Machine learning-based observation-constrained projections reveal elevated global socioeconomic risks from wildfire

A new study develops a machine learning framework to observationally constrain CMIP6-simulated fire carbon emissions, finding a weaker increase in 21st-century global fires but higher increase in their socioeconomic risks than previously thought. Reliable projections of wildfire and associated socioeconomic risks are crucial for the development of efficient and effective adaptation and mitigation strategies. The lack of or limited observational constraints for modeling outputs impairs the credibility of wildfire projections. Here, we present a machine learning framework to constrain the future fire carbon emissions simulated by 13 Earth system models from the Coupled Model Intercomparison Project phase 6 (CMIP6), using historical, observed joint states of fire-relevant variables. During the twenty-first century, the observation-constrained ensemble indicates a weaker increase in global fire carbon emissions but higher increase in global wildfire exposure in population, gross domestic production, and agricultural area, compared with the default ensemble. Such elevated socioeconomic risks are primarily caused by the compound regional enhancement of future wildfire activity and socioeconomic development in the western and central African countries, necessitating an emergent strategic preparedness to wildfires in these countries.

Automated summary

A new study develops a machine learning framework to observationally constrain CMIP6-simulated fire carbon emissions, finding a weaker increase in global fires but higher increase in their socioeconomic risks than previously thought. The study highlights the importance of reliable projections of wildfires and associated socioeconomic risks for the development of effective adaptation and mitigation strategies. Additionally, it emphasizes the need for observational constraints in modeling outputs to enhance the credibility of wildfire projections. The study presents a machine learning framework to constrain future fire carbon emissions and indicates a weaker increase in global fire carbon emissions but higher increase in global wildfire exposure in population, gross domestic production, and agricultural area.