Global assessment of population exposure to multiple climate-related hazards from 2003 to 2021: a retrospective analysis

Background

The climate crisis is increasingly recognised as a health crisis, driven in part by the growing frequency and intensity of climate-related hazards, such as heatwaves and wildfires. These hazards can coincide, potentially leading to compound impacts. However, little is known about where and how often such combinations occur globally. This study aims to map historical population exposure to multiple interacting climate-related hazards and identify regions that have been most affected.

Methods

In this retrospective study, we analysed global data from the 2024 Lancet Countdown on health and climate change, International Best Track Archive for Climate Stewardship, The Inter-Sectoral Impact Model Intercomparison Project, and WorldPop from 2003 to 2021 at a 0·25° resolution to examine the population that was exposed to combinations of six climate hazards: heatwaves, droughts, wildfires (PM2·5), extreme precipitation, river floods, and tropical cyclones. We identified administrative level 1 regions based on several hotspot definitions and did detailed case studies in the most exposed regions.

Findings

We detail how frequently people at each grid point have been exposed to various combinations of hazards, revealing both regular patterns and rare, region-specific occurrences. Our analysis indicates an increase in per-person exposure to many hazards during the study period, with a more pronounced rise in multihazard exposure than single-hazards. Between 2003–12 and 2012–21, per-person exposure to three or more hazards increased by 69%. Heatwaves were the most common hazard and also showed the clearest trend, largely driving the observed increases in both single-hazard and multihazard exposure. Multihazard hotspots vary depending on the specific definition used. For instance, exposure to multiple hazards is explained by the seasonality of hazards, which leads them to co-occur in the same months. Additionally, incorporating specific vulnerable age groups into our analysis reveals hotspots that consider the sociodemographic characteristics of the regions.

Interpretation

These findings highlight the importance of incorporating multihazards into climate and health risk assessments. Our study enables an examination of historical events to deepen our understanding of interactions between hazards. Given the rarity of many hazard combinations, traditional epidemiological methods might fall short. Thus, alternative approaches, such as storyline development, are essential to enhance our preparation for future multihazard occurrences. This work finally serves as a crucial baseline future multihazard risk assessment under different climate and socioeconomic scenarios.