Multi-Generational CMIP Consensus on Regional Climate Risk

The identification of regional climate change “hotspots” (areas projected to experience pronounced and impactful changes) is a critical step in prioritizing adaptation resources and policy interventions. The Regional Climate Change Index (RCCI) provides a standardized framework for classifying regional climate risk (low, medium, high) by synthesizing changes in mean precipitation, surface air temperature, and their interannual variability. To move beyond single-model generation assessments and quantify the robustness of these risk classifications, this study introduces a novel Risk Reliability Index (RRI). The RRI is calculated by cross-multiplying RCCI risk classifications (1=low, 2=medium, 3=high) across three generations of the Coupled Model Intercomparison Project (CMIP3, CMIP5, and CMIP6), summing the products of all unique pairwise combinations for each grid cell. This yields an index ranging from 3 to 27, where higher values indicate not only a higher classified risk but also stronger agreement across model generations, enhancing confidence in the projected regional signal. Analysis of the resulting global Risk Reliability Matrix reveals distinct geographical patterns of multi-generational consensus. The highest RRI values (indicating higher risk with stronger model agreement) are consistently identified in the Mediterranean Basin, Sahara and Sahel regions, Arabian Peninsula, parts of the Amazon region, Northeast and Central Brazil, Southern Africa, and high-latitude Northern Hemisphere regions, including the Arctic and Siberia. These areas emerge as persistent climate hotspots where successive model generations robustly project compounded changes. Conversely, the lowest RRI values (indicating lower risk with stronger model agreement) are found over extensive oceanic regions, particularly the Southern Ocean and parts of the eastern tropical Pacific, southern portion of South America, as well as some continental interiors. While not risk-free, these regions show the most consistent inter-model projection of relatively lower magnitude changes across the three CMIP ensembles. This work underscores that regions with high RRI values represent priority targets for climate adaptation due to both the severity of projected changes and the high confidence across model generations. The Risk Reliability Index provides a simple, transparent metric for integrating multi-model, multi-generational evidence into climate risk assessments, offering a valuable tool for scientists and policymakers to identify regions where climate change signals are most robust and actionable.