利用LMDI模型解构了2010—2019年甘肃省13个细分行业碳排放影响因素及其作用效应，运用Tapio脱钩模型分析了经济增长与碳排放的脱钩关系，在此基础上，检验了各因素对脱钩做出的努力程度。结果表明：（1） 2010—2019年甘肃省细分行业碳排放总量增加3843.13×10⁴ t，主要集中在石油制造业、化工制造业、钢铁制造业以及电力行业等高能耗行业；能源消费结构的高碳化特征显著，能源消费强度呈下降趋势。改善高能耗产业能源消费结构、推动高能耗产业转型升级是未来甘肃省碳减排的重点。（2） 经济增长和人口规模对碳排放产生增量效应，而能源强度、能源结构对碳排放产生减排效应，产业结构对部分行业产生减排效应。（3） 各行业碳排放与经济增长的脱钩情况趋于向好，除电力行业仍为弱脱钩外，其他行业均由2010—2016年的负脱钩或弱脱钩转变为2016—2019年的强脱钩或衰退脱钩。（4） 能源强度效应的脱钩努力最高，能源结构和产业结构效应的脱钩努力尽管较小但逐渐增强，人口规模效应的脱钩努力不明显。

Accurate identification of specific focus points of industry carbon reduction is crucial to realize China’s goal of “carbon peak by 2030 and carbon neutral by 2060”. This study used the Logarithmic Mean Divisia Index (LMDI) method to decompose the influencing factors and their effects on the carbon emission of 13 subsectors (from 2010 to 2019) in Gansu Province. The Tapio decoupling model was used to analyze the relationship between carbon emission and economic growth. Accordingly, a decoupling effort model of influencing factors, excluding economic factors, was constructed to analyze the efforts made by other factors to decoupling. The following results are obtained. (1) From 2010 to 2019, carbon emissions for subsectors in Gansu Province increased by 3843.13×10⁴ t, mainly in petroleum, chemical, steel, and power industries. Specifically, the energy consumption structure of Gansu Province was characterized by high carbon emissions. Coal consumption made up 64.89% of the entire fossil energy consumption in 2019. Energy consumption intensity emerged a decreasing trend, whereas energy efficiency kept improving. (2) Economic growth and population scale exhibited an incremental effect caused by the economic growth effect. Energy intensity and structure demonstrated a reduction effect, and the reduction effect of energy intensity was more significant. However, the influence direction of the industrial structure effect fluctuated greatly in different time periods and industries. The industrial structure effect on chemical and construction industries had a relatively significant reduction, whereas that on steel and power industries was increased carbon emissions. (3) The decoupling effect of carbon emissions from the economic growth of 13 subsectors improved. From 2010 to 2013, all industries exhibited a weak decoupling effect, except for mining and light manufacturing that showed negative decoupling and expansion connection. From 2013 to 2016, some industries, such as agriculture, chemical, and steel manufacturing, underwent a strong decoupling effect. From 2016 to 2019, all sectors changed to strong or recessionary decoupling, except for the power sector, which remained weak. (4) The energy intensity effect played the most important role in decoupling. Particularly, the decoupling effect of energy and industrial structures was small but gradually increasing, whereas that of the population scale was not evident. Evidently, reducing energy consumption intensity and improving energy use efficiency are crucial points to accelerate the process of carbon emission reduction and effectively enhance the decoupling level in Gansu Province. On the basis of this finding, the following should be proposed. First, governments and enterprises should actively introduce low-carbon production technologies and high-efficiency energy-saving equipment, encourage innovation, and focus on the development and optimization of energy-saving and environmental protection technologies. Second, governments should comprehensively consider the characteristics of local industrial structures, carbon emission levels, and emission reduction potentials of subsectors and then formulate differentiated quota schemes for industrial carbon emissions for high- and low-energy industries.