梳理1992—2018年中亚五国碳排放变化过程,运用碳排放库兹涅茨曲线模拟人均碳排放与人均国内生产总值（GDP）的关系,基于中亚五国人均GDP、工业增加值、研发投入、自华进口中间品和资本品的技术溢出存量等指标,构建面板数据固定效应模型以探究碳排放影响机制。结果表明：（1） 中亚五国碳排放先降后升,各国碳排放的变化率波动较大,其中哈萨克斯坦碳排放占中亚五国碳排放比重最大。（2） 就人均碳排放与人均GDP拟合的碳排放库兹涅茨曲线关系而言,哈萨克斯坦和土库曼斯坦表现为倒N型,乌兹别克斯坦、塔吉克斯坦、吉尔吉斯斯坦和中亚五国表现为N型。（3） 中亚五国自华进口额呈现较快的增长趋势,其中自华进口的中间品和资本品合计约占自华进口总额的3.920%~10.976%,中亚五国整体及各国中间品技术溢出存量均明显大于资本品技术溢出存量。（4） 在影响中亚五国碳排放的诸多因素中,人均GDP和工业增加值占GDP比重的回归系数显著为正,人均GDP、工业增加值占GDP比重每上升1%,碳排放分别增加0.337%和0.343%,研发投入、交互项LnZJ_it×LnS_it（中间品技术溢出存量的自然对数×工业增加值占GDP比重的自然对数）的回归系数显著为负,研发投入每增加1%,碳排放可减少约0.432%,在降低因工业增加值占GDP比重上升导致的碳排放方面,中间品技术溢出存量的影响极小。

The five Central Asian countries are the commercial throat of the Land Silk Road, and their strategic position is important. The five Central Asian countries and China depend on each other geographically and promote each other’s economy and trade. The five Central Asian countries are the gateway and transit point of China’s land corridor connecting Europe, while China is the support and shortcut for the five Central Asian countries to move to the Pacific Ocean. Recently, the pace of economic and trade cooperation between China and the five Central Asian countries has accelerated, and remarkable achievements have been made in trade, investment, economic and technological cooperation, and financial cooperation. Between 1992 and 2018, processing the changes in carbon emissions in the five Central Asian countries using Kuznets curve of carbon emissions to simulate the relationship between carbon emissions per capita and gross domestic product (GDP) per capita, then combing with indicators of GDP per capita, the total value added of industry, R&D investment and technology spillover stock of imported intermediate and capital goods from China to construct a panel data with fixed-effect model to explore the impact mechanism of carbon emissions in the five Central Asian countries. (1) The study finds the carbon emissions of the five Central Asian countries, which decreased and then increased; the rate of change of carbon emissions has fluctuated. The carbon emissions in Kazakhstan accounted for the largest proportion among the five Central Asian countries. (2) In relation to the Kuznets curve of carbon emission and GDP per capita, Kazakhstan and Turkmenistan exhibit an inverted N-shaped model. Uzbekistan, Tajikistan, Kyrgyzstan, and the five Central Asian countries as a whole are N-shaped models. (3) The five Central Asian countries’ imports from China have grown, and the intermediate and capital goods have accounted for 3.920% to 10.976% of the total imports from China. In addition, the technology spillover stock of imported intermediate goods from China has exceeded that of imported capital goods in the five Central Asian countries. (4) Considering the factors that affect the carbon emissions of the five Central Asian countries, the regression coefficients of GDP per capita and the proportion of industrial added value in GDP are positive. For every 1% increase in GDP per capita and the proportion of industrial added value in GDP, carbon emissions will increase by 0.337% and 0.343%, respectively. The regression coefficients of R&D investment and the interaction term of LnZJ_it×LnS_it are negative. For every 1% increase in R&D investment, carbon emissions can decrease by about 0.432%. For carbon emission reduction, caused by the increase in the proportion of industrial added value in GDP, the impact of technology spillover stock of intermediate goods is deemed minimal. This study explains the potential law of carbon emissions in five Central Asian countries and provides a scientific reference for Chinese technology export to five Central Asian countries.