Global patterns of human risk from hot and dry climate extremes, wildfires and poor air quality: insights from multi-hazard and compound analyses

Climate change is intensifying the frequency, severity, and interactions of extreme heat, drought, wildfires, and air pollution, increasing risks to both ecosystems and human populations worldwide. These risks can emerge from the accumulation of multiple hazards over time (multi-hazard risk), but also from their simultaneous co-occurrence (compound events). Here, we present a global, spatially explicit assessment of human risk from wildfires and air quality associated with hot and dry extremes, explicitly integrating multi-hazard and compound risk representations, following a hazard, exposure, and vulnerability perspective.

Using global datasets at 0.75° spatial resolution for the period 2003–2022, hazards are quantified based on the number of hot days per month (derived from exceedances of daily maximum temperature above the 90th percentile of a 1991–2020 climatology), drought occurrence (as depicted by the 6-month Standardized Precipitation–Evapotranspiration Index, SPEI), wildfire activity (characterized using MODIS Fire Radiative Power, FRP), and the number of days with PM_2.5concentrations exceeding World Health Organization air quality thresholds. Human exposure is represented exclusively by gridded population density, while vulnerability is characterized using indicators capturing human sensitivity and adaptive capacity, e.g., the Human Development Index (HDI) and Water Stress Index (WSI).

Human risk is quantified by combining hazard intensity, population exposure, and vulnerability, following both a multi-hazard and a compound formulation. In a multi-hazard formulation, hazards are aggregated without requiring temporal co-occurrence, capturing the cumulative burden of climate extremes. In parallel, compound risk is assessed by explicitly accounting for the co-occurrence of hazards within the same temporal windows, enabling a direct comparison between cumulative and compound representations of risk. In addition, we quantify the global population affected by different risk classes. Our estimates indicate that approximately half of the world’s population is currently exposed to high to very high risk, while a substantially smaller fraction resides in low or extremely low risk conditions. High and very-high risk classes together account for several billion people, underscoring the widespread nature of climate-related human risk. When aggregated at the country level, risk levels exhibit a clear socioeconomic gradient, with higher average risk values concentrated in lower-income countries, low life expectancy at birth, and high infant mortality rate.

The results illustrate how a compound events perspective can alter the spatial distribution and relative intensity of human risk compared to a multi-hazard one, highlighting regions where hazard interactions may further amplify societal impacts. This work provides a generalized framework for global human risk assessment, offering new insights into how different representations of climate extremes shape risk patterns and supporting the development of more effective adaptation and risk reduction strategies.

This work is supported by FCT, I.P./MCTES through national funds (PIDDAC): LA/P/0068/2020 - https://doi.org/10.54499/LA/P/0068/2020, UID/50019/2025, https://doi.org /10.54499/UID/PRR/50019/2025, UID/PRR2/50019/2025. This work was performed under the scope of project https://doi.org/10.54499/2022.09185.PTDC (DHEFEUS) and the Horizon Europe research and innovation programmes under grant agreement number 101081661 (WorldTrans).