Assessing Climate Change Impacts on Wind Energy Resources over China Based on CMIP6 Multimodel Ensemble

Assessing how wind energy potential will change in the context of global warming is fundamental to local energy development and planning. Twenty-two CMIP6 GCM outputs under three emission scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5) are fed into the convolutional neural networks based on efficient channel attention (ECA-Net) to generate wind energy density projections. This study demonstrates that the ECA-Net model can accurately capture the regional features of wind speed over China. Results show that the wind energy resource potential of China exhibits a significant (p < 0.01) decreasing trend of 0.74% decade^–1, 0.99% decade^–1, and 1.36% decade^–1 during 2015–2100 under SSP1-2.6, SSP2-4.5, and SSP5-8.5, respectively. Compared with the baseline period (1985–2014), China’s average annual wind energy resource potential will decrease by 3.55%, 0.06%, and 2.24% (5.73%, 5.02%, and 8.84%) during 2031–2060 (2071–2100) under SSP1-2.6, SSP2-4.5, and SSP5-8.5, respectively. The results also highlight increased inter- and intra-annual variability of wind energy resources in areas such as parts of the Tibetan plateau, which poses a challenge for regional energy deployment and management. These findings suggest that the sustainability of China’s wind energy development may be challenged by climate change.