Provincial-level carbon emission drivers and emission reduction strategies in China: Combining multi-layer LMDI decomposition with hierarchical clustering

Abstract

Understanding provincial-level driving mechanisms of carbon emissions is critical for achieving China's national action targets on climate change mitigation. The paper combines two-layer logarithmic mean divisia index (LMDI) decomposition with Q-type hierarchical clustering to systematically evaluate the contributions of related drivers from 30 provinces to the growths in China's national carbon emissions, and accordingly puts forward more targeted emissions reduction strategies for each cluster of provinces based on the features of CO2 emissions and underlying emissions driving forces. It was concluded that the contributions of various provinces to national CO2 emissions growths and their respective driving mechanisms differed considerably and changed dynamically over time. From 1996 to 2013, Shandong, Hebei, Inner Mongolia, Jiangsu and Guangdong were the top five provincial drivers behind carbon emissions growths in China, and provincial economic expansion and energy intensity reduction played dominant roles, respectively, in raising and abating carbon emissions. Across the sub-periods, the driving effects of economic activity, energy intensity and energy structure changes on carbon emissions and their evolving trends clearly varied by province. Certain provinces, such as Beijing and Shanghai, have presented significantly decreased contributions to national carbon emissions growths and even became negative contributors; while other provinces, such as Hebei, Shanxi and Anhui, have made increased contributions. Therefore, the formulation of China's emissions reduction strategies should consider both the features of provincial carbon emissions and the underlying causes shaping them, and should be adjusted in a timely manner to respond to different development stages.