Likelihood of observed fire weather extremes in Europe increases nonlinearly from preindustrial to 3°C warming

The European continent has experienced severe wildfire activity in multiple regions as a result of extreme drought and heat events in recent years, particularly in 2003, 2017 and 2018. Quantifying how climate change has altered the likelihood of extreme wildfire occurrence and its accompanying fire weather conditions remains challenging due to strong internal climate variability and short observational records. 

Here, we quantify changes in the probability of extreme wildfire conditions considering four fire weather indicators: the Canadian Fire Weather Index (FWI), drought conditions (i.e. 3-month Standardized Precipitation Evapotranspiration Index; SPEI-3M), heat (maximum temperature; Tmax) and atmospheric moisture demand (vapor pressure deficit; VPD). First, we assess the return periods of the four fire weather indicators during these extreme wildfire periods under observed climate using CERRA reanalysis data (2001-2020). Second, we quantify how the likelihood of conditions that describe observed extreme wildfire periods changes between preindustrial, present, 2°C and 3°C global warming levels, by bootstrapping data from the 100-member Community Earth System Model Large Ensemble (CESM2-LE). 

We show that the probability of fire weather conditions during observed extreme wildfire periods increases nonlinearly with global warming. The probability of the FWI as found during these extreme wildfire periods doubled from preindustrial to present levels and is projected to increase three- and seven-fold under 2°C and 3°C of global warming, respectively. For SPEI-3M, VPD and Tmax we find even stronger increases. Our results highlight the substantial benefits of limiting global warming to well below 2°C for reducing wildfire-relevant climate extremes.