Spatio-temporal dynamic relationship of renewable energy penetration and CO2 emissions in China: new evidence from PMG-ARDL model and ESTDA

This study examines the complex relationship between renewable energy penetration (REP) and CO 2 emissions in China, aiming to provide insights into deep decarbonization strategies. Using the advanced and robust pooled mean group-autoregressive distributed lag (PMG-ARDL) estimation, this paper analyzes the long-run equilibrium and short-run dynamics of REP and CO 2 emissions based on panel data from 30 Chinese provinces (1997–2021). Wavelet coherence analysis (WCA) and exploratory spatio-temporal data analysis (ESTDA) are also employed to explore time-frequency relationships and spatial correlation dynamics. The findings indicate that: (1) A 1% increase in REP reduces long-term and short-term carbon emissions by 0.05% and 0.26%, respectively. (2) A more pronounced coherence in the 1–2 year time-frequency band. (3) REP and CO 2 exhibit opposing spatial distribution patterns (−0.224), with distinct north-south clustering, dominated by low-high and low and high clustering, respectively. (4) Most provinces’ LISA time paths demonstrate strong spatiotemporal stability, while Central, South, and Southwest China exhibit a more complex spatial variability process, and the spatiotemporal transitions, predominantly of Type IV, reflect strong locking effects. This paper offers policy suggestions for China’s low-carbon energy system and lessons for other economies.