Climate change and compound events: More frequent hot-dry and hot-wet conditions in a warming world

Global mean temperature has increased in recent decades and is projected to continue rising throughout the 21st century, affecting climate extremes. In particular, compound extreme events, the combination of two or more drivers or hazards (not necessarily extreme individually) whose interaction can amplify impacts, are expected to become more frequent, making their thorough assessment a hot topic for research.

This work presents a global-scale characterization, evaluation and projection of temperature–precipitation compound extreme events and assesses their frequency as well as the spell-related metrics. A threshold-based classification is introduced to define and quantify compound-event occurrence across multiple categories, allowing a robust intercomparison through different climate regimes without fixing a unique “extreme” definition. Based on daily temperature and precipitation over the three-month window centered around the climatologically hottest month for each location, four categories were defined to cover hot-dry, very hot-dry, hot-wet and very hot-wet conditions. 

To  comprehensively assess these compound-event categories, we consider the Regional Climate Model (RCM) simulations from the CORDEX-CORE ensemble, which provides historical simulations and future projections developed on, approximately, a 25km grid for most continental domains worldwide. To reduce the systematic model biases in temperature and precipitation while preserving long-term trends, we apply the ISIMIP bias-adjustment approach (trend-preserving quantile mapping). Results are synthesized over IPCC AR6 reference regions, using CORDEX domains as additional spatial benchmarks.

Future projections are presented for three Global Warming Levels (GWLs): +1.5, +2 and +3 °C relative to pre-industrial conditions. We find an increase in the occurrence of compound events across all categories as warming intensifies, with regionally varying patterns. In addition, the mean number of spells tends to increase with warming, particularly for the hottest categories (very hot–dry and very hot–wet), suggesting that the most extreme compound conditions occur more often in repeated episodes under higher GWLs. These results provide a policy-relevant perspective on how the frequency and persistence of hot–dry and hot–wet compound events might evolve with increasing warming.

This work is part of grant PID2023-149997OA-I00 funded by MICIU/AEI/10.13039/501100011033 and by ERDF/EU.mi

Keywords: compound events, regional climate models, CORDEX, Global Warming Levels, climate change, extreme climate.