Europe faces up to tenfold increase in extreme fires in a warming climate

Wildfires, while a natural part of flammable ecosystems, have seen a surge in destructive events in recent years, causing significant social disruption and economic losses. In Europe, human activity accounts for 96% of ignitions. The intensity and spread of wildfires are primarily influenced by weather conditions, vegetation, and topography. The past six years have witnessed three of the worst fire years in Europe since records began in 1980, with 2021 seeing nearly double the burnt area compared to the previous year. Globally, a shift in the scale and persistence of wildfires is observable, with fires occurring in new locations, outside traditional fire seasons, and with greater intensity. Megafires, characterized by high intensity, account for the majority of burnt areas. Their secondary impacts include erosion, deforestation, carbon stock depletion, and threats to human safety and infrastructure. These events pose significant challenges to wildland fire management, often requiring fires to burn until natural extinction due to the difficulty in suppressing megafires. While some studies have examined climate-induced trends in fire danger regionally, a comprehensive quantitative analysis of how temperature and precipitation influence the probability of extreme wildfire events across Europe is lacking. This study addresses this gap by performing a sensitivity analysis of how changes in precipitation and temperature influence fire danger at a pan-European scale. The study uses the Fire Weather Index (FWI), calculated from the ERA5 reanalysis, to construct two-dimensional impact response surfaces (IRS). FWI, based on daily weather variables including temperature and precipitation, is a widely used metric for estimating fire weather globally. While not a direct measure of fire activity, FWI correlates with burnt areas, especially for significant events, and indicates the suppression effort required. The study assumes relative stability in fire ignitions and no significant changes in fire suppression measures or land use. The study utilizes CMIP6 projections under the SSP2-4.5 scenario to understand future alterations in fire weather patterns.

The introduction adequately summarizes relevant literature on wildfire trends, impacts, and the influence of climate change. It cites studies examining regional fire danger trends and the limitations of existing research on a pan-European scale. The review highlights the use of the Fire Weather Index (FWI) and its correlation with fire activity, noting the limitations of using FWI as a sole proxy for fire events. The reliance on the assumption of relative stability in fire ignitions and the lack of significant change in fire suppression or land use is also acknowledged.

The study employs a multi-faceted approach using reanalysis data from ERA5 and climate change projections from CMIP6 multi-model datasets. ERA5 provides high-resolution atmospheric data from 1940 to near real-time. The CMIP6 simulations under the SSP2-4.5 scenario provide projections of future climates. The FWI, calculated from ERA5 data using the GEFF model, is the primary metric for assessing fire danger. A cluster analysis (k-means algorithm) is performed on the historical daily FWI data (1981-2010) to identify five distinct fire regions in Europe (boreal, temperate, semi-arid, alpine, and Mediterranean). For each region, a 20-year return period FWI value is determined to define an extreme fire event. The duration of the fire season is defined as the number of days exceeding half the 20-year return period FWI value. Two-dimensional impact response surfaces (IRS) are used to quantify the impact of projected temperature and precipitation changes on the probability of extreme fire events and the duration of the fire season. The IRS link climate change outputs (CMIP6 projections) to potential impacts on FWI. Monthly mean changes in temperature and precipitation from CMIP6 are upsampled to daily values using ERA5 data. Incremental curves are derived by perturbing ERA5 data with monthly factors from CMIP6 projections, ranging from 0-6°C temperature increase and -40% to +60% precipitation change. These perturbed data are used to calculate FWI and create the IRS. The probability density functions (PDFs) of future climate variables from the CMIP6 multi-model simulations (SSP2-4.5 scenario) are used to estimate the likelihood of different climate change outcomes. The methodology also discusses the limitations of using FWI as a proxy for fire activity, emphasizing that the correlation is stronger for extreme fire weather scenarios.

The study's key findings indicate a substantial increase in the probability of extreme fire events across Europe under future climate scenarios. Specifically: * **Regional Variation:** The impact of temperature and precipitation changes on fire risk varies significantly across the five identified fire regions. Mediterranean (MED) regions show the highest sensitivity to temperature increases, with a 1°C rise nearly doubling the probability of an extreme fire event. Boreal, temperate, and alpine regions exhibit a less pronounced response to temperature changes until exceeding the 2°C threshold. Precipitation changes have a more pronounced effect on boreal and temperate regions. * **Probability Increase:** Under the CMIP6 SSP2-4.5 scenario, the Mediterranean region shows a potential tenfold increase in the probability of extreme fire events, while more than half of Europe's land area is expected to see at least a doubling in the probability (from 5% to 10%). * **Extended Fire Season:** A significant increase in the length of fire seasons is projected across Europe, with an average extension of one week under a +2°C temperature increase. The Mediterranean region is expected to experience the most substantial extension, with some areas potentially seeing an extension of up to 20 days by the end of the century. Northern Europe could see an increase of up to 30% in fire assistance days. * **Spatial Distribution:** The Mediterranean region (including parts of Turkey, Greece, Italy, the Iberian Peninsula, Southern France, and North Africa), along with parts of the Atlantic coast of central and northern Europe, are projected to experience the most significant increases in fire risk. Northern Europe, while showing less pronounced increases, remains vulnerable to prolonged droughts.

The findings highlight the significant impact of climate change on the risk of extreme wildfires across Europe. The Mediterranean region is identified as the most vulnerable, facing a potential tenfold increase in extreme fire events. This increased risk, coupled with longer fire seasons, poses substantial challenges to firefighting capabilities and resource allocation. The study's results emphasize the need for proactive adaptation strategies, including improved fire management practices, enhanced early warning systems, and increased international cooperation. While the study focuses on FWI as a proxy for fire activity, the strong correlation during extreme fire events supports the conclusions. The limitation of not considering vegetation dynamics (fuel availability) and the assumption of no significant changes in fire suppression efforts are acknowledged. The results are consistent with other studies indicating an increased risk of megafires due to climate change.

The study demonstrates the substantial and regionally varying impacts of climate change on extreme fire weather in Europe. The Mediterranean region is projected to experience a dramatic increase in fire risk, while other regions will also face heightened challenges. The findings underscore the urgent need for strengthened fire management strategies and adaptation measures to cope with the escalating threat of extreme wildfires in a warming climate. Future research could incorporate vegetation dynamics and the impact of mitigation efforts to refine these projections further.

The study acknowledges several limitations. First, it assumes relative stability in fire ignitions and no significant changes in fire suppression capabilities or land-use practices. Second, it does not explicitly model vegetation dynamics, particularly post-fire vegetation recovery and changes in fuel availability. Third, the analysis assumes a continued lack of substantial climate change mitigation efforts. These limitations could influence the interpretation and generalizability of the results, and further research incorporating these factors is needed for a more comprehensive understanding.