LIU Wei ,JIA Zhijie ,DU Meng ,DONG Zhnfeng ,PAN Jieyu ,LI QinruiPAN LinynChris UMOLE
a College of Resources and Environment,Chengdu University of Information Technology,Chengdu,610225,China
b Environmental Planning Institute,Ministry of Ecology and Environment,Chengdu,610225,China
Keywords：CO2 emissions Final energy consumption Logarithmic mean Divisia index (LMDI)model Industrial structure Grey relation analysis Sichuan Province
ABSTRACT Within the context of CO2 emission peaking and carbon neutrality,the study of CO2 emissions at the provincial level is few.Sichuan Province in China has not only superior clean energy resources endowment but also great potential for the reduction of CO2 emissions.Therefore,using logarithmic mean Divisia index (LMDI) model to analysis the influence degree of different influencing factors on CO2 emissions from final energy consumption in Sichuan Province,so as to formulate corresponding emission reduction countermeasures from different paths according to the influencing factors.Based on the data of final energy consumption in Sichuan Province from 2010 to 2019,we calculated CO2 emission by the indirect emission calculation method.The influencing factors of CO2 emissions originating from final energy consumption in Sichuan Province were decomposed into population size,economic development,industrial structure,energy consumption intensity,and energy consumption structure by the Kaya–logarithmic mean Divisia index (LMDI)decomposition model.At the same time,grey correlation analysis was used to identify the correlation between CO2 emissions originating from final energy consumption and the influencing factors in Sichuan Province.The results showed that population size,economic development and energy consumption structure have positive contributions to CO2 emissions from final energy consumption in Sichuan Province,and economic development has a significant contribution to CO2 emissions from final energy consumption,with a contribution rate of 519.11%.The industrial structure and energy consumption intensity have negative contributions to CO2 emissions in Sichuan Province,and both of them have significant contributions,among which the contribution rate of energy consumption structure was 325.96%.From the perspective of industrial structure,secondary industry makes significant contributions and will maintain a restraining effect;from the perspective of energy consumption structure,industry sector has a significant contribution.The results of this paper are conducive to the implementation of carbon emission reduction policies in Sichuan Province.
Global climate change is a major issue facing humanity today,and greenhouse gas (GHG) emissions are a major driving force affecting global climate change.GHG emissions are determined by a combination of economic policies,energy policies,environmental policy requirements and fuel combustion technologies (Mahony,2013;Tursun et al.,2015;Mao et al.,2016;Nam et al.,2016;Gu et al.,2019;Alajmi,2021).At the same time,energy consumption is closely related to the economic development,population size,living standard,industrial structure,energy consumption structure,energy consumption intensity,and energy use technology level.
Based on the Kaya–logarithmic mean Divisia index (LMDI) model,Yang et al.(2019) examined the influencing factors of CO2emissions in China from 1996 to 2016 and found that economic activity is the largest driving force of CO2emissions,while energy consumption intensity is the largest restraining factor.Through the use of the Kaya–LMDI model to extend the Kaya identity to decompose CO2emission change into predetermined influencing factors,the energy-related CO2emission increments in Pakistan from 1990 to 2014 were studied (Lin and Ahmad,2016).Ma et al.(2018) employed the LMDI model based on the Sankey diagram in energy and CO2emission analysis of the contribution of various influencing factors to energy-related CO2emission growth at the national level.The results reveal that gross domestic product (GDP) per capita growth is the main factor driving CO2emission growth,while the reduction in energy consumption intensity,improvement in the energy supply efficiency,and introduction of nonfossil energy sources in cogeneration inhibited CO2emission growth.An extended LMDI decomposition model was used to decompose the influencing factors of CO2emissions in China’s manufacturing industry into eight effects,and the results indicated that iron metal smelting and rolling are the largest sources of CO2emissions.The industrial activity is the most important factor leading to an increase in CO2emissions in manufacturing.The energy consumption intensity is the most important factor contributing to a decrease in CO2emissions in manufacturing,and there are differences in the drivers of CO2emissions in manufacturing between different periods (Liu et al.,2019).This situation is consistent with the international and domestic economic development environment and relevant policies of the Chinese government regarding the close relationship between energy savings and emission reduction.However,the factors affecting CO2emission reduction varies among industries (Liu et al.,2019).
With the use of system dynamics (SD) and the LMDI model to investigate the change in CO2emissions in Shanghai City of China and the selection of additional factors,such as travel,private car ownership and urban structure,and relatively new influencing factors such as income level,Gu et al.(2019) revealed that per capita GDP is the main driving factor of CO2emission growth,and the energy consumption intensity is the main influencing factor of CO2emission reduction.Moreover,private car ownership and income are positively correlated with CO2emissions,while the economic structure and residential energy consumption intensity are negatively correlated with CO2emissions(Gu et al.,2019).
Based on the LMDI model,we studied the influencing factors of CO2emissions of transportation industry in China and quantified the intensity of the pollutant discharge,energy consumption structure,transportation intensity,technology,labour input and capital effects,and the results revealed that the capital investment effect is a key factor in promoting CO2emissions,while the technology state is the main factor limiting CO2emissions (Liu et al.,2021).Through the application of the LMDI decomposition model in factor decomposition of industrial energy consumptionrelated CO2emissions in Dalian City of China,Chen et al.(2016) demonstrated that the economic growth effect is the most important factor influencing industrial CO2emissions,while the industrial structure is the most notable negative factor of industrial CO2emission increase.In terms of energy,coal remains the largest contributor to China"s energy-related CO2emissions;among end-use sectors,industry is the biggest contributor,and emissions reductions should focus on high-energy consumption sectors such as steel,while emissions reductions for construction and transportation systems should focus on energy consumption of electrical appliances and passenger vehicle emissions(Yang et al.,2020).
Sichuan Province is the largest energy-consuming region in Southwest China and a base of high-quality clean energy in China,with the total energy consumption increased from 6353.00×104t of standard coal in 1990 to 20791.00×104t of standard coal in 2019 (Energy Statistics Division,National Bureau of Statistics,2020).In general,Sichuan Province is facing with difficulties and challenges such as insufficient space for energy savings and carbon reduction,the difficult task of industrial low-carbon transformation,and a high demand for fossil energy consumption,but this province also exhibits a relatively low CO2emissions level,continuous optimization of its energy structure,and a high potential of ecological carbon sinks.Specifically,the industrial low-carbon development process in Sichuan Province is effective.During the 13thFive-Year Plan period,the percentage of the three industries (primary,secondary,and tertiary industries) in Sichuan Province was optimized and adjusted from 12.10%,43.50%,and 44.40%,respectively in 2016 to 11.40%,36.20%,and 52.40%,respectively in 2020.By the end of 2020,the CO2emission of per capita GDP reached approximately 0.68 t and the CO2emission per capita reached 3.30 t;the former is in 6thand the latter is in 10th,respectively,among the ten major economic provinces in China.
To sum up,previous research selected countries,cities and industrial as the research boundary,and applied the LMDI model to study the influencing factors of CO2emissions from the perspectives of economic development,population size,industrial structure,energy consumption,and energy consumption structure.At present,there are few studies on the contribution of CO2emissions originating from final energy consumption (abbreviated as CO2emissions) at the provincial scale.Sichuan Province has superior endowment of clean energy resources and great potential for CO2emission reduction (National Development and Reform Commission,2021).Therefore,this study used the LMDI decomposition model to analyse CO2emissions in Sichuan Province to find the key influencing factors.Specifically,we analysed the correlation between CO2emissions and possible influencing factors in Sichuan Province,and decomposed the influencing factors of CO2emissions.The results have important practical research significance to promote the implementation of the “double control” target and CO2emission peaking target by 2030 for Sichuan Province.
2.1.Study area
Sichuan Province (26°03′–34°19′N,97°21′–108°33′E) is located in the hinterland of Southwest China and in the upper reaches of the Yangtze River.The western Sichuan Province is plateau and mountainous regions,with altitude of higher than 3000 m.The eastern Sichuan Province is basins and hills,and the altitude ranges from 500 to 2000 m.The characteristics that the west of Sichuan Province is high and the east of Sichuan Province is low are particularly obvious.The eastern Sichuan Province,including the Sichuan Basin and the surrounding mountains,belongs to the central subtropical humid climate zone and is characterized by an oceanic climate;it is warm and humid throughout the year,with annual average temperatures of 16°C–18°C.The mountainous area of southwestern Sichuan Province is a subtropical semi-humid climate zone,with annual average temperatures of 12°C–20°C and annual precipitation values of 900–1200 mm.Northwestern Sichuan Province belongs to the alpine plateau climate zone,with annual average temperatures of 4°C–12°C and annual precipitation values of 500–900 mm (The People’s Government of Sichuan Province,2021).
Due to the significant energy advantages of Sichuan Province in water resources and natural gas,and the continuous exploration of natural gas,Sichuan Province ranked the first in natural gas consumption in China in 2015,at 17.10×109m3.With the comprehensive development of eastern Sichuan Province,central Sichuan Province,northwestern Sichuan Province,and southern Sichuan Province,the reserves of proven natural gas in Sichuan Province are increasing.At present,the recoverable resources of conventional gas,tight gas,and shale gas have reached 26.45×1012m3,also ranking the first in China,and there is still a greater room for growth.The electricity energy consumption in Sichuan Province is generally developing,because the province has large precipitation and hydropower contributes the most (Chen,2018).
2.2.Data sources
The data in this paper were obtained from the Sichuan Statistical Yearbook in 2020 (Energy Statistics Division,National Bureau of Statistics,2021) and China Energy Statistical Yearbook from 2011 to 2020 (Energy Statistics Division,National Bureau of Statistics,2012–2021),where the GDP and industry output values were converted at comparable prices in 2010.Because there is no financial industry or real estate industry in the industry classification of the China Energy Statistical Yearbook (Energy Statistics Division,National Bureau of Statistics,2012–2021),this paper combined the industry classification methods of the China Energy Statistical Yearbook from 2011 to 2020(Energy Statistics Division,National Bureau of Statistics,2012–2021) and Sichuan Statistical Yearbook 2020 (Energy Statistics Division,National Bureau of Statistics,2021) and classified the financial industry and real estate industry into other service sectors.At the same time,wholesale and retail trade and hotels and catering services were combined into one sector.Finally,six industries were obtained,agriculture,forestry,animal husbandry and fishery sector,industry sector,construction sector,transportation,storage and post sector,wholesale,retail trade,hotels and catering service sector,and other service sectors.Energy consumption was determined as the sum of each energy source converted into standard coal.We use resident population and each energy consumption according to the original data.
2.3.Emission factor method
The calculation of CO2emissions can be divided into direct and indirect emission calculation methods.The direct emission calculation method in the 2019 Refinement of the 2006 Intergovernmental Panel on Climate Change (IPCC)Guidelines for National Greenhouse Gas Inventories,compiled and revised by the Intergovernmental Panel on Climate Change (IPCC,2019),is a commonly used method for countries or regions to account for CO2emissions.
In this work,we chose the direct emission calculation method to calculate related CO2emissions in Sichuan Province.The average low-level heat of energy (B) and standard coal coefficient (S) were obtained from the General Rules for the Calculation of the Comprehensive Energy Consumption (GB/T 2589-2020) (China National Institute of Standardization et al.,2020),and the carbon content per unit calorific value (A) and carbon oxidation rate (R) were retrieved from the Guidelines for the Preparation of Provincial Greenhouse Gas Inventories (Trial) (NCSC,2011),as indicated in Table 1.
Table 1 Various reference coefficients for the different fuel types.
Electricity consumption will not generate CO2,but fossil energy burning in the process of electricity production in coal-,oil-,and gas-fired power plants can emit CO2,so CO2emissions originating from electricity consumption were considered in this paper.We selected the emission factor for electricity of 0.52527 t CO2/MWh referring to the Greenhouse Gas Emission Report of the State Grid Sichuan Electric Power Company (State Grid Sichuan Electric Power Company,2018),and chose the electricity-standard coal conversion coefficient of 0.1229 kgce/(kW•h) based on the General Rules for the Calculation of the Comprehensive Energy Consumption (GB/T 2589-2020).
Based on the 2019 Refinement of the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC,2019),the equation for CO2emission accounting is as follows:
whereCdenotes CO2emissions (t);tdenotes the year;idenotes the sector;jdenotes the energy type;Edenotes the energy consumption (kg or m3);andKdenotes the CO2emission factor (kg CO2/kg or kg CO2/m3).
The CO2emission factors for the different energy sources can be calculated based on each of the reference factors listed in Table 2 with the following equation:
whereAdenotes the average low-level heat generation(t C/TJ) ;Bdenotes the carbon content per unit calorific value(kJ/kg or kJ/m3) ;andRdenotes the carbon oxidation rate.
2.4.Kaya–logarithmic mean Divisia index (LMDI) model
According to a related study by Ang (2005),the idea of the LMDI model is to view decomposed factor variables as continuous differentiable functions of time,and the time is then differentiated to decompose the contribution of each factor variable to the target change (Tursun et al.,2015;Ortega-Ruiz et al.,2020).This model provides the following advantages (Ang,2005;Fan et al.,2007): (1) the model exhibits a solid theoretical foundation;(2) this model is simple to use and adaptable;(3) the decomposition term equation is independent and simple and achieves a suitable ability to explain the target;(4) the decomposition results do not contain unexplained residual terms;and (5)the problem of zero and negative values can be resolved.In this paper,we chose the additive form of the LMDI model to study the contribution of influencing factors to the amount of variation in CO2emissions and extended the LMDI model according to the study of Ang (2005).
The influencing factors of CO2emissions originating from final energy consumption in Sichuan Province were decomposed into population size,economic development,industrial structure,energy consumption intensity,and energy consumption structure by the LMDI decomposition model.
We first obtained the deformation expansion based on the Kaya’s constant equation (Jiang et al.,2016;Shen et al.,2016):
The amount of variation in CO2emissions between the base year (0) and the target year (t) can be decomposed into 6 effects shown in Equation 4:
Based on the additive LMDI model of Ang (2005) and the fact that the unwanted residual term can be decomposed into a weighting factor via the logarithmic mean of two positive numbers,we calculated the above 6 effects as follows:
The definitions of the variables in Equations 3–10 are provided in Table 2.
Table 2 Definition of variables in Equations 3–10.
2.5.Grey relation analysis
In grey relation analysis,examination and comparison processes are performed according to the development trend,and this method exhibits low requirements for the number of samples and the correlation among sample sequences;notably,no typical distribution pattern is required (Pan et al.,2011;Chai et al.,2012;Huang et al.,2019).There are many influencing factors of CO2emissions,but there exists no obvious distribution pattern.Through grey relation analysis,the relationship between CO2emissions and influencing factors can be obtained to determine whether the selection of influencing factors is reasonable.According to the equation of the LMDI decomposition model and considering the definitions of the variables in Table 2,this paper identified 20 influencing factors of CO2emissions in Sichuan Province,as listed in Table 3.
In this paper,Deng’s grey relation analysis method was used to investigate the influencing factors of CO2emissions,and the specific steps are as follows (Liu et al.,2019):
(1) Determination of the data sequence.
It is assumed that the sequence of features can be denoted asX0=(X0(1),X0(2),…,X0(n)),and compare sequencesXi=(Xi(1),Xi(2),…,Xi(n)),wherei=1,2,…,m,andk=1,2,…,n.
(2) Performance of data dimensionless processing.
The dimensionless processing approach for the original data adopts the initial values and the following calculation equation:
whereX" iis the initial value ofXi.
(3) Calculation of the absolute difference sequence of the feature and comparison sequences.
Absolute difference calculation equation is as follows.
(4) Determination of the maximum and minimum absolute difference values.
Table 3 Influencing factors of CO2 emissions originating from final energy consumption (abbreviated as CO2 emissions) in Sichuan Province.
whereMdenotes the maximum value,andmdenotes the minimum value.
(5) Calculation of the number of grey relation coefficients
whereξis the discrimination coefficient,ranging from 0 to1,usually chosen as 0.5;andγis the number of grey relation coefficients.
(6) Calculation of the grey relation degree.
The average value of the grey relation coefficient is the requested correlation degree,which serves as a quantitative indicator of the magnitude of the relationship between the characteristic and comparative series,and can be calculated with the following equation:
3.1.Grey relation analysis of CO2 emissions
We used grey relation analysis to ensure the relationship between CO2emissions and the selected 20 influencing factors.As indicated in Table 4,the correlation degree between CO2emissions and the selected 20 influencing factors was higher than 0.6000,and the correlation degree between CO2emissions and 18 influencing factors was higher than 0.9000.It is noted that only the fuel oil and liquefied petroleum gas (LPG) consumption exhibited low correlation degrees with CO2emissions.Therefore,the selected 20 influencing factors exhibited a suitable correlation with CO2emissions,which could be used to analyze the CO2emission contribution in Sichuan Province with the LMDI model.
Table 4 Grey relation analysis results for the correlation degree between CO2 emissions and the selected 20 influencing factors in Sichuan Province.
3.2.Decomposition analysis of CO2 emissions
3.2.1.Decomposition analysis of different industrial structure sectors to CO2 emissions
Figure 1 shows CO2emissions of various industrial structure sectors in Sichuan Province from 2010 to 2019.The results revealed that CO2emissions in Sichuan Province exhibited an overall rising-falling-rising change trend from 2010 to 2019.Compared to 2010,CO2emissions in 2019 increased by 4672.63×104t,an increase of 20.72%.CO2emissions in Sichuan Province peaked at 27,604.99×104t in 2012,while the value also peaked in the same year for the industry sector.The trends of CO2emission for the industry sector,agriculture,forestry,animal husbandry and fishery sector,and construction sector remained stable.In contrast,the trends of CO2emission for the wholesale,retail trade,hotels and catering service sector,transportation,storage and post sector,and other service sectors were increasing.It could be found that CO2emissions originating from industry sector were the main contribution source of CO2emissions in this region.
Fig.1.CO2 emissions originating from final energy consumption (abbreviated as CO2 emissions) for different industrial structure sectors in Sichuan Province from 2010 to 2019.
Figure 2 shows the contribution degree of all industrial structure sectors to CO2emissions in Sichuan Province from 2010 to 2019.As shown in Figure 2,the contribution degree of industry sector to CO2emissions was the highest,accounting for 81.00% of the total emissions in 2012.Transportation,storage and post sector ranked the second,with the contribution degree of 12.00%.The contribution of wholesale and retail trade,hotels and catering service sector and other service sectors to CO2emissions varied between 4.00% and 7.00%.Furthermore,the contribution degrees of agriculture,forestry,animal husbandry and fishery sector and construction sector to CO2emissions were relatively low,with the values varying between 1.00% and 3.00%.
3.2.2.Decomposition analysis of differentdecomposed influencing factors to CO2 emissions
Based on the established LMDI decomposition model,we decomposed the contribution amount and degree of the population size,economic development,industrial structure,energy consumption intensity,energy consumption structure and CO2emission factor to the variation in CO2emissions in Sichuan Province (Fig.3).According to Equation 10,when the CO2emission coefficients of different energy sources remain unchanged,the contribution amount and degree of CO2emission factors to CO2emission variation were zero.
As shown in Figure 3,the variation in CO2emissions caused by the decomposed influencing factors in Sichuan Province exhibited an increasing tend from 2011 to 2012,a decreasing trend from 2012 to 2016,and an increasing trend again from 2016 to 2019.During 2011–2019,population size and economic development played a continuous promotion role in CO2emissions,and contribution of economic development was much larger than that of population size.From 2011 to 2019,the industrial structure played a continuous restraining role in CO2emissions,while energy consumption intensity also showed a continuous restraining role (except in 2012).The contribution of energy consumption intensity was generally larger than that of industrial structure.The effect of energy consumption structure on CO2emissions fluctuated,and the contribution was relatively small from 2011 to 2019.
The cumulative contribution degrees of population size,economic development,industrial structure,energy consumption intensity,and energy consumption structure to the variation in CO2emissions reached 22.43%,519.11%,–124.84%,–325.96%,and 9.30%,respectively.The cumulative contributions of economic development,industrial structure,and energy consumption intensity to the variation in CO2emissions were much greater than those of population size and energy consumption structure.To sum up,economic development,industrial structure,and energy consumption intensity significantly impacted CO2emissions in Sichuan Province.
Fig.2.Contribution degree of different industrial structure sectors to CO2 emissions in Sichuan Province from 2010 to 2019.
Fig.3.Variation in CO2 emissions caused by the decomposed influencing factors (population size,economic development,industrial structure,energy consumption intensity,and energy consumption structure) as well as the annual difference of CO2 emissions in Sichuan Province from 2011 to 2019.The values of the left axis refer to the difference of CO2 emissions between the adjacent two years,while the values of the right axis refer to the difference between positive and negative values in the current year.
Through the empirical study of panel quantile regression,Liu et al.(2012) showed that the impact of economic development level on CO2emissions was greater than that of population factor.Through the factor decomposition model,Li et al.(2019) concluded that economic growth was the main driving factor of CO2emissions in Sichuan Province.By analyzing the characteristics of CO2emissions from fossil energy consumption in typical cities,Zheng et al.(2020) pointed out that both economic development and population size were promoting factors for the growth of CO2emissions in four types of cities: city of rapid economic development,city of high CO2emissions,city of low CO2emissions,and city of low CO2emissions but rapid economic development,but the cumulative contribution of economic development to CO2emissions was greater than that of population size.According to the results of this study,population size and economic development played a role in promoting CO2emissions in Sichuan Province,and the contribution of economic development was higher than that of population size,which is consistent with the empirical analysis.Through the empirical study of factor decomposition,it is concluded that there are many influencing factors of CO2emissions in China,and economic development,industrial structure,and energy consumption intensity are key influencing factors (Ning et al.,2012).
(1) Population size effect
Population size is closely related to CO2emissions,and a large population base is an important driving factor of China becoming the largest GHG emitter in the world (Chen,2017).Population size can affect the regional final energy consumption from several aspects,thereby affecting GHG emissions,including CO2emissions.These factors can often be divided into two categories,i.e.,positive and negative factors.The positive impact can be summarized as follows: with increasing resident population in a given area,the final energy demand increases with increasing resident population,eventually leading to increased CO2emissions.Negative influencing factors can be summarized as the increase in the number of permanent residents in a certain region,the continuous improvement of the regional urbanization level,improved urban energy supply system,public transport network and other factors to improve energy efficiency,as well as the continuous improvement of low carbon awareness of residents and other factors,to a certain extent,decrease final energy consumption,and ultimately reduce CO2emission level.
As shown in Figure 4,the change trend of the growth rate of resident population was consistent with the variation in CO2emissions,indicating that the impact of resident population growth on CO2emissions is greater than that of urbanization level enhancement.At the same time,the rural population’s awareness of environmental protection is poor;some rural areas cannot use clean energy,and the implementation of green low-carbon concepts and clean lifestyle patterns is ineffective,which can affect the CO2emission reduction in these rural areas (Chen,2017).
Fig.4.Variation in CO2 emissions caused by the resident population growth and growth rate of resident population from 2011 to 2019.Variation in CO2 emissions refers to the difference of CO2 emissions between the adjacent two years.
Using the improved Stochastic Impacts by Regression on Population,Affluence,and Technology model (STIRPA),Zhang et al.(2014) showed that the total number of households and household consumption were the main factors for the rapid increase of CO2emissions.According to the results of this study,population size played a continuous role in promoting CO2emission in Sichuan Province,and the cumulative CO2emission change is 1047.97×104t.The influence of population size on energy consumption is complex.Coupled with the relatively uneven distribution of population in various regions of Sichuan Province,the growth of permanent population will accelerate the urbanization process,and the demand for infrastructure will increase,so the energy consumption of infrastructure construction process will also increase significantly.At the same time,because of the increase in infrastructure construction and the continuous improvement of urbanization level,it has contributed to the reduction of energy consumption intensity to a certain extent.Further decomposing population factors and studying the contribution of different population factors to CO2emissions will help Sichuan Province complete the transition to green and lowcarbon development and complete the task of carbon emission reduction.
(2) Economic development effect
The per capita GDP can not only reflect the economic development level in a country or region but also reflect the wealth of residents.It is generally acknowledged that upon improvement in people’s living standards,the total amount of energy consumption increases,which exerts an increasingly significant impact on the environment.
As indicated in Table 5,the absolute values of the contribution of economic development to CO2emissions were greater than 100.00% during 2011–2019 (except for 2015 and 2016),and economic development was the leading influencing factor of the variation in CO2emissions in Sichuan Province.
Table 5 Contribution amount and degree of economic development to CO2 emissions.
According to the LMDI decomposition results,economic development continuously impacted CO2emissions in Sichuan Province,and the variation in cumulative CO2emissions from 2011 to 2019 was 24,255.98×104t.As shown in Figure 5,the change trend of the contribution of economic development to CO2emissions from 2011 to 2019 was consistent with that of the growth rate of per capita GDP,which indicated that the growth rate of per capita GDP promotes CO2emissions in Sichuan Province.Under the influences of the economic scale,labour cost,population structure,resource environment,and other factors,the growth rate of per capita GDP exhibited a downward trend in Sichuan Province.The economic development level in Sichuan Province lags behind national development and structure adjustment levels (Sichuan Provincial Bureau of Statistics,2021).To accelerate development,it is suggested to highlight the comprehensive promotion of service industry development level,facilitate the development of service industry convergence with other service sectors,accelerate investment structure adjustment,increase investments driving consumer upgrades,implement measures to promote a dynamic balance between investment and consumption levels,and mitigate CO2emission peaking and carbon neutrality policy pressures.All these can inhibit energy consumption and CO2emissions in Sichuan Province.
Fig.5.Variation in CO2 emissions caused by the economic development and growth rate of per capita GDP from 2011 to 2019.Variation in CO2 emissions refers to the difference of CO2 emissions between the adjacent two years.
(3) Industrial structure effect
The secondary industry and tertiary industry play an important role in the economic development process of Sichuan Province and comprise the main sources of CO2emissions.Adjustment and change in the structure of different industries and within each industry can affect CO2emissions.As indicated in Table 6,the cumulative CO2emission variation in industrial structure reached–5834.73×104t during 2011–2019.Agriculture,forestry,animal husbandry and fishery sector and industry sector continuously played an inhibiting role in cumulative CO2emission variation.Industry sector was the main source of CO2emission reduction,and the cumulative CO2emission variation reached as high as–6684.59×104t.Construction sector,wholesale and retail trade,hotels and catering service sector,and other service sectors,mainly the real estate and finance industries,were the main contributors to cumulative CO2emission variation.
Table 6 Kaya–logarithmic mean Divisia index (LMDI) decomposition model results of industrial structure effect on the variation of CO2 emissions.
As shown in Figure 6,the proportion of industry output value in Sichuan Province continued to decline from 2010 to 2019,while the proportion of the output value of other service sectors (such as the real estate and financial industries) with a low energy consumption continued to increase,which is consistent with the change trend of CO2emissions (Table 6) in the industry sector and other service sectors.
Fig.6.Proportion of output value in different industrial structure sectors in Sichuan Province from 2010 to 2019.
Among all sectors,agriculture,forestry,animal husbandry and fishery sector exhibited inhibition effects on CO2emissions with a relatively limited contribution,which is related to the characteristics of a low energy consumption and certain carbon sequestration capacity of the industry in Sichuan Province.As the industrial structure in Sichuan Province continues to be optimized and adjusted,industry sector and construction sector can increasingly alleviate the problems of an excess capacity and consumption of high-aggregation performance sources,which can reduce the intensity of CO2emissions and thus inhibit CO2emissions (Chen and Li,2021).Moreover,with continuous improvement in the quality of life of residents,the energy consumption in the other service sectors continue to grow,thus eventually promoting CO2emissions.During the study period,industrial CO2emissions in Sichuan Province exhibited a downward trend,but the industrial energy consumption-related CO2emissions remained the main source of CO2emissions,accounting for 71.00% of the total emissions in 2019.Therefore,further optimization of the industrial structure and reduction in industrial CO2emissions should remain the focuses of CO2emission reduction work in Sichuan Province.
(4) Energy consumption intensity effect
Energy consumption intensity is one of the most commonly used indicators to compare the comprehensive energy utilization efficiency among different countries or regions.The lower the energy consumption intensity,the higher the energy utilization efficiency and the more developed the energy utilization technology (Chen,2017).
According to the LMDI decomposition model results,the cumulative CO2emission variation from 2011 to 2019 reached–15231.02×104t,and the cumulative contribution degree was–325.96%,energy consumption intensity effect is an important influencing factor in mitigating the rapid growth of CO2emissions in Sichuan Province.Energy consumption intensity of the various industries mainly played an inhibiting role in regard to CO2emissions,among which the cumulative CO2emission variation under the energy consumption intensity effect in industry sector reached–11,160.20×104t,accounting for 73.27% of the total amount,followed by the cumulative CO2emission variation in transportation,storage and post sector,at–1383.50×104t,accounting for 9.08% of the total amount (Table 7).
Table 7 LMDI decomposition model results of energy consumption intensity effect on the variation of CO2 emission.
As shown in Figure 7,the energy consumption intensity of different sectors in Sichuan Province decreased year by year from 2010 to 2019.Agriculture,forestry,animal husbandry and fishery sector,construction sector,wholesale,retail trade,hotels and catering service sector and other service sectors,which are dominated by service industries,all maintained a relatively low energy consumption intensity,and the energy consumption intensity of these four industrial sectors was lower than 0.30 t of standard coal per 104CNY.In contrast,the energy consumption intensity of transportation,storage and post sector and industry sector was relatively high,showing a decreasing trend from 2010 to 2019.The energy consumption intensity of transportation,storage and post sector decreased from 2.02 t of standard coal per 104CNY to 1.05 t of standard coal per 104CNY,a reduction of 92.38%.The decline of energy consumption intensity in industry was relatively small,with a decrease of 0.40 t of standard coal per 104CNY.In conclusion,the change trend of the energy consumption intensity of all industries is consistent with that of the contribution of the energy consumption intensity effect of all industries to CO2emissions.In terms of energy consumption intensity changes,the variation in transportation,storage and post sector was greater than that in industry,but in terms of all energy consumption types and CO2emissions,the consumption in industry sector was much higher than that in transportation,storage and post sector.Therefore,the cumulative contribution of industry sector in the energy consumption intensity effect was much greater than the cumulative contribution of transportation,storage and post sector.
(5) Energy consumption structure effect
The changes in the overall energy consumption structure and energy consumption structure of high energyconsuming industries can impact CO2emissions (Chen,2017).The consumption of fossil fuels such as coal and oil directly impacts CO2emissions because CO2emissions of fossil energy are much higher than those of other energy sources,and the combustion process produces a large amount of CO2.Therefore,a reasonable energy consumption structure could promote regional CO2emission reduction.
As shown in Figure 8,the energy consumption structure in Sichuan Province mainly includes raw coal,coke,gasoline,diesel oil,natural gas,and electricity.The proportion of raw coal consumption continued to decline from 2010 to 2017 but sharply increased and then decreased in 2018,with an average annual proportion of 21.51% in the total energy consumption structure.From 2010 to 2019,the proportion of coke consumption first increased and then decreased and peaked at 18.29% in 2015,with an annual average proportion of 15.44%.The proportion of gasoline consumption first increased and then decreased from 2010 to 2019,with a narrow variation range (between 6.85%and 9.61%) and an annual average proportion of 7.84%.Diesel oil consumption maintained a steady growth from 2010 to 2019,with an average annual growth rate of 4.38% and an average annual proportion of 11.38%.Natural gas consumption decreased from 2010 to 2012 and increased from 2013 to 2019,accounting for 13.09% on average.The overall electricity consumption exhibited a continuous growth trend during the study period,with an average annual growth rate of 4.32% and an average annual proportion of 20.28%.
Fig.7.Energy consumption intensity of different sectors in Sichuan Province from 2010 to 2019.
Fig.8.Proportion of energy consumption structure in Sichuan Province from 2010 to 2019.LPG,liquefied petroleum gas.
Industry accounted for the largest proportion of energy consumption structure (including 11 energy types) in Sichuan Province (Fig.9).More than 95.00% of raw coal,coal washed and dressed,coke,and crude oil were used in the industry sector.The consumption of industry sector in fuel oil,LPG,natural gas and electricity also accounted for 88.52%,31.67%,80.72%,and 78.41%,respectively.The energy consumption of gasoline and kerosene occurred in transportation,storage and post sector,accounting for 41.21% and 86.97%,respectively.The consumption of diesel oil occupied the largest proportion of 37.90% in the construction sector,followed by the consumption in transportation,storage and post sector,accounting for 36.38%.
As shown in Figure 10,the energy consumption of agriculture,forestry,animal husbandry and fishery sector mainly involved diesel oil,accounting for 84.20% of the total energy consumption,followed by electricity,occupying 6.94%of the total amount.The energy consumption of industry mainly included raw coal,coke,electricity,and natural gas,totally accounting for 86.83% of the total industrial energy consumption in this sector,among which raw coal occupied the largest proportion,at 28.87%.The energy consumption of construction sector largely included gasoline,electricity,diesel oil,and natural gas,accounting for 81.49% of the total energy consumption in construction sector,in which gasoline accounted for the largest proportion,at 27.98%.Gasoline,kerosene,and diesel oil accounted for 87.18% of the total energy consumption of transportation,storage and post sector,while electricity and natural gas occupied 10.68% of the total energy consumption in this sector.The energy consumption of wholesale,retail trade,hotels and catering service sector was dominated by gasoline,diesel oil,natural gas,and electricity,accounting for up to 91.82% of the total energy consumption.Moreover,the energy consumption of other service sectors mostly involved gasoline,diesel oil,and electricity,accounting for 84.92% of the total energy consumption,in which electricity had the largest proportion,at 34.13%.
Fig.9.Proportion of industrial structure in different energy consumption structure types in Sichuan Province.
Fig.10.Proportion of energy consumption structure in different industrial structure types in Sichuan Province.
According to the LMDI decomposition model results,the effect of energy consumption structure on the cumulative CO2emission variation reached 434.44×104t,with a cumulative contribution degree of 9.30% (Fig.11).From the perspective of energy,the cumulative CO2emission variation in electricity energy consumption reached 2782.12×104t,with a contribution degree of 640.39%,which is the main energy source promoting CO2emissions,followed by natural gas,diesel oil,and LPG,with the contribution degrees of 160.85%,109.12%,and 16.19%,respectively.The cumulative CO2emission variation in raw coal energy consumption reached–2218.39×104t,with a contribution degree of–510.63%,which was the main energy source suppressing CO2emissions,followed by gasoline,coal washed and dressed,kerosene,and coke.Their contribution degrees reached–139.28%,–105.75%,–35.99%,and–33.78%,respectively.The contribution of crude oil and fuel oil to the cumulative CO2emission variation was very small and could be ignored.
Fig.11.Cumulative CO2 emission variation in different energy consumption structure types in Sichuan Province.Cumulative CO2 emission variation refers to the accumulation of the difference CO2 emissions between the adjacent two years from 2010 to 2019.
Fig.12.Cumulative CO2 emission variation in different industrial structure types in Sichuan Province.Cumulative CO2 emission variation refers to the accumulation of the difference in CO2 emissions between the adjacent two years from 2010 to 2019.
From the perspective of industrial structure,Figure 12 shows that the cumulative CO2emission variation in industry sector reached 205.12×104t,with a contribution degree of 47.21%,which was the main contribution source of the cumulative CO2emission variation,followed by other service sectors,wholesale,retail trade,hotels and catering service sector,transportation,storage and post sector,and construction sector,with the contribution degrees of 20.16%,18.04%,12.30%,and 7.19%,respectively.The cumulative CO2emission variation in agriculture,forestry,animal husbandry and fishery sector reached–21.31×104t,with a contribution degree of–4.91%.
Based on the final energy consumption data in Sichuan Province from 2010 to 2019,the present work calculated CO2emissions by the direct emission calculation method and determined the influencing factors using grey relation analysis.Economic development and energy consumption intensity were the key factors promoting and inhibiting the growth in CO2emissions,respectively.Cumulative contributions of population size,economic development,and energy consumption structure promoted CO2emissions,among which the contribution degree of economic development reached as high as 519.11%,while the contribution degree of population size and energy consumption structure were only 22.43% and 9.30%,respectively.The contributions of industrial structure and energy consumption structure suppressed CO2emissions in Sichuan Province,and the contribution degrees of these two effects reached–124.84% and–325.96%,respectively.In terms of industrial structure effect,as Sichuan Province further promotes the optimization and upgrading of the industrial structure,the proportion of the secondary industry was continuously declining,so the industrial structure played a suppressing role on CO2emissions.From the perspective of energy consumption intensity,the reduction in energy consumption intensity could greatly contribute to CO2emission reduction,especially in industries with a high energy consumption and relatively high energy consumption intensity.
Under the effect of energy policies and CO2emission peaking goals,energy consumption intensity will inhibit CO2emissions by further improving the energy utilization efficiency and continuously improving the level of new energy development and utilization.The current inhibitory effect of energy consumption structure is not stable and obvious.Optimization and adjustment of energy consumption structure usually requires a long period.Renewable energy sources such as solar energy,hydropower and wind energy in Sichuan Province constitute national high-quality clean energy base.With the development and utilization of renewable energy sources,the future clean and low-carbon energy consumption structure will exert an increasingly obvious inhibiting effect on CO2emissions in Sichuan Province.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was financially supported by the National Natural Science Foundation of China (41771535) and the National Social Science Foundation Major Project (20&ZD092).