基于LMDI模型和Q型聚类的中国城镇生活碳排放因素分解分析

为研究城镇居民生活碳排放特征及影响因素，基于LMDI模型从全国和省级层面研究了我国30个省、自治区、直辖市（不含港澳台及西藏自治区）2006-2015年的城镇生活碳排放，将城镇生活碳排放分解为生活能源消费结构效应、生活能源强度效应、消费倾向效应、人均可支配收入效应和城镇人口规模效应，分析各效应逐年和累积效应贡献度以及区域差异，并基于LMDI模型的计算结果对我国30个省、自治区、直辖市进行Q型聚类分析.结果表明:①从全国层面看，人均可支配收入、城镇人口规模是刺激因素，其中，人均可支配收入的影响效应最为显著，而消费倾向、生活能源消费结构、生活能源强度抑制了生活碳排放的增长.②从省级层面看，人均可支配收入、城镇人口规模的累积效应均为正，而消费倾向、生活能源消费结构、生活能源强度对各省、自治区、直辖市生活碳排放的影响效应有正有负，显示出显著差异.因此，政府应引导城镇人口合理增长，并积极制定相应政策优化居民生活能源消费结构.在制定碳减排战略时，要将省级生活碳排放的表面特征与其潜在驱动力相结合，根据不同区域有针对性地实施碳减排政策，同时应及时作出调整，以应对不同的发展阶段. 

Factor Decomposition of Chinese Urban Household Carbon Emissions-Empirical Analysis based on LMDI Model and Q-Clustering

In order to study the characteristics and influencing factors of urban household carbon emissions, this study investigates China''s urban household carbon emissions from 2006 to 2015 at the national and provincial level (excluding Hong Kong,Ma Cao,Taiwan and Tibet Autonnomous Region) by the Logarithmic Mean Divisia Index (LMDI) method. The urban household carbon emissions are divided into five effects:household energy consumption structure, household energy intensity, consumption tendency, the per capita disposable income and the urban population size. The regional difference of each effect and the contribution degree are also analyzed. In addition, based on the result of LMDI model, Q-clustering analysis is carried out for 30 provinces, autonomous regions and municipalities in China. The results show that the per capita disposable income and the urban population size increase urban household carbon emissions at the national level, while the energy consumption structure, household energy intensity and consumption tendency inhibit the growth of household carbon emissions. Among them, the per capita disposable income is the main affecting factor. From the provincial level, the per capita disposable income and the urban population size have a positive impact on urban household carbon emissions. The effects of energy structure, household energy intensity and consumption tendency on the household carbon emission reflect significant differences between provinces, autonomous regions and municipalities. Therefore, the government should guide the reasonable growth of urban population and actively formulate corresponding policies to optimize the structure of residential energy consumption. In terms of formulating carbon emission reduction strategies, the government should combine the surface features of provincial household carbon emissions with its potential driving factors and implement the corresponding carbon emission reduction policy according to different regions. Besides, it is suggested to adjust the reduction policy in a timely manner to cope with different development stages.