Determinants of spatial disparities of petroleum terminal utilization carbon emissions in China
FANGYebing1,2,3,, WANGLimao1,2,, MOUChufu1,2, ZHANGHong4, QUQiushi1,2 1. Institute of Geographic Sciences and Natural Resources Research,Chinese Academy of Sciences,Beijing 100101,China2. College of Resources and Environment,University of Chinese Academy of Sciences,Beijing 100049,China3. College of Territorial Resources and Tourism,Anhui Normal University,Wuhu 241003,China 4. School of geospatial information,People's Liberation Army information engineering university,Zhengzhou 450001,China 通讯作者:通讯作者:王礼茂,E-mail:lmwang@igsnrr.ac.cn 收稿日期:2017-09-7 修回日期:2017-11-25 网络出版日期:2017-12-31 版权声明:2017《资源科学》编辑部《资源科学》编辑部 基金资助:教育部人文社会科学研究青年基金项目(13YJCZH037) 作者简介: -->作者简介:方叶兵,男,安徽芜湖人,副教授,主要研究领域为资源经济和能源地缘政治。E-mail:fyb11@sina.com
关键词:石油终端利用;碳排放;空间分异;主导因素;地理探测器;中国 Abstract It is well known that China is the world's second largest petroleum consumer and petroleum is China's second largest source of CO2 emissions. Here we diagnose the dominant factors of differentiation of carbon emissions and reveal the dynamic mechanisms of carbon emissions using the coefficient of variation and geographical detector model. We found that the CO2 emissions in China's petroleum terminal in the east is higher than that in the west,and the number of provinces of eastern China with higher CO2 emission is much more than that of western China in 2015. However,the CO2 emission intensity in China's petroleum terminal in the west is higher than that in the east,and the number of provinces in western China with higher CO2 emission is much more than that of eastern China in 2015. CO2 emissions from petroleum terminals are significantly higher at the national level and at the three major regional levels than CO2 emissions from petroleum terminals. In the three regions,the CO2 emission and emission intensity of the petroleum terminal showed the strongest differentiation in the eastern region and the smallest in the central region. At the national scale,dominant factors influencing regional differences in CO2 emissions in the petroleum terminal consumption are the diesel consumption ratio,population size and per capita GDP. While dominant factors affecting regional differentiation of CO2 emission intensity are industry accounted for GDP,petroleum consumption intensity of transportation industry and gasoline Consumption ratio. There are significant differences in the five types of carbon emissions,namely,energy structure oriented region mainly affected by diesel consumption ratio and gasoline consumption ratio,energy intensity oriented region mainly affected by petroleum consumption intensity of transportation industry,industrial structure oriented region mainly affected by industry accounted for GDP,the level of economic development oriented region mainly affected by per capita GDP,and population oriented region mainly affected by population size.
本文主要采用变异系数、自然断点法和地理探测器等研究方法。利用变异系数测度全国、三大地带终端石油CO2排放量和CO2排放强度的区域差异状况(图1);利用自然断点法将(结构、效益、规模等三个维度)能源结构、能源强度、产业结构、经济发展和人口规模等要素划分为3级,对各影响因子进行分层;基于省级数据,地理探测器用于提取全国层面石油终端利用CO2排放量和排放强度分异的主导因素,并对三大区域的主导因素分析进行进一步区分,阐述其作用机制。根据三大地带的决定力大小将30个省划分五种碳排放地域类型。通过碳排放与主导因素的完全耦合程度,发现三大地带中各主导因素影响下的典型省份。最后根据象限图所划分的区域类型,对不同省份的石油终端CO2排放量和CO2排放强度提出优先减排和针对性的区域碳减排政策。 显示原图|下载原图ZIP|生成PPT 图12015年中国石油终端利用CO2排放量和CO2排放强度空间分布 -->Figure 1The spatial distribution of CO2 emissions and CO2 emissions intensity from the terminal petroleum utilization in China in 2015 -->
(1)碳排放总量计算。根据各年份柴油、煤油、汽油、燃料油等四种油品的实物消费量与其相对应的单位实物量CO2排放系数乘积加总求和,得出各地带相应的CO2排放量。公式如下: (1) 式中 为第j省区的碳排放总量;i为四种油品的某一种类; 为第j省区在某个时期对第i种油品的实物消费数量; 为第i种油品的单位实物量CO2排放系数(表1)。 Table 1 表1 表1各油品单位实物量CO2排放系数 Table 1Carbon dioxide emission in unit quantity of different petroleum products carbon dioxide emission factor
学界对碳排放的影响因素主要归结为能源结构、能源强度、产业结构、经济产出和人口规模。因此,本文遵照已有文献研究成果和数据可获得性原则,遴选出13个因子,如表2所示,旨在从石油消费结构、石油消费强度、产业结构、经济产出和人口规模因素的共同作用影响下,探测石油终端利用碳排放增长的地域分异机制。进行变异系数和地理探测器分析时,数据均取对数处理。采用自然断点法进行区域量化分级得分,从低分到高分在1~3范围内取值。 Table 2 表2 表2石油终端利用CO2排放探测因子体系 Table 2The spatial influencing factors of carbon emissions of the terminal petroleum use
整体看,石油终端CO2排放量空间与CO2排放强度空间分布基本不一致(图1)。 对2015年30省CO2排放量数据取对数,采用自然断点法将其分为三类。从石油终端CO2排放量的空间分布看,中国石油终端利用CO2排放量的空间分异基本呈东高西低、东多西少的分布特征(图1a)。2015年全国石油终端利用高碳排放量的主要省份为:东部地带的辽宁、山东、江苏、浙江和广东等5个省,中部地带的河南和湖北等2个省,西部地带的四川省。 同理,对CO2排放强度数据取对数后进行自然断点分类。从石油终端CO2排放强度空间分布看,总体呈西高东低,西多东少的分布格局(图1b)。2015年全国石油终端利用高碳排放强度的主要省份为西部地带的新疆、甘肃、宁夏、青海、四川、贵州和云南等7个省,东部地带的上海、山东、辽宁和海南等4个省,中部地带的有黑龙江省。总体看,低碳排放强度主要集中于东部省份。 从变异系数分析结果看,2015年,从全国层面看,石油终端CO2排放量变异系数(0.80)要高于CO2排放强度变异系数(0.29),这表明CO2排放量区域差异较大,远高于CO2排放强度的区域差异(图2)。从区域层面看,东、中、西三大地带的CO2排放量变异系数分别为0.73、0.38和0.71,也均高于同地区的CO2排放强度变异系数值,分别为0.35、0.18和0.23。从地区横向比较看,东部地带在CO2排放量和CO2排放强度的变异系数均最高,其次为西部地带,变异系数最低为中部地带。 显示原图|下载原图ZIP|生成PPT 图2中国石油终端利用CO2排放量和CO2排放强度的变异系数 -->Figure 2The coefficient of variation of CO2 emissions and CO2 emission intensity from petroleum terminal utilization in China -->
3.2 中国碳排放地域分异的主导因素分析
采用自然断点法将所有探测因子值从低到高依次分为1–3三个等级进行量化。利用地理探测器测算方法,分别计算反映各探测因子对影响能力的 值。 (1)CO2排放量地域分异主导因素。石油终端利用碳排放受能源结构、产业结构、能源强度、社会经济发展水平等方面的综合影响,本文选取13项指标(表2),对影响中国石油终端利用碳排放的主导因素进行因子探测(Factor Detector)。将13项指标,分别与石油终端利用碳排放进行空间因子探测分析,计算得到各因素对石油终端利用碳排放的决定力 。 根据地理探测器因子探测分析,将13项因子分别与CO2排放量进行空间探测,计算得到各因素对碳排放量的决定力 ,各因素决定力大小依次为:柴油消费比 、人口规模 、人均GDP 、燃料油消费比 、汽油消费比 、交通运输业占GDP比重 、工业占GDP比重 、农业占GDP比重 、工业石油强度 、交通运输业石油强度 、城市化率 、煤油消费比 、农业石油强度 。其值依次分别为0.426、0.356、0.299、0.230、0.161、0.103、0.103、0.086、0.063、0.057、0.055、0.045、0.013。取前50%的因子,它们对全国层面的有重要意义。柴油消费比 、人口规模 和人均GDP 的决定力之和占影响石油终端利用CO2排放量的总决定力之和的54.2%,因此,将柴油消费比、人口规模和人均GDP 作为影响全国石油终端CO2排放量分异的主导因素(图3)。 显示原图|下载原图ZIP|生成PPT 图3中国石油终端利用CO2排放量和CO2排放强度决定力 -->Figure 3The power determinant values of CO2 emissions and CO2 emission intensity from petroleum terminal utilization in China -->
东、中、西三大地带是国家制定区域政策和学界研究全国尺度区域差异时采用的地理单元。所以本文以省域为基本地理单元,从全国、三大地带不同空间尺度运用变异系数和地理探测分析,借助ArcGIS10.2软件,探测全国层面、三大地带等不同区域尺度的终端石油CO2排放空间分异格局和影响因素。 通过探测影响全国石油终端CO2排放分异的主导因素,进一步分析各主导因素对区域石油终端CO2排放分异的作用机制,为各区域针对性地采取碳减排政策提供参考依据(图4)。 显示原图|下载原图ZIP|生成PPT 图42015年中国石油终端利用CO2排放主导因素分级 -->Figure 4The classification of the dominant factors of CO2 emissions from petroleum terminal utilization in China in 2015 -->
将影响石油终端利用CO2排放量的柴油消费比、人均GDP和人口规模等三个主导因素在三大地带进行决定力对比(图5)。2015年CO2排放量决定力分别为: 东部地带 (0.294)> (0.251)> (0.217);中部地带 (0.708)> (0.273)> (0.255);西部地带 (0.946)> (0.736)> (0.004)。 显示原图|下载原图ZIP|生成PPT 图5中国三大地带石油终端利用CO2排放量主导因素的决定力对比 -->Figure 5The power determinant value of dominant factors of CO2 emissions from petroleum terminals utilization in Three Economic Regions of China -->
再将影响石油终端利用CO2排放强度的汽油消费比、交通运输业石油强度和工业占GDP比重等三个主导因素在三大地带进行决定力对比(图6)。碳排放强度决定力分别为: 东部地带 (0.477)> (0.469)> (0.314);中部地带 (0.577)> (0.447)> (0.088);西部地带 (0.683)> (0.496)> (0.025)。 显示原图|下载原图ZIP|生成PPT 图6中国三大地带石油终端利用CO2排放强度主导因素的决定力对比 -->Figure 6The power determinant value of dominant factors of CO2 emissions intensity from petroleum terminals utilizationin in Three Economic Regions of China -->
按照三大地带分层,利用地理探测器进行多级探测。根据决定力大小将30个省份划分为能源结构指向型、能源强度指向型、产业结构指向型、经济发展水平指向型和人口指向型等5种石油终端利用碳排放地域类型。根据自然断点法所获得13个探测因子的类别值分别与石油终端消费CO2排放量、石油终端消费CO2排放强度的类别值进行空间耦合匹配(图7和图8见第2241页),将探测因子分类等级与碳排放分类完全匹配的省份视为典型省份。 显示原图|下载原图ZIP|生成PPT 图72015年中国石油终端CO2排放量与影响因素耦合匹配 -->Figure 7The coupling map of regional petroleum terminal carbon emissions and influencing factors in China in 2015 -->
显示原图|下载原图ZIP|生成PPT 图82015年中国石油终端CO2排放强度与影响因素耦合匹配 -->Figure 8The coupling map of regional petroleum terminal CO2 emissions intensity and influencing factors in China in 2015 -->
进一步利用石油终端利用CO2排放量和石油终端利用CO2排放强度构建象限图,以平均值作为原点(图9)。 显示原图|下载原图ZIP|生成PPT 图92015年中国各省石油终端利用CO2排放强度和排放量的象限分析 注:图中碳排放量和碳排放强度均为取ln后的数值。 -->Figure 9The quadrant of petroleum terminal utilization of CO2 emissions and emission intensity in China in 2015 -->
由于市级行政单元是碳减排政策的主要执行行政区域和研究碳排放地域分异的主要区域,本研究限于数据,仅以省域为研究单元,对市、县等次一级区域的考虑不够。此外,在计算油类消费量时人为忽视了城镇和乡村的消费差别。地理探测器模型的运用可为典型地区和省份的碳减排方向提供了简便方法,可就典型省份进行多空间层次进行深入研究,用更多样本数据充实到研究中,将城乡差别等因素也纳入到探测因子中,为科学提出减排政策提供正确方向。 The authors have declared that no competing interests exist.
[National Information Center.Carbon Emission Control Target and Realization Path in the 13th Five-Year Plan Period[EB/OL]. (2017-05-08)[2017-06-06]. URL [本文引用: 1]
[ZouY,Wang GS,Yu WJ,et al. An analysis of sectorial oil consumption track in typical countries and its implications to China's trend [J]. ,2010,31(5):666-672.] [本文引用: 1]
[Qin YC,Rong PJ,Yang QT,et al. Research progress of impact of urbanization on carbon emissions [J]. ,2014,33(11):1526-1534.] [本文引用: 1]
[6]
ShenL,Sun YZ.Review on carbon emissions,energy consump-tion and low-carbon economy in China from a pers-pective of global climate change [J]. ,2016,26(7):855-870. [本文引用: 2]
[ Li LN,Loo B P Y.Carbon dioxide emissions from passenger transport in China:geographical characteristics and future challenges [J]. ,2016,35(7):1230-1242.] [本文引用: 2]
[ Chen JL,Li PX,Gao JL.Spatiotemporal patterns and influencing factors of carbon emissions in the Pan-Yangtze River Delta region,1990-2014 [J]. ,2016,35(12):1472-1482.] [本文引用: 3]
[YuanY,Xi QM,Sun TS,et al. The impact of the industrial structure on regional carbon emission:empirical evidence across countries [J]. ,2016,35(1):82-94.] [本文引用: 1]
[12]
Zhang LQ,Chen SP,Zhu YW,et al. The measurement of carbon emission effect of construction land changes in Anhui Province based on the extended LMDI model [J]. ,2013,4(2):186-192. [本文引用: 1]
[WangJ,YangJ.Regional carbon emission intensity disparities in China based on the decomposition method of shapley value [J]. ,2014,36(3):557-566.] [本文引用: 2]
[Tao YG,Huang ZF,Wu LM,et al. Measuring carbon dioxide emissions for regional tourism and its factor decomposition:a case study of Jiangsu province [J]. ,2014,69(10):1438-1448.] [本文引用: 2]
[WangH,Chen CC,PanT,et al. County scale characteristics of CO2 emission's spatial-temporal evolution in the Beijing-Tianjin-Hebei metropolitan region [J]. ,2014,35(1):385-393.] [本文引用: 1]
[Deng JX,LiuX,WangZ.Characteristics analysis and factor decom-position based on the regional difference changes in China’s CO2 emission [J]. ,2014,29(2):189-200.] [本文引用: 4]
[Wang CJ,Zhang XL,Zhang HO,et al. Influencing mecha-nism of energy-related carbon emissions in Xinjiang based on IO-SDA model [J]. ,2016,71(7):1105-1118.] [本文引用: 2]
[Cheng YQ,Wang ZY,Zhang SZ,et al. Spatiotemporal dynamics of carbon intensity from energy consumption in China [J]. ,2014,24(4):631-650.] [本文引用: 1]
[21]
ZhangW,Shen YM,Zhou YY.Increased CO2 emissions from energy consumption based on three-level nested I-O structural decomposition analysis for Beijing [J]. ,2014,5(2):115-122. [本文引用: 1]
[Zhao YT,Huang XJ,Zhong TY,et al. Spatial pattern evolu-tion of carbon emission intensity from energy consumption in China [J]. ,2011,32(11):3145-3152.] [本文引用: 1]
[Wu XR,Zhang JB,TianY,et al. Provincial agricultural carbon emis-sions in China:calculation,performance change and influencing factors [J]. ,2014,36(1):129-138.] [本文引用: 1]
[25]
GuoJ,Liu ZX,SunP.Carbon emissions from industrial sectors in China:driving factors and the potential for emission reduction [J]. ,2013,4(2):132-140. [本文引用: 1]
[LiX,Qin YC,Zhang LJ,et al. Research progress on carbon emissions in accommodation industry [J]. ,2013,32(3):408-415.] [本文引用: 1]
[27]
Li YM,Zhao JF,Liu GS.Decomposition analysis of carbon emissions growth of tertiary industry in Beijing [J]. ,2015,6(5):324-330. [本文引用: 1]
[28]
Wang JF,Li XH,ChristakosG,et al. Geographical detectors-based health risk assessment and its application in the neural tube defects study of the Heshun Region,China [J]. ,2010,24(1):107-127. [本文引用: 2]
[DingY,Cai JM,Ren ZP,et al. Spatial disparities of economic growth rate of China's National-level ETDZs and their determinants based on geographical detector analysis [J]. ,2014,33(5):657-666.] [本文引用: 1]
[ZhuH,Liu JM,TaoH,et al. Temporal-spatial pattern and contributing factors of urban RBDs in Beijing [J]. ,2015,70(8):1215-1228.] [本文引用: 1]
[32]
HuY,WangJ,LiX,et al. Geographical detector-based risk assessment of the under-five mortality in the 2008 Wenchuan Earthquake,China [J]. ,2011,6(6):e21427. [本文引用: 1]