Environment: Global social risks from wildfire assessed

 
Recent machine learning studies suggest that wildfire exposure and increased socio-economic risk of wildfires may accelerate in the 21st century, but the increase in wildfire carbon emissions worldwide may slow. A paper reporting that it was done will be published in Nature Communications.The findings of this study may help to better understand the potential for regional disparities in wildfire exposure and help strategically prepare for future wildfires.

Obtaining reliable forecasts of wildfires and the associated socio-economic risks is important in developing strategies to address and mitigate the effects of climate change.However, the current Earth system model used to predict long-term wildfires remains somewhat uncertain.

Now, Yan Yu, Jiafu Mao and colleagues use machine learning to set constraints on the simulation of wildfire carbon emissions and socio-economic risks using the Earth system model constructed in the Phase 6 Combined Model Mutual Comparison Program. He also studied carbon emissions and socio-economic risks. Yu et al. Found that wildfire increases in carbon emissions slowed in the 21st century, but global exposure to wildfires increased in terms of population, gross domestic product (GDP) and agricultural areas. It shows the idea of ​​accelerating.Yu et al. Also indicate that forests and savanna in Africa, northern Australia, and eastern South America, where wildfires have been common, may continue to have frequent wildfires in the 21st century.Furthermore, it was found that increased wildfire activity and socio-economic development (population, GDP, agriculture, etc.) in West Africa and Central African countries may increase socio-economic risks in these regions.

 
* This article is reprinted from "Nature Japan Featured Highlights".
Reprinted from: "Environment: Assessing the socio-economic risks of wildfires around the world'