Dynamic analysis of agricultural carbon emissions efficiency in Chinese provinces along the Belt and Road

Hua Zhang; Sidai Guo; Yubing Qian; Yan Liu; Chengpeng Lu

doi:10.1371/journal.pone.0228223

Abstract

To better understand the agricultural resources and environmental problems of the provinces along The Belt and Road in China, it is critical to investigate their agricultural carbon emission efficiency and evolutionary trends. Based on the panel data of 18 key provinces and cities between 2006 and 2015, this paper evaluated the agricultural carbon emission efficiency with the data envelopment analysis–Malmquist model and further explored their dynamic evolutionary trends. There were several main findings. First, the efficiency levels of agricultural carbon emissions showed significant regional differentiation among the areas, with that along the 21st-Century Maritime Silk Road being much higher than that along the Silk Road Economic Belt. Second, technical efficiency was the key factor that restricted the improvement of the comprehensive efficiency of agricultural carbon. Third, most provinces invested in too many redundant and unreasonably allocated resources, showing a trend of diminishing returns to scale. Last, According to dynamic evolution analysis, the total productivity still demonstrated a diminishing trend. This paper provides some suggestions for effectively improve the efficiency of agricultural carbon emissions in China, such as optimize the agricultural industrial structure, increasing the investment of carbon emission reduction technology, and implementing a carbon emission quota clearing system. This paper contributes to the improvement of the environment in China.

Citation: Zhang H, Guo S, Qian Y, Liu Y, Lu C (2020) Dynamic analysis of agricultural carbon emissions efficiency in Chinese provinces along the Belt and Road. PLoS ONE 15(2): e0228223. https://doi.org/10.1371/journal.pone.0228223

Editor: Bing Xue, Institute for Advanced Sustainability Studies, GERMANY

Received: October 30, 2019; Accepted: January 9, 2020; Published: February 7, 2020

Copyright: © 2020 Zhang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data for the variables were obtained from the Chinese Rural Statistical Yearbook (2007–2016) [57], the Annals of China’s Environmental Statistics (2007–2016) [58], the National Bureau of Statistics of China [59], and the Chinese economic database [60].

Funding: This study is sponsored by the National Natural Science Foundation of China (No. 41701142, 71774026, and 41471116), Soft Science Project of Science and Technology Department of Sichuan (No. 18RX0986), Research Center of Sichuan Rural Development (No. CR1813) and Project for Serving Gansu Province’s Economic and Social Development of Lanzhou University (No. 2019-FWZX-04). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Global warming involves the essential task of worldwide environmental governance. The Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC) indicated that the global temperature has raised by 0.85 °C from 1880 to 2012, and the increase in surface temperature every 10 years is higher than that of the prior 10-year period [1]. Global warming has triggered glacier melting, sea level rise, deterioration of freshwater resources, the slow speed of air flow, and the formation of haze, all of which also affect the survival and development of human beings. Therefore, solving the problem of global warming is critical for the long-term stability and well-being of the earth. By 2400, the existing atmospheric greenhouse gas (GHG) components will increase the global temperature by 1 °C on average, and new GHG emissions will produce an additional increase of 2 to 6 °C, resulting in a 10–25-cm rise in the sea level per century. Carbon emission is one of the primary sources of GHGs, and therefore a reduction in carbon emissions has become the principal means of global governance of GHGs. Since the industrial revolution, the massive exploitation and use of fossil energy have rapidly increased industrial carbon emissions [2]. However, the carbon emissions of agricultural activities, such as agricultural land use, crop planting, and livestock farming, cannot be ignored. Given widespread agricultural activities, agriculture has become an important source of GHG emissions [3, 4] while being the most vulnerable to climate change [5, 6]. Therefore, many scholars have extensively researched carbon dioxide emissions from agriculture [7, 8].

The issue of agricultural carbon emissions has been investigated by researchers from various perspectives based on different countries and regions, such as the causality of agricultural CO₂ emissions [9–13]. Samuel and Phebe investigated the relationship between CO₂ and agriculture in Ghana between 1961 and 2012, their findings showed that CO₂ emissions affect the production of coarse grain, cocoa bean, fruits, and vegetables; they suggested the integration of climate change measures into Ghana’s national strategies to strengthen the country’s effort to achieve sustainability [10, 14]. Some scholars studied the relationship between carbon dioxide and economic growth in China’s agricultural sector using the carbon dioxide decoupling theory [15, 16], investigating the structure and features of agricultural production from 1993 to 2013 in Shanghai and measuring the increase in agricultural carbon emissions, and suggested that improving energy efficiency and reducing the amount of energy consumption will be possible solutions for further emissions reduction during agricultural development. Luo et al. stated that, “fertilizer, in-season rice cultivation, and cattle generated the most CO₂ emissions in the categories of agricultural production activities, farming, and livestock husbandry, respectively.”, and also explored the spatial-temporal characteristics of CO₂ emissions and their intensity in China’s agricultural sector [17].

The carbon footprint of crop cultivation patterns and animal husbandry forms and the carbon balance in the agroecosystem are key topics, such as in the fields of food crops, animal husbandry, vegetables, and fruits during climate change [18–20]. Scholars have estimated the carbon footprint of some products to comprehensively assess agricultural carbon emissions [21–26], and this has provided a reliable data foundation for formulating effective carbon dioxide reduction strategies. Fargione et al. [27] and Arevalo et al. [28] researched the change in the carbon effect in two cases of reclaiming biomass energy crops and transforming agricultural land into fast-growing shortcut woody crops, respectively. Lal suggested that soil erosion is the most widespread form of soil degradation and indicated that soil erosion has a significant impact on the global carbon cycle, and this issue must be considered when assessing the global carbon budget [29]. Kindler et al. found that loss of soil carbon through leaching significantly affects the carbon balance of the agricultural system [30].

The many studies that have been conducted to investigate the critical characteristics of agricultural carbon dioxide emissions not only enriched the carbon emission research system but also laid a solid foundation for the development of this study. However, some research gaps exist and need to be filled in: (1) Most of the related literature is based on a single perspective, lacking the necessary macro coordination. Few scholars have systematically analyzed and evaluated the agricultural carbon emission efficiency of the provinces along the B&R areas. (2) Most authors studied carbon emission efficiency from a macro level or focused on the technical perspective [31–35]. Few scholars have discussed the situation, evolution trend, and convergence of agricultural carbon emission efficiency. (3) From the data perspective, the cross-section data from one year are usually selected as a comparison point, whereas few scholars have used panel data to explore the dynamic distribution and evolutionary trends of agricultural carbon emission efficiency. The paper aims to fill these research gaps. As China is in a critical period of transformation in its agricultural economy, agricultural carbon emissions, as an essential part of China’s carbon emissions, should be included in the government’s energy saving and emission reduction work. The B&R initiative represents a bright new aspect of China’s opening to the world. Efficient measurement of the agricultural carbon emission efficiency in the provinces along the B&R can provide the necessary theoretical basis for the formulation of various agricultural carbon emission reduction policies.

Since its reform and opening, China has made remarkable economic achievements, but rapid growth inevitably results in a massive amount of consumption of resources, making China the largest emitter of carbon in the world [36]. Determining how to effectively reduce carbon emissions has become one of the hot spots for academia in China. In September and October 2013, Chinese Government introduced the Silk Road Economic Belt and the 21st-Century Maritime Silk Road policies. These two strategies are combined as The Belt and Road (B&R) in this study. This B&R development strategy has attracted considerable attention from the international community. In 2017, the amount of trade between China and the countries along the B&R reached 7.4 trillion Chinese Yuan (CNY) and increased by 17.8% over 2016[37], which shows that economic and trade cooperation has produced remarkable results. In the new B&R scheme, China and the countries involved are completing agricultural connections under a dual or multilateral cooperation mechanism. The field of cooperation has been expanding, and the economic benefit of agriculture is increasing in some countries. The total agricultural output value along the B&R reached 2916.72 billion yuan in 2017, accounting for 7.59% of the gross domestic product (GDP). In particular, the total agricultural output value reached 1990.15 billion yuan in the Silk Road Economic Belt, accounting for 11.32% of GDP [38]. In addition, the proportion of agricultural land area is high, but the per capita cultivated land area is lower. This shows that agriculture generally occupies a high proportion in the national economic composition along the B&R and plays a particularly important role in the development of the national economy. With the introduction of policies such as “Guidance on Promoting Green Belt and Road” and “The Belt and Road Ecological and Environmental Cooperation Plan,” the research on agricultural carbon emissions is conducive to strengthening the protection of regional agricultural ecological security and improving the ecological environment of regional agricultural production. Therefore, to understand the agricultural resources and environmental problems of the provinces along the B&R in China, it is critical to investigate their agricultural carbon emission efficiency and evolutionary trends, which provide an essential reference and suggestions to formulate an agricultural carbon emissions reduction policy.

Based on the panel data of 18 key provinces and cities in China between 2006 and 2015, including agricultural input, expected output and unexpected output, we evaluated agricultural carbon emission efficiency with a data envelopment analysis (DEA)-Malmquist model and further explored their dynamic evolutionary trends.

Methods and data source

Models

Data envelopment analysis model.

The data envelopment analysis (DEA) model was developed to help compare the relative performance of decision-making units [39]. As a nonparametric estimation method, DEA is a quantitative analysis method that is used to evaluate the relative effectiveness of different sectors based on the linear programming method, which is based on multiple input and output indicators [40]. DEA generates an efficiency score between 0 and 1 for each of the units, indicating how effectively they are at managing their resources [41]. Charnes et al. [39] proposed the CCR model, and Banker et al. [42] further developed CCR into the BCC model by changing the assumption that the scale returns would change. Many studies have been conducted to analyze carbon emissions using the DEA model [43–46]. Under the assumption of stable returns of scale, the input–BCC model is formulated as follows:

Consider a set of n DMUs that is associated with m inputs and s outputs. Each DMU_j (j = 1,2, ⋯, n) consumes amount X_ij (>0) of input i and produces amount Y_ij (>0) of output r. Consider the observed input X_j = (X_1j, X_2j, ⋯, Xmj)^T≥0, X_j≠0, and the observed input Y_j = (Y_1j, Y_2j, ⋯, Ymj)^T≥0, X_j≠0 for DMU_j.

Generally, the Production Possibility Set (PPS) of the BCC model and the corresponding variable returns to scale can be defined as follows: (1)

To evaluate the efficiency of DMU_j0 related to the Production Possibility Set t, the linear programming model can be defined as follows: (2) where is the input factor of this DMU, is the output factor, and λ_j stands for the weight. The model can be used to evaluate the relative efficiency of DMU_j0. Eq (2) uses as few input factors (minθ, ) as possible under the condition of . When θ = 1, the input cannot be reduced since the output does not decrease through the weight combination, and the DMU is effective at this time. Otherwise, there is a fictitious DMU (θ < 1); that is, the same or even more outputs can be achieved with less input.

After introducing the relaxation variables S⁻, S⁻ and non-Archimedes infinity ε, Eq (2) is transformed into a CCR model. (3) where E₁^T and E₂^T are m-dimension and s-dimension unit vectors respectively, and non-Archimedes infinity ε is a value less than any positive number but greater than 0. Let λ*, S*⁻, S*⁺, and θ* represent an optimal solution for the CCR model, and the decision theorem of DEA validity is as follows:

If θ* = 1, then DMU_j0 is weakly valid.

If θ* = 1, S*⁻ = 0 and S*⁺ = 0, then DMU_j0 is valid.

To evaluate the pure technical efficiency of regional agricultural carbon emission, the BCC model with non-Archimedes infinity ε is as follows: (4)

Compared with the CCR model, the BCC model has only one more constraint . Let λ*, S*⁻, S*⁺, and θ* represent an optimal solution for the BCC model, and the decision theorem of DEA validity is as follows:

If θ* = 1, then DMU_j0 is weakly valid.

If θ* = 1, S*⁻ = 0 and S*⁺ = 0, then DMU_j0 is valid.

The BCC model clearly distinguishes the comprehensive technical efficiency (CTE), pure technical efficiency (PTE), and scale efficiency (SE) of each decision-making unit. CTE is used to measure the ability of maximum output given a fixed investment or minimum input given a fixed outcome. SE is used to measure the extent to which economies of scale can function when the scale reaches an effective point. PTE is used to measure the efficiency level after eliminating the scale factor in the comprehensive technical efficiency [47]. Therefore, for each decision-making unit, CTE is a combination of PTE and SE.

Malmquist index model.

The Malmquist index was first introduced by Malmquist in 1953, and it was initially used to analyze the movement effect of consumption bundles on different indifference curves. Later, this index was applied to production analysis and the measurement of productivity changes in two successive periods [48–50], combining the nonparametric linear programming method and the DEA method, and then the Malmquist index was gradually widely applied.

The distance function is the basis of Malmquist index. It is a method of studying multiple inputs and outputs without any assumptions about producer behavior. The distance function can mathematically be expressed as the reciprocal of the optimal solution of the CCR model and BCC model. (5) (6) where stands for the input and output vector of region i at time t, respectively; S stands for the possible production set at time t; C and V stand for the constant returns to scale and variable returns to scale, respectively; and represents the distance functions, taking technology standard at time t. The higher the value of , the more effective the production is and the higher the technical efficiency is.

Since the Malmquist index is defined by benchmark technology, the output-based Malmquist index referring to technology at time t is as follows: (7)

Similarly, the output-based Malmquist index referring to technology at time t+1 is as follows: (8)

The Malmquist production index can be defined as the geometric average of Eqs (7) and (8) according to the ideal index construction method [43]. (9) (10) where x_i^t and x_i^{t + 1} stand for the input vector of region i at times t and t + 1, respectively; y_i^t and y_i^{t + 1} stand for the output vector of region i at times t and t + 1, respectively; and D_i^t (x_i^t, y_i^t) and D_i^t + 1 (x_i^{t + 1}, y_i^{t + 1}) represent the distance functions taking technology standard at times t and t + 1, respectively, as reference points. Eq (10) is the deformation of Eq (9), which refers to the separation of technological change from technical efficiency change. is the technical efficiency change from time t to time t + 1, and the rest is TC, which is the technical change from time t to time t + 1.

(11)

Eq (11) changes the original assumption that the scale remuneration is fixed and is suitable when the return of scale is changeable. This function further divides the comprehensive technical efficiency change (CTEC) into pure technical efficiency change (PTEC) and scale efficiency change (SEC) [41]. Subscript v shows the circumstance of changeable return of scale, whereas subscript c shows the circumstance of the eternal return of scale. The first part of Eq (11) shows the PTEC change. The second part is the change of SEC, whereas the third part is the same as Eq (10), which shows the CTEC change.

The Malmquist index method uses multiple input and output variables to analyze efficiency. It divides the cause of productivity change into technology change (TC) and CTEC, whereas CTEC can be subdivided into PTEC and SEC. Thus, the Malmquist index can be shown as: ML = CTEC × TC = PTEC × SEC × TC.

Data and variables

Agricultural carbon emissions from carbon sources.

Agriculture has a general and narrow definition. From a broad perspective, agriculture contains five kinds of industrial forms: planting, forestry, animal husbandry, fishery, and sideline industry, whereas the narrow sense refers to only the planting industry. The sources of agricultural carbon emissions can be subdivided into three sections. The first section includes the carbon emissions that are caused by agricultural land use, including chemical fertilizers, pesticides, and agricultural film. Carbon emissions from the use of diesel by agricultural machinery, the loss of organic carbon produced by soil destruction when tilling the soil, and the carbon release from the indirect consumption of fossil fuels during the irrigation process are all included in this section. Second is the production of methane and other greenhouse gases during rice growth. The third is carbon emissions from ruminant farming, including methane production from intestinal fermentation and excrement management. Agricultural carbon emissions are the total amount of carbon emissions that are produced by various types of carbon sources. According to the IPCC and existing literature, the carbon emission coefficients of various agricultural carbon sources are shown in Table 1.

Download:

Table 1. Carbon emission coefficients of various agricultural carbon sources.

https://doi.org/10.1371/journal.pone.0228223.t001

According to Table 1, the total carbon emissions mainly include three parts. First, the carbon emissions of chemical fertilizers, pesticides, agricultural film and diesel oil are the product of the usage amount and carbon emission coefficient. The formula for calculating carbon emissions from agricultural land use is as follows: C₁ = fertilizer use*0.8956 + pesticide use*4.9341 + agricultural film use*5.18 + diesel oil use* 0.5927. Second, the carbon emissions of agricultural irrigation, tillage and rice are the product of area and carbon emission coefficient. The formula for calculating carbon emissions from rice planting is as follows: C₂ = irrigation area * 266.48 +soil destruction area * 312.6 +rice planting area* 49.57. Then, the carbon emissions of ruminants are divided into two parts, where one is the product of the number of animals and the coefficient of intestinal fermentation and the other is the product of the number of animals and the coefficient of fecal management. The formulas for calculating carbon emissions from ruminant farming are as follows: C₃ = cow number* (416.02+122.76) + buffalo number* (375.1+13.64) +other cattle number* (320.54+6.82) + horse number *(122.76+11.18) +donkey number*(68.2+6.14) + mule number *(68.2+6.14) +pig number (6.82+27.28) +goat number *(34.1+1.16) +sheep number *(34.1+1.02). Therefore, the total carbon emissions from carbon sources are C = C₁+ C₂+ C₃.

Selection of input and output indicators.

According to previous research [3,7,11,15,19] and based on traditional macroeconomic theory, the input variables of the agricultural carbon emission efficiency of the provinces and cities along the B&R can be summarized with the following four items: (1) labor input, measured by the number of agricultural workers; (2) land input, measured by the cultivated land area; (3) capital investment, estimated by the annual investment in agricultural fixed assets and the PIM index that is used to convert the flow index into the stock index; and (4) input of agricultural material, which contains fertilizer input, diesel fuel, agricultural film input, and pesticide input. Fertilizer input is measured by the actual amount of fertilizer application, including nitrogen, phosphorus, potassium, and compound fertilizers. Diesel fuel is measured by the annual use of diesel in the agricultural sector. Agricultural film input is measured by the annual consumption of the membrane. Pesticide input is measured by the annual use of pesticides.

The total output values of agriculture, forestry, animal husbandry and fisheries were selected as the expected outputs and agricultural carbon emissions as the undesirable outputs. To eliminate the interference of prices, this study calculated the total output values of agriculture, forestry, animal husbandry, and fisheries in 2006. When using the DEA model to evaluate the environmental efficiency, the efficiency is usually presented in the form of the evaluation index. The greater the index value, the higher the efficiency level. Therefore, higher efficiency requires less input and more output. The input and output indicators can be seen in Table 2.

Download:

Table 2. Input and output indicators of agricultural carbon emissions.

https://doi.org/10.1371/journal.pone.0228223.t002

The DEA model requires both the input and output data of each decision unit to be positive. As a negative output index, carbon dioxide emissions cannot meet the operating conditions of the DEA model, and the selection of the BCC model results in evaluation invalidity. Thus, CO₂ emission indicators must be transformed accordingly. The current applicable methods are curved measure evaluation, pollutant input treatment, data conversion function processing, and the directional distance function. The research selected the data conversion function method [56], as this method is regarded as one of the most satisfactory efficiency assessment methods and includes different types of negative outputs and linear and nonlinear data conversions. Since the linear data conversion method is more advantageous in the model with variable returns to scale analysis (VRS), this method is used to transform the environmental pollutants index data. It can be measured as: Y^’_i = -Y+C, where C represents a large vector to ensure that all converted output data are positive.

Data sources.

The data for the variables were obtained from the Chinese Rural Statistical Yearbook (2007–2016) [57], the Annals of China’s Environmental Statistics (2007–2016) [58], the National Bureau of Statistics of China [59], and the Chinese economic database [60].

Results

Based on the steps, the descriptive statistical results of variables are calculated and listed in Table 3.

Download:

Table 3. Descriptive statistical results of variables.

https://doi.org/10.1371/journal.pone.0228223.t003

Measurement of agricultural carbon emission efficiency based on BCC model

With the use of DEAP 2.0 (the Computer Vision and Systems Laboratory at University Laval, Quebec City, Canada) and the BCC model, this paper evaluated the agricultural carbon emission efficiency of 18 key provinces and cities (there are 16 provinces and two municipalities in total; for ease of reference, the paper refers to them all as provinces) along the B&R. The empirical model determined the comprehensive technical efficiency, pure technical efficiency, and the change of economies of scale.

Comprehensive technical efficiency of agricultural carbon emissions.

Among the three efficiency indicators, the comprehensive technical efficiency most directly reflects the efficiency values of the 18 provinces within 10 years, which were affected by both the pure technical efficiency and scale efficiency. Additionally, efficiency here is a relative measure (relative to other provinces) rather than an absolute one. Therefore, the paper selected the comprehensive technical efficiency value as a measurement, and the result is shown in Table 4.

Download:

Table 4. Comprehensive technical efficiency of agricultural carbon emissions of the key provinces along the Belt and Road (B&R) within 2006–2015.

https://doi.org/10.1371/journal.pone.0228223.t004

Table 3 shows that the agricultural carbon emission efficiencies of the five provinces in Xinjiang, Tibet, Shanghai, Zhejiang and Hainan have long reached the effective frontier of comprehensive technology. There are five provinces with high efficiencies (0.8 < x < 1): Qinghai, Heilongjiang, Liaoning, Fujian and Guangdong. The remaining eight provinces are classified in the middle level: Shaanxi, Gansu, Ningxia, Inner Mongolia, Jilin, Guangxi, Yunnan, and Chongqing. Forty-five percent of the provinces remain in the medium efficiency level, which highlights the importance of improving carbon emission efficiency. Carbon emission reduction strategies need to be further implemented.

Second, from the fluctuation of the comprehensive technical efficiency of each province over the last 10 years, except for the five provinces of Xinjiang, Tibet, Shanghai, Zhejiang and Hainan which have already reached the effective frontier, the fluctuation range of Shaanxi, Gansu, Ningxia, Qinghai, Liaoning, Chongqing, Fujian and Guangdong is basically between 0.1 and 0.2, which is a small range. Additionally, the fluctuation range of efficiency in Inner Mongolia, Heilongjiang, Jilin and Guangxi is between 0.2 and 0.3. Since the launch of China’s the Belt and Road (B&R) initiative in 2013, the carbon emission efficiency of all other provinces has not achieved sustained growth except for Guangdong.

Pure technical efficiency and economies of scale of agricultural carbon emission.

The comprehensive technical efficiency is affected by both the pure technical efficiency and scale efficiency, which can be calculated as: comprehensive technical efficiency = pure technical efficiency × scale efficiency. The paper further discusses how the latter two efficiencies affect the comprehensive technical efficiency of each province. The details are provided in Table 5.

Download:

Table 5. Average value of the efficiencies of key provinces along the B&R, 2006–2015.

https://doi.org/10.1371/journal.pone.0228223.t005

Table 4 shows that five provinces—Xinjiang, Tibet, Shanghai, Zhejiang, and Hainan—have long reached comprehensive technical efficiency. Their pure technical efficiency and efficiency are both 1, proving that they have reached the frontiers from both perspectives. Shaanxi, Gansu, Ningxia, Qinghai, Inner Mongolia, Jilin, Yunnan, and Chongqing are the provinces whose pure technical efficiencies are less than their scale efficiencies. For these provinces, the lag in carbon emission reduction technology is restricting the efficiency of their agricultural carbon emissions. The provinces with lower scale efficiency than pure technical efficiency are Heilongjiang, Guangxi, and Guangdong, showing that their scale of resource allocation is unreasonable and that appropriate adjustments should be undertaken.

Increase/decrease in economies of scale of agricultural carbon emission.

For the majority of provinces, the lag in technology is the key factor restricting the efficiency of their agricultural carbon emission. For the scale efficiency indicator, only five provinces reached the best production frontier, and considerable improvements are needed in the other 13 provinces, indicating that the allocation of the input resources needs to be improved in those 13 provinces. Therefore, the paper further investigated the increase/decrease in the scale remuneration of the agricultural carbon emissions from 2006 to 2015 in different provinces (Table 6).

Download:

Table 6. Increase/decrease in economies of scale of agricultural carbon emissions of key provinces along the B&R, 2006–2015.

https://doi.org/10.1371/journal.pone.0228223.t006

Shanxi, Gansu, Jilin, Guangxi, and Yunnan demonstrated a long-term decreasing trend. In these areas, radial movement should be reduced, and the rational allocation of resources should be optimized to improve the efficiency of agricultural carbon emissions. Similar strategies also apply to Inner Mongolia and Chongqing. The four efficient provinces of Guangdong, Fujian Liaoning, and Heilongjiang have some years of diminishing returns on the scale. Thus, governments should strive to improve scale efficiency and achieve the frontier of production in comprehensive technical efficiency. For Ningxia and Qinghai, which maintained increasing returns to scale in most of the years, a possible strategy involves increasing the input materials to improve carbon emission efficiency. For Xinjiang, Tibet, Shanghai, Zhejiang, and Hainan, the scale of remuneration remained unchanged for 10 years.

Regional comparison of agricultural carbon emissions.

The Belt and Road Initiative is divided into the 21st-Century Maritime Silk Road and The Silk Road Economic Belt. Each contains different provinces, and a detailed classification is outlined in Tables 4–6. The paper further compared the three average values of the efficiencies in those regions to provide useful advice on how to strengthen exchange and cooperation between regions. The results are shown in Figs 1–3.

Download:

Fig 1. Comprehensive technical efficiency of different regions, 2006–2015.

https://doi.org/10.1371/journal.pone.0228223.g001

Download:

Fig 2. Pure technical efficiency of different regions, 2006–2015.

https://doi.org/10.1371/journal.pone.0228223.g002

Download:

Fig 3. Scale efficiency of different regions, 2006–2015.

https://doi.org/10.1371/journal.pone.0228223.g003

The above figures show that for any of the three efficiencies, the overall efficiency of the B&R always falls between the higher values for the 21st-Century Maritime Silk Road provinces and the lower values of The Silk Road Economic Belt provinces. Therefore, the governance of agricultural carbon emissions should focus on the provinces along The Silk Road Economic Belt. Those provinces along the 21st-Century Maritime Silk Road benefit from a small proportion of agricultural industry, fewer carbon emissions from agriculture, high total economic volume, and massive investment in environmental management. They have achieved an ideal state of agricultural carbon emissions. It is highly recommended that those provinces should strengthen communication and cooperation with provinces along the Silk Road Economic Belt, providing experience and financial support for carbon emission reduction.

Dynamic evolution analysis of agricultural carbon emission efficiency based on the Malmquist index model

With the use of the DEA-Malmquist index, the paper could test the dynamic change in the agricultural carbon emission efficiency in various provinces. In the process of dynamic evolution, it can determine whether the efficiency has been promoted since the B&R initiative in 2013 to measure the impact of the initiative on China’s environmental protection scheme.

Temporal dynamic evolution of agricultural carbon emission efficiency.

Table 7 shows that the total factor productivity of the key provinces in China fluctuated during the period of 2006 to 2015. The main trend was downward, except for an increase of 6.5% from 2010 to 2011. Even after 2013, the total factor productivity still demonstrated a diminishing trend, which further indicates that the B&R initiative has not improved China’s carbon emission efficiency. From the decomposition of the Malmquist index, the reduction in productivity was mainly due to the mobile efficiency of the production frontier or technological change. During the descending periods, this change led to a 5.45% average decrease in the total factor productivity. Thus, the paper concludes that the reduction of the technology utilization level is the critical issue restricting the efficiency of agricultural carbon emissions. Even after the introduction of the B&R initiative in 2013, technology still causes a downward trend in total factor productivity, and there was no substantial change thereafter.

Download:

Table 7. Temporal dynamic evolution in agricultural carbon emission efficiency of the provinces along B&R, 2006–2015.

https://doi.org/10.1371/journal.pone.0228223.t007

In addition to the temporal analysis of the dynamic changes in the agricultural carbon emission efficiency of the provinces, the paper also analyzed the specific conditions of each province, as shown in Table 8. First, the results revealed that the agricultural carbon emission efficiencies of the key provinces were trending downward. Only 3 of the 18 provinces, Ningxia, Heilongjiang, and Shanghai, had a Malmquist greater than 1. Second, the technical improvement index for all the other provinces was less than 1, except for Shanghai, which verifies that the lag in technology is the key factor in reducing agricultural carbon emissions.

Download:

Table 8. Dynamic evolution of agricultural carbon emission efficiency of each province along the B&R, 2006–2015.

https://doi.org/10.1371/journal.pone.0228223.t008

Spatial dynamic evolution of agricultural carbon emission efficiency.

The paper also discussed the spatial difference and evolution of the agricultural carbon emission efficiency of key provinces along the B&R with the use of the Malmquist index and ArcGIS (Environmental Systems Research Institute, RedLands, and The United States). The spatial distribution was measured in different periods. The paper selected the years 2007, 2011, and 2015 to observe the spatial differences in agricultural carbon emissions at different time points and then discussed their evolution characteristics.

Figs 4–6 reflect the spatial evolution at three key time points from 2006 to 2015. It regards 2011 as the best time point, as the Malmquist of most provinces (16) was greater than 1, with only one in the 0.9–0.95 and another in the 0.8–0.9 range. In 2007, the Malmquist index increased in six provinces, with a value larger than 1. Five of the other provinces remained within 0.95–1, while the others had smaller values. For 2015, the performance was even more unsatisfactory, with two provinces experiencing an increase in the Malmquist index. The other cities were categorized into groups of 0.95–1, 0.9–0.95, 0.8–0.9, or less than 0.8. This difference further illustrates the conclusion provided in the previous section.

Download:

Fig 4. Spatial distributions of the Malmquist index of key provinces along the B&R in 2007.

https://doi.org/10.1371/journal.pone.0228223.g004

Download:

Fig 5. Spatial distributions of the Malmquist index of key provinces along the B&R in 2011.

https://doi.org/10.1371/journal.pone.0228223.g005

Download:

Fig 6. Spatial distributions of the Malmquist index of key provinces along the B&R in 2015.

https://doi.org/10.1371/journal.pone.0228223.g006

Discussions and conclusions

Exploring the efficiency is the basic premise in further research on China’s agricultural carbon emissions [61], which helps explain agricultural “water-land-energy-carbon” (WLEC) nexus and improve the efficiency of agricultural resources use [62]. Based on the panel data of 18 key provinces and cities along the B&R between 2006 and 2015, the paper evaluated agricultural carbon emissions efficiency using the DEA-BCC model and further explored their dynamic evolution trend with the DEA-Malmquist model. The results showed that: (1) the level of agricultural carbon emissions in various provinces was uneven. Xinjiang, Tibet, Shanghai, Zhejiang, and Hainan were at the forefront of agricultural carbon emissions, with the five provinces of Qinghai, Heilongjiang, Liaoning, Fujian, and Guangdong being highly efficient. The remaining eight provinces demonstrated medium efficiency. For most provinces, the efficiency of agricultural carbon emissions must be improved. (2) Technical efficiency is the key factor that restricts the promotion of comprehensive efficiency of agricultural carbon. (3) Although the impact of scale efficiency on comprehensive efficiency is less than that of technical efficiency, except for five provinces at the frontier of production, the scale efficiency of the other provinces requires improvement. Through the increase or decrease in scale remuneration, most provinces demonstrated downward trends in their economies of scale, implying that their investment is too redundant, and that the allocation of resources is not reasonable. (4) The efficiency level of agricultural carbon emissions presented significant regional differentiation among the areas, with those along the 21st-Century Maritime Silk Road being much higher than those included in the Silk Road Economic Belt. (5) According to dynamic evolution analysis, the total factor productivity still demonstrated a diminishing trend. Therefore, although the B&R Initiative has promoted the exchange and development of China with the surrounding countries in the economic, cultural, and political arenas, the initiative has contributed little to the environmental amelioration of our country, which is contrary to the concept of sustainable development, which should be seriously considered by the Chinese government.

From the above research results, the paper provides some suggestions to effectively enhance the efficiency of agricultural carbon emissions in China. (1) The government should optimize the agricultural industrial structure and accelerate agricultural modernization in the Belt and Road region. (2) A low-carbon agricultural development mechanism should be constructed with the basis of the modern agricultural industrial technology system, and scientific research should be integrated and applied to the whole process of agricultural production. (3) Each province should increase their investment in low-carbon agriculture and establish a complete set of low-carbon agricultural ecological compensation technology system to play China's model as a leading role in low-carbon agriculture along the Belt and the Road. (4) Due to the significant regional differences in agricultural carbon emission efficiencies, the focus of agricultural energy conservation and emission reduction work should be on striving to realize the transformation of agricultural production to a low-carbon direction in the Silk Road Economic Belt areas.

There are also some deficiencies in this paper in the availability of data and deficiencies at the author's own level, such as the lack of deep excavation about the causes and influencing factors for the differences in regional agricultural carbon emissions and the agricultural carbon emission efficiencies of various countries or regions along the Belt and Road region. Then, the cooperation and incentive mechanisms of agricultural carbon emission reduction are not involved. Next, the authors will further research these problems that urgently need to be solved. There are also some deficiencies in this research in the availability of data and the deficiencies at the author's own level, such as the lack of deep excavation about the causes and influencing factors in the differences in regional agricultural carbon emissions and the agricultural carbon emission efficiencies of various countries or regions along the Belt and Road region. Then, the cooperation and incentive mechanisms of agricultural carbon emission reduction are not involved. Next, the authors will further research these problems that urgently need to be solved.

References

1. IPCC (Intergovernmental Panel on Climate Change). Climate Change 2014 Synthesis Report; Intergovernmental Panel on Climate Change: Geneva, Switzerland, 2014.
2. Shao S.; Yang L.; Gan C.; Cao J.; Geng Y.; Guan D. Using an extended LMDI model to explore techno-economic drivers of energy-related industrial CO₂ emission changes: A case study for Shanghai (China). Renew. Sustain. Energy Rev. 2016, 55, 516–536.
- View Article
- Google Scholar
3. Fei R.; Lin B.Q. Technology gap and CO₂ emission reduction potential by technical efficiency measures: A meta-frontier modeling for the Chinese agricultural sector. Ecol. Indic. 2017, 73, 653–661.
- View Article
- Google Scholar
4. Dogan E.; Sebri M.; Turkekul B. Exploring the relationship between agricultural electricity consumption and output: New evidence from Turkish regional data. Energy Policy 2016, 95, 370–377.
- View Article
- Google Scholar
5. Wang W.; Xie H.; Zhang N.; Dong X. Sustainable water use and water shadow price in China’s urban industry. Resour. Conserv. Recycl. 2018, 128, 489–498.
- View Article
- Google Scholar
6. Xue B. Honing the Climate Change Message. Science, 2015, 348(6237):872.
- View Article
- Google Scholar
7. Sebri M.; Abid M. Energy use for economic growth: A triradiate analysis from Tunisian agriculture sector. Energy Policy 2012, 48, 711–716.
- View Article
- Google Scholar
8. Nayak D.; Saetnan E.; Cheng K.; Wang W.; Koslowski F.; Cheng Y.F.; Zhu W.Y.; Wang J.K.; Liu J.X.; Moran D.; et al. Management opportunities to mitigate greenhouse gas emissions from Chinese agriculture. Agric. Ecosyst. Environ. 2015, 209, 108–124.
- View Article
- Google Scholar
9. Ramirez-Villegas J.; Salazar M.; Jarvis A.; Navarro-Racines C.E. A way forward on adaptation to climate change in Colombian agriculture: Perspectives towards 2050. Clim. Chang. 2012, 115, 611–628.
- View Article
- Google Scholar
10. Samuel A.; Phebe A. The relationship between carbon dioxide and agriculture in Ghana: a comparison of VECM and ARDL model. Environmental Science and Pollution Research. 2016, 23(11), 10968–10982. pmid:26898935
- View Article
- PubMed/NCBI
- Google Scholar
11. Rafiq S.; Salim R.; Apergis N. Agriculture, trade openness and emissions: An empirical analysis and policy options. Aust. J. Agric. Resour. Econ. 2016, 60, 648–365.
- View Article
- Google Scholar
12. Jebli M.B.; Youssef S.B. Renewable energy consumption and agriculture: Evidence for cointegration and Granger causality for Tunisian economy. Int. J. Sustain. Dev. World Ecol. 2016, 24, 149–158.
- View Article
- Google Scholar
13. Jebli M.B.; Youssef S.B. The role of renewable energy and agriculture in reducing CO₂ emissions: Evidence for North Africa countries. Ecol. Indic. 2017, 74, 295–301.
- View Article
- Google Scholar
14. Lu C.P.; Xue B.; Lu C.Y.; Wang T.; Jiang L.; Zhang Z.L.; Ren W.X. Sustainability investigation of resource-based cities in northeastern China. Sustainability. 2016, 8, 1058.
- View Article
- Google Scholar
15. Zhang Z.Y. Decoupling elasticity analysis on low carbon agriculture in Shanghai. Adv. Econ. Bus. 2016, 4, 117–123.
- View Article
- Google Scholar
16. Xiong C.H.; Yang D.G.; Huo J.W.; Zhao Y.N. The relationship between agricultural carbon emissions and agricultural economic growth and policy recommendations of a low-carbon agriculture economy. Pol. J. Environ. Stud. 2016, 25, 2187–2195.
- View Article
- Google Scholar
17. Luo Y.; Wu C.; Zhang J. Decoupling CO₂ emissions from economic growth in agricultural sector across 30 Chinese provinces from 1997 to 2014. J. Clean. Prod. 2017, 159, 220–228.
- View Article
- Google Scholar
18. Rotz C.A.; Montes F.; Chianese D.S. The carbon footprint of dairy production systems through partial life cycle assessment. J. Dairy Sci. 2010, 93, 1266–1282. pmid:20172247
- View Article
- PubMed/NCBI
- Google Scholar
19. Cheng K.; Yan M.; Nayak D.; Smith P.; Pan G.X.; Zheng J.W. Carbon footprint of crop production in China: An analysis of national statistics data. J. Agric. Sci. 2014, 153, 422–431.
- View Article
- Google Scholar
20. Rebolledo-Leiva R.; Angulo-Meza L.; Iriarte A.; González-Araya M.C. Joint carbon footprint assessment and data envelopment analysis for the reduction of greenhouse gas emissions in agriculture production. Sci. Total Environ. 2017, 593–594, 36–46. pmid:28342416
- View Article
- PubMed/NCBI
- Google Scholar
21. Cheng K.; Pan G.X.; Smith P.; Luo T.; Li L.Q.; Zheng J.W.; Zhang X.H.; Han X.J.; Yan M. Carbon footprint of China’s crop production-an estimation using agro-statistics data over 1993–2007. Agric. Ecosyst. Environ. 2011, 142, 231–237.
- View Article
- Google Scholar
22. Flachowsky G.; Kamphues J. Carbon footprints for food of animal origin: What are the most preferable criteria to measure animal yields? Animal 2012, 2, 108–126.
- View Article
- Google Scholar
23. Hillier J.; Brentrup F.; Wattenbach M.; Walter C.; Suarez T.G.; Canals L.; Smith P. Which cropland greenhouse gas mitigation options give the greatest benefits in different world regions? Climate and soil-specific predictions from integrated empirical models. Glob. Chang. Biol. 2012, 18, 1880–1894.
- View Article
- Google Scholar
24. Vergé X.P.C.; Dyer J.A.; Worth D.E.; Smith W.N.; Desjardins R.L.; Mcconkey B.G. A greenhouse gas and soil carbon model for estimating the carbon footprint of livestock production in Canada. Animals 2012, 2, 437–454. pmid:26487032
- View Article
- PubMed/NCBI
- Google Scholar
25. Wang W.; Koslowski F.; Nayak D.R.; Smith P.; Saetnan E.; Ju X.T.; Guo L.P.; Han G.D.; Perthuis C.D.; Lin E.; et al. Greenhouse gas mitigation in Chinese agriculture: Distinguishing technical and economic potentials. Glob. Environ. Chang. 2014, 26, 53–62.
- View Article
- Google Scholar
26. Yue Q.; Xu X.R.; Hillier J.; Cheng K.; Pan G.X. Mitigating greenhouse gas emissions in agriculture: From farm production to food consumption. J. Clean. Prod. 2017, 149, 1011–1019.
- View Article
- Google Scholar
27. Fargione J.; Hill J.; Tilman D.; Polasky S.; Hawthorne P. Land clearing and the biofuel carbon debt. Science 2008, 319, 1235–1238. pmid:18258862
- View Article
- PubMed/NCBI
- Google Scholar
28. Arevalo C.M.; Bhatti J.S.; Chang S.X.; Sidders D. Land use change effects on ecosystem carbon balance: From agricultural to hybrid poplar plantation. Agric. Ecosyst. Environ. 2011, 141, 342–349.
- View Article
- Google Scholar
29. Lal R. Soil erosion and the global carbon budget. Environ. Int. 2003, 29, 437–450. pmid:12705941
- View Article
- PubMed/NCBI
- Google Scholar
30. Kindler R.; Siemens J.; Kaiser K.; Walmsley D.C.; Bernhofer C.; Buchmann N.; Cellier P.; Eugster W.; Gleixner G.; Grünwald T.; Heim A.; Ibrom A.; Jones S.K.; Jones M.; Klumpp K.; Kutsch W.; Larsen S.K.; Lehuger S.; Loubet B.; McKenzie R.; Moors E.; Osborne B.; Pilegaard K.; Rebmann C.; Saunders M.; Schmidt W.M.; Schrumpf M.; Seyfferth J.; Skiba U.; Soussana J.; Sutton A.M.; Tefs C.; Vowinckel B.; Zeeman J.M.; Kaupenjohann M. Dissolved carbon leaching from soil is a crucial component of the net ecosystem carbon balance. Global Change Biology, 2011, 17, 1167–1185.
- View Article
- Google Scholar
31. Li J.Y.; Fu B.J.; Liu S.G.; Dargush P.; Gao G.G.; Liu J.B.; Wei F.L. Vegetation restoration changes topsoil biophysical regulations of carbon fluxes in an eroding soil landscape. Land Degrad. Dev. 2018, 29, 4061–4070.
- View Article
- Google Scholar
32. Zhang J.; Zeng W.; Wang J.; Yang F.; Jiang H. Regional low-carbon economy efficiency in China: Analysis based on the Super-SBM model with CO2 emissions. J. Clean. Prod. 2015, 163, 202–211.
- View Article
- Google Scholar
33. Yan D.; Lei Y.L.; Li L.; Song W. Carbon emission efficiency and spatial clustering analyses in China’s thermal power industry: Evidence from the provincial level. J. Clean. Prod. 2017, 156, 518–527.
- View Article
- Google Scholar
34. Zhu S.; Jiang Z.; Zhang H.; Tian G.; Wang Y. A carbon efficiency evaluation method for manufacturing process chain decision-making. J. Clean. Prod. 2017, 148, 665–680.
- View Article
- Google Scholar
35. Lu C.P.; Ren W.X.; and Xue B. Modelling impact of climate change and air pollution in cities. Proceedings of the Institution of Civil Engineers—Engineering Sustainability. 2017,170:3, 133–140.
- View Article
- Google Scholar
36. Oke A.E.; Aigbavboa C.O.; Dlamini S.A. Carbon Emission Trading in South African Construction Industry. Energy Procedia 2017, 142, 2371–2376.
- View Article
- Google Scholar
37. Ministry Of Commerce Of The People's Republic Of China: http://www.mofcom.gov.cn/article/i/dxfw/nbgz/201802/20180202707466.shtml
38. National Bureau of statistics of China: http://www.stats.gov.cn/
39. Charnes A.; Cooper W.W.; Rhodes E. Measuring the efficiency of decision-making units. Eur. J. Oper. Res. 1978, 2, 429–444.
- View Article
- Google Scholar
40. Ke S.F.; Chen Z.C.; Robson M.; Wen Y.L.; Tian X.H. Evaluating the implementation efficiency of the Natural Forest Protection Program in ten provinces of western China by using Data Envelopment Analysis (DEA). International Forestry Review. 2015, 17(4), 469–476.
- View Article
- Google Scholar
41. Ali S.S.; Mohammad T.; Almas H. Assessing the sustainability of high-, middle-, and low-income countries: A network DEA model in the presence of both zero data and undesirable outputs. 2019.
42. Banker R.D.; Charnes A.; Cooper W. Some models for estimating technical and scale inefficiency in Data Envelopment Analysis. Manag. Sci. 1984, 30, 1078–1092.
- View Article
- Google Scholar
43. Guo X.D.; Zhu L.; Fan Y.; Xie B.C. Evaluation of potential reductions in carbon emissions in Chinese provinces based on environmental DEA. Energy Policy 2011, 39, 2352–2360.
- View Article
- Google Scholar
44. Khoshnevisan B.; Rafiee S.; Omid M.; Mousazadeh H. Applying data envelopment analysis approach to improve energy efficiency and reduce GHG (greenhouse gas) emission of wheat production. Energy 2013, 58, 588–593.
- View Article
- Google Scholar
45. Perez K.; González-Araya M.C.; Iriarte A. Energy and GHG emission efficiency in the Chilean manufacturing industry: Sectoral and regional analysis by DEA and Malmquist indexes. Energy Econ. 2017, 66, 290–302.
- View Article
- Google Scholar
46. Dariush A. Ranking all DEA-Efficient DMUs based on cross efficiency and analytic hierarchy process Methods. Journal of Optimization. 2015, 594727, 1–10.
- View Article
- Google Scholar
47. Isidoro G.; Carmelo R. Productivity and efficiency change and shareholder value: evidence from the Spanish banking sector. Applied Economics. 2008, 40(15), 2037–2044.
- View Article
- Google Scholar
48. Caves D.W.; Christensen L.R.; Diewert W.E. Multilateral Comparisons of Output, Input, and Productivity Using Superlative Index Numbers. Econ. J. 1982, 92, 73–86.
- View Article
- Google Scholar
49. Färe R.; Grosskopf S.; Norris M.; Zhang Z.Y. Productivity growth, technical progress, and efficiency change in industrialized countries. Am. Econ. Rev. 1994, 84, 66–83.
- View Article
- Google Scholar
50. Fisher I. The Making of Index Numbers: A Study of Their Varieties, Tests, and Reliability. Houghton, muffin Co. Boston and New York, 1922, pp.526.
51. West T.O.; Marland G. A synthesis of carbon sequestration, carbon emissions, and net carbon flux in agriculture: Comparing tillage practices in the United States. Agric. Ecosyst. Environ. 2002, 91, 217–232.
- View Article
- Google Scholar
52. Tian Y.; Zhang J.B.; Li B. Research on Spatial-Temporal Characteristics and Factor Decomposition of Agricultural Carbon Emission Based on Input Angle-Taking Hubei Province for Example. Res. Agric. Mod. 2011, 32, 752–755.
- View Article
- Google Scholar
53. Intergovernmental Panel on Climate Change (IPCC). Climate Change 2007: The Fourth Assessment Report of The Intergovernmental Panel on Climate Change; Intergovernmental Panel on Climate Change: Geneva, Switzerland, 2007.
54. Wu F.L.; Li L.; Zhang H.L.; Chen H. Net Carbon Emissions of Farmland Ecosystem Influenced by Conservation Tillage. Chin. J. Ecol. 2007, 26, 2035–2039.
- View Article
- Google Scholar
55. Wang X.K.; Li C.S.; Ou Y.; Zhi Y. Greenhouse gas emissions and Chinese grain production. Ecol. Environ. 2003, 12, 379–383.
- View Article
- Google Scholar
56. Seiford L.M.; Zhu J. Modeling Undesirable Factors in Efficiency Evaluation. Eur. J. Oper. Res. 2002, 142, 16–20.
- View Article
- Google Scholar
57. Chinese Rural Statistical Yearbook. China Statistical Publishing House, Beijing 2007–2014.
58. Annals of China’s Environmental Statistics. China Statistical Publishing House, Beijing 200–2014.
59. National Bureau of statistics of China: http://www.stats.gov.cn/
60. The Chinese economic database: https://www.ceicdata.com/zh-hans/products/china-economic-database.
61. Tian Y.; Zhang J.B.; He Y.Y. Research on Spatial-Temporal Characteristics and Driving Factor of Agricultural Carbon Emissions in China. Journal of Integrative Agriculture. 2014, 6(13), 1393–1403.
- View Article
- Google Scholar
62. Liu X.Y.; Zhang S.; Bae J. The impact of renewable energy and agriculture on carbon dioxide emissions: Investigating the environmental Kuznets curve in four selected ASEAN countries. Journal of Cleaner Production. 2017, 164, 1239–1247.
- View Article
- Google Scholar

[ref1] 1. IPCC (Intergovernmental Panel on Climate Change). Climate Change 2014 Synthesis Report; Intergovernmental Panel on Climate Change: Geneva, Switzerland, 2014.

[ref2] 2. Shao S.; Yang L.; Gan C.; Cao J.; Geng Y.; Guan D. Using an extended LMDI model to explore techno-economic drivers of energy-related industrial CO₂ emission changes: A case study for Shanghai (China). Renew. Sustain. Energy Rev. 2016, 55, 516–536.
View Article
Google Scholar

[3] View Article

[4] Google Scholar

[ref3] 3. Fei R.; Lin B.Q. Technology gap and CO₂ emission reduction potential by technical efficiency measures: A meta-frontier modeling for the Chinese agricultural sector. Ecol. Indic. 2017, 73, 653–661.
View Article
Google Scholar

[6] View Article

[7] Google Scholar

[ref4] 4. Dogan E.; Sebri M.; Turkekul B. Exploring the relationship between agricultural electricity consumption and output: New evidence from Turkish regional data. Energy Policy 2016, 95, 370–377.
View Article
Google Scholar

[9] View Article

[10] Google Scholar

[ref5] 5. Wang W.; Xie H.; Zhang N.; Dong X. Sustainable water use and water shadow price in China’s urban industry. Resour. Conserv. Recycl. 2018, 128, 489–498.
View Article
Google Scholar

[12] View Article

[13] Google Scholar

[ref6] 6. Xue B. Honing the Climate Change Message. Science, 2015, 348(6237):872.
View Article
Google Scholar

[15] View Article

[16] Google Scholar

[ref7] 7. Sebri M.; Abid M. Energy use for economic growth: A triradiate analysis from Tunisian agriculture sector. Energy Policy 2012, 48, 711–716.
View Article
Google Scholar

[18] View Article

[19] Google Scholar

[ref8] 8. Nayak D.; Saetnan E.; Cheng K.; Wang W.; Koslowski F.; Cheng Y.F.; Zhu W.Y.; Wang J.K.; Liu J.X.; Moran D.; et al. Management opportunities to mitigate greenhouse gas emissions from Chinese agriculture. Agric. Ecosyst. Environ. 2015, 209, 108–124.
View Article
Google Scholar

[21] View Article

[22] Google Scholar

[ref9] 9. Ramirez-Villegas J.; Salazar M.; Jarvis A.; Navarro-Racines C.E. A way forward on adaptation to climate change in Colombian agriculture: Perspectives towards 2050. Clim. Chang. 2012, 115, 611–628.
View Article
Google Scholar

[24] View Article

[25] Google Scholar

[ref10] 10. Samuel A.; Phebe A. The relationship between carbon dioxide and agriculture in Ghana: a comparison of VECM and ARDL model. Environmental Science and Pollution Research. 2016, 23(11), 10968–10982. pmid:26898935
View Article
PubMed/NCBI
Google Scholar

[27] View Article

[28] PubMed/NCBI

[29] Google Scholar

[ref11] 11. Rafiq S.; Salim R.; Apergis N. Agriculture, trade openness and emissions: An empirical analysis and policy options. Aust. J. Agric. Resour. Econ. 2016, 60, 648–365.
View Article
Google Scholar

[31] View Article

[32] Google Scholar

[ref12] 12. Jebli M.B.; Youssef S.B. Renewable energy consumption and agriculture: Evidence for cointegration and Granger causality for Tunisian economy. Int. J. Sustain. Dev. World Ecol. 2016, 24, 149–158.
View Article
Google Scholar

[34] View Article

[35] Google Scholar

[ref13] 13. Jebli M.B.; Youssef S.B. The role of renewable energy and agriculture in reducing CO₂ emissions: Evidence for North Africa countries. Ecol. Indic. 2017, 74, 295–301.
View Article
Google Scholar

[37] View Article

[38] Google Scholar

[ref14] 14. Lu C.P.; Xue B.; Lu C.Y.; Wang T.; Jiang L.; Zhang Z.L.; Ren W.X. Sustainability investigation of resource-based cities in northeastern China. Sustainability. 2016, 8, 1058.
View Article
Google Scholar

[40] View Article

[41] Google Scholar

[ref15] 15. Zhang Z.Y. Decoupling elasticity analysis on low carbon agriculture in Shanghai. Adv. Econ. Bus. 2016, 4, 117–123.
View Article
Google Scholar

[43] View Article

[44] Google Scholar

[ref16] 16. Xiong C.H.; Yang D.G.; Huo J.W.; Zhao Y.N. The relationship between agricultural carbon emissions and agricultural economic growth and policy recommendations of a low-carbon agriculture economy. Pol. J. Environ. Stud. 2016, 25, 2187–2195.
View Article
Google Scholar

[46] View Article

[47] Google Scholar

[ref17] 17. Luo Y.; Wu C.; Zhang J. Decoupling CO₂ emissions from economic growth in agricultural sector across 30 Chinese provinces from 1997 to 2014. J. Clean. Prod. 2017, 159, 220–228.
View Article
Google Scholar

[49] View Article

[50] Google Scholar

[ref18] 18. Rotz C.A.; Montes F.; Chianese D.S. The carbon footprint of dairy production systems through partial life cycle assessment. J. Dairy Sci. 2010, 93, 1266–1282. pmid:20172247
View Article
PubMed/NCBI
Google Scholar

[52] View Article

[53] PubMed/NCBI

[54] Google Scholar

[ref19] 19. Cheng K.; Yan M.; Nayak D.; Smith P.; Pan G.X.; Zheng J.W. Carbon footprint of crop production in China: An analysis of national statistics data. J. Agric. Sci. 2014, 153, 422–431.
View Article
Google Scholar

[56] View Article

[57] Google Scholar

[ref20] 20. Rebolledo-Leiva R.; Angulo-Meza L.; Iriarte A.; González-Araya M.C. Joint carbon footprint assessment and data envelopment analysis for the reduction of greenhouse gas emissions in agriculture production. Sci. Total Environ. 2017, 593–594, 36–46. pmid:28342416
View Article
PubMed/NCBI
Google Scholar

[59] View Article

[60] PubMed/NCBI

[61] Google Scholar

[ref21] 21. Cheng K.; Pan G.X.; Smith P.; Luo T.; Li L.Q.; Zheng J.W.; Zhang X.H.; Han X.J.; Yan M. Carbon footprint of China’s crop production-an estimation using agro-statistics data over 1993–2007. Agric. Ecosyst. Environ. 2011, 142, 231–237.
View Article
Google Scholar

[63] View Article

[64] Google Scholar

[ref22] 22. Flachowsky G.; Kamphues J. Carbon footprints for food of animal origin: What are the most preferable criteria to measure animal yields? Animal 2012, 2, 108–126.
View Article
Google Scholar

[66] View Article

[67] Google Scholar

[ref23] 23. Hillier J.; Brentrup F.; Wattenbach M.; Walter C.; Suarez T.G.; Canals L.; Smith P. Which cropland greenhouse gas mitigation options give the greatest benefits in different world regions? Climate and soil-specific predictions from integrated empirical models. Glob. Chang. Biol. 2012, 18, 1880–1894.
View Article
Google Scholar

[69] View Article

[70] Google Scholar

[ref24] 24. Vergé X.P.C.; Dyer J.A.; Worth D.E.; Smith W.N.; Desjardins R.L.; Mcconkey B.G. A greenhouse gas and soil carbon model for estimating the carbon footprint of livestock production in Canada. Animals 2012, 2, 437–454. pmid:26487032
View Article
PubMed/NCBI
Google Scholar

[72] View Article

[73] PubMed/NCBI

[74] Google Scholar

[ref25] 25. Wang W.; Koslowski F.; Nayak D.R.; Smith P.; Saetnan E.; Ju X.T.; Guo L.P.; Han G.D.; Perthuis C.D.; Lin E.; et al. Greenhouse gas mitigation in Chinese agriculture: Distinguishing technical and economic potentials. Glob. Environ. Chang. 2014, 26, 53–62.
View Article
Google Scholar

[76] View Article

[77] Google Scholar

[ref26] 26. Yue Q.; Xu X.R.; Hillier J.; Cheng K.; Pan G.X. Mitigating greenhouse gas emissions in agriculture: From farm production to food consumption. J. Clean. Prod. 2017, 149, 1011–1019.
View Article
Google Scholar

[79] View Article

[80] Google Scholar

[ref27] 27. Fargione J.; Hill J.; Tilman D.; Polasky S.; Hawthorne P. Land clearing and the biofuel carbon debt. Science 2008, 319, 1235–1238. pmid:18258862
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref28] 28. Arevalo C.M.; Bhatti J.S.; Chang S.X.; Sidders D. Land use change effects on ecosystem carbon balance: From agricultural to hybrid poplar plantation. Agric. Ecosyst. Environ. 2011, 141, 342–349.
View Article
Google Scholar

[86] View Article

[87] Google Scholar

[ref29] 29. Lal R. Soil erosion and the global carbon budget. Environ. Int. 2003, 29, 437–450. pmid:12705941
View Article
PubMed/NCBI
Google Scholar

[89] View Article

[90] PubMed/NCBI

[91] Google Scholar

[ref30] 30. Kindler R.; Siemens J.; Kaiser K.; Walmsley D.C.; Bernhofer C.; Buchmann N.; Cellier P.; Eugster W.; Gleixner G.; Grünwald T.; Heim A.; Ibrom A.; Jones S.K.; Jones M.; Klumpp K.; Kutsch W.; Larsen S.K.; Lehuger S.; Loubet B.; McKenzie R.; Moors E.; Osborne B.; Pilegaard K.; Rebmann C.; Saunders M.; Schmidt W.M.; Schrumpf M.; Seyfferth J.; Skiba U.; Soussana J.; Sutton A.M.; Tefs C.; Vowinckel B.; Zeeman J.M.; Kaupenjohann M. Dissolved carbon leaching from soil is a crucial component of the net ecosystem carbon balance. Global Change Biology, 2011, 17, 1167–1185.
View Article
Google Scholar

[93] View Article

[94] Google Scholar

[ref31] 31. Li J.Y.; Fu B.J.; Liu S.G.; Dargush P.; Gao G.G.; Liu J.B.; Wei F.L. Vegetation restoration changes topsoil biophysical regulations of carbon fluxes in an eroding soil landscape. Land Degrad. Dev. 2018, 29, 4061–4070.
View Article
Google Scholar

[96] View Article

[97] Google Scholar

[ref32] 32. Zhang J.; Zeng W.; Wang J.; Yang F.; Jiang H. Regional low-carbon economy efficiency in China: Analysis based on the Super-SBM model with CO2 emissions. J. Clean. Prod. 2015, 163, 202–211.
View Article
Google Scholar

[99] View Article

[100] Google Scholar

[ref33] 33. Yan D.; Lei Y.L.; Li L.; Song W. Carbon emission efficiency and spatial clustering analyses in China’s thermal power industry: Evidence from the provincial level. J. Clean. Prod. 2017, 156, 518–527.
View Article
Google Scholar

[102] View Article

[103] Google Scholar

[ref34] 34. Zhu S.; Jiang Z.; Zhang H.; Tian G.; Wang Y. A carbon efficiency evaluation method for manufacturing process chain decision-making. J. Clean. Prod. 2017, 148, 665–680.
View Article
Google Scholar

[105] View Article

[106] Google Scholar

[ref35] 35. Lu C.P.; Ren W.X.; and Xue B. Modelling impact of climate change and air pollution in cities. Proceedings of the Institution of Civil Engineers—Engineering Sustainability. 2017,170:3, 133–140.
View Article
Google Scholar

[108] View Article

[109] Google Scholar

[ref36] 36. Oke A.E.; Aigbavboa C.O.; Dlamini S.A. Carbon Emission Trading in South African Construction Industry. Energy Procedia 2017, 142, 2371–2376.
View Article
Google Scholar

[111] View Article

[112] Google Scholar

[ref37] 37. Ministry Of Commerce Of The People's Republic Of China: http://www.mofcom.gov.cn/article/i/dxfw/nbgz/201802/20180202707466.shtml

[ref38] 38. National Bureau of statistics of China: http://www.stats.gov.cn/

[ref39] 39. Charnes A.; Cooper W.W.; Rhodes E. Measuring the efficiency of decision-making units. Eur. J. Oper. Res. 1978, 2, 429–444.
View Article
Google Scholar

[116] View Article

[117] Google Scholar

[ref40] 40. Ke S.F.; Chen Z.C.; Robson M.; Wen Y.L.; Tian X.H. Evaluating the implementation efficiency of the Natural Forest Protection Program in ten provinces of western China by using Data Envelopment Analysis (DEA). International Forestry Review. 2015, 17(4), 469–476.
View Article
Google Scholar

[119] View Article

[120] Google Scholar

[ref41] 41. Ali S.S.; Mohammad T.; Almas H. Assessing the sustainability of high-, middle-, and low-income countries: A network DEA model in the presence of both zero data and undesirable outputs. 2019.

[ref42] 42. Banker R.D.; Charnes A.; Cooper W. Some models for estimating technical and scale inefficiency in Data Envelopment Analysis. Manag. Sci. 1984, 30, 1078–1092.
View Article
Google Scholar

[123] View Article

[124] Google Scholar

[ref43] 43. Guo X.D.; Zhu L.; Fan Y.; Xie B.C. Evaluation of potential reductions in carbon emissions in Chinese provinces based on environmental DEA. Energy Policy 2011, 39, 2352–2360.
View Article
Google Scholar

[126] View Article

[127] Google Scholar

[ref44] 44. Khoshnevisan B.; Rafiee S.; Omid M.; Mousazadeh H. Applying data envelopment analysis approach to improve energy efficiency and reduce GHG (greenhouse gas) emission of wheat production. Energy 2013, 58, 588–593.
View Article
Google Scholar

[129] View Article

[130] Google Scholar

[ref45] 45. Perez K.; González-Araya M.C.; Iriarte A. Energy and GHG emission efficiency in the Chilean manufacturing industry: Sectoral and regional analysis by DEA and Malmquist indexes. Energy Econ. 2017, 66, 290–302.
View Article
Google Scholar

[132] View Article

[133] Google Scholar

[ref46] 46. Dariush A. Ranking all DEA-Efficient DMUs based on cross efficiency and analytic hierarchy process Methods. Journal of Optimization. 2015, 594727, 1–10.
View Article
Google Scholar

[135] View Article

[136] Google Scholar

[ref47] 47. Isidoro G.; Carmelo R. Productivity and efficiency change and shareholder value: evidence from the Spanish banking sector. Applied Economics. 2008, 40(15), 2037–2044.
View Article
Google Scholar

[138] View Article

[139] Google Scholar

[ref48] 48. Caves D.W.; Christensen L.R.; Diewert W.E. Multilateral Comparisons of Output, Input, and Productivity Using Superlative Index Numbers. Econ. J. 1982, 92, 73–86.
View Article
Google Scholar

[141] View Article

[142] Google Scholar

[ref49] 49. Färe R.; Grosskopf S.; Norris M.; Zhang Z.Y. Productivity growth, technical progress, and efficiency change in industrialized countries. Am. Econ. Rev. 1994, 84, 66–83.
View Article
Google Scholar

[144] View Article

[145] Google Scholar

[ref50] 50. Fisher I. The Making of Index Numbers: A Study of Their Varieties, Tests, and Reliability. Houghton, muffin Co. Boston and New York, 1922, pp.526.

[ref51] 51. West T.O.; Marland G. A synthesis of carbon sequestration, carbon emissions, and net carbon flux in agriculture: Comparing tillage practices in the United States. Agric. Ecosyst. Environ. 2002, 91, 217–232.
View Article
Google Scholar

[148] View Article

[149] Google Scholar

[ref52] 52. Tian Y.; Zhang J.B.; Li B. Research on Spatial-Temporal Characteristics and Factor Decomposition of Agricultural Carbon Emission Based on Input Angle-Taking Hubei Province for Example. Res. Agric. Mod. 2011, 32, 752–755.
View Article
Google Scholar

[151] View Article

[152] Google Scholar

[ref53] 53. Intergovernmental Panel on Climate Change (IPCC). Climate Change 2007: The Fourth Assessment Report of The Intergovernmental Panel on Climate Change; Intergovernmental Panel on Climate Change: Geneva, Switzerland, 2007.

[ref54] 54. Wu F.L.; Li L.; Zhang H.L.; Chen H. Net Carbon Emissions of Farmland Ecosystem Influenced by Conservation Tillage. Chin. J. Ecol. 2007, 26, 2035–2039.
View Article
Google Scholar

[155] View Article

[156] Google Scholar

[ref55] 55. Wang X.K.; Li C.S.; Ou Y.; Zhi Y. Greenhouse gas emissions and Chinese grain production. Ecol. Environ. 2003, 12, 379–383.
View Article
Google Scholar

[158] View Article

[159] Google Scholar

[ref56] 56. Seiford L.M.; Zhu J. Modeling Undesirable Factors in Efficiency Evaluation. Eur. J. Oper. Res. 2002, 142, 16–20.
View Article
Google Scholar

[161] View Article

[162] Google Scholar

[ref57] 57. Chinese Rural Statistical Yearbook. China Statistical Publishing House, Beijing 2007–2014.

[ref58] 58. Annals of China’s Environmental Statistics. China Statistical Publishing House, Beijing 200–2014.

[ref59] 59. National Bureau of statistics of China: http://www.stats.gov.cn/

[ref60] 60. The Chinese economic database: https://www.ceicdata.com/zh-hans/products/china-economic-database.

[ref61] 61. Tian Y.; Zhang J.B.; He Y.Y. Research on Spatial-Temporal Characteristics and Driving Factor of Agricultural Carbon Emissions in China. Journal of Integrative Agriculture. 2014, 6(13), 1393–1403.
View Article
Google Scholar

[168] View Article

[169] Google Scholar

[ref62] 62. Liu X.Y.; Zhang S.; Bae J. The impact of renewable energy and agriculture on carbon dioxide emissions: Investigating the environmental Kuznets curve in four selected ASEAN countries. Journal of Cleaner Production. 2017, 164, 1239–1247.
View Article
Google Scholar

[171] View Article

[172] Google Scholar

Figures

Abstract

Introduction

Methods and data source

Models

Data envelopment analysis model.

Malmquist index model.

Data and variables

Agricultural carbon emissions from carbon sources.

Selection of input and output indicators.

Data sources.

Results

Measurement of agricultural carbon emission efficiency based on BCC model

Comprehensive technical efficiency of agricultural carbon emissions.

Pure technical efficiency and economies of scale of agricultural carbon emission.

Increase/decrease in economies of scale of agricultural carbon emission.

Regional comparison of agricultural carbon emissions.

Dynamic evolution analysis of agricultural carbon emission efficiency based on the Malmquist index model

Temporal dynamic evolution of agricultural carbon emission efficiency.

Spatial dynamic evolution of agricultural carbon emission efficiency.

Discussions and conclusions

References