Global Multi-model Projections of Urban Heating and Air-conditioning Energy Demand

Abstract:

Densely populated urban areas consume most of the world’s delivered energy. How future urban heating and air conditioning (HAC) energy demand responds to climate change, given the uncertainties of the climate system, remains largely unclear. This compromises the stability of existing power grids and challenges future energy infrastructure planning. Dynamic modeling of HAC energy demand at large scales, however, is challenging given the near-universal absence of urban representation in the state-of-the-art Earth system models (ESMs). Here we develop a novel framework that combines big-data methods with process-based climate modelling to deliver the first global dynamic multi-model projections of urban HAC energy demand under future climates, with quantitative characterization of robustness and uncertainty. The framework is trained and validated with Community Earth System Model version 2 (CESM2) outputs and performs sufficiently well in projecting location-specific HAC energy demand on a global scale. We find that for CMIP6-projected climates under SSP3-RCP7 scenario, global urban air-conditioning energy demand rises by ~102 to 233% while heating energy demand decreases by ~26 to 61% over the century. Combined HAC energy demand globally is dominated by heating subject to model assumptions but varies largely between regions and climate zones. The sensitivity of HAC energy demand to temperature change is non-linear and spatially diverse. The results demonstrate the importance of urban energy and climate interactions and may help inform climate-conscious urban planning and building designs towards sustainable urban development.