Urbanization amplifies continental- to regional-scale warming

Urbanization has traditionally been overlooked while estimating past changes in large-scale climate and is not resolved in future climate projections. This is due to the small fraction of Earth's surface historically covered by cities, the lack of representation of urban areas in most climate and Earth system models, and observational practices that try to minimize the influence of urban heat islands on the climate signal. In this study, we integrate global land surface temperature observations, which avoid many of the sampling pitfalls of ground-based weather station data, with historical urban area estimates to reveal that the urban contribution to continental- and regional-scale warming has become more significant over time, particularly in rapidly urbanizing regions and countries in Asia. Our findings suggest that anticipated urban expansion over the next century will further amplify the urban influence on large-scale surface climate, with projections indicating an approximate increase of 0.16 K for North America and Europe under a high-emission scenario by 2100. Consequently, we propose that urbanization, akin to other forms of land use/land cover change, must be explicitly included in climate change assessments. This inclusion necessitates the integration of dynamic urban extent and biophysical processes into current-generation Earth system models, enabling the quantification of potential urban feedback on the climate system across various scales.