Projected climate extremes and demographic change provide a spatial context for linguistic vulnerability in Multilingual Yunnan

Abstract Climate change may reshape the demographic and socioeconomic conditions under which minority languages are maintained, although projected population change alone cannot demonstrate migration, language shift, or language loss. Focusing on multi-ethnic and multilingual Yunnan, China, this study integrates outputs from eight CMIP6 climate models with 1-km gridded demographic projection of resident population to examine changes in very-wet-day precipitation (R95p), warm-day frequency (TX90p), consecutive dry days (CDD), and resident population from 2020 to 2100 under the SSP5–8.5 scenario. The ensemble results indicate substantial increases in R95p and TX90p, which reach 91% and 275% above their respective 2020 levels by 2100, whereas CDD changes by only 1.23%. The projected population rises initially, peaks at 51.63 million in 2033, and then declines steadily. By 2100, resident populations in the western–northwestern and southern analytical regions are projected to decrease by 52.9% and 54.0%, respectively, relative to 2020. Spatial correlation analysis shows that R95p and TX90p are negatively associated with population across many grid cells, while relationships between CDD and population are generally weak. These results indicate scenario-based spatial and temporal covariation rather than evidence that climate extremes directly cause migration, demographic decline, or minority-language loss. Nevertheless, the geographical coincidence of intensifying climatic stress and long-term population contraction provides a basis for screening locations where speaker networks and community institutions supporting minority-language maintenance may face additional pressure. Because actual language outcomes depend on broader socioeconomic, familial, institutional, and adaptive conditions, household- and community-level field investigations are indispensable for evaluating these linguistic vulnerability hypotheses and informing coordinated governance.