Evolution of Logistics Industry Carbon Emissions in Heilongjiang Province, China

Abstract

This paper analysed the spatio-temporal distribution characteristics and impact mechanism of the logistics carbon emissions in China’s Heilongjiang Province. Different levels of logistics carbon emissions during 2000–2020 were calculated using an energy-based emission factor method, and the Tapio decoupling model was used to analyse the decoupling effect between carbon emissions and economics. Then, a hotspot analysis and a spatial autocorrection analysis were used to study the spatial-temporal patterns of logistics carbon emissions. Finally, the influencing factors were analysed by using the logarithmic mean divisia index. Results showed that logistics carbon emissions in Heilongjiang increased from 1.66 million tons in 2000 to 6.21 million tons in 2020. The overall decoupling state was expansion negative decoupling, and it turned into strong decoupling in the last five years. The hotspots of logistics carbon emissions were mainly concentrated in developed cities in the south, and the coldspots were mainly in undeveloped regions in the north. Economic development was the main factor driving the increase of logistics carbon emissions, and logistics development was the main factor inhibiting the growth of logistics carbon emissions. The study results can provide references for formulation region-specific carbon reduction measures, which are beneficial to the low-carbon sustainable development of the logistics industry in this region.

1. Introduction

The excessive emission of greenhouse gases (GHG) is the main contributor to global warming, which has been gaining immense attention globally [1]. In order to reduce greenhouse gas emissions and mitigate climate change, the United Nations Framework Convention on Climate Change (UNFCCC) was established in 1992 as the basis for a global response to tackle the climate change issue. After that, a series of international agreements were put forward including the Kyoto Protocol (1997), the Copenhagen Accord (2009), and the Paris Agreement (2016), of which the Kyoto Protocol and the Paris agreement are legally binding [2,3]. The former mandated that the obligated industrialized countries cut their GHG by 5.2% compared to the year 1990, and the latter included commitments from industrialized and developing countries to cut their emissions and work together to adapt to the impacts of climate change, and called on countries to strengthen their commitments over time [2,3]. Carbon dioxide is the main component of greenhouse gases, so reducing carbon emissions has become a major concern for countries all over the world [4,5]. China is the largest emitter of carbon dioxide gas in the world, with 9.9 billion metric tons in 2019 [6]. On 22 September 2020, the Chinese President Xi Jinping announced at the General Debate of the 75th Session of the United Nations General Assembly that China will scale up its intended nationally determined contributions by adopting more vigorous policies and measures and strive to peak carbon dioxide emissions by 2030 and achieve carbon neutrality by 2060. In order to achieve the “double carbon” goal, the national key industries and enterprises should play a leading role in reaching carbon emission peak and realizing carbon neutrality by adopting scientific and efficient system solutions.

The logistics industry is an important industry for the national economy, which connects production, manufacturing, and consumption. It has played an indispensable role in promoting economic development, adjusting the economic structure, expanding domestic demand, and improving social welfare. However, the logistics industry consumes a mass of energy in transportation, warehousing, distribution, and other links, resulting in massive carbon dioxide emissions. It is considered to be the only industry in which carbon emissions continue to rise [7]. According to the statistics, the carbon emissions of the transportation, warehousing, and postal services in China was 204.9 million tons in 2000, while this number climbed up to 732.5 million tons in 2019, an increase of 257% [8]. In addition, the development of China’s logistics industry in different regions was unbalanced due to the internal and external factors [9]. Therefore, it is of great significance to analyse the characteristics and contributing factors of regional logistics carbon emissions in order to formulate targeted carbon emission reduction measures.

Located in Northeast China, Heilongjiang Province is an important channel from Asia and the Pacific to Russia and the European continent, which is an important border province of China [10]. Currently, the studies on carbon emissions in Heilongjiang Province mainly focused on traditional industries such as the agriculture industry and animal husbandry, while the studies on the carbon emissions of the fast-growing logistics industry were rarely reported. In addition, the economic development of different cities in this province was uneven, and the temporal and spatial distribution features of logistics carbon emissions and the decoupling effect between logistics carbon emissions and economics were unknown. The factors driving or inhibiting the provincial and city-level logistics carbon emissions were not clear as well. Therefore, the purpose of this paper is to analyse the spatial-temporal characteristics and impact mechanism of logistics carbon emissions in Heilongjiang Province in order to obtain the hotspots of logistics carbon emissions and identify the major driving and restraining factors of carbon emissions in the logistics industry. Then, appropriate carbon reduction countermeasures can be formulated, which will promote the sustainable development of the logistics industry in Heilongjiang Province.

Specifically, the objectives of this paper are to: (1) calculate the provincial-level and city-level logistics carbon emissions and carbon emission intensity from 2000 to 2020 using the energy-based emission factor method, (2) analyse the decoupling state of logistics industry carbon emissions in Heilongjiang Province based on the Tapio decoupling model, (3) study the spatial distribution patterns of logistics carbon emissions based on a hotspot analysis and spatial autocorrelation, and (4) decompose the influencing factors of logistics carbon emissions using the LMDI model and identify the driving factors and inhibitors of logistics carbon emissions.

The main framework of this study is as follows: the first section is the introduction, which consists of the background, the significance, and the scientific contribution of the study. The second section is the literature review, which aims to clarify the research and contribution of relevant scholars on carbon emission accounting, the spatial analysis of logistics carbon emissions, and the influencing factors, and provide the corresponding theoretical references for this study. The third section elaborates the relevant methods used in this study in detail. The fourth and fifth sections are the results and the discussion, respectively, which mainly present the research results and make an in-depth discussion from different perspectives. The last section is the conclusion of this study.

2. Literature Review

Currently, several studies have been carried out to account for the carbon emissions of the logistics industry at the national or provincial scale. The main analysis methods include an emission factor method and an energy coefficient method [11]. For example, Gao [12] applied the emission factor method to calculate the carbon emissions of the logistics industry in the Shanxi Province of China. She found that the carbon emissions of diesel and gasoline were far greater than those of other energy sources. Yuan [13] measured the carbon emissions of the logistics industry in China by using the energy coefficient method and found that the carbon emissions from oil and coal energy accounted for more than 95% of the total, while the carbon emissions of natural gas only accounted for less than 1%. Zhang [14] analysed the logistics industry carbon emissions in the Beijing–Tianjin–Hebei region of China and found that the difference between Beijing and Hebei was small, with its carbon emissions twice that of Tianjin. Qian [15] calculated the logistics industry carbon emissions in Northeast China from 2010 to 2017 by using the IPCC (Intergovernmental Panel on Climate Change) energy carbon emission coefficient, and found that Liaoning Province had the most carbon emissions, followed by Heilongjiang Province and Jilin Province.

The analyses on the spatial distribution characteristics of carbon emissions and their evolution over time are helpful to identify the hot and coldspots of carbon emissions and formulate targeted carbon reduction measures. Claeke and Qin [16] applied the coefficient of variation, the Thiel coefficient, and the Gini coefficient to study the spatial layout of carbon emissions in China from 1997 to 2007, and found that there were significant differences in carbon dioxide emissions among different regions. Hu [17] used a spatial autocorrelation analysis to analyse the spatial distribution pattern of carbon emission intensity of 30 provinces and cities in China, and found that the carbon emission intensity of the eastern coastal region with good economic development was the lowest, while that of the western region was the highest. Wang [18] analysed the spatial distribution pattern of carbon emissions at the provincial level in China from 2000 to 2015, and found that high-concentration areas were mainly concentrated in economically developed and resource-rich provinces such as Liaoning, Shandong, and Jiangsu Provinces. Hu [19] analysed China’s carbon emissions at the county level from 1997 to 2017 by using global and local Moran indices, and found that with the development of the economy, the number of high–low concentrations of carbon emissions at the county level increased significantly, while the number of low–low concentration areas decreased significantly. Xu [20] studied the provincial-level carbon emission efficiency of the logistics industry in China by constructing a super-PEBM (Pearson Epsilon-Based Measure) model, and found that the carbon emission efficiency of the logistics industry in Eastern China was the highest, while that in Western China was the lowest.

It is known that logistics carbon emissions are the result of various influencing factors such as carbon emission intensity, energy factors, logistics factors, economic development factors, population factors, etc., and the commonly used analysis methods of influencing factors include the STIRPAT (Stochastic Impacts by Regression on Population, Affluence, and Technology) model [21], the grey correlation analysis method [22,23], the LMDI (Logarithmic Mean Divisia Index) decomposition method [24], and so on. For example, Timilsina and Shrestha [25] decomposed the carbon emissions influencing factors of the transportation industry in 12 Asian countries using the LMDI method and found that energy intensity and economic factors were the main drivers of carbon emissions. Yao et al. [26] applied the STIRPAT model to study the influencing factors of logistics carbon emissions in the rural area of China and found that the regional economy had the greatest impact on rural logistics carbon emissions. Liu and Li [27] used the LMDI model to analyse the influencing factors of logistics carbon emissions in 30 provinces and cities in China. They found that both the energy intensity factor and the energy structure factor had suppressive effects on carbon emissions, and the output scale had an absolute driving effect on carbon emissions. Hu et al. [28] analysed the major driving factors of logistics carbon emissions in the Yangtze River economic belt of China using the grey correlation method and found that the carbon emission efficiency had the most impact on logistics carbon emissions, while the logistics scale had the least impact. In all, due to the diversity of influencing factors and the imbalance in the development of the logistics industry among different regions, the effect of the same factor on the carbon emissions of the logistics industry was different; therefore, it is necessary to study the influencing factors of carbon emissions in the regional logistics industry in a targeted manner.

In summary, the current methods for accounting logistics carbon emissions and influencing factors analysis are relatively mature, and a certain amount of related research results have been achieved. However, most studies focused on the analysis of national-scale or provincial-scale logistics carbon emissions, which rarely explored the spatial distribution features of logistics carbon emissions within the scope of a province due to the limitation of data availability. In this study, the city-level and county-level logistics carbon emissions were reallocated in consideration of the logistics output and population factors, and the spatio-temporal pattern evolution and impact mechanism of regional logistics carbon emissions were analysed by combining the hotspot analysis, Moran index, and LMDI methods, which is beneficial to formulate corresponding carbon reduction measures for a specific region and provide a theoretical reference for the sustainable development of the regional logistics industry.

3. Materials and Methods

3.1. Calculation of Carbon Emissions and Carbon Emission Intensity of Logistics Industry

Generally, the carbon emission accounting methods can be classified into three categories, including an emission factor method, an actual measurement method, and a mass balance method [29,30]. Emission factors can reflect the average GHG intensity per unit of activity data for a given source, which can be obtained from authoritative statistics or agencies (e.g., IPCC), therefore the energy-based emission factor method was used in this study to account for the carbon emissions in the logistics industry in Heilongjiang Province. Carbon emission intensity refers to the amount of carbon emissions for each unit of GDP (Gross Domestic Product) over a certain period [31]. This indicator is mainly used to reflect the relationship between the national economy and carbon emissions. A region is in a state of low-carbon development when the economic output increases and the carbon emissions per unit of GDP declines.

In this study, the carbon emissions of the logistics industry in Heilongjiang Province from 2000 to 2020 were calculated based on the energy consumption, standard coal coefficients of different fuels, and carbon emission coefficients. Due to the limitation of data availability, the carbon emissions were reallocated according to the proportion of the output value of the logistics industry in the 12 prefecture-level cities in Heilongjiang Province. Meanwhile, the county-level logistics carbon emissions were reallocated based on the proportions of the population in the 121 districts and counties in Heilongjiang Province. The calculation formulas for carbon emissions and carbon emission intensity are shown in Equations (1) and (2) [32], where $C^t$ is the quantity of logistics carbon emissions in year t; $E_i^t$ is the consumption of energy i in year t; ${\theta }_i$ is the standard coal coefficient of energy i; ${\delta }_i$ is the carbon emission coefficient of energy i; $Y^t$ is the carbon emission intensity in year t; and $G^t$ is the GDP in year t.

(1)$C^t=E_i^t\times {\theta }_i\times {\delta }_i$

(2)$Y^t=C^t/G^t$

The energy consumption data of the province during the period of 2000 and 2020 was sourced from the China Energy Statistics Yearbook (2001–2021). Since the current statistical yearbook in China does not involve the logistics industry, the data of the transportation, warehousing, and postal industries was combined to represent the energy consumption data of the logistics industry. Various conversion coefficients of standard coal for different types of energy and carbon emission coefficients are shown in Table 1.

3.2. Decoupling Effect of Regional Logistics Carbon Emissions

Decoupling theory is the basic theory of the connection between economic growth and resource consumption or environmental pollution proposed by the Organization for Economic Co-operation and Development (OECD) [35]. It was initially applied by the OECD in agricultural policy research, and later used in the field of environment to study the relationship between environmental quality and economic development [36]. The logistics industry has an indispensable role in the country and regional economic development, however, the carbon emissions of the logistics industry due to energy consumption have significant impacts on the environment, which is one of the global warming driving factors, therefore, it is necessary to study the decoupling effect of the logistics industry carbon emissions and analyse the stage of the logistics industry carbon emissions and economy.

There are two main methods to measure the decoupling state, namely the OECD decoupling model and the Tapio decoupling model. The former mainly relies on the ratio of carbon emissions to economic development at the beginning and at the end of the period to construct the decoupling index and decoupling factor, and then judge the decoupling state between economic development and carbon emissions. The latter was developed by Tapio on the basis of the OECD decoupling model [37]. Compared with the OECD model, the Tapio model takes the total changes and relative changes of carbon emissions and economic output into consideration, and eliminates the influence of the data dimension by introducing a decoupling elastic index, which can divide the decoupling state more accurately, and thus be applied to a more flexible study of the relationship between environment and economy [38].

In this study, the Tapio decoupling model [37] was used to study the decoupling effect of logistics carbon emissions in Heilongjiang Province. The calculation formula of the decoupling elasticity index is expressed by Equation (3), where %ΔC is the percentage of change in carbon emissions, %ΔV is the percentage of change in the output value of the logistics industry.

(3)$DE=\frac{\%\Delta C}{\%\Delta V}$

Based on the results of Equation (3), the decoupling states under Tapio can be divided into three categories: connection, decoupling, and negative decoupling, as shown in Table 2. Under the state of connection, growth connection means that both carbon emissions and logistics output are growing at a close rate; the recession connection means that both carbon emissions and logistics output are falling with similar declining rates. Under the state of decoupling, strong decoupling refers to the state of carbon emissions decreasing and logistics output growing, which is an ideal state; weak decoupling means that both carbon emissions and GDP increase, but carbon emissions grow at a slower rate than that of logistics output; recession decoupling means that both carbon emissions and logistics output decline, but the falling rate of logistics output is less than that of carbon emissions. Under the state of negative decoupling, strong negative decoupling means that carbon emissions grow and logistics output falls, which is the worst case; weak negative decoupling means that both carbon emissions and logistics output decline, but the falling rate of carbon emissions is less than that of logistics output; expansionary negative decoupling means that both carbon emissions and logistics output increase, but the increment of carbon emissions is greater than that of logistics output.

3.3. Spatial Analysis of Carbon Emissions of Regional Logistics

3.3.1. Hotspot Analysis

A hotspot analysis is a spatial analysis and mapping technique, which was proposed by Getis and Ord [39]. It can be used to identify statistically significant hotspots and coldspots by aggregating points of occurrence into polygons or converging points that are in proximity to one another based on a calculated distance [40]. This method is to calculate the $G_i^{*}$ value of each dataset element in the data, and define the high and low clusters with the help of the standard normal distribution. The $G_i^{*}$ value returned by each element is the Z score, and based on the Z score and P value, it is possible to know where the high and low values are spatially clustered. A higher Z score indicates the presence of hotspot clusters in this area; on the contrary, a smaller Z score indicates the presence of cold spot aggregation in this area [41]. $G_i^{*}$ can be calculated as Equation (4) [39], where $G_i^{*}$ is the Z score, $B_j$ is the value of the element j, $W_{ij}$ is the weight matrix, $\overline{B}$ is the mean of the elements attribute value, S is the standard deviation, n is the total number of elements.

(4)$G_i^{*}=\left({\displaystyle \sum _{j=1}^n{W}_{ij}{B}_j-\overline{B}{\displaystyle \sum _{j=1}^n{W}_{ij}}}\right)/\left(S\sqrt{\frac{n}{n-1}{\displaystyle \sum _{j=1}^n{W}_{ij}^2-\frac{1}{n-1}{\left({\displaystyle \sum _{j=1}^n{W}_{ij}}\right)}^2}}\right)$

In this study, the city-level and county-level logistics carbon emissions were used as input elements, respectively, and the Jenks Natural breaks classification method in the hotspot analysis was used to divide logistics carbon emissions in Heilongjiang Province into hotspots, sub-hotspots, sub-coldspots, and coldspots on using ArcGIS (hotspot analysis (Getis-Ord $G_i^{*}$), Environmental Systems Research Institute, Inc., Redlands, CA, USA). The annual data of 2000, 2010 and 2020 was used for the hotspot analysis on logistics carbon emissions. In order to discuss the evolution trend of logistics carbon emissions, the longitudinal timeline data in the three years were ordered from small to large and divided into four categories, including coldspots, sub-coldspots, sub-hotspots, and hotspots. The hotspot analysis was also conducted at the same time node (2000, 2010, 2020) in order to obtain the transverse spatial pattern of logistics carbon emissions. It is noted that the Songling District and Jiagedaqi District of the Greater Khingan Mountains belong to Heilongjiang Province but are located in Inner Mongolia, so these regions were not included in the map drawing of Heilongjiang Province.

3.3.2. Spatial Autocorrelation Analysis

A spatial autocorrelation analysis is mainly used to study the correlation degree of an attribute value in some regions connected by geographical position, which is divided into a global spatial autocorrelation analysis and a local spatial autocorrelation analysis. Global spatial autocorrelation is to analyse the correlation of the whole research area, which can answer the question of whether there is correlation in the research area [42], but does not clearly indicate the agglomeration characteristics between research units, while a local spatial autocorrelation can solve this problem well. Local spatial autocorrelation is used to reveal the correlation between a research unit and its adjacent units for a specific attribute value, and to indicate the specific spatial location of aggregation.

In this study, the global Moran’s index was used to study whether there is spatial autocorrelation of logistics carbon emissions in each district and county of Heilongjiang Province, and the local Moran’s index was used to study the specific location where the agglomeration of logistics carbon emissions occurs. The global Moran’s index and local Moran’s index can be calculated by Equations (5) and (6) [42], where $x_i$ and $x_j$ are the observed carbon emission values of unit I and J in the study area, $\overline{x}$ is the average of the observed carbon emissions of all research units, $M_{ij}$ is the weight matrix, m is the total number of elements. The value range of the global Moran’s index I is between [−1, 1]. When I is greater than 0, it indicates that the spatial autocorrelation of carbon emissions from the logistics industry in all districts and counties of Heilongjiang Province is positive; when I is less than 0, it indicates that the spatial autocorrelation is negative. The greater the absolute value of the Moran’s index, the stronger the autocorrelation is

(5)$I=\frac{m{\displaystyle \sum _{i=1}^m{\displaystyle \sum _{j=1}^m{M}_{ij}(x_i-\overline{x})(x_j-\overline{x})}}}{{\displaystyle \sum _{i=1}^m{\displaystyle \sum _{j=1}^m{M}_{ij}{\displaystyle \sum _{i=1}^m{(x_i-\overline{x})}^2}}}}$

(6)$I_i=\frac{m(x_i-\overline{x}){\displaystyle \sum _j{M}_{ij}(x_j-\overline{x})}}{{\displaystyle \sum _i{(x_i-\overline{x})}^2}}$

3.4. Analysis of Influencing Factors of Logistics Carbon Emissions

In this study, the LMDI model was used to decompose the influencing factors of carbon emissions growth in the logistics industry in Heilongjiang Province. The logarithmic mean divisia index was proposed and used by Ang to discover the effect of various driving factors in relation to the total effect [43]. It can provide a residual-free decomposition and accommodate the presence of zero values in the data [44]. By constructing an extended Kaya identity and deforming accordingly, the contribution of each influencing factor with a complex relationship structure can be separated by the LMDI decomposition model, which can provide a basis for further analysis and related policy decision making [45]. According to the proportion of the contribution of each influencing factor, the impact of each influencing factor can be determined. LMDI has been widely used in assessing the effects of socio-economic drivers and characteristics of various industry sections (e.g., food industry, iron and steel industry, construction industry, and electricity) on CO₂ emissions [46,47,48].

In this study, the energy structure, energy efficiency, cargo turnover required per unit of GDP, per capita GDP, and population size were selected as the decomposition indicators in LMDI model, in which energy structure and energy efficiency represented energy factor, cargo turnover required per unit of GDP represented the development factor of the logistics industry, the per capita GDP represented the economic development factor, and the population size represented the population factor. The LMDI decomposition model [43] for the analysis of the influencing factors of carbon emissions in the logistics industry is expressed as Equation (7), where C is the logistics carbon emissions in the current year, $C_i$ is the carbon emissions of energy I in the year, E is the annual energy consumption, $E_i$ is the consumption of energy I, T is the annual regional cargo turnover, G is the regional GDP in the current year, P is the total population at the end of the year in the region; $C_i/E_i$ is the carbon emissions generated per unit of energy i consumed, $E_i/E$ is the proportion of energy i consumed in total energy, denoted as S, which represents the energy structure factor, $E/T$ is the energy consumption per unit of cargo turnover, denoted as defined as I, which represents energy efficiency factor, $T/G$ is the amount of cargo turnover required per unit of GDP, denoted as L, which represents the logistics industry development factor, and $G/P$ is GDP per capita, denoted as R, which represents the economic development factor.

(7)$C={\displaystyle \sum _i\frac{C_i}{E_i}}\times \frac{E_i}{E}\times \frac{E}{T}\times \frac{T}{G}\times \frac{G}{P}\times P$

The changes of logistics carbon emissions can be obtained by subtracting the carbon emissions at the beginning of the period from that at the end of the period. Using the LMDI model, the variation of carbon emissions between the current year and the base year 2000 is decomposed into five parts (Equation (8)), namely the contribution of energy structure ($\Delta C_S$), energy efficiency ($\Delta C_I$), logistics industry development factor ($\Delta C_L$), economic development factor ($\Delta C_R$), and population factor ($\Delta C_P$). The calculation formulas for the contribution value of energy structure, energy efficiency, logistics industry development factor, economic development factor, and demographic factor are expressed as Equations (9)–(13), where $W_i=\left(C_i^t-C_i^0\right)/\left(\ln C_i^t-\ln C_i^0\right)$ [43].

(8)$\Delta C_{tot}=C_t-C_0={\displaystyle \sum _i(\Delta C_S}+\Delta C_I+\Delta C_L+\Delta C_R+\Delta C_P)$

(9)$\Delta C_S={\displaystyle \sum _i{W}_i\ln \frac{{\left(E_i/E\right)}^t}{{\left(E_i/E\right)}^0}}$

(10)$\Delta C_I={\displaystyle \sum _i{W}_i\ln \frac{{\left(E/T\right)}^t}{{\left(E/T\right)}^0}}$

(11)$\Delta C_L={\displaystyle \sum _i{W}_i}\ln \frac{{\left(T/G\right)}^t}{{\left(T/G\right)}^0}$

(12)$\Delta C_R={\displaystyle \sum _i{W}_i}\ln \frac{{\left(G/P\right)}^t}{{\left(G/P\right)}^0}$

(13)$\Delta C_P={\displaystyle \sum _i{W}_i}\ln \frac{{\left(P\right)}^t}{{\left(P\right)}^0}$

The contribution rate of influencing factors of carbon emissions can be expressed as the product of contribution rate of all influencing factors, as shown in Equation (14) [43], where, $D_{\mathrm{t}}$ represents the total effect of all influencing factors, and $D_S$, $D_I$, $D_L$, $D_R$ and $D_P$ represent the contribution rate of energy structure, energy efficiency, logistics development factor, economic development factor, and population size to the carbon emission of logistics industry in Heilongjiang Province, respectively. When 0 < D < 1, it indicates that this factor has an inhibitory effect on carbon emissions; when D > 1, it indicates that this factor has a pulling effect.

(14)${{\displaystyle D}}_{\mathrm{t}}=\frac{{{\displaystyle C}}_t}{{{\displaystyle C}}_0}={{\displaystyle D}}_S\times {{\displaystyle D}}_I\times {{\displaystyle D}}_L\times {{\displaystyle D}}_R\times {{\displaystyle D}}_P$

The calculation formulas for the contribution rate of energy structure, energy efficiency, logistics industry development factor, economic development factor, and demographic factor are expressed as Equations (15)–(19) [43].

(15)${{\displaystyle D}}_S=\exp \left[{\displaystyle \sum _i\frac{{{\displaystyle W}}_i}{\left({{\displaystyle C}}^{\mathrm{t}}-{{\displaystyle C}}^0\right)/\left(\ln {{\displaystyle C}}^{\mathrm{t}}-\ln {{\displaystyle C}}^0\right)}\ln \frac{{{\displaystyle \left(E_i/E\right)}}^{\mathrm{t}}}{{{\displaystyle \left(E_i/E\right)}}^0}}\right]$

(16)${{\displaystyle D}}_I=\exp \left[{\displaystyle \sum _i\frac{{{\displaystyle W}}_i}{\left({{\displaystyle C}}^{\mathrm{t}}-{{\displaystyle C}}^0\right)/\left(\ln {{\displaystyle C}}^{\mathrm{t}}-\ln {{\displaystyle C}}^0\right)}\ln \frac{{{\displaystyle \left(E/T\right)}}^{\mathrm{t}}}{{{\displaystyle \left(E/T\right)}}^0}}\right]$

(17)${{\displaystyle D}}_L=\exp \left[{\displaystyle \sum _i\frac{{{\displaystyle W}}_i}{\left({{\displaystyle C}}^{\mathrm{t}}-{{\displaystyle C}}^0\right)/\left(\ln {{\displaystyle C}}^{\mathrm{t}}-\ln {{\displaystyle C}}^0\right)}\ln \frac{{{\displaystyle \left(T/G\right)}}^{\mathrm{t}}}{{{\displaystyle \left(T/G\right)}}^0}}\right]$

(18)${{\displaystyle D}}_R=\exp \left[{\displaystyle \sum _i\frac{{{\displaystyle W}}_i}{\left({{\displaystyle C}}^{\mathrm{t}}-{{\displaystyle C}}^0\right)/\left(\ln {{\displaystyle C}}^{\mathrm{t}}-\ln {{\displaystyle C}}^0\right)}\ln \frac{{{\displaystyle \left(G/P\right)}}^{\mathrm{t}}}{{{\displaystyle \left(G/P\right)}}^0}}\right]$

(19)${{\displaystyle D}}_P=\exp \left[{\displaystyle \sum _i\frac{{{\displaystyle W}}_i}{\left({{\displaystyle C}}^{\mathrm{t}}-{{\displaystyle C}}^0\right)/\left(\ln {{\displaystyle C}}^{\mathrm{t}}-\ln {{\displaystyle C}}^0\right)}\ln \frac{{{\displaystyle \left(P\right)}}^{\mathrm{t}}}{{{\displaystyle \left(P\right)}}^0}}\right]$

4. Results

4.1. Temporal Evolution Characteristics of Logistics Carbon Emissions in HeilongjiangProvince

4.1.1. Total Carbon Emissions of Logistics Industry

The logistics carbon emissions of Heilongjiang Province and its prefecture level cities during the period of 2000 to 2020 are shown in Figure 1. The overall carbon emissions of the logistics industry in Heilongjiang Province generally showed an upward trend, increasing from 1.66 million tons in 2000 to 6.21 million tons in 2020, an increase of 274%. From 2000 to 2010, the overall carbon emissions increased steadily. There was a great leap in logistics industry carbon emissions from 2010 to 2011, an increase of nearly 80%, which was due to the substantial increase of raw coal energy consumption. From 2011 to 2015, the carbon emissions of the logistics industry continued to increase, with an average annual increase of 0.3690 million tons. After 2015, the carbon emissions showed a decreasing trend, with an average annual decrease of 0.1927 million tons. The main reason was that the consumption of clean energy such as natural gas had been counted and recorded in the statistical yearbook of Heilongjiang Province since 2016, and the consumption of raw coal energy was reduced. It is also noted that the cities of Harbin and Daqing contributed the most to the carbon emissions of the logistics industry in Heilongjiang Province, followed by Qiqihar, Suihua, and Mudanjiang. Corresponding to the degree of carbon emissions contribution, these cities have relatively good economic development owing to their geographical locations and featured industries.

4.1.2. Carbon Emissions of the Logistics Industry by Energy Sources

The distribution of carbon emissions of different energy in the logistics industry of Heilongjiang Province are shown in Figure 2. It can be seen that from 2000 to 2010, the average proportions of carbon emissions from raw coal, gasoline, kerosene, and diesel were 14%, 35.5%, 2.4%, and 37.9%, respectively, accounting for 89.8% of the total carbon emissions. From 2011 to 2015, the average proportions of carbon emissions from raw coal, gasoline, kerosene, and diesel were 31.1%, 26.3%, 7.8%, and 25.7%, respectively, accounting for 90.9% of the total. The proportions of the major energy sources were very close during the two periods, but the proportion of coal energy increased by nearly 17%, and the proportion of oil energy decreased by nearly 16%. During the period of 2016 to 2020, the proportions of carbon emissions from the four energy sources dropped to 80.5% of the total carbon emissions, a 5.3% decrease in coal energy, and a 5.1% decrease in oil energy, compared with the previous period. The consumption of natural gas gradually increased during 2016 to 2020. Heilongjiang Province is located in the northeast of China. As the traditional industrial base of China, the fossil energy accounts for a higher proportion of the total energy consumption all the time. With the development trend of green logistics and low-carbon logistics, Heilongjiang Province has made some adjustments in its energy structure by decreasing the proportion of fossil energy and increasing the use of clean energy gradually, but there is still a large space for optimization.

4.2. Carbon Emission Intensity of Logistics Industry and Decoupling Effect Analysis

4.2.1. Carbon Emission Intensity of the Logistics Industry

The carbon emission intensity of the logistics industry in Heilongjiang Province from 2000 to 2020 shown in Figure 3. During the period of 2000 to 2010, the carbon emission intensity was in a state of fluctuation, with an average of 0.122 kg/CNY (Chinese Yuan). There was a marked increase after 2010, an increase of 65%. The carbon emission intensity has started to decline since 2016 with slight fluctuation, and maintained at a relatively lower level. The average logistics carbon intensity of Heilongjiang Province during the period of 2011–2020 was 0.208 kg/CNY, which was still higher than that of the logistics industry in the western provinces of China [49]. However, compared with the carbon emission intensity (0.2807 kg/CNY) of the logistics industry in the Yangtze River Delta region of China (including Shanghai, Jiangsu, Zhejiang, Anhui) [50], the logistics carbon emission intensity of Heilongjiang Province was relatively low. Since the carbon emission intensity of the logistics industry can be determined by both carbon emissions and logistics output, it is necessary to conduct a decoupling effect analysis on the logistics industry carbon emissions in Heilongjiang Province to propose suitable countermeasures for carbon intensity reduction.

4.2.2. Decoupling Effect of Logistics Carbon Emissions and Economics

The average growth rates of carbon emissions and output value of the logistics industry, decoupling elasticity index, and decoupling state of logistics industry in Heilongjiang Province and 12 prefecture-level cities from 2000 to 2020 are presented in Table 3. The logistics output of Heilongjiang Province over the years was reduced by constant price in 2000. During the period of 2000 to 2020, six of the 12 prefecture-level cities were in the state of expansion negative decoupling, which means that they will face greater pressure to reduce carbon emissions. Jixi city, Qiqihar City and Jiamusi city were in the state of growth connection, indicating that the growth rate of logistics output was close to that of carbon emissions. Shuangyashan city, Mudanjiang city, Heihe city and Suihua city were in the weak decoupling state, indicating that the carbon emissions and logistics output in these cities were in a growing state, but the growth rate of carbon emissions was less than that of logistics output.

The overall logistics industry carbon emissions and logistics output in Heilongjiang Province showed a state of expansion negative decoupling from 2000 to 2020, indicating that both the logistics output and the carbon emissions increased in the long term, and the growth rate of carbon emissions was higher than that of logistics output. However, the carbon emissions of logistics industry and the logistics output in Heilongjiang Province began to show a strong decoupling state since 2016, indicating that carbon emissions were decreasing and the output value of logistics was increasing in the short run, which was a relatively ideal state. The reasons for this phenomenon were that Heilongjiang Province started to replace some fossil fuel energy with cleaner energy such as natural gas since 2016 and the local government effectively implemented cost reduction and efficiency increase in logistics industry and enlarged the construction of logistics infrastructure facilities, including perfecting the logistics infrastructure network, logistics enterprise equipment innovation, and improvement of logistics operation modes.

4.3. Spatial Distribution Characteristics of Logistics Carbon Emissions in Heilongjiang Province

4.3.1. Hotspot Analysis on Logistics Carbon Emissions

The hotspot analysis on carbon emissions from the logistics industry in Heilongjiang Province is shown in Figure 4. It is noted that the number of coldspots and sub-coldspots gradually decreased with time, while the number of hotspots and sub-hotspots increased. There were only coldspots and sub-coldspots in 2000. Twenty-two coldspots turned into sub-coldspots and 10 sub-coldspots turned into sub-hotspots from 2000 to 2010; eight sub-hotspots turned into hotspots and 12 sub-coldspots and coldspots turned into sub-hotspots from 2010 to 2020. The sub-hotspots were mainly concentrated in Harbin in 2010. In 2020, the hotspots were mainly distributed in Harbin, and the sub-hotspots were mainly distributed in Harbin, Daqing, and Qiqihar.

The hotspot analysis on logistics carbon emissions in Heilongjiang Province based on the transverse timeline data is shown in Figure 5. The proportions of hotspots accounted for 9.9%, 8.3%, and 2.5% of the total in 2000, 2010, and 2020. The proportions of sub-hotspots were 17.4%, 24.8%, and 14.9%, respectively. The coldspot regions accounted for nearly one third of the total; 33.1% in 2000, 33.9% in 2010, and 33.9% in 2020. The proportions of sub-coldspots were 31.4%, 33.1%, and 48.8%, respectively. It is noted that from 2000 to 2020, the hotspots of carbon emissions were mainly concentrated in Harbin and Suihua, while the coldspots were mainly concentrated in the Greater Khingan Mountains region, Heihe and Yichun. The sub-hotspots and sub-coldspots were widely distributed and accounted for a large proportion. The number of hotspots decreased significantly in 2020, dropping from 12 to only 3. All the three hotspots were located in Harbin. The reason for this phenomenon was that the economic output of Harbin was far greater than other regions and the development of logistics industry was more prosperous, resulting in more carbon emissions.

4.3.2. Spatial Autocorrelation Analysis of Carbon Emissions in Logistics Industry

In this study, the spatial autocorrelation analysis on the district- and county-level logistics carbon emissions in Heilongjiang Province in 2000, 2010, and 2020 was conducted by using a global and local Moran’s index (Equations (5) and (6)). Results showed that the global Moran’s index in 2000, 2010, and 2020 were 0.458, 0.518, and 0.532, respectively, and the Z values were 8.046, 9.576, and 9.723, correspondingly, all passing the 99% significance test (Figure 6), which indicated that there was a significant positive spatial autocorrelation between the district- and county-level logistics carbon emissions, and the districts and countries with similar levels of logistics carbon emissions in Heilongjiang Province showed aggregate distribution, and the correction gradually increased as time. Therefore, it is necessary to conduct a local Moran’s index analysis in order to obtain whether the logistics carbon emissions of each research unit have spatial agglomeration with its neighboring districts and counties.

The results of the local spatial autocorrelation analysis are shown in Figure 7. It is noted that the spatial aggregation trend of logistics carbon emissions in Heilongjiang Province was relatively stable from 2000 to 2020, mainly showing high–high aggregation and low–low aggregation. The occurrence of high–high concentration was mainly concentrated in the economically developed cites of Harbin, Daqing, Qiqihar, and Suihua. The low–low concentrations were mainly concentrated in the underdeveloped regions such as the Greater Khingan Mountains region, and the cites of Hegang, Yichun, and Heihe. Zeng [51] also reported a similar result that the high–high aggregation areas of carbon emissions from China’s logistics industry were mainly concentrated in the economically developed eastern coastal cities, while the low–low aggregations are mainly in the underdeveloped western regions.

In 2020, the number of high–high concentration areas in Heilongjiang Province was 12, of which 10 districts/counties were located in Harbin city, and 2 were located in Suihua City. The high–high concentrated area was mainly concentrated in Harbin, the capital city of Heilongjiang Province, which also conformed to the polarization–diffusion effect in economics [52]. As a major contributor to the carbon emissions of the logistics industry in Heilongjiang Province, Harbin had a relatively developed economy, which consumed more resources and generated a large amount of carbon emissions, thus driving the carbon emissions of the surrounding countries. There was little change in the low–low concentration regions from 2000 to 2020.

4.4. Impacts of Different Influencing Factors on the Growth of Logistics Carbon Emissions

Based on Equation (7), the carbon emissions in the logistics industry of Heilongjiang Province were decomposed into five parts and the corresponding contribution values were calculated based on Equations (9)–(13). The results were successfully verified by using Equation (8). The contribution rates of different influencing factors on the growth of logistics carbon emissions were figured out by using Equations (15)–(19), as shown in Figure 8. It is noted that the growth of logistics carbon emissions in Heilongjiang Province was mainly driven by the economic development. In the early stage, the rapid economic growth was driven by the measures with high input and high consumption, while the environmental cost was also high due to low energy efficiency. After 2015, due to the improvement of energy efficiency and the inhibitory effect of population size, the carbon emissions of the logistics industry decreased year by year. According to the contribution rates to the carbon emissions of the logistics industry in Heilongjiang Province, the major influencing factors that had pulling effects were economic development, energy structure, and energy efficiency, and the order was as follows: economic development > energy efficiency > energy structure. Logistics industry development factor mainly played an inhibiting effect on the carbon emissions of the logistics industry. The factor of population size showed a pulling effect on the logistics industry in the early stage, and then gradually turned into an inhibiting effect.

4.4.1. Impacts of Energy Structure on Logistics Carbon Emissions

During the study period, the contribution rates of the energy structure effect were all greater than 1.0, ranging between 1.005 and 1.418 with an average of 1.144, which showed a pulling effect on logistics carbon emissions. From 2000 to 2010, gasoline and diesel accounted for about 84% of the total energy consumed in the logistics industry, followed by raw coal (10%) and kerosene (3%). The energy structure of the logistics industry in Heilongjiang Province was relatively stable at that time, and there was no significant difference in logistics carbon emissions among years. During the period of 2011 to 2016, the average consumption of raw coal climbed to 3.9391 million tons, an increase of 237.43% compared with the average consumption from 2005 to 2010. The energy consumption of gasoline and kerosene increased slightly. In addition, the average annual consumption of heat energy reached 7.7596 million gigajoules. As a result, the logistics carbon emissions increased significantly due to the large increase of raw coal and heat energy consumption. After 2016, the decline of raw coal consumption year by year and the use of natural gas clean energy had improved the carbon emissions of the logistics industry. In all, energy structure factor presented a pulling effect on the logistics carbon emissions in Heilongjiang Province. This is mostly due to the fact that raw coal, gasoline, kerosene, diesel and other high-carbon petrochemical energy accounted for 90% of the energy consumption in the logistics industry in Heilongjiang Province, which belonged to the energy structure of high energy consumption. The proportion of clean energy was relatively small. Therefore, there is still considerable room for improvement in adjusting the energy structure.

4.4.2. Impacts of Energy Efficiency on Logistics Carbon Emissions

The factor of energy efficiency mainly had a pulling effect on the growth of carbon emissions of logistics industry in Heilongjiang Province from 2000 to 2020, which can be divided into two stages. In the first stage of 2000–2015, the contribution rate of energy efficiency to logistics carbon emissions increased from 1.004 to 2.019, while in the second stage of 2015–2020, the contribution rate showed a downward trend, dropping from 2.019 to 1.521. Therefore, the energy efficiency factor played a pulling role in the growth of logistics carbon emissions in the long run, but the pulling effect gradually reduced after 2015, owing to the maturity of logistics industry, the upgrading of logistics facilities and the equipment and optimization of industry management.

4.4.3. Impacts of Logistics Development on Logistics Carbon Emissions

The factor of logistics development had an inhibitory effect on the growth of carbon emissions of the logistics industry. During the period of 2000 to 2020, the maximum contribution rate of logistics development factor to carbon emissions was 0.966, the minimum value was 0.664, and the average value was 0.780. All the values were less than 1.0, which indicated that logistics development factor restrained the growth of logistics carbon emissions. This can be explained by the fact that the increase of freight turnover in Heilongjiang Province was far less than that of logistics output, which indicated that the turnover of goods required per unit of logistics output dropped with time. With the progress of science and technology, the automation and modernization level of the logistics industry was continuously improved, the logistics facilities and equipment were constantly updated, the energy loss in the process of goods turnover decreased, and the energy efficiency effectively improved. As a result, the development factor of the logistics industry had a restraining effect on carbon emissions of the logistics industry.

4.4.4. Impacts of Economic Development on Logistics Carbon Emissions

The factor of economic development was the main driver of the growth of carbon emissions in the logistics industry. During the period of 2000 to 2020, the contribution rate showed an increasing trend year by year, ranging between 1.056 and 3.102, with an average of 2.200. Therefore, the factor of economic development pulled the carbon emissions of the logistics industry and the degree of effect increased with time. This conclusion is consistent with the research conclusion of most scholars on the study of influence factors of the regional logistics industry in China. With the rapid development of science and technology, the continuous progress of society, people begin to pursue a high-quality life after their basic living needs are met, which has significantly increased the turnover of logistics goods, thus increasing the carbon emissions of the logistics industry.

4.4.5. Impacts of Demographic Factor on Logistics Carbon Emissions

The impact of the demographic factor on the growth of carbon emissions in the logistics industry can be divided into two stages. In the first stage of 2000–2012, demographic factors played a driving role in the growth of carbon emissions of the logistics industry, but the driving degree was not significant, and the average contribution rate was only 1.001. The population of Heilongjiang Province increased slowly year by year during this period, resulting in the slow increase of logistics demand and logistics carbon emissions. After 2012, the population of Heilongjiang Province started to decline and approximately one seventh of the population moved to the other more developed provinces due to the geographical location and development level of this province, which made the demographic factor become the inhibiting factor of logistics carbon emissions, but the contribution rate was only 0.913, so the inhibiting effect was weak.

5. Discussions

The logistics industry is a basic, strategic, and pioneering industry supporting the development of the national economy [49]. The analyses on the spatio-temporal distribution features and driving factors of regional logistics carbon emissions are of great significance for the formulation of specific logistics policies and carbon reduction measures. In this study, different levels of logistics carbon emissions in Heilongjiang, China were calculated. The total logistics carbon emissions of this province increased from 1.66 million tons in 2000 to 6.21 million tons in 2020, and the period of 2010–2015 experienced a significant growth. Compared to the logistics carbon emission in the provinces of East China during the period of 2010–2019 [53], the logistics carbon emissions of Heilongjiang Province was less than those of the developed regions such as Shanghai, Jiangsu, and Shandong, but the growth rate was far greater than all the east provinces except Jiangxi. Similar to some other Chinese provinces, the logistics industry carbon emissions and logistics output in Heilongjiang Province generally showed a state of expansion negative decoupling. Mu et al. [49] analysed the carbon emission decoupling of the logistics industry in the western provinces of China and found that the expansion negative decoupling state accounted for a certain proportion (9.09–36.36%) among the 11 western provinces during 2012–2017. Therefore, Heilongjiang Province as a border province should adopt more effective energy saving and emission reduction measures in the development of the logistics industry.

It is very important to identify the hotpots of logistics carbon emissions and understand the spatial autocorrelation relationships among different hotspot classification groups. Most previous studies focused on the large-scale spatial analysis of logistics carbon emissions [23,24,25,44,53], however few studies were conducted on the spatial analysis of district or county-level logistics carbon emissions due to the data constraints. In this study, the city-level and county-level logistics carbon emissions were reallocated based on the provincial-level logistics carbon emissions, logistics output and population. Similarly, DE Azevedo et al. [54] used population density as the weight factor to reallocate the national-scale carbon emissions in Brazil and Liu et al. [55] used the energy consumption method in combination of allocation by population density to calculate the county-level carbon emissions in China. According to the longitudinal timeline data of logistics carbon emissions, the number of coldspots and sub-coldspots gradually decreased with time, while the number of hotspots and sub-hotspots increased (Figure 4). According to the transverse timeline data, the proportions of hotspots decreased from 9.9% in 2000 to 2.5% in 2020 (Figure 5), which indicated that the gaps of logistics carbon emissions between the developed districts were narrowing. The capital city of Harbin was always the hotspot. This phenomenon was similar to the findings based on the provincial-level logistics carbon emissions, in which the hotspots were mainly concentrated in the developed regions [18].

In addition, the study on the influencing factors of logistics carbon emissions can facilitate the formulation of corresponding measures against the driving factors and help reduce the contribution of carbon emissions from these factors, which is beneficial to the low carbon development of the logistics industry. Our study results showed that the major pulling factors of logistics carbon emissions in Heilongjiang Province were economic development, energy structure, and energy efficiency, the logistics development was the main factor inhibiting the growth of logistics carbon emissions, and the impact of demographic factor was converted from pulling effect to inhibitory effect. These conclusions are similar to those of the other scholars. For example, Xu et al. [56] found that the economic development had the greatest pulling effect on China’s regional logistics carbon emissions and logistics development effect had the characteristics of first driving and then restraining in the three major regions. Ma and Wang [57] used the LMDI decomposition model and the data of 1991–2010 to analyse the influencing factor of carbon emissions per capita in China’s logistics industry. They found that economic growth was the most important pulling factor compared to energy structure and energy efficiency, and logistics development exhibited a significant inhibitory effect on per capita carbon emissions. Wang et al. [58] analysed the influencing factors of CO₂ emissions in the logistics industry in the northeastern provinces of China and found that with the outflow of population, the demographic factor was gradually changed from pulling factor to inhibitory factor.

It should be noted that there are also some limitations in this study. Due to the constraints of statistical data availability, the city-level and county-level logistics carbon emissions had to be estimated based on logistics output and population factors, which may affect the accuracy of carbon emission accounting. In addition, only five influencing factors were considered in the LMDI decomposition model and the impacts of other factors were not verified. Future research can be conducted by adopting different allocation methods in accounting the city-level and county-level logistics carbon emissions, comparing different methods for analysing the influencing factors, and adding more influencing factors in the models in order to increase the robustness of the study results.

6. Conclusions

In this study, the logistics carbon emissions in Heilongjiang Province, its 12 prefecture-level cities, and 121 districts and counties were calculated, the decoupling effect between logistics carbon emissions and logistics output was analysed, the spatial distribution features along the longitudinal and transverse timelines were identified, and the decomposition analysis of the influencing factors was conducted. The following conclusions are drawn from the study:

• (1). During the period of 2000 to 2020, the carbon emissions of the logistics industry in Heilongjiang Province of China generally showed an upward trend, which increased slowly in the first 10 years, then increased sharply due to the massive use of raw coal energy in the next five years, and finally showed a slight downward trend due to the use of clean energy natural gas. The logistics carbon emissions from the consumption of traditional fossil fuel energy accounted for a substantial proportion. With the use of clean energy, the logistics carbon emissions started to decline, which indicated that the implementation of energy structure optimization had made some achievements, but there was still some room for improvement in adjusting the energy structure.

• (2). The overall decoupling state of the logistics industry carbon emissions and logistics output in Heilongjiang Province was expansion negative decoupling from 2000 to 2020, and a strong decoupling state in the last five years, which was an ideal state. Among the 12 prefecture-level cities, 5 cities showed expansion negative decoupling, indicating that these cities will face great pressure of emission reduction in the future. Three cities showed growth connection, indicating that the carbon emissions of logistics industry and output value of logistics industry had similar growth rate. Four cities showed a weak decoupling state, indicating that the growth rate of carbon emissions was less than that of the output value of the logistics industry.

• (3). The hotspots and coldspots of logistics carbon emissions changed over time. The hotspots were mainly concentrated in developed regions, while the coldspots were mainly in undeveloped regions. There was a significant positive spatial autocorrelation between the district- and county-level logistics carbon emissions. The high–high agglomeration areas were concentrated in the southern region, and low–low agglomeration areas were concentrated in the north

• (4). Energy structure, energy efficiency, and economic development factors had pulling effects on the growth of carbon emissions of the logistics industry in Heilongjiang Province. The economic development factor had the most impact on carbon emissions, followed by energy efficiency and energy structure. Logistics development factor had an inhibitory effect on the carbon emissions of the logistics industry. The demographic factor showed a pulling effect on carbon emissions of the logistics industry in the early stage, while with the loss of population, the demographic factor turned into a restraining factor.

Author Contributions

Conceptualization, Z.C. and J.W.; methodology, Z.C.; software, Z.C.; validation, Z.C. and J.W.; formal analysis, Z.C.; investigation, Z.C.; writing—original draft preparation, Z.C.; writing—review and editing, J.W.; supervision, J.W.; project administration, J.W.; funding acquisition, J.W. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data used to support the findings of this study are available from the corresponding author upon request.

Acknowledgments

The authors would like to acknowledge the Natural Science Foundation of Heilongjiang Province for providing support to this study.

Conflicts of Interest

The authors declare no conflict of interest. The funder had no role in the design, execution, interpretation, or writing of the study.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures and Tables

Figure 1. Development trends of logistics carbon emissions in Heilongjiang province and 12 prefecture-level cities from 2000 to 2020.

Figure 2. Carbon emissions of the logistics industry by energy sources in Heilongjiang Province.

Figure 3. Carbon emission intensity of logistics industry in Heilongjiang Province.

Figure 4. Results of hotspot analysis of logistics carbon emissions based on the longitudinal timeline data in Heilongjiang Province.

Figure 5. Analysis on carbon emission hotspot of logistics industry based on the transverse timeline data in Heilongjiang Province.

Figure 6. The global spatial autocorrelation of district- and county-level logistics carbon emissions in Heilongjiang Province.

Figure 7. LISA cluster map of carbon emissions from logistics industry in Heilongjiang province.

Figure 8. Contribution rates of various influencing factors to logistics carbon emissions.

Table 1

Conversion coefficients of standard coal and carbon emission coefficients for different types of energy.

Energy Type	Conversion Coefficients of Standard Coal (Kilogram of Coal Equivalent) 3	Carbon Emission Coefficients (Tonne of Carbon per Tonne of Coal Equivalent) 4
Coal (kg)	0.7143	0.7559
Coke (kg)	0.9714	0.8031
Washed coal (kg)	0.9	0.7286
Gasoline (kg)	1.4714	0.5538
Kerosene (kg)	1.4714	0.5714
Diesel (kg)	1.4571	0.5921
Fuel oil (kg)	1.4286	0.6053
LPG 1 (kg)	1.7143	0.4934
Natural gas (m3)	1.33	0.4434
LNG 2 (kg)	1.7572	0.4726
Heat (MJ)	0.03412	0.8793
Electricity (kW·h)	0.1229	2.2132

¹ LPG-liquefied petroleum gas; ² LNG-liquefied natural gas; ³ The coefficients are from the literature [33]; ⁴ The coefficients are from the literature [34].

Table 2

Classification of decoupling states.

Decoupling State		ΔC	ΔV	DE
connection	Growth connection	>0	>0	[0.8, 1.2]
	Recession connection	<0	<0	[0.8, 1.2]
decoupling	Strong decoupling	<0	>0	(−∞, 0]
	Weak decoupling	>0	>0	[0, 0.8]
	Recession decoupling	<0	<0	[1.2, +∞)
negative decoupling	Strong negative decoupling	>0	<0	(−∞, 0]
	Weak negative decoupling	<0	<0	[0, 0.8]
	Expansion negative decoupling	>0	>0	[1.2, +∞)

Table 3

Decoupling of carbon emissions from logistics industry and logistics output in Heilongjiang Province during the period of 2000 to 2020.

Cities	%ΔC	%ΔV	DE	Decoupling State
Harbin	0.086	0.049	1.767	Expansion negative decoupling
Qiqihar	0.073	0.061	1.186	Growth connection
Jixi	0.094	0.110	0.856	Growth connection
Hegang	0.090	0.071	1.272	Expansion negative decoupling
Daqing	0.006	0.055	8.567	Expansion negative decoupling
Yichun	0.070	0.041	1.705	Expansion negative decoupling
Qitaihe	0.067	0.037	1.779	Expansion negative decoupling
Shuangyashan	0.125	0.164	0.759	Weak decoupling
Kiamusze	0.087	0.077	1.130	Growth connection
Mudanjiang	0.100	0.131	0761	Weak decoupling
Heihe	0.396	0.675	0.587	Weak decoupling
Suihua	0.183	0.310	0.591	Weak decoupling
Total	0.0794	0.047	1.687	Expansion negative decoupling

References

1. Reilly, J.M.; Jacoby, H.D.; Prinn, R.G. Multi-Gas Contributors to Global Climate Change-Climate Impacts and Mitigation Costs of Non-CO₂ Gases; Pew Center Global Climate Change, Massachusetts Institute of Technology: Cambridge, MA, USA, 2003; pp. 1-3.

2. Freeland, S. The Kyoto protocol: An agreement without a future?. UNSW Law J.; 2001; 24, pp. 532-542.

3. The Paris Agreement. United Nations Framework Convention on Climate Change. The Paris Agreement. 2016; Available online: https://unfccc.int/sites/default/files/resource/parisagreement_publication.pdf (accessed on 30 July 2022).

4. The, S.H.; Wiedmann, T.; Castel, A. Hybrid life cycle assessment of greenhouse gas emissions from cement, concrete and geopolymer concrete in Australia. J. Clean. Prod.; 2017; 152, pp. 312-320.

5. Wich-Konrad, T.; Luke, W.; Oles, M. Assessment of industrial greenhouse gas reduction strategies within consistent system boundaries. Chem. Ing. Tech.; 2020; 92, pp. 1393-1402. [DOI: https://dx.doi.org/10.1002/cite.202000059]

6. Tang, Z.P.; Yu, H.J.; Chen, M.X. The scenario analysis of China’s carbon emission peak and carbon neutrality based on extremum condition of function. J. Nat. Resour.; 2022; 37, pp. 1247-1260. [DOI: https://dx.doi.org/10.31497/zrzyxb.20220510]

7. Zhang, L.G. The dynamic change and regional difference analysis on the driving factors of changes in carbon dioxide emission in China’s logistic industry. China Circ. Econ.; 2016; 30, pp. 29-39.

8. Pan, X.H.; Liang, S.; Zhang, M.H. Study on the distribution characteristics and dynamic evolution of carbon emissions in China. Chin. Eng. Cons; 2021; 9, pp. 27-34.

9. Wen, X.Y. Analysis of the imbalance and agglomeration degree of regional logistics industry development -based on data from six provinces in central China. J. Commer. Econ.; 2021; 3, pp. 94-97.

10. Zhang, L. SWOT Analysis of the development of modern logistics industry in Heilongjiang province. China J. Commer.; 2017; 35, pp. 38-40.

11. Liu, M.D.; Meng, J.J.; Liu, B.H. Progress in the studies of carbon emission estimation. Trop Geogr.; 2014; 34, pp. 248-258.

12. Gao, Y. Study on calculation and energy structure of logistics industry carbon footprint in Shaanxi. Logist. Technol.; 2018; 37, pp. 87–89+155.

13. Yuan, J. Research on Regional Difference Measurement and Influencing Factors of Carbon Emission in China’s Logistics Industry. Master’s Thesis; Chengdu University of Technology: Chengdu, China, 2019.

14. Zhang, J.L.; Li, C. Carbon emissions differences of logistics industry in Beijing-Tianjin-Hebei region. J. North China Electr. Power Univ. Soc. Sci. Ed.; 2019; 1, pp. 53-61.

15. Qian, K. Carbon emission trend analysis of logistics industry in Northeast China. Econ. Soc. Dev. Obs.; 2021; 18, pp. 105-106.

16. Clarke-Sather, A.; Qin, W. Carbon inequality at the sub-national scale: A case study of provincial-level inequality in CO₂ emissions in China 1997–2007. Energy Policy; 2011; 39, pp. 5420-5428. [DOI: https://dx.doi.org/10.1016/j.enpol.2011.05.021]

17. Hu, Y.; Liu, G.C.; Kong, X.Z. Temporal-spatial variance of China’s carbon dioxide emission industry. Resour. Ind.; 2016; 18, pp. 67-75.

18. Wang, Y.; He, Y.F. Spatiotemporal dynamics and influencing factors of provincial carbon emissions in China. World Reg. Stud.; 2020; 29, pp. 512-522.

19. Hu, M.R. Spatial Pattern and Influencing Factors of Carbon Emissions at Country Level in China. Master’s Thesis; Lanzhou University: Lanzhou, China, 2021.

20. Xu, W.G. Efficiency measurement of China’s regional carbon emission efficiency in logistics industry: A method based on Super-PEBM model. Econ. Res. Gui; 2022; 13, pp. 38-42.

21. Liddle, B.; Lung, S. Age-structure urbanization and climate change in developed countries: Revisiting STRPAT for disaggregated population and consumption-related environmental impacts. Popul. Environ.; 2010; 31, pp. 317-343. [DOI: https://dx.doi.org/10.1007/s11111-010-0101-5]

22. Deng, J.L. Basis of Grey Theory; 3rd ed. Huazhong University of Science and Technology Press: Wuhan, China, 2002.

23. Liu, S.F.; Yang, Y.J. Grey System Theory and Its Application; 3rd ed. Science Press: Beijing, China, 2014.

24. Ang, B.W. The LMDI approach to decomposition analysis: A practical guide. Energy Policy; 2003; 33, pp. 867-871. [DOI: https://dx.doi.org/10.1016/j.enpol.2003.10.010]

25. Timilsina, G.R.; Shrestha, A. Factors affecting transport sector CO₂ emissions growth in Latin American and Caribbean countries: An LMDI decomposition analysis. Int. J. Energy Res.; 2010; 33, pp. 396-414. [DOI: https://dx.doi.org/10.1002/er.1486]

26. Yao, G.X.; Zhao, Z.Q.; Hu, B.L. A study on the decoupling trend of rural logistics carbon emissions and regional economic growth in China. East China Econ. Manag.; 2017; 31, pp. 51-56.

27. Liu, Y.; Li, L. Research on decoupling and its influencing factors of the carbon emission for logistics industry in China. Environ. Sci. Technol.; 2018; 41, pp. 177-181.

28. Hu, X.F.; Wang, X.H.; Wu, S. Research on carbon emission and the driving factors of logistics industry in Yangtze River Economic Belt. Ecol. Econ.; 2019; 35, pp. 49-55.

29. Shen, W.T. China logistics industry carbon emission measurement. Manag. Admin.; 2013; 12, pp. 63-66.

30. Li, F.; Zhang, Y.Q.; Zhu, X.C. Carbon emission calculation of construction machinery during the expressway construction period. IOP Conf. Ser. Earth. Environ. Sci.; 2021; 657, 12087. [DOI: https://dx.doi.org/10.1088/1755-1315/657/1/012087]

31. Yu, W.M.; Zhang, T.T.; Shen, D.J. Country-level spatial pattern and influencing factors evolution of carbon emission intensity in China: A random forest model analysis. China Environ. Sci.; 2022; 42, pp. 2788-2798.

32. Hu, X.Y. Research on the Scale and Influence Factors of Hubei Logistics Industry’s Carbon Emissions. Master’s Thesis; Hubei University: Wuhan, China, 2018.

33. National Bureau of Statistics Department of Energy. China Energy Statistical Yearbook; China Statistics Publishing House: Beijing, China, 2021.

34. Wu, Y.C. Study on the Influencing Factors of Carbon Emission and Industrial Association of China’s Logistics Industry. Master’s Thesis; Inner Mongolia University of Technology: Huhhot, China, 2021.

35. Chen, Q. Research on the Low-Carbon Development Path of China’s Logistics Industry. Master’s Thesis; North China Electric Power University: Beijing, China, 2019.

36. Tachibana, J. A method for regional-scale material flow and decoupling analysis: A demonstration case study of Aichi Prefecture, Japan. Transp. Policy; 2008; 22, pp. 112-154. [DOI: https://dx.doi.org/10.1016/j.resconrec.2008.08.003]

37. Tapio, P. Towards a theory of decoupling: Degrees of decoupling in the EU and the case of road traffic in Finland between 1970 and 2001. Transp. Policy; 2005; 12, pp. 137-151. [DOI: https://dx.doi.org/10.1016/j.tranpol.2005.01.001]

38. Zhang, W.L. Research on Decoupling of Carbon Emissions of Logistics Industry in Jing-Jin-Ji Based on LMDI Model. Master’s Thesis; Tianjin University of Technology: Tianjin, China, 2016.

39. Liedtke, C.; Baedeker, C.; Kolberg, S. Resource intensity in global food chains: The Hot Spot Analysis. Br. Food J.; 2010; 112, pp. 1138-1159. [DOI: https://dx.doi.org/10.1108/00070701011080267]

40. Ord, J.K.; Getis, A. Local spatial autocorrection statistics: Distributional issues and an application. Geogr. Anal.; 1995; 27, pp. 286-306. [DOI: https://dx.doi.org/10.1111/j.1538-4632.1995.tb00912.x]

41. Zhu, W.Q. Research on the Judgment of Key Traffic Violations Based on Time and Space Hotspot Analysis. Master’s Thesis; Southeast University: Nanjing, China, 2019.

42. Wei, S.H.; Li, J.; Li, Z.Y. Spatial-temporal evolution and its influencing factors of China’s agricultural competitiveness. Acta GeoL. Sin.; 2020; 75, pp. 1287-1300.

43. Ang, B.W. Decomposition analysis for policy making in energy: Which is the preferred method?. Energy Policy; 2004; 32, pp. 1131-1139. [DOI: https://dx.doi.org/10.1016/S0301-4215(03)00076-4]

44. Guo, J. Research on Low Carbon Port Formation Mechanism and Investment Optimization. Master’s Thesis; Dalian Maritime University: Dalian, China, 2020.

45. He, H.; Myers, R.J. Log Mean Divisia Index decomposition analysis of the demand for building materials: Application to concrete, dwellings, and the U.K. Environ. Sci. Technol.; 2021; 55, pp. 2767-2778. [DOI: https://dx.doi.org/10.1021/acs.est.0c02387]

46. Kastner, T.; Rivas, M.J.I.; Koch, W. Global changes in diets and the consequences for land requirements for food. Proc. Natl. Acad. Sci. USA; 2012; 109, pp. 6868-6872. [DOI: https://dx.doi.org/10.1073/pnas.1117054109]

47. Tian, Y.; Zhu, Q.; Geng, Y. An analysis of energy-related greenhouse gas emissions in the Chinese iron and steel industry. Energy Policy; 2013; 56, pp. 352-361. [DOI: https://dx.doi.org/10.1016/j.enpol.2012.12.068]

48. Wang, L.; Wang, Y.; He, H. Driving force analysis of the nitrogen oxides intensity related to electricity sector in China based on the LMDI method. J. Clean. Prod.; 2020; 242, 118364. [DOI: https://dx.doi.org/10.1016/j.jclepro.2019.118364]

49. Mu, X.Y.; Wang, L.; Xu, R. Study on carbon decoupling and influencing factors of logistics industry in Western Provinces. Environ. Sci. Technol.; 2020; 43, pp. 214-219.

50. Yu, Q. Low Carbon Development Evaluation and Countermeasure Research of Logistics Industry in Yangtze River Delta. Master’s Thesis; Anhui University of Science and Technology: Huainan, China, 2020.

51. Zeng, J. Study on Spatial Differences of Carbon Emissions in China’s Logistics Industry. Master’s Thesis; Nanchang Hang Kong University: Nanchang, China, 2019.

52. Yan, Z.H.; Ren, Y.L.; Liu, Y.Q. Spatio-temporal patterns and influencing factors of carbon emissions in Zhejiang province. Resour. Environ. Yangtze Basin; 2017; 26, pp. 1427-1435.

53. Liu, X.; Meng, X.R.; Wang, X.Q. Analysis of influencing factors on carbon emission of logistics industry in east China. J. Hebei Univ. Environ.; 2022; 32, pp. 1-6.

54. De Azevedo, T.R.; Costa, C., Jr.; Brandão, A., Jr.; dos Santos Cremer, M.; Piatto, M.; Tsai, D.S.; Barreto, P.; Martins, H.; Sales, M.; Galuchi, T. et al. SEEG initiative estimates of Brazilian greenhouse gas emissions from 1970 to 2015. Sci. Data; 2018; 5, 180045. [DOI: https://dx.doi.org/10.1038/sdata.2018.45]

55. Liu, C.; Su, Y.; Li, L.L. The estimation and spatial distribution of the carbon emissions of county units in China. Environ. Pollut. Control; 2020; 42, pp. 113-119.

56. Xu, G.Y.; Zhao, T.; Wang, R. Decomposition and decoupling analysis of factors affecting carbon emissions in China’s regional logistics industry. Sustainability; 2022; 14, 6061. [DOI: https://dx.doi.org/10.3390/su14106061]

57. Ma, Y.Y.; Wang, W.G. Character of carbon emission of logistics industry in China and its affecting factors decomposition analysis: Based on LMDI technique. Math. Pract. Theory; 2013; 43, pp. 31-42.

58. Wang, X.Y.; Wang, J.Y. On influence factors and decoupling effect on CO₂ emissions in logistics industry in northeastern provinces of China. J. Hengyang Normal Univ.; 2021; 42, pp. 59-64.