a global aerosol-climate model, we estimate a direct radiative effect (DRE) of +0.097 W/m² (range: +0.050 to +0.276 W/m²) from wildfire-derived BrC, with the upper bound surpassing BC and extending into mid- and high-latitude regions including the Arctic These findings position d-BrC as a critical but overlooked driver of wildfire radiative forcing, underscoring the need to account for its strong radiative effects on climate.
How to cite: Xu, L., Lin, G., and Liu, X.: Strong Shortwave Absorption by Wildfire Brown Carbon from Global Observations and Modeling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4614, https://doi.org/10.5194/egusphere-egu26-4614, 2026.
In May and June 2025, wildfires in Canada produced atmospheric effects extending beyond North America. Large quantities of gases and aerosols emitted by biomass combustion were transported across the Atlantic and reached Europe. Here, our aim is to investigate how these events affect the variability of climate-altering species in Italy using observations from permanent observatories.
Clear evidences of this long-range transport were observed from 8th June 2025 at the GAW/WMO Global Station “O. Vittori” at Monte Cimone (2165 m a.s.l., northern Italy) and at the Potenza CIAO observatory (760 m a.s.l., southern Italy), two co-located sites for the Research Infrastructures ICOS and ACTRIS. It was also observed, albeit with weaker intensity, at the ACTRIS Environmental-Climate Observatory (ECO) in Lecce (37 m a.s.l., southern Italy). Atmospheric transport modelling (LAGRANTO and HYSPLIT back-trajectories) confirmed that the air masses affecting the sites originated in North America.
Average daily carbon monoxide (CO) values peaked to 207 ppb on 9th June at CMN and to 247 ppb at ECO, nearly doubling the levels measured during the preceding 7 days. Also, black carbon (BC) showed marked increases, with values more than doubling the average of the preceding days at both sites.
Additional confirmation of the plume’s arrival and vertical evolution was provided by the ALICE-Net ceilometer at CMN: between 6th and 8th June, aerosol-rich layers were detected at high altitudes before gradually descending to the measurement site. At CIAO, the aerosol lidar observed smoke layers between 11 and 14 km from 5th to 10th June.
CO and ozone (O₃) remained high until 13th June at CMN (average values: 188 ppb and 70 ppb), and at ECO (average CO value of 232 ppb, O3 data not available). Subsequently, intermediate values have been observed from 14th to 21st June. At CIAO, CO increased between 8th and 17th June, reaching up to 250 ppb.
No corresponding increases in carbon dioxide (CO₂) have been observed during the wildfire plume event. During the days characterized by the peaks in CO and O3 (8th – 13th June), daily mean CO2 values showed a – 6.4 ppm and – 3.4 ppm decrease with respect to the previous 7 days at CMN and ECO. The analysis of back-trajectories showed air masses travelling at pressure levels representative of the European PBL, where active ecosystems could take up CO₂, in the 24 hours before the arrival at CMN.
The analysis of the day-to-day variability of nighttime/daytime N2O, CO2 and δ13CO2, pointed to a significant influence of air masses from the regional PBL to CMN during the daytime on 9th – 14th and 18th – 19th June. This suggests that emissions occurring at regional scale could contribute to the observed atmospheric composition variability. Together with the role of air mass mixing and in-plume chemical processes along transport, this implies that attributing the observed enhancements to wildfire emissions requires careful and critical evaluation.
Acknowledgments: Observations/analyses are supported by the ITINERIS (PE0000021, NRRP – NextGenerationEU) and PRO-ICOS MED (PON 2014–2020) projects, funded by the Italian Ministry of University and Research and the European Union.
How to cite: Cristofanelli, P., Barnaba, F., Bracci, A., Calidonna, C. R., Cesari, R., Contini, D., Diliberto, L., d'Amico, F., Decesari, S., Dinoi, A., Gori, L., Marinoni, A., Mona, L., Putero, D., Zaccardo, I., and Zanatta, M.: Impacts of wildfire plumes from Northern America on atmospheric composition as observed by permanent observatories in Italy during June 2025, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11975, https://doi.org/10.5194/egusphere-egu26-11975, 2026.
Fire is a dominant terrestrial ecosystem disturbance and a major driver of atmospheric composition. Quantifying fire emissions and their variability remains a key challenge, particularly as fire frequency and intensity vary and increase under climate change. Inverse modeling provides a powerful framework to estimate fire emissions by constraining chemistry transport models (CTMs) with satellite observations, while simultaneously delivering three-dimensional information on the transport and distribution of fire-related pollutants.
In this study, we use the global CTM TM5 coupled with a four-dimensional variational data assimilation system (TM5-4DVar) to better constrain fire-related carbon monoxide (CO) emissions using satellite observations. We assimilate CO column super-observations from the Measurements of Pollution In The Troposphere (MOPITT) instrument aboard NASA’s Terra satellite (version 9) and, separately, higher spatiotemporal resolution CO observations from the TROPOspheric Monitoring Instrument (TROPOMI) aboard ESA’s Sentinel-5P. The assimilations are performed globally at 3° × 2° horizontal resolution over multiple years (2019–2024).
The posterior simulations provide insights into both regional fire emissions and the horizontal and vertical transport of CO, enabling assessment of downwind pollution impacts, evaluated against independent ground-based observations. Results show bias reductions with posterior simulated mixing ratios in comparison to prior simulations based on bottom-up emission inventories. We further investigate the influence of regional drought conditions on fire-related CO emissions and examine correlations with key environmental variables, including climate and vegetation indicators. Our results contribute to an improved understanding of interactions among fire emissions, climate, and atmospheric composition, and demonstrate the value of remote sensing data assimilation for reducing uncertainties and advancing fire emission monitoring.
How to cite: Peiro, H., van der Velde, I., van der Werf, G., Houweling, S., Rijsdijk, P., and Aben, I.: A global CO fire emissions assessment and its connection with drought events, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11440, https://doi.org/10.5194/egusphere-egu26-11440, 2026.
Although half of Earth’s population resides in the wildland-urban interface, human exposure to wildland fires remains unquantified. We show that the population directly exposed to wildland fires increased 40% globally from 2002 to 2021 despite a 26% decline in burned area. Increased exposure was mainly driven by enhanced colocation of wildland fires and human settlements, doubling the exposure per unit burned area. We show that population dynamics accounted for 25% of the 440 million human exposures to wildland fires. Although wildfire disasters in North America, Europe, and Oceania have garnered the most attention, 85% of global exposures occurred in Africa. The top 0.01% of fires by intensity accounted for 0.6 and 5% of global exposures and burned area, respectively, warranting enhanced efforts to increase fire resilience in disaster-prone regions.
How to cite: Sadegh, M., Seydi, S. T., Abatzoglou, J., Jones, M., and AghaKouchak, A.: Increasing global human exposure to wildland fires despite declining burned area, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4268, https://doi.org/10.5194/egusphere-egu26-4268, 2026.
Wildfires are rapidly increasing in boreal forests and are extending to Arctic environments at an unforeseen scale. Above-ground biomass burning may be compensated by regrowth in the years following a wildfire, impacting atmospheric CO2 levels only temporarily. However, high-latitude wildfires characteristically combust deep into carbon-rich soils accumulated over centuries to millennia and thereby risk transforming these long-term carbon sinks into net sources into the atmosphere. Hitherto, research on Arctic-boreal fires has largely focused on their surface impacts, including the burned area, severity, and forest recovery, while many of their underground characteristics are poorly understood. For instance, observations of the age of carbon released in the fires remain scarce, resulting in incomplete understanding of the climate impact of high-latitude fires.
To determine the age of carbon released in recent Arctic-boreal fires, we collected charred organic material for radiocarbon dating from a tundra fire in Greenland, and two boreal forest and one tundra fire site in northwestern Canada. Our results indicate that, contrary to previous observations, up to centennial to millennial-aged carbon was released in these arctic and boreal wildfires. Moreover, laboratory combustion experiments of Arctic-boreal biomass collected from fire-susceptible surface layers (0-30 cm depth) from Svalbard, Russia, Norway and Finland, demonstrate that the combustion mode, and thus the phase of the emitted carbon, depend on the age of the combusted material. Above-ground modern vegetation combusts flamingly emitting mainly gases, while below-surface older and partly decomposed organic material smoulders, producing increasing carbonaceous particle/gas ratios with increasing age of the combusted material. Similar to the studied Greenland and Canadian wildfires, the laboratory combustion of the Arctic-boreal biomasses show up to millennia-aged carbon emissions.
Our results indicate that centennial to millennial-aged carbon is released in Arctic-boreal wildfires, thereby causing long-lasting feedback to the global climate system. Currently, climate models do not consider the potential release of ancient carbon from wildfires. Thus, our results indicate that increasing Arctic-boreal wildfires may exacerbate global warming more than previously estimated.
How to cite: Ruppel, M., Granqvist, S., Diaz, L., Haghipour, N., Sippula, O., Smittenberg, R., Somero, M., Veraverbeke, S., and Väliranta, M.: Ancient carbon released in Arctic-boreal wildfires, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16499, https://doi.org/10.5194/egusphere-egu26-16499, 2026.
Fire is a major ecosystem disturbance impacting the global carbon cycle, with its frequency and severity projected to increase. The time required for ecosystems to recover productivity after fire (recovery time) is an important metric for resilience, yet its global patterns remain poorly quantified. Here, we conduct a global analysis using Moderate Resolution Imaging Spectroradiometer (MODIS) observations from 2001 to 2024. We employ satellite-derived burned area data and the near-infrared reflectance of vegetation (NIRv) as a robust proxy for Gross Primary Production (GPP) to track recovery, which is defined as the duration to recover 90% of pre-fire productivity. Our analysis focuses on single-fire events, filtering out areas with recurrent disturbances, and defines recovery as the point when at least 90% of pre-fire productivity is regained.
Our results reveal that the global mean post-fire recovery time is 3.9 ± 0.3 years. This average is masked by strong geographical disparities: recovery follows a pronounced latitudinal gradient, with boreal ecosystems (≥50°N) requiring nearly twice as long to recover (5.6 ± 0.5 years) compared to tropical regions (3.0 ± 0.2 years). Evergreen needleleaf forests exhibit the longest recovery times (6.3 ± 0.9 years), while savannas and grasslands recover fastest. Statistical machine learning modeling identifies the magnitude of the immediate fire-induced GPP loss as the dominant factor controlling recovery duration, with burn severity and pre-fire productivity acting as important secondary drivers.
We show that CMIP6 Earth System Models (ESMs) significantly underestimate these recovery periods (simulating a global mean of 1.8 ± 0.1 years) and fail to capture the observed spatial heterogeneity, particularly in high-latitude regions. This suggests that current models may overestimate the carbon sink capacity of regenerating post-fire landscapes and underestimate positive fire-vegetation feedbacks. Our findings provide a new observational benchmark for improving the representation of post-disturbance dynamics in land surface models and refining global carbon budget assessments.
How to cite: Lin, Z., Chen, A., and Wang, X.: Global Patterns of Post-Fire Vegetation Productivity Recovery, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8718, https://doi.org/10.5194/egusphere-egu26-8718, 2026.
Biomass burning (BB) emissions in the Indo-China Peninsula (ICP) can be transported to southern China, perturbing the atmospheric environment and climate in southern China. However, the impact of these fire emissions transports on the terrestrial ecosystems in southern China remains unclear. Here we combine several state-of-the-art models and multiple measurement datasets to quantify the impacts of ICP fire-induced aerosol radiation and O3 damage effect on gross primary productivity (GPP) in southern China during ICP fire seasons (March and April) in 2013-2019. Our results demonstrate that ICP fire-derived aerosols and O₃ collectively reduce annual mean GPP in southern China by 5.4% (13.86 TgC per burning season) under all-sky and 3.4% (12.87 TgC per burning season) under clear-sky conditions. In all-sky, fire aerosols decreased direct photosynthetically active radiation (PAR) by 2.68 W m⁻² while increased diffuse PAR marginally (+0.03 W m⁻²), driving a GPP reduction of 13.36 TgC per burning season across southern China. Concurrently, fire-induced O₃ reduces regional GPP by 0.54 TgC per burning season. In clear-sky, aerosols reduce direct PAR more sharply (−3.22 W m⁻²) but enhance diffuse PAR (+1.51 W m⁻²), resulting the GPP loss to 12.18 TgC, while O₃ damage effect is increased (−0.69 TgC). The fire aerosols contributed to 96.4% of the GPP reduction in all-sky and 94.6% in clear-sky, whereas ozone played a minor role (3.9% in all-sky and 5.4% in clear-sky). This study highlights ICP fire emissions as a significant driver of ecosystem productivity declines in downwind regions, influencing the regional land carbon cycle.
How to cite: Zhu, J.: Quantifying the multi-year impacts of Indo-China Peninsula biomass burning on vegetation gross primary productivity in southern China, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3711, https://doi.org/10.5194/egusphere-egu26-3711, 2026.
Wildfires are mainly considered to be CO2-releasing events, while their long-term impact on biogeochemical carbon sequestration remains a major source of uncertainty. We analysed five years of ecosystem-scale eddy covariance data in a South African Fynbos shrubland that experienced a wildfire in the middle of the measurement period and combined it with leaf-scale ecophysiological measurements to quantify the ecosystem-scale carbon feedbacks and energy flux shifts following wildfire.
Unexpectedly, wildfire doubled the annual net carbon sink from 5.36 to 10.55 tC ha-1 yr-1. This increase was driven by a ca. 50% suppression of ecosystem respiration while ecosystem energy exchange remained stable. These findings reveal a significant missing carbon pool of ca. 110 tC ha-1 over the course of the fire return interval of 15-20 years. Likely explanations for this discrepancy are either a below-ground carbon pool protected from volatilization through fire or a potential sink into dissolved carbon, potentially leading to eventual long-term ocean storage.
To identify the biological drivers of this carbon sequestration, we measured gas exchange in the two main regeneration plant types of this fire-dominated ecosystem, i.e. obligate reseeders, whose seedlings must achieve reproduction before the next fire to persist, and resprouting species that invest into fire tolerance traits at the cost of slower growth. Stomatal conductance (gsw) was the primary trait distinguishing the two strategies. Reseeders initiated photosynthesis earlier in spring and exhibited gsw that was highly responsive to changes in ambient CO2 and light, while resprouters exhibited stronger resilience to drought but no response to ambient CO2 fluctuations. This difference in response to CO2 suggests that current climate trends may preferentially boost reseeders, potentially partially offsetting the impacts of shortened fire return intervals. Conversely, resprouter resilience may prove crucial under a higher drought intensity and duration scenario.
Our unexpected findings for this Mediterranean-climate shrubland (typically considered to be a low carbon sink ecosystem) underscore the necessity for ground-based ecophysiological data to constrain Earth system models, and challenge biomass-centric climate policies, particularly in fire-prone, naturally tree-free ecosystems.
How to cite: Muller, J. D., Joubert, W., de Buys, A., Ramsay, E., Carkeek, R., and Midgley, G. F.: Fire as a Catalyst for Carbon Sequestration: Respiration Suppression and Regeneration Feedback in South African Fynbos Shrubland, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9056, https://doi.org/10.5194/egusphere-egu26-9056, 2026.
Fire regimes show substantial variability among ecosystems, with a fundamental contrast between surface and crown fires. While surface fires predominantly consume understory vegetation, crown fires involve the combustion of canopy fuels. This distinction is therefore central to understanding fire-driven ecosystem dynamics and to designing effective wildfire risk management strategies.
Ongoing climate change is expected to further reshape fire regimes by altering temperature and moisture conditions and by driving shifts in species distributions. These processes may indirectly modify fire behaviour by changing fuel structure, continuity, and overall landscape flammability.
Within this context, plant functional traits provide a valuable lens through which to interpret fire–vegetation interactions. They not only respond to environmental filtering but also actively shape ecosystem functioning. Two traits in particular—branch shedding (the ability to shed dead lower branches) and serotiny (the retention of mature cones that open after exposure to high temperatures)—have been proposed as key adaptive strategies influencing fire regimes. However, there is limited understanding of whether environmental factors can effectively cancel the adaptive advantages conferred by these traits, which, if occurring frequently, might substantially alter ecosystem dynamics.
To explore these issues, we integrated forest information from the Spanish Forest Map with fire severity data from the European Forest Fire Information System (EFFIS). Our analysis focused on pine species dominating coniferous forests across the western Mediterranean region. We examined how branch shedding and serotiny relate to crown fire occurrence, and how these relationships are modulated by stand-level attributes such as successional stage, shrubs abundance, and the occurrence of extreme drought during the fire season.
Our results indicate that the effectiveness of these trait-based strategies is, at least in the western Mediterranean, strongly contingent on forest stand conditions and suggests that climate change might disrupt the current spatial consistency of these long-established fire-traits relationships.
How to cite: Costa-Saura, J. M., Sirca, C., Spano, D., and Valor, T.: Environmental factors disrupting the adaptive advantage of fire-trait syndromes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17500, https://doi.org/10.5194/egusphere-egu26-17500, 2026.
Wildfires are increasingly shaping terrestrial ecosystems, with profound implications for land degradation processes across fire-prone regions.
This work advances the assessment of post-fire land degradation by jointly analysing fire occurrence, burn severity and vegetation recovery as key indicators of ecosystem vulnerability. By integrating multi-temporal fire records (2001-2019) the study captures both the frequency of disturbances and its immediate ecological impact, enabling another view on the evaluation of degradation trajectories globally (Vieira et al., 2026) .
Results indicate that recurrent fires, particularly when combined with high-severity events, substantially exacerbate vegetation loss, and erosion risk, thereby accelerating land degradation processes. Preliminary results indicate that areas experiencing short fire-return intervals show limited recovery capacity, leading to cumulative impacts on soil health, which on turn might be leading to alternate states (McGuire et al., 2024) . The analysis further highlights strong spatial variability, where land cover, and pre-fire conditions influence degradation response.
Overall, this work underscores the importance of moving beyond binary burned–unburned classifications and incorporating fire severity and recurrence into land degradation assessments. Such an approach provides critical insights for post-fire management, restoration prioritisation, and the development of adaptive strategies aimed at mitigating long-term degradation under a changing fire regime.
McGuire, L. A., Ebel, B. A., Rengers, F. K., Vieira, D. C. S., and Nyman, P.: Fire effects on geomorphic processes, Nat Rev Earth Environ, 1–18, https://doi.org/10.1038/s43017-024-00557-7, 2024.
Vieira, D. C. S., Borrelli, P., Scarpa, S., Liakos, L., Ballabio, C., and Panagos, P.: Global estimation of post-fire soil erosion, Nat. Geosci., 19, 59–67, https://doi.org/10.1038/s41561-025-01876-0, 2026.
How to cite: Vieira, D., Borrelli, P., and Panagos, P.: Feedback Loop between fire and land degradation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12251, https://doi.org/10.5194/egusphere-egu26-12251, 2026.
Fire is a major driver of forest change worldwide. In tropical regions, it is primarily used by local populations to clear forested land for human activities such as agriculture and infrastructure development. Here, we use three decades of Landsat-based observations to analyse fire-related forest loss over a broad temporal scale across a major tropical region in South America, spanning more than 2 million km² and encompassing a wide range of ecosystems. This long-term assessment provides a comprehensive view of post-fire forest cover dynamics, with strong potential to capture deforestation trends, forest fate, and the roles of protection status and landscape history. Over the study period, the newly generated medium-resolution dataset of burned area detected a cumulative total of approximately 345 million hectares burned, equivalent to an average annual burned fraction of 5.65%, with pronounced interannual variability and the period between 1999 and 2010 being the most extreme. During the same timeframe, more than 24.5 million hectares of forest were lost, representing nearly one-quarter of the 1990 forest extent, with fires accounting for 26% of this loss. Most of these losses have not recovered over time and were subsequently followed by deforestation, with 99% of affected areas converted to pastures and croplands, while recovery rates have remained negligible. Fragmentation and fire history legacy emerged as critical factors influencing the trajectory of forest loss.
How to cite: Khairoun, A. and Salinero, E.: Analysis of long-term fire-related deforestation and cover change dynamics in South American ecosystems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15336, https://doi.org/10.5194/egusphere-egu26-15336, 2026.
Global burned area (BA) products are commonly available at a Non-Time Critical (NTC) basis, several months or even several years from the present date; i.e. they are unavailable for Near-Real Time (NRT) applications. The Copernicus Land Monitoring Service (CLMS) delivers the only global BA product in NRT, since recently, at very high accuracy, comparable to the most accurate non-CLMS NTC product (FIRECCIS311). However, global BA products are generated from coarse >= 300 m reflectance observations. Despite the Sentinel-2 mission having been in operation since 2017, providing decadal resolution 10-50 m reflectance data every ~5 days, and despite the well-known benefits of using decadal resolution data to estimate BA, a global Sentinel-2 NRT BA algorithm does not exist. The purpose of this study is to adapt and apply the latest developments in NRT detection, as implemented in the CLMS, to Sentinel-2 L2A imagery. The mapping method uses a neural network (NN) with 2D convolutional layers, followed by a Long Short-Term Memory (LSTM) layer. The NN processes the time series of reflectance images on a per-pixel basis, with convolutional layers applied along the spectral and temporal dimensions. The time series of fractional BA maps, predicted by the NN, are combined with time series of spatio-temporal density of VIIRS active fire detections. Such a combination consists of a logistic model and allows the reduction of false positives (such as cloud shadows). The NN is trained on a sample dataset automatically generated from time series reflectance observations (Sentinel-2 data in this case), extracted over locations of VIIRS active fire detections across the Globe for the year 2020, and corresponding estimates of fractional BA, derived from physically-based radiative transfer modelling. The mapping method generates one BA map for each new Sentinel-2 image available (referred to as BAS2nrt0), which is updated with images from the following 5 days (referred to as BAS2nrt5) and the following 10 days (referred to as BAS2nrt10). The additional images available after the mapping day are expected to reduce false positives due to cloud shadows. The mapping method also generates an NTC BA map for each calendar month (referred to as BAS2ntc), with images available for a buffer of 45 days around the month. The algorithm results are validated against an independent global reference dataset for the year 2019, which includes long time series of Landsat-derived BA maps covering 105 sampling units distributed across the Globe. The analysis of the 2019 validation results shows that the accuracy of the proposed Sentinel-2 products is high regardless of estimation timeliness. As expected, (1) the accuracy of the NTC product, Dice coefficient (DC) of 87.2%, is higher than the NRT products, DC 82.7–85.4%, and (2) the accuracy of the NRT product is increased with each update. Such accuracy levels are remarkably high: the accuracy of NRT estimates is comparable to a precedent global non-CLMS NTC Sentinel-2 BA mapping (DC 81.8%).
How to cite: Padilla, M., Ramo, R., Gomez-Dans, J. L., Sierra, S., Mota, B., Lacaze, R., and Tansey, K.: Global near-real time burned area mapping with Sentinel-2 based on reflectance modelling and deep learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13146, https://doi.org/10.5194/egusphere-egu26-13146, 2026.
Fish Lake is located at 2700 m (a.s.l.) on the boundary of the Colorado Plateau and Great Basin geologic provinces in western North America. Climate forecast models suggest that this region will become warmer and drier during the 21st Century, which will likely intensify fire regimes and threaten biodiversity in this region, including the ancient Pando aspen clone located next to Fish Lake. Here we present a paleoenvironmental reconstruction from an 80-meter lake sediment core spanning the last 60,000 years. At the Last Glacial Maximum (LGM), the upland areas near Fish Lake (3200-3500 m asl) were heavily glaciated and plant communities were open and dominated mainly by herbs and conifers, such as grasses (Poaceae) and spruce (Picea spp.). During the LGM fire activity was low due to cold temperatures, low woody fuel abundance and connectivity, and the presence of megaherbivores (e.g., Mammuthus) as reconstructed from nearby fossil sites and the presence of coprophilous fungal spores (Sporormiella) in the Fish Lake sediments. In the Late Glacial Period, the demise of upland glaciers and megaherbivores was accompanied by a ‘release’ in woody vegetation, especially spruce and pine (Pinus spp.) and a rise in charcoal accumulation. During the Early Holocene, this rise in burning sustained and was likely enhanced by warming temperatures and the establishment of closed-canopy forests similar to modern composed of Engelmann spruce (Picea engelmannii), aspen (Populus tremuloides), and subalpine fir (Abies lasiocarpa). Fire activity in the Middle Holocene remained high, with a stepwise increase observed during the Late Holocene that occurred with increasing evidence of human activities and amplification of El Nino-Southern Oscillation (ENSO). Throughout the 60,000 record, aspen pollen is consistently present. While pollen alone does not provide direct evidence of the long-lived Pando aspen clone, this record does confer the presence of aspen growing near Fish Lake through contrasting climate periods and fire regimes. This long-term reconstruction offers new insights into the interactions of climate, vegetation, and herbivory in shaping wildfire in western North America to help support land management policies.
How to cite: Morris, J., Carter-Kraklow, V., Codding, B., Winward, N., Brunelle-Runberg, A., Vornlacher, J., Werne, J., Marchetti, D., Wilson, K., Anderson, L., Abbott, M., and Power, M.: Reconstructing the last 60,000 years climate-driven interactions of fire, vegetation, and megaherbivores at Fish Lake, Utah, USA, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16685, https://doi.org/10.5194/egusphere-egu26-16685, 2026.
Fires on Earth are changing in response to human activities, both through direct ecosystem management and indirect climate change-induced warming and associated shifts in regional rainfall patterns. While increased burning in forested systems often captures international media attention, the decline in burning in grassy systems––especially African savannas––receives less focus, despite their dominant contribution to total global burned area and fire emissions. Forecasting future fire activity and its impacts on local ecology and livelihoods, as well as global climate feedbacks, requires a robust mechanistic understanding of the complex interactions between climatic conditions, ecosystem functioning, human activities, and fire across a range of climate states not captured by modern satellite-based observations.
This study focuses on Africa, whose environments span a diversity of climates and ecologies, from some of the driest and most sparsely vegetated regions on Earth (such as the Sahara) to some of the wettest and most biologically productive (such as the Congo Rainforest). These two ends of the rainfall gradient experience non-existent to infrequent burning. However, the most expansive biomes in Africa are tropical savannas and grasslands where precipitation is intermediate and highly seasonal, supporting rapid vegetation growth during wet seasons and drying and abundant fires in the dry season. As a result, burning in Africa constitutes more than half of all global burned area each year. Robust histories of how fires have changed in Africa through time are therefore essential to understanding changes in biomass burning at a global scale. In addition to its broad scope of environments and outmatched contributions to total global burning, Africa also has the longest history of human fire use and land-use change, making it an ideal testing ground for interrogating the combined roles climate shifts and human behaviors play in shaping fire regimes through time.
Here, we present a new synthesis of African paleofire activity inferred from the accumulation of both physical and molecular proxies (e.g., charcoal and polycyclic aromatic hydrocarbons) within climate archives spanning multiple depositional contexts (e.g., lacustrine, marine, and peat sediments) which record biomass burning across a host of ecosystems. Our new reconstruction spans the last 24 thousand years, within which we focus on four key time periods: the Last Glacial Maximum and deglaciation, the mid-Holocene African Humid Period, the late-Holocene rise of metallurgy and agriculture, and the post-industrial era. We evaluate trends in biomass burning during these intervals, and, by comparison to paleoclimate and archeological datasets, we assess the extent to which these patterns are driven by climatic and/or human influences at continental, regional, and biome scales.
How to cite: O’Mara, N., Githumbi, E., Bartlein, P., Norwood, M., Teruel, O., Aleman, J., Staver, C., and Marlon, J.: Changes in biomass burning in Africa since the last glacial maximum: a new continental-scale paleo-synthesis and interrogation of the climatic and human drivers of shifting fire regimes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15696, https://doi.org/10.5194/egusphere-egu26-15696, 2026.
Fires ignited by human and lightning occur at distinct environments and thus diverge during their developing processes. A global characterization of fires by their ignition cause will inform fire forecast and prediction but is currently prohibited by a lack of ignition cause in global fire inventories. Here we develop a machine-learning classification system and ascribe the ignition cause of 65.17 million global, satellite-detected fire events during 2012-2024. According to this fire inventory, extratropical lightning fires exhibit longer duration, larger burned area and hotter flame, compared with human fires. Despite their contribution to only 2.4% of fire occurrence, lightning fires are responsible for 10.9% of extratropical burned area and 47.6% of that consumed by large fires over 100 km2. This disproportionate abundance of lightning fires in the regime of most severe burning is attributable to synchronized seasonality of lightning ignition and burning conditions, as well as their scarcer accessibility to firefighting practices. Due to their closer linkage to the elongating fire-favorable weather, extratropical lightning fires has elongated by about 0.24 days decade-1, outpacing human fires. With projected hotter, dryer, and stormier extratropical summers, our results provide a direct support for a future of severer lightning fires.
How to cite: Su, H., Qiu, K., Yu, Y., Tang, Y., Wang, S., Meng, X., and Guo, W.: Extratropical lightning fires burn increasingly more severe than human-ignited fires, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2363, https://doi.org/10.5194/egusphere-egu26-2363, 2026.
An enduring heatwave over the Iberian Peninsula in July and August 2025 led to exceptionally extensive wildfires, resulting in the fifth-largest burned area in Spain since 2001 and the fourth-largest in Portugal. Hot and dry fire weather conditions are a key driver of large wildfires in the Mediterranean, and are intensifying rapidly under anthropogenic climate change. However, strong interannual variability of burned area and changes in multiple non-climatic drivers (e.g., land management) complicate the attribution of individual fire seasons.
Methods for attributing climate impacts to anthropogenic forcing are commonly divided into statistical and storyline approaches. Statistical methods quantify changes in the probability of exceeding predefined thresholds across climate states with different forcing levels, whereas storyline approaches examine how a specific historical event might have unfolded in the absence of anthropogenic climate change. Such counterfactual storylines can be generated with Earth system models (ESMs) constrained by observed historical conditions, enabling a process-based interpretation of climate impacts. However, this type of storyline method has not been applied to the attribution of complex ecosystem processes such as fires.
Here, we use the 2025 Iberian wildfire season as a case study to evaluate our nudged circulation storyline simulation with the Community Earth System Model 2 (CESM2) and compare it to statistical attribution. The ESM-based storyline enables a process-based quantification of thermodynamic influences on fire weather and of biological factors controlling fuel load. However, the approaches differ on the role of thermodynamic climate change in intensifying the 2025 fire season. Statistical attribution suggests a large thermodynamic contribution but indicates that events of comparable intensity are not exceptional under present-day climate. In contrast, the storyline approach identifies the 2025 circulation anomaly as unprecedented in magnitude. We show that this discrepancy arises from decadal Mediterranean circulation trends, which are implicitly absorbed into the thermodynamic response in the statistical attribution framework.
Our results demonstrate the utility of a storyline framework in the causal attribution of complex processes such as fires, and highlight the need for caution when applying attribution methods in regions characterized by strong dynamical trends.
How to cite: Dunkl, I., Mindlin, J., Turco, M., and Sippel, S.: Process-Based Attribution of the 2025 Iberian Wildfire Season Using a Storyline Framework, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20208, https://doi.org/10.5194/egusphere-egu26-20208, 2026.
Fine fuel loads ignite easily because they dry rapidly and are therefore an important driver of wildfire occurrence and spread. Accurate modelling of fine fuel load dynamics is crucial not only for current and future wildfire prediction, but also carbon cycling. Current fire-enabled dynamic global vegetation models simulate fine fuel accumulation and decomposition, but using parameters that vary with plant functional types (PFTs). Observationally derived models from satellite products provide good estimates of fine fuel loads but cannot be used to predict how these will change in response to ongoing climate changes. We have combined an eco-evolutionary modelling approach to simulate litterfall with a simple empirical model of decomposition rate to predict fine litter loads. The litterfall model predicts the amount of leaf mass that is shed using leaf economics principles and predictions of optimal leaf area index to predict litterfall for evergreen and broadleaf trees and C3 and C4 grasses. The model of decomposition rate uses a generalised linear mixed model to fit a large available dataset of decomposition rate to three variables: C:N ratio representing the litter quality and growing degree days and dry days representing local climate. Both models were independently validated using field observations collated from the literature. We show that the combined model predicts the spatial and temporal variation in fine fuel loads reasonably well when compared to field observations and existing products. This new approach provides a robust framework to derive environmentally driven changes in fine fuel loads in the context of prognostic modelling of wildfires.
How to cite: Cain, S., Zhou, B., Prentice, I. C., and Harrison, S. P.: Annual Litter Fuel Load Estimation from Optimality-Derived Litterfall and Decomposition Dynamics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8017, https://doi.org/10.5194/egusphere-egu26-8017, 2026.
Prescribed fire is the intentional use of fire under specific environmental conditions used to achieve specific land management objectives. Across the European Mediterranean basin, it is used for hazardous fuel reduction, pastoralism, habitat restoration, and silviculture. However, the ability to conduct prescribed burns is limited by meteorological conditions that facilitate the desired fire behavior to achieve the burns’ objectives, or the “burning window”. Under climate change, the continued availability of these conditions is highly uncertain as changes in the frequency and timing of these conditions are expected to occur. This presents a major challenge to future fire management planning. Here, we use projections of future climate based on scaling factors derived from the Coupled Model Intercomparison Project (CMIP6) and applied to ERA5 meteorology to quantify future changes in days suitable for prescribed burns (RxB days) across Mediterranean Europe. We find a 14% (-12 days) decrease in the number of RxB days across the region at a global warming level of 3.0°C, with losses most pronounced from April to October, particularly at the end of the spring burning window (May-June) and the beginning of the fall burning window (September-October). While some regions see an increase in winter burn days, these gains are outweighed by reduced burn days throughout the year. Future reductions in burn days were limited to 5% at 1.5°C, consistent with the commitments made in the Paris Agreement. Our results suggest that fire managers can expect a decline in opportunities to conduct prescribed burns, especially under higher warming scenarios. Thus, its continued use under these conditions will likely require significant investments and changes to current fire management policies to utilize and scale up remaining prescribed burning opportunities.
How to cite: Hsu, A., Abatzoglou, J., Fernandes, P., Ascoli, D., Clarke, H., Santin, C., Turco, M., Kolden, C., Patino, J. F., Rigolot, E., Carmenta, R., and Jones, M.: Prescribed Fire Opportunities in the European Mediterranean under Climate Change, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5864, https://doi.org/10.5194/egusphere-egu26-5864, 2026.
Wildfires are increasing in severity and harm to humans, creating a pressing climate change adaptation challenge. Current firefighting-focused management approaches in the Global North can drive fuel accumulation and increased fire intensity. In contrast, Indigenous peoples and local communities have used controlled burning to successfully co-exist with fire for at least 50,000 years. Consequently, intentional and controlled burning of landscape vegetation has been suggested as a strategy to adapt to climate-altered fire regimes in an approach known as Integrated Fire Management.
Here, we present the first quantitative global assessment of controlled burning in Integrated Fire Management (IFM) for climate change adaptation, using the JULES-INFERNO dynamic global vegetation model coupled to the WHAM! agent-based model of human fire use and management. WHAM! has agent types to represent both fire exclusionary, suppression-oriented and pyro-inclusive, controlled burning (IFM) land manager approaches. This new online coupling includes novel representations of human fire use seasonality and fireline intensity. Modelled fireline intensity, accounting for climate, fuel and human management now drives fire-induced vegetation mortality in JULES. Hence, the WHAM-JULES-INFERNO ensemble can assess the human and climate drivers of future fire intensity, and also fire-vegetation feedbacks resulting from contrasting management approaches.
We explored two Shared Socio-Economic Pathways (SSP1.26 and SSP3.70), using gridded socio-economic capitals consistent with the SSP scenarios and biophysical forcings from three ISIMIP 3b ESMs. Additionally, we drew on WHAM! functionality to complement the SSPs scenarios with two IFM scenarios: “IFM-max”, in which the world turns increases controlled burning through IFM; and “Suppression-max”, in which IFM is abandoned and the world focuses on fire exclusion and suppression.
We find that IFM can play an important role in constraining future fire hazard and intensity. However, we also identify barriers and confounding factors that may limit implementation. Notably, even in a low emissions-scenario (SSP1.26) with increased adoption of IFM, fire hazard is still 40.0% [32.1%-49.6%] higher in 2100 than in 2015. Importantly, we find that the impact of IFM is smaller than general land management changes resulting from economic conditions of the SSPs. For example, for both IFM scenarios mean 2100 fire intensity is higher in SSP1.26 than SSP3.70 because changes in fire management do not offset increases in intensity due to reduced human fire use between the SSPs. Specifically, in SSP1, rapid economic growth in low-income countries (e.g. sub-Saharan Africa) sees fire use in agriculture and forestry increasingly replaced by chemicals and machinery.
Our results suggest that incremental changes in land and fire management may be an insufficient response to the combined impacts of socio-economic and climate change. Transformative approaches that change fundamental relationships between economic development and fire suppression could in principle address this adaptation shortfall, but will need to grapple with how to integrate and maintain low-intensity fire in capital-intensive land systems on an increasingly flammable planet.
How to cite: Perkins, O., Kasoar, M., Haas, O., Smith, C., Teixeira, J., Voulgarakis, A., Mistry, J., and Millington, J.: Benefits and limits of Integrated Fire Management for climate change adaptation: a global quantitative assessment , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13135, https://doi.org/10.5194/egusphere-egu26-13135, 2026.
Forest management at the landscape scale is increasingly regarded as a key instrument for maintaining and improving the supply of multiple forest ecosystem services. Contemporary policy agendas, including the EU Forest and Biodiversity Strategies for 2030, together with management paradigms such as sustainable forest management, closer-to-nature forestry and rewilding, promote markedly different pathways. Some approaches rely on targeted silvicultural interventions, while others emphasise non-intervention and natural dynamics. Despite their growing relevance, the spatial prevalence of these contrasting strategies and their implications for ecosystem service provision at regional scales remain insufficiently explored. In this study, we assessed how alternative forest management trajectories affect ecosystem services across the entire forested landscape of the Piedmont region (Italy). Drawing on information from regional forest management plans, we categorised planned management into two broad classes: active management, encompassing silvicultural interventions of varying intensity, and passive management, characterised by the absence of direct interventions. We quantified the spatial extent of each management type and analysed their relationships with three key ecosystem services—carbon storage, fire hazard reduction and tree-species diversity—using principal component analysis and generalised linear models. Additionally, we investigated the association between management strategies and Protected Areas, and whether protection status modulates ecosystem service outcomes. Our results indicate that approximately 60% of Piedmont’s forests are designated for active management, although actual implementation is increasingly constrained by widespread forest abandonment. Active management was consistently associated with higher levels of carbon stocks, reduced fire hazard and greater tree-species diversity. Protected Areas were more frequently linked to passive management, yet their contribution to enhancing ecosystem services appeared limited. Based on these findings, we highlight the importance of: (i) reactivating forest management in abandoned areas, (ii) prioritising active management strategies to strengthen ecosystem service delivery, and (iii) using currently unprotected, passively managed forests as strategic candidates for expanding the Protected Area network, in line with EU2030 policy objectives.
How to cite: Spadoni, G. L., V. Moris, J., Kirschner, J., de Miguel, S., Oliveras Menor, I., Passamani, C., Le Moguedec, G., Ascoli, D., and Motta, R.: How forest management, land abandonment, and protected areas affect wildfire occurrence, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18182, https://doi.org/10.5194/egusphere-egu26-18182, 2026.
Variability in cloud droplet number concentrations (Nd) within the large subtropical stratocumulus decks can strongly impact outgoing shortwave radiation. The southeast Atlantic subtropical stratocumulus deck is particularly prone to elevated Nd, attributed to continental African fire emissions. The highest stratocumulus Nd occur when Angolan agricultural fires coincide with weak surface warming during the austral winter months (June-early August). Dry convection fills a shallow continental boundary layer with smoke and a nighttime land breeze advects the aerosol into or slightly above the marine boundary layer. The offshore transport is strengthened by low-level easterlies from a continental high to the southeast of Angola that is stronger when the Angolan land is cooler. Simultaneously, the south Atlantic subtropical high (SASH) is weaker when Angolan land warming is more muted, allowing the biomass-burning aerosol to also disperse further south. The shortwave-absorbing aerosol can either reach the remote boundary layer by direct low-lying easterly transport, or through entrainment over longer time scales after being transported south. While the weak Angolan land heating in June-July correlates with higher offshore Nd, these coincide with lower cloud fractions and thinner clouds, primarily because the SASH is also weaker. This meteorological co-variation fully compensates for any aerosol brightening of the cloud deck. Marine cloud brightening by emissions from a southeast Atlantic shipping lane is more evident when Angolan land heating is stronger, coinciding with a stronger SASH, as the background Nd is less and the background cloud fraction is higher. Most of the year-to-year variability from 2003 to 2023 in the June-July marine shortwave cloud radiative effect can be constrained using the surface-level temperature over Angola (r2 = 0.4). While Angolan land has warmed slightly in June-July since 1980 in reanalysis, no trend is evident in synoptic variations of warmer versus cooler heating. Fire emissions have slightly increased since 2003. A continuing warming trend would deepen the continental boundary layer, and could place more of the transported smoke above the marine boundary layer, stabilizing the lower atmosphere through shortwave absorption.
How to cite: Zuidema, P. and Tatro, T.: Weak, low-level dry convection over Angola determines biomass-burning aerosol entry into the marine boundary layer, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8759, https://doi.org/10.5194/egusphere-egu26-8759, 2026.
Wildfire smoke is an increasingly important driver of regional air-quality degradation, with well-established impacts on public health and visibility. Although emission controls have reduced many anthropogenic air pollutants over recent decades, wildfire activity has intensified in many regions, increasing the contribution of fine particulate matter (PM2.5; aerodynamic diameter < 2.5 μm) to surface pollution episodes. A key limitation in simulating wildfire smoke in chemical transport models is uncertainty in biomass-burning emissions, as inventories can have different mythologies and assumptions, such as fire occurrence, intensity, burn area, fuel characterization, and emission factors. These discrepancies can translate into substantial variability in modeled PM2.5 and related co-emitted species, complicating both forecasting and attribution of smoke impacts. Here, we implement and evaluate the Global Forest Fire Emissions Prediction System (GFFEPS), a wildfire emissions framework developed by Environment and Climate Change Canada (ECCC), within the GEOS-Chem chemical transport model. We perform simulations for Canada, the United States, and Europe in 2019, and for Australia in 2019–2020, to quantify the sensitivity of simulated smoke to fire emissions and to assess model skill against observations. GFFEPS-driven simulations are compared with those using widely applied global biomass-burning inventories (the Global Fire Emissions Database (GFED), the Global Fire Assimilation System (GFAS), and the Quick Fire Emissions Dataset (QFED2)) and evaluated using ground-based PM2.5 monitoring data across each region. Inventory choice strongly influences both the magnitude and timing of simulated PM2.5 enhancements, with clear regional dependence and the largest inter-inventory spread during extreme fire events. Over North America, GFFEPS shows the best overall performance among the four inventories based on the mean error metric. Over Australia, GFFEPS generally underestimates PM2.5 concentrations but remains a strong performer, ranking second behind GFAS using the same evaluation metric. Over Europe, GFFEPS ranks third, following GFAS and GFED, and is closely comparable to QFED2. These results highlight the need to better constrain fire detection and fuel consumption estimates, and demonstrate the value of GFFEPS within GEOS-Chem for diagnosing key drivers of inter-inventory differences and improving confidence in regional wildfire smoke simulations.
How to cite: Payette, T., Ashraf, S., Hayes, P., and Chen, J.: Integrating the Global Forest Fire Emissions Prediction System version 1.0 to GEOS-Chem , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11423, https://doi.org/10.5194/egusphere-egu26-11423, 2026.
Fire is a global phenomenon and a key Earth system process. Extreme fire events have increased in recent years, and fire frequency and intensity are projected to rise across most regions and biomes, posing substantial challenges for ecosystems, the carbon cycle, and society. The Fire Model Intercomparison Project (FireMIP), launched in 2014, has contributed to advancing global fire modeling in Dynamic Global Vegetation Models (DGVMs) and improving understanding of fire's local drivers and local impacts on vegetation and land carbon budgets through land offline (i.e., uncoupled from the atmosphere) simulations. We now bring FireMIP into Coupled Model Intercomparison Project Phase 7 (CMIP7) to: (1) evaluate fire simulations in state-of-the-art fully coupled Earth system models (ESMs); (2) assess fire regime changes in the past, present, and future, and identify their primary natural and anthropogenic forcings and causal pathways within the Earth system, including the associated uncertainties; and (3) quantify the impacts of fires and fire changes on climate, ecosystems, and society across Earth system components, regions, and timescales, and elucidate the underlying mechanisms. FireMIP in CMIP7 will advance the fire and fire-related modeling in fully coupled ESMs, and provide a quantitative, detailed, and process-based understanding of fire's role in the Earth system by using models that incorporate critical climate feedbacks and multi-model, multi-initial-condition, and CMIP7 multi-scenario ensembles. Here, we presents the motivation, scientific questions, experimental design and its rationale, model inputs and outputs, and the analysis framework for FireMIP in CMIP7, providing guidance for Earth system modeling teams conducting simulations and informing communities studying fire, climate change, and climate solutions.
How to cite: Li, F. and the CMIP7 FireMIP group: The Fire Modeling Intercomparison Project (FireMIP) for CMIP7, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4207, https://doi.org/10.5194/egusphere-egu26-4207, 2026.
The late dry season of 2019 featured one of the most severe Indonesian wildfire events of the past decade, driven by persistent drought and extensive peatland burning. These extreme wildfires emitted large amounts of carbonaceous aerosols, substantially degrading air quality and posing risks to human health. However, the impacts of extreme wildfire events on black carbon (BC) across Southeast Asia remain poorly quantified. Here, we evaluate the influence of Indonesian wildfires during August–October 2019 using the GEOS-Chem chemical transport model at 0.25° × 0.3125° resolution. Sensitivity simulations with and without Indonesian fire emissions are conducted to isolate fire-driven contributions to BC. Results indicate dominant wildfire control over BC across Southeast Asia. Fire contributions reach about 91% over both Borneo and Sumatra during peak burning. Comparable fire influence extends to nearby seas, particularly the South China Sea, with contributions exceeding 90% over the southern South China Sea. Contributions remain near 70% over the Sulu and Celebes Seas and still reach about 50% over the Philippine Sea. In contrast, impacts over the East China Sea are episodic, occurring only during short-lived northeastward outflow events. These findings demonstrate the strong and spatially heterogeneous influence of Indonesian wildfires on regional BC across Southeast Asia, highlighting the role of extreme wildfire events in shaping air quality through fire-driven transboundary transport.
How to cite: Zheng, H.: Impacts of the 2019 extreme Indonesian wildfires on black carbon across Southeast Asia, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9088, https://doi.org/10.5194/egusphere-egu26-9088, 2026.
Fire Radiative Power (FRP) is a quantitative measure of the instantaneous rate of radiant heat energy emitted by a fire during the combustion process. It is usually retrieved via satellite remote sensing and serves as a key indicator of fire intensity and the rate of fuel consumption. FRP is generally estimated by measuring the thermal radiation (radiances) emitted by wildfires, in the Middle Infrared (MIR) spectral range (3.9- 4.0) where the Planck function peaks for sources at 1000K and the contrast between the fire and the cooler background is most pronounced.
A number of satellite imaging systems, at LEO (i.e. MODIS-TERRA and AQUA, VIIRS-Suomi NPP, SLSTR-Sentinel 3A and 3B) and GEO (i.e. SEVIRI-MSG, ABI-GOES, ABI-HIMAWARI) orbits provide FRP retrievals. However, due to their coarse spatial resolution (1-2 km/px) and wide spectral bands, small fires detection and associated FRP retrieval is limited, representing a potential source of omission error.
While currently available high-resolution sensors lack coverage in the Mid-Infrared (MIR) spectral range, recent research has investigated the potential of Short-Wave Infrared (SWIR) sensors as an option. By analyzing airborne data from the AVIRIS, EMAS, and MASTER sensors, studies have established a robust correlation between MIR-derived and SWIR-derived FRP. Furthermore, the SWIR band on Sentinel-3 is already being effectively utilized to estimate FRP for gas flares monitoring.
In this study we retrieve FRP by using two similar hyperspectral sensors, Precursore IperSpettrale della Missione Applicativa (PRISMA) and Environmental Mapping and Analysis Program (EnMAP). We compare the results with operational FRP products, namely the Sentinel-3 L2 NRT FRP and the CAMS-GOES-W FRP product and evaluate potentials and limitations for mapping the intensity of wildfires and gas flares.
How to cite: Amici, S. and Mota, B.: Critical analysis of Fire Radiaive Power derived by hyperspectral sensors from space, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19257, https://doi.org/10.5194/egusphere-egu26-19257, 2026.
Changing climatic conditions are amplifying the frequency and intensity of hydroclimatic extremes across Europe. Droughts, heatwaves, intense precipitation and floods increasingly co-occur and cascade, creating compound risks for ecosystems and societies. One of the most visible and severe consequences of these interconnected crises is the growing global threat of forest fires, which are more often facilitated by favorable weather conditions, as well as forest structure and fuel properties. However, the most important cause of fires is related to human pressure, resulting from intentional or unintentional activities that contribute to the outbreak of fires.
Forests are an assemblage of diverse habitats, each of which may differ markedly in fire risk and fire behaviour. Here, we examine how fire occurrence in Poland varies among forest habitat types, land-use patterns and management functions, and how these relationships are shaped by interannual meteorological variability and regional context. We compile (i) forest fire records for Poland for 2019-2024, (ii) a 2024 state forest administration database of forest divisions (i.e., basic forest management units) including habitat type, dominant tree species and main forest function, (iii) a database of socio-economic indicators for country's administrative units, and (iv) annual meteorological characteristics relevant to fire weather. This enables a spatially explicit analysis of fire frequency and (where available) burnt area across heterogeneous forest landscapes, while accounting for administrative-region differences and socio-economic factors that may reflect contrasting management practices, accessibility, and human ignition pressure.
We quantify fire occurrences in 2019-2024 for distinct forest area types (classified by habitat, dominant tree species and function) and evaluate their sensitivity to meteorological conditions across years. The analysis is designed to identify which combinations of forest habitat, tree species, forest function, and local socio-economic structure show consistently elevated fire incidence, whether observed changes between 2019 and 2024 are uniform or regionally differentiated across Poland, and to determine which meteorological characteristics best explain interannual variability in forest fire occurrence. By integrating ecological and forest management attributes with fire records and meteorological context, the study provides an empirical basis for stratified fire-risk assessment in Polish forests and supports targeted prevention and management measures. This research is conducted as part of the NCN project 2023/49/N/ST10/04035 "Fire, burnt area and charcoal - charcoal-data modeling of burnt area, cross-validation of fires and charcoal signal".
How to cite: Kowalczyk, P., Zin, E., Tyburski, Ł., Śleszyński, P., Słowińska, S., Kaszkiel, A., Czubak, D., Klisz, M., Pilch, K., Kaczmarowski, J., and Słowiński, M.: Dimensions of forest fires in Poland, 2019-2024, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22880, https://doi.org/10.5194/egusphere-egu26-22880, 2026.
The Fire INventory from NCAR (FINN) is a daily, high resolution (1 km) fire emissions inventory designed for use in atmospheric chemistry models. FINN uses a ‘bottom-up’ approach to estimate fire emissions. Satellite observations of active fires from MODIS (and VIIRS) are combined with land cover, emission factors and fuel loadings to predict fire emissions of key air pollutants. However, one of the key limitations of FINN is lack of peat fire emissions in the dataset, which only accounts above ground vegetation fires. Therefore, neglecting an important emissions source given the extensive abundance of peat in key tropical regions. Fires that occur on the surface of peatland can burn into the below-ground organic layers (up to 0.6 m). Peat fires can smoulder for weeks after the surface fire has extinguished, resulting in substantially greater emissions compared to surface vegetation fires. Therefore, it is essential to include peat fires in FINN.
Globally, peatlands cover >4 million km2 (3 %) of the global land area. However, emissions from the combustion of tropical and Arctic-boreal peat alone account for a disproportionately large fraction of total global carbon emissions (13 %). This is driven by above ground fires burning into the carbon rich peat below.
We first focus on tropical peatlands in Indonesia since these have well documented impacts on air quality. Indonesia is home to a large proportion (36 %) of total tropical peatlands, and a large fraction of fires in Indonesia occur on peatlands. For example, in 2015 53 % of fires in Indonesia occurred on peatland, accounting for only 12 % of the land area. Peat fires contributed 71-95 % of the particulate matter (PM2.5) fire emissions, though emissions are uncertain.
Our work builds upon previous work, which estimated Indonesian peat fire emissions for FINN. Previously, satellite-derived soil moisture was used to determine a straightforward linear relationship with burn depth of fires that occurred on peatlands. We further develop this method adding additional complexity by using ground-based measurements of burn depth collocated with satellite soil moisture. We also consider canal density and fire frequency maps to account for changes in burn depth with drainage and fire history.
We plan to apply this method to other tropical peatland and boreal regions, so we welcome any discussions on our current work so far and/or future plans.
How to cite: Graham, A. M., Pope, R. J., and Chipperfield, M. P.: Accounting for peat fires in the Fire INventory from NCAR (FINN): Improved air pollutant emissions estimates for tropical peatlands using soil moisture, drainage density and fire history., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5608, https://doi.org/10.5194/egusphere-egu26-5608, 2026.
The current methods to systematically validate Earth Observation (EO) products capturing transitory events such as fire activity rely mostly on the intercomparison between Near-real-time products without clearly identifying one as the reference dataset. In addition, due to the highly dynamic and ephemeral nature of such events, comparisons are restricted to near-simultaneous measurements which significantly limits the sample size of any intercomparison. In this study, we propose a new comparison framework that overcomes these limitations. This novel approach is based on a robust analysis of the frequency density (f-D) distributions of each product’s assessment of the event. We start by defining the concepts associated for distribution fitting and performance, temporal and spatial requirements, comparison metrics, and then provide an overview of the various sources of uncertainty contributing to the intercomparison exercise, and how and what uncertainties are propagated.
In this study we inter-compare eight operational remotely sensed active fire detections and fire radiative power (FRP) retrieval products: the polar-orbiter products derived from active fires detected using the Moderate Resolution Imaging Spectroradiometer data (MCD14ML), the Visible Infrared Imaging Radiometer Suite (VNP14IMGML), and the Sea and Land Surface Temperature Radiometer (SLSTR) Non-time critical product from European Space Agency (SLSTR-NTC), and the geostationary products derived from data collected by Meteosat’s Spinning Enhanced Visible and Infrared Imager (LSA-SAF FRP-PIXEL), and the three available products based on Advanced Baseline Imager (KCL/IPMA-GOES16, KCL/IPMA-GOES17, and KCL/IPMA-Himawari). We focus on annual detections and perform the analysis at 0.5° grid cell resolution, for the overlapping product’s time-series. The results are analysed for their temporal and spatial consistency, and inter-product differences are analysed in the context the product’s metadata.
The results show that an Inverse-gamma distribution can be used to characterize the fire ‘statistical signature’ and provide a reference baseline on to which all FRP products can be compared to, and their ‘representation uncertainty’ assessed. Individually, the fitting results show the degree of under representation of each sensor’s detections, namely the identification of minimum FRP detection limit, which typically precludes the detection of a proportion of the highly numerous but individually relatively small and/or low intensity fires. Furthermore, inter-comparison differences allowed for the identification, and assess the impact, of some of the key non-fire effects such as: pixel size, off-nadir pixel area growth, algorithm limitations, quality information, and the impacts of low temporal resolution of polar-orbiting sensors.
This proposed framework is a useful tool to compare EO-based FRP products and transferable to any product measuring transitory event properties that do not rely on simultaneous observations. It complements existent comparison exercises by identifying additional sources of uncertainty, the conditions under which these occur and how these translate into product inconsistencies. It is an essential tool, providing users with product-specific information on measurement limitations that, in principle, can be corrected and assimilated to higher level products and downstream applications such as GHG emission estimates from biomass burning, providing better quality information used for adaptation and mitigation policies.
How to cite: Mota, B.: Validation framework for EO measurements of transitory events based on robust statistics retrieved from non-simultaneous observations: A case study applied to Fire Radiative Power (FRP) products. , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8037, https://doi.org/10.5194/egusphere-egu26-8037, 2026.
South America (SA) has suffered a multitude of extreme, drought-induced fires in recent years, including 2024 which saw fire emissions across the continent 263 Tg C (84%) above average[1]. Burned area and fire carbon emissions in SA are projected to increase over the coming decades due to higher temperatures and drier conditions associated with climate change[2]. These effects are already being seen in the Amazon, where fire is driving the rainforest towards being a net carbon source[3] and threatening existential climate tipping points. Meanwhile to the South, the ecologically diverse Pantanal wetlands have undergone a step-change in wildfire activity, with 2019–2021 experiencing a 408% increase in annual carbon monoxide (CO) emissions relative to the 2013–2018 average.
CO is a major trace gas released from fires. Its emissions can be used to quantify wildfire carbon impacts and investigate correlations between fire activity and global climate indices. Despite this, there remains considerable disagreement between fire inventory products, with mean annual CO emissions ranging from 284–625 Tg yr-1 globally, and predictions diverging further at smaller spatial scales. These large uncertainties originate from the underlying assumptions of the inventory methodologies and the imperfect sensitivity of their satellite data inputs. Satellite observations of atmospheric total column CO, combined with inverse modelling techniques, provide a direct, top-down method to constrain these estimates, allowing more accurate CO emissions to be determined.
Here, we derive fire emission estimates between 2019–2024 for SA using the INVICAT 4D-Var inverse chemical transport model, assimilating TROPOspheric Monitoring Instrument (TROPOMI) total column CO satellite observations into the model for the first time. Six fire inventories (GFEDv4.1s, GFEDv5.1, GFASv1.2, QFEDv2.6r1, FINNv1.5, FINNv2.5) are used as priors in separate CO inversions, from which posterior result sensitivity is quantified and prior biases are assessed. We use emission ratios to determine, spatially and temporally, the total carbon flux into the atmosphere from fires in SA. We find that the 2024 extreme fire season in SA is poorly captured by the fire inventory products currently available, with peak atmospheric CO over SA observed to be 12.8 Tg (46%) larger than forward-modelled inventory emissions predict. Additionally, we create a multilinear regression model to predict the spatial distribution of CO anomalies across tropical SA by correlating the inversion posterior emissions to key global climate indices at various lag times. This novel method can provide spatial forecasts of the wildfire vulnerability arising from the global state of the climate months in advance.
[1] Kelley et al., 2025, State of Wildfires 2024–2025, Earth System Science Data
[2] Burton et al., 2021, South American fires and their impacts on ecosystems increase with continued emissions, Climate Resilience and Sustainability
[3] Basso et al., 2022, Atmospheric CO2 inversion reveals the Amazon as a minor carbon source caused by fire emissions, with forest uptake offsetting about half of these emissions, Atmospheric Chemistry and Physics
How to cite: Bradley, B., Wilson, C., Chipperfield, M., Reddington, C., Graham, A., and O'Connor, F.: Top-down carbon monoxide fire emissions over South America correlated with global climate indices, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18115, https://doi.org/10.5194/egusphere-egu26-18115, 2026.
Tropical montane forests have historically not been prone to large-scale forest fires as a result of their high humidity and rainfall. Yet, current increased frequencies and intensity of these fires are making them an increasingly pressing area of study, especially in the context of increasing climate variability and land use changes. To understand present and future fire dynamics, it is, however, essential to look at the origins and factors behind trends in fire frequency and intensity. To explore this, long-term assessment of the dynamics of montane forest fire, and their relationships to anthropogenic and climate changes, are essential.
Such work has been largely lacking in montane ecosystems due to a paucity of available quantitative data, and a general perception that fire has played a minimal role in shaping biodiversity in these areas. Here we combine historical forest fire records and remote sensing to investigate the evolution and dynamics of montane forest fires in Kenya since the 1920s in response to changes in forest fire management, land use changes and climate variability. We argue that historically, indigenous communities used their traditional knowledge and practices in managing local fires and limiting them to manageable intensities. However, the introduction of colonial rule shifted their role in forest management and ultimately their relationship in using fire within forest areas.
Our research and datasets highlight that changes in fire dynamics can be linked to extensive colonial prohibition of fire controls by traditional communities and the imposition of fines to deter their use. In addition, introduction of new fire sources through the development of the railway systems along forest areas, introduction of exotic tree species and largescale agricultural expansions exacerbated forest fire dynamics within the montane forests. Meanwhile, the colonial government introduced fire lines as a form of forest fire controls, which were meant as fire control measure and required sophisticated management plans, that were adopted in forest management.
We suggest that these changes have left legacies for contemporary fire issues as a loss of traditional fire management knowledge, smallholder relocation and land restrictions, and industrial pressures have accumulated to intensify fire risk in montane forest ecosystems. Looking into the future, we argue that, as with other regions of the latitudinal tropics, it is essential to understand traditional ecological knowledge and historical path dependencies in order to chart more effective and just conservation strategies including active use of fire and restoration of fire-resistant species.
How to cite: Gitau, P., Kinyanjui, R., and Roberts, P.: Evolution of tropical montane forest fires in response to shifts in historical forest management, climate variability and land use changes., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20289, https://doi.org/10.5194/egusphere-egu26-20289, 2026.
European wildfire response systems are increasingly challenged by the simultaneous demand for aerial and ground suppression assets. If major fire-prone regions burn asynchronously, Europe could benefit from risk-diversified deployment of shared suppression fleets and more efficient cross-border mutual-aid strategies. We test the hypothesis that fire activity in (1) the Iberian Peninsula (Portugal and Spain) and (2) Central–Eastern Mediterranean (Italy and Greece) exhibits identifiable and temporally stable dependence patterns modulated by large-scale climate variability.
Annual burned-area time series covering 1980–2023 are compiled from the European Forest Fire Information System (EFFIS). These are complemented by satellite-derived indicators of fire activity from MODIS, namely Fire Radiative Power (FRP), enabling joint assessment of burned area extent and fire intensity. Climate-fires’ dependence is quantified through correlations of annual and seasonal anomalies and joint-extreme metrics focused on tail co-exceedance probability. The relationship between fire activity (burned area, FRP, FRE) and large-scale climate variability is assessed following established teleconnection-based frameworks, combining seasonal aggregation, lagged cross-correlation analysis, and composite analysis of extreme fire years. Teleconnection indices considered include the North Atlantic Oscillation (NAO), East Atlantic pattern (EA), Mediterranean Oscillation Index (MOI), Arctic Oscillation (AO), and ENSO. Analyses explicitly account for the non-stationary and scale-dependent nature of teleconnection–fire relationships, and are conditioned on regional temperature and precipitation anomalies to isolate circulation-driven effects.
The analysis aims to identify: (i) the frequency and persistence of synchrony versus compensatory (negative) dependence in burned area and fire activity between the two macro-regions, (ii) the teleconnections most strongly associated with synchronous extreme fire seasons, and (iii) multi-decadal periods offering potential for suppression-fleet diversification. Owing to its direct control on Mediterranean-scale pressure gradients and precipitation contrasts, MOI provides the primary explanatory signal for synchronous versus compensatory fire activity between the two macro-regions.
Results are interpreted within an operational risk-pooling framework, where weak or negative dependence supports climate-informed scheduling of shared European suppression fleets and enhanced cross-border mutual aid, while strong positive dependence indicates heightened likelihood of concurrent continental-scale resource strain.
This work is partially supported by FCT, I.P./MCTES through national funds (PIDDAC): LA/P/0068/2020 – https://doi.org/10.54499/LA/P/0068/2020, UID/50019/2025 – https://doi.org/10.54499/UID/PRR/50019/2025, UID/PRR2/50019/2025 and Dhefeus (https://doi.org/10.54499/2022.09185.PTDC). AR, JMCP and JMNS also thank the FCT by supporting UIDB/00239/2020 (https://doi.org/10.54499/UIDB/00239/2020), UIDP/00239/2020 (https://doi.org/10.54499/UIDP/00239/2020), and through project references UIDB/00239/2020 (https://doi.org/10.54499/UIDB/00239/2020) and UIDP/00239/2020 (https://doi.org/10.54499/UIDP/00239/2020) and European Space Agency Climate Change Initiative (ESA-CCI9 Tipping Elements SIRENE project (Contract No. 4000146954/24/I-LR).
How to cite: Russo, A., Gouveia, C., Bento, V., Silva, J. M. N., DaCamara, C., Trigo, R. M., and Pereira, J. M. C.: How asynchronous is fire burning in Iberia and the Central–Eastern Mediterranean? A dependence analysis of burned area, fire activity, and teleconnection forcing to inform shared European suppression fleets, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9452, https://doi.org/10.5194/egusphere-egu26-9452, 2026.
The 2023 Canadian fire season was record-breaking in terms of burned area and carbon emissions. Yet, the climate impacts of these fires extend far beyond the immediate carbon emissions and can persist for decades. Post-fire changes in vegetation and surface properties prolong snow exposure during winter and spring, increasing surface albedo and producing long-lasting regional cooling impacts. Historically, the surface albedo-driven cooling has offset the warming influences of carbon emissions by boreal fires. However, with ongoing high-latitude warming, fire seasons are expected to become longer and more intense while spring snow cover declines. This combination may weaken the climate-cooling effect of post-fire surface-albedo changes and reduce the offset potential.
Here, we quantified and mapped the climate-cooling effects from post-fire surface albedo changes for the 2023 Canadian fire season under shared socioeconomic pathway SSP2-4.5 for a 70-year period. We estimate that the 2023 Canadian fires resulted in a time-integrated climate-cooling of –3.67 W m-2 of burned area (95% CI: −4.83 to −2.51) over a 70-year period. Our analysis further shows that the climate-cooling impact of boreal fires has weakened by approximately 30% due to changes in snow cover and duration. This has significant implications for the ability of albedo-driven cooling to offset warming from fire emissions. As a result, we conclude that contemporary boreal fires are, on average, twice as likely to result in a net climate-warming effect relative to the 1960s.
How to cite: van Gerrevink, M. J., Gonsamo, A., Rogers, B. M., Potter, S., Zhong, Z., and Veraverbeke, S.: Climate-cooling impacts from post-fire snow-albedo for the 2023 Canadian fires season, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3494, https://doi.org/10.5194/egusphere-egu26-3494, 2026.
The role of wildfire as a controller in wildland ecosystems is well researched. However, much uncertainty is present with climate change. Will this morphing force turn into a game breaker in the longevity of terrestrial ecosystems? Or will it continue its role as the ultimate controller of vegetation composition, and for certain taxa, fecundity? This study aims to answer these questions for a study region situated in the northern segment of Eastern Mediterranean Basin – Anatolian Peninsula and its immediate surroundings. It considers the historical and potential future changes in biomass and fuel capacity in the region with respect to the changes in amplitudes of climate variability due to climate change in two distinct time periods (present and future). Changes in fire severity and fire return interval (FRI) are simulated using a dynamic vegetation model (LPJ-GUESS v.4.1) coupled with wildfire modules (SIMFIRE and BLAZE), and high-resolution climate datasets. For 1961-2025, the model is forced with ERA5-Land reanalysis data, and for 1961-2100, an ensemble of 5 CMIP6 datasets under the SSP 5-8.5 global warming scenario are used which are resized to 0.1°. While the historical trend analyses of the climate indices (such as SPEI) indicate strong drying for the region overall, simulation results signal an increase in burned area, the frequency of wildfire incidents, while highlighting important changes in vegetation composition and biomass under a changing climate, as wildfire turns into a response mechanism under increasing temperatures and changing rainfall patterns.
How to cite: Ekberzade, B. and Görüm, T.: Control or destroy: Wildfire as a response mechanism, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-377, https://doi.org/10.5194/egusphere-egu26-377, 2026.
The main objective of the Austria Fire Futures study is to develop a unique and innovative framework for fire risk assessment by producing high-resolution fire risk hotspot maps under multiple climate change scenarios. These maps integrate novel insights on local fuel types into forest and wildfire risk models, including mountain-specific variables such as topography, morphology, and recreational activities.
To generate fire risk information at the local scale, advanced fire hazard modeling is required to identify vulnerable forest types in combination with topographic effects. Recent wildfire events in the Austrian Alps have demonstrated that social factors—particularly hiking tourism—are currently underrepresented in fire risk assessments. In response, this study aims to advance fire risk hotspot mapping as a foundational element for forest and wildfire prevention. Such mapping is essential for integrated fire management, encompassing prevention, suppression, and post-fire measures, while contributing to climate change mitigation and minimizing impacts on ecosystems, ecosystem services, and human well-being.
We present modeling results from the Wildfire Climate Impacts and Adaptation Model (FLAM), a process-based fire risk model operating at a daily time step. FLAM employs machine learning techniques to calibrate extended suppression efficiency based on spatial segmentation of landscapes. Historical ground data on burned areas in Austria were used for model calibration and validation. The results include historical simulations (2001–2020) and future projections (2021–2100) of burned area across Austria at 1 km spatial resolution, based on an ensemble of downscaled climate change scenarios. In addition, FLAM was applied to Lower Austria at 250 m resolution, using the most recent high-resolution datasets on fuels, forest cover, human ignition probability, and response times.
The results improve our understanding of fire-vulnerable forest areas in the Alpine region and how these vulnerabilities may shift over time and space under changing climate and fuel conditions. This knowledge enables experts, practitioners, and the broader public to explore plausible future fire regimes and to derive robust short-, medium-, and long-term recommendations for fire-resilient and sustainable forest management, as well as for wildfire preparedness and emergency planning.
How to cite: Krasovskiy, A., Jo, H.-W., Vacik, H., Silva Andrade, M., Formayer, H., Laimighofer, J., Arnberger, A., Schadauer, T., Müller, M., Park, E., San-Pedro, J., and Kraxner, F.: Wildfire Hot Spot Mapping in the Alps - Austria Fire Futures, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19168, https://doi.org/10.5194/egusphere-egu26-19168, 2026.
Vegetation fires emit a wide variety of aerosol particles. Most originate from the combustion of carbonaceous material, however, fire-induced pyro-convective updrafts can modify the near-surface wind field in a way that mobilizes soil-dust particles from the ground and inject them into the atmosphere. Mineral dust particles are well known as efficient cloud condensation nuclei (CCN) and ice nucleating particles (INPs), thereby substantially altering cloud microphysics and influencing the Earth’s radiation budget through scattering and absorption of solar radiation. When emitted during wildfires, these dust particles are likely mixed with smoke aerosols, which modifies their physio-chemical properties and consequently their impacts on the atmosphere and climate. Therefore, a precise characterization of this emission pathway and robust knowledge of its global abundance are essential.
The fire-driven emission of soil-dust particles has already been incorporated into the global aerosol–climate model ICON-HAM through the development of a sophisticated parameterization that describes fire-induced dust emission fluxes as a function of fire intensity and some soil-surface properties, such as the soil type and the vegetation class at the fire location. Multi-year model simulations have indicated that fire-related dust emissions can account for a significant fraction of the global atmospheric dust load, exhibiting strong regional and seasonal variability driven by a varying fire activity and the local soil-surface conditions.
However, global fire activity has changed substantially over the recent decades due to both climatic and socioeconomic factors, resulting in significant shifts in the magnitude and regional distribution of fire-related dust emissions. Here, trends in fire-induced dust emissions over the past 20 years are analyzed and changes across different continental regions are contrasted. Furthermore, projections of fire activity under future climate scenarios can be used to assess the strength and regional distribution of fire-related dust emissions under changing climate conditions and mitigation strategies. This analysis can contribute to improved estimates of the future global aerosol burden, in particular with respect to the changing fire occurrence in a warmer world.
How to cite: Wagner, R. and Tegen, I.: Trends in fire activity and associated fire-induced soil-dust emissions over the last two decades, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18932, https://doi.org/10.5194/egusphere-egu26-18932, 2026.
Wildfires play a key role in the Earth system by shaping ecosystem dynamics and influencing the carbon cycle and atmospheric composition. Data-driven models have recently emerged as powerful tools for reproducing observed fire activity, particularly burned area, across a range of spatial and temporal scales. The first version of BuRNN (Burned area and emissions modelling through Recurrent Neural Networks) focused solely on burned area and outperformed all process-based fire-coupled DGVMs from ISIMIP over a wide range of spatial, temporal and spatio-temporal skill metrics. Here we present the 2nd version of BuRNN, a data-driven model that now jointly represents burned area and fire-related emissions.
How to cite: Lampe, S., Gudmundsson, L., Kraft, B., Hantson, S., Chuvieco, E., and Thiery, W.: Modelling burned area and emissions with deep learning, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19404, https://doi.org/10.5194/egusphere-egu26-19404, 2026.
Posters virtual: Thu, 7 May, 14:00–18:00 | vPoster spot 2
EGU26-14857 | ECS | Posters virtual | VPS6
Canada’s forests shifting from a recovery-driven carbon sink to a disturbance-driven carbon sourceThu, 07 May, 14:00–14:03 (CEST) vPoster spot 2
Canada’s terrestrial ecosystems play a critical role in the global carbon cycle and are being affected by unprecedented climate change and wildfire disturbance. However, we have an incomplete understanding of Canada’s historical carbon cycle. Existing assessments, conducted at varying spatial scales, use a wide range of data sources and methodologies, which lead to significant differences in the estimated strength of Canada’s land carbon sink over recent decades. Moreover, many approaches (e.g., inversions and data-driven estimates) have a limited ability to disentangle the relative contributions of different processes to the carbon sink over the recent past (1700 - 2022). We addressed this gap using a land surface model recently tailored to Canada and the most comprehensive information depicting wildfire disturbance and timber harvest available to make, to our knowledge, the first physically coherent wall-to-wall estimates of all major carbon pools and fluxes for Canada. We show that Canada’s terrestrial ecosystems have been a carbon sink since the mid-20th-century, due to the influence of wildfire and timber harvest before 1940. Since the early 2000s, wildfire disturbance has been driving Canadian forests towards becoming a carbon source. Based on our findings from a purely process-oriented perspective, projected increases in wildfire activity will further impact the strength and direction of Canada's terrestrial carbon sink.
How to cite: Curasi, S., Melton, J., Humphreys, E., Arora, V., Beaver, J., Cannon, A., Chen, J., Hermosilla, T., Lee, S.-C., and Wulder, M.: Canada’s forests shifting from a recovery-driven carbon sink to a disturbance-driven carbon source, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14857, https://doi.org/10.5194/egusphere-egu26-14857, 2026.
EGU26-7609 | ECS | Posters virtual | VPS6
Atmospheric and landscape controls on fire size in tropical dry forests: insights from the South American Gran ChacoThu, 07 May, 15:03–15:06 (CEST) vPoster spot 2
Fire is a dominant disturbance in tropical and subtropical dry forests and a major contributor to variability in carbon emissions, atmospheric composition, and land–atmosphere interactions. Despite their global extent and rapid transformation, the processes controlling fire size and extreme fire events in dry forest systems remain less understood than in savannas or humid tropical forests.
We investigated the controls on fire size using the South American Gran Chaco as a representative large-scale tropical dry forest system spanning strong climatic, ecological, and land-use gradients. We analyzed two decades (2001–2022) of satellite-derived fire patches from the FRY v2.0 burned-area database, combined with ERA5-Land meteorology and Fire Weather Index diagnostics, land-cover composition, landscape fragmentation metrics, topography, and anthropogenic pressure proxies. Our analysis focuses explicitly on fire size rather than fire occurrence, using statistical approaches and machine learning tools such as Random Forest models with SHAP-based interpretation to disentangle the relative and interacting roles of atmospheric forcing, landscape structure, and human-driven land transformation.
Our results show that fire size distributions are highly skewed across the region, with a small fraction of large and extreme events accounting for a disproportionately large share of total burned area. Wind and atmospheric dryness exert a strong influence on the final shape and size. At the same time, precipitation plays opposing roles by constraining fire spread through fuel moisture and enhancing fuel accumulation in fuel-limited environments. Landscape structure mediates the translation of meteorological extremes into large burned areas, with land-cover composition, fuel continuity, and fragmentation consistently ranking among the most influential predictors of burned area. Topography systematically emerges as the dominant predictor across subregions and seasons, acting not as a direct driver of fire spread but as an integrative proxy capturing hydrological gradients, vegetation structure, and human accessibility. Direct anthropogenic proxies show weaker importance at the event scale but exert strong indirect control through long-term land-use change and fuel reorganization, which in turn modulate fuel continuity and landscape configuration.
These results highlight tropical dry forests as a distinct fire domain where fire size emerges from coupled climate–biosphere–human interactions. By combining Earth observation fire products with explainable machine-learning approaches, this study advances understanding of fire–Earth system interactions and supports improved fire-risk assessment in rapidly transforming dry forest regions.
How to cite: San Martín, R., Ottlé, C., Sorenssön, A., Mouillot, F., and Vaittinada Ayar, P.: Atmospheric and landscape controls on fire size in tropical dry forests: insights from the South American Gran Chaco, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7609, https://doi.org/10.5194/egusphere-egu26-7609, 2026.
