Climate change is a global phenomenon, yet its fingerprints varysubstantially across regions. This talk highlights a range of theseregional patterns using observational records and climate modelsimulations, analyzed with machine learning and complementarystatistical tools.The first part of the talk examines the magnitude of climatechange across temporal and spatial scales, showing how longtermwarming reshapes seasonal and diurnal temperature cyclesin different regions.The second part examines how quickly climate mitigation signalscan be detected against regional climate variability, highlightingwhere the effects of emission reductions are likely to emergesooner or later across the globe.The final part of the talk addresses the question of climatechange acceleration. Despite rapidly increasing greenhouse gasemissions, recent studies suggest that the global mean warmingrate remains linear. We revisit this issue by shifting the focusfrom global averages to regional scales, where we detectsignificant acceleration in warming across a substantial fractionof the world.
