The Spatiotemporal Evolution of Arctic Climate Types and Its Attribution

Abstract The Arctic climate has exhibited a warming and wetting trend in recent decades. Based on Arctic temperature and precipitation data (1940–2023), this study employed climate classification methods to examine Arctic climate states and their variations. Since traditional rule-based classification schemes inadequately represent the climate over the Arctic Ocean, we employed a data-driven k -means approach. This method objectively identified four distinct Arctic climate types: cold-dry, semi-cold-dry, semi-warm-wet, and warm-wet. Spatiotemporal analysis reveals strong contrasts between polar day and night. During the polar day, the semi-cold-dry climate shrank progressively over the high-latitude waters of the Arctic Ocean. In contrast, during the polar night, this climate type expanded and became dominant across the Eurasian Sea, from the East Siberian Sea to the Kara Sea. Transitions among these Arctic climate types were accompanied by the strengthened synchronization between temperature and precipitation. Attribution analyses indicate that polar-day transitions were driven primarily by local warming feedback and secondarily by external forcing. Conversely, polar-night changes were dominated by external forcing. The Coupled Model Intercomparison Project phase 6 (CMIP6) projections show that under high-emission scenarios, the semi-warm-wet type is projected to become the dominant climate over the Arctic Ocean by the late twenty-first century. This study examines Arctic climate classification and evolutionary patterns, contributing to the understanding and prediction of global climate change.