**1. Introduction**

The global increasing carbon emission under high carbon emission mechanisms has triggered a series of environmental problems, e.g., global climate anomalies, sea level rise, and frequent extreme weather events, which have greatly affected the production and living of human beings in recent decades [1,2]. Previous studies suggested that the land use system serves as a vital link between the human socio-economic system and the natural ecological environment, and the carbon emissions caused by land use change have been one of the important influencing factors of global warming [1–3]. In fact, various human activities, e.g., social construction, economic development, industrial arrangement, urban expansion, and energy consumption, are all closely related to the carbon emissions, all of which are ultimately implemented in different land use practices [4,5], and relevant land use change has been considered as the second most important influencing factor of the global increasing atmospheric CO<sup>2</sup> content [6,7]. The 14th Five-Year Plan of China proposed to achieve the "peak carbon dioxide emissions" by 2030 and "carbon neutrality" by 2060,

**Citation:** Yan, H.; Guo, X.; Zhao, S.; Yang, H. Variation of Net Carbon Emissions from Land Use Change in the Beijing-Tianjin-Hebei Region during 1990–2020. *Land* **2022**, *11*, 997. https://doi.org/10.3390/ land11070997

Academic Editors: Li Ma, Yingnan Zhang, Muye Gan and Zhengying Shan

Received: 1 June 2022 Accepted: 29 June 2022 Published: 30 June 2022

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

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

that the low-carbon economy should serve as a new economic growth mode in the future, and that new green energy sources should be developed to reduce the dependence of economic growth on the major fossil energy sources such as coal and oil [4]. It is, therefore, of practical significance to explore the regional carbon emissions from land use change to achieve low carbon land use, promote low carbon economic development, and establish a resource-conserving and environment-friendly society in this context [8,9].

Previous studies on carbon emissions from land use change at home and abroad were primarily concentrated on the spatiotemporal variation of different land use types and their relevant effects on carbon emissions, carbon emission accounting, influencing factors of land use change, and carbon emissions [10–13]. The Guidelines for the National Greenhouse Gas Inventories prepared by the Intergovernmental Panel on Climate Change (IPCC) provided a valuable methodological reference for accounting carbon emissions from land use change [13,14]. Besides, some scholars also proposed the emission coefficient method for the carbon emission accounting of cropland, forest land, grassland, and built-up area and explored the effects of land use change on carbon emissions [14–16]. In addition, most scholars have generally explored the influencing factors of carbon emissions with various econometric methods, such as the factor decomposition method [17,18], and some other scholars explored the drivers of carbon emissions with the Laspeyres decomposition method [19–21]. Moreover, more scholars carried out decomposition analyses based on the Logarithmic Mean Divisia Index (LMDI) model to reveal the influencing factors of carbon emissions in various relevant fields, e.g., carbon emissions per capita, carbon emissions due to industrial combustion energy, carbon emissions in the manufacturing industry, and drivers of carbon emissions from energy consumption for different time durations in China [22,23]. The results of these studies provided valuable methodological references for exploring a series of issues related to carbon emissions from land use change [24,25]. However, these existing studies focused more on some major land use types, such as cropland and built-up area, with less consideration of the carbon emissions from some other land use types [25,26]. In particular, there are relatively fewer quantitative studies on the influencing factors of carbon emissions from land use change, especially the studies on the influencing factors of carbon emissions from land use change with the LMDI model [6,27,28].

The Beijing-Tianjin-Hebei region, as one of the major urban agglomerations in China, accounted for 11–12% of the national total carbon emissions, which is higher than the national proportions of both gross domestic product (GDP) and the population of this region [5,12]. In particular, the goal of peak carbon dioxide emissions and carbon neutrality has put forward new and higher requirements for the synergistic development of this region [16,18,29]. This study estimated the net carbon emissions from land use change in the Beijing-Tianjin-Hebei region from the perspective of land use and decomposed the influencing factors of carbon emissions from land use change to quantitatively reveal the effects of these influencing factors on the carbon emissions from land use change, aiming to provide a firm scientific basis for improving the regional land use planning, promoting the low-carbon economic development, and guiding the development of the Beijing-Tianjin-Hebei region into the capital economy circle.

#### **2. Materials and Methods**

#### *2.1. Study Area*

The Beijing-Tianjin-Hebei region is located in the northern part of China (36◦5 0–42◦400 N, 113◦270–119◦500 E), which is one of the three major urban agglomerations in China [18,26,30]. There is very complex terrain in this region, where there are mainly higher mountains and plateaus in the northern and western parts and flatter plains in the southern and eastern parts (Figure 1). There is a warm-temperate continental monsoon climate in this region, with higher temperatures and precipitation in the summer. The Beijing-Tianjin-Hebei region is the political, economic, cultural, and scientific center of China and plays a strategically important role in the economic development of China [18]. The BeijingTianjin-Hebei region takes up approximately 2% of the total national land area, but it accounts for approximately 8.1% of the total national population and 9.44% of the total national GDP. However, there is very high energy consumption intensity along with rapid urbanization in the Beijing-Tianjin-Hebei region, leading to very high carbon emission. For example, the carbon emission in the Beijing-Tianjin-Hebei region reached 1.085 billion tons in 2018, accounting for about 1/9 of the total national carbon emissions of China [26,30]. Exploring the long time-series variation of net carbon emissions from land use change can provide valuable information for addressing the pressure of carbon emission reduction in the Beijing-Tianjin-Hebei region, especially in the context of the coordinated development of the Beijing-Tianjin-Hebei region [26,27]. jing-Tianjin-Hebei region takes up approximately 2% of the total national land area, but it accounts for approximately 8.1% of the total national population and 9.44% of the total national GDP. However, there is very high energy consumption intensity along with rapid urbanization in the Beijing-Tianjin-Hebei region, leading to very high carbon emission. For example, the carbon emission in the Beijing-Tianjin-Hebei region reached 1.085 billion tons in 2018, accounting for about 1/9 of the total national carbon emissions of China [26,30]. Exploring the long time-series variation of net carbon emissions from land use change can provide valuable information for addressing the pressure of carbon emission reduction in the Beijing-Tianjin-Hebei region, especially in the context of the coordinated development of the Beijing-Tianjin-Hebei region [26,27].

mountains and plateaus in the northern and western parts and flatter plains in the southern and eastern parts (Figure 1). There is a warm-temperate continental monsoon climate in this region, with higher temperatures and precipitation in the summer. The Beijing-Tianjin-Hebei region is the political, economic, cultural, and scientific center of China and plays a strategically important role in the economic development of China [18]. The Bei-

*Land* **2022**, *11*, x FOR PEER REVIEW 3 of 15

**Figure 1.** Location of the Beijing-Tianjin-Hebei region. **Figure 1.** Location of the Beijing-Tianjin-Hebei region.

#### *2.2. Data and Processing 2.2. Data and Processing*

The spatial data used in this study includes the 1-km land use data extracted from the Land Use Remote Sensing Monitoring Data of China in 1990, 1995, 2000, 2005, 2010, 2015, and 2020 (http://www.resdc.cn, accessed on 31 October 2021), which were reverted from Landsat TM/ETM images. The land types were classified into cropland, forest land, grassland, water area, built-up area, and barren land [12,26], based on which this study explored the land use transfer matrix of the Beijing-Tianjin-Hebei region during 1990– 2020. Besides, the non-spatial data used in this study mainly included the socioeconomic data (e.g., population and GDP), regional energy consumption data, and carbon emission coefficients of the Beijing-Tianjin-Hebei region, which were obtained from various issues of the China Statistical Yearbook, the China Energy Statistical Yearbook, IPCC reports, and the relevant literature. Finally, these data in different parts of the study area were combined to obtain the regional energy consumption amount and the carbon emission coefficients of different land use types, and these regional data were further summed up to obtain the relevant total data of the whole Beijing-Tianjin-Hebei region. The spatial data used in this study includes the 1-km land use data extracted from the Land Use Remote Sensing Monitoring Data of China in 1990, 1995, 2000, 2005, 2010, 2015, and 2020 (http://www.resdc.cn, accessed on 31 October 2021), which were reverted from Landsat TM/ETM images. The land types were classified into cropland, forest land, grassland, water area, built-up area, and barren land [12,26], based on which this study explored the land use transfer matrix of the Beijing-Tianjin-Hebei region during 1990–2020. Besides, the non-spatial data used in this study mainly included the socioeconomic data (e.g., population and GDP), regional energy consumption data, and carbon emission coefficients of the Beijing-Tianjin-Hebei region, which were obtained from various issues of the China Statistical Yearbook, the China Energy Statistical Yearbook, IPCC reports, and the relevant literature. Finally, these data in different parts of the study area were combined to obtain the regional energy consumption amount and the carbon emission coefficients of different land use types, and these regional data were further summed up to obtain the relevant total data of the whole Beijing-Tianjin-Hebei region.

#### *2.3. Carbon Emission Accounting Model and Carbon Emission Coefficients*

The carbon emissions can be categorized into direct and indirect carbon emissions [27]. The former refers to carbon emissions caused by the processes of maintenance and conversion and specific land types, while the latter refers to carbon emissions generated by the land serving as a carrier of production and living processes of human beings [28,29]. A

large number of studies have shown that some land use types may be both carbon sources and sinks, and the intensity of carbon sources and sinks generally varies greatly among different land use types [30,31]. This study primarily focused on the carbon emission effects of land use change caused by human activities, i.e., the amounting of carbon emissions and the sequestration of cropland, forest land, grassland, water body, built-up area, and barren land under the influence of human activities, and it finally summarized the carbon emission amount of different land use types. The major crops in the study area include wheat and maize, and these crops on cropland can absorb CO<sup>2</sup> in the air through photosynthesis in general, but most of the crop biomass is then decomposed in the soil and released back into the air in the short term, so there are generally insignificant effects of crop biomass as a carbon sink. Meanwhile, the effects of cropland inputs and soil emissions on the carbon emissions can also be reflected with the carbon emission coefficient of cropland [30]. By contrast, carbon emissions from energy consumption and industrial activities such as housing, mining, and manufacturing and transportation are the main sources of carbon emissions. Thus, the built-up area and cropland generally serve as the carbon sources, with positive carbon emission coefficients, while the forest land, grassland, water body, and barren land generally serve as carbon sinks, which are carbon absorbers with negative carbon emission coefficients. The regional carbon emission can be estimated based on the carbon emission coefficients according to the guidelines of IPCC as follows:

$$E\_{\mathbb{C}} = \sum e\_i = \sum A\_i \times \xi\_i \tag{1}$$

where *E<sup>c</sup>* is the total carbon emission (or absorption) amount, *e<sup>i</sup>* is the carbon emission (or absorption) amount from the *i*th land use type, *A<sup>i</sup>* is the area of the *i*th land use type, and *δi* is the carbon emission (or absorption) coefficient of the *i*th land use type.

The carbon emission (or absorption) coefficients of different land use types, which were assumed to keep stable during the study period, were determined as follows. The cropland that provides both carbon emission and carbon absorption serves as both a carbon source and carbon sink [29,30]. It is therefore necessary to take into account the greenhouse gas produced during the crop production and CO<sup>2</sup> absorption of crops during the reproductive period, and the difference between the two can be used to estimate the net carbon emission coefficient of the cropland [22,24]. Previous studies have shown that the carbon emission coefficient and carbon sequestration coefficient of cropland are approximately 0.422 t/hm<sup>2</sup> and 0.007 t/hm<sup>2</sup> , respectively [31,32], so this study took the difference between the two as the net carbon emission coefficient of cropland, i.e., 0.415 t/hm<sup>2</sup> . In addition, the forest land and grassland are the most important carbon sink and carbon sequestration systems in the terrestrial ecosystem, and previous studies have shown that the carbon emission coefficients of the forest land and grassland were <sup>−</sup>0.623 t/hm<sup>2</sup> and <sup>−</sup>0.144 t/hm<sup>2</sup> , respectively [33,34], which were also adopted in this study. In addition, previous studies showed that there is very limited carbon absorption of the water body and barren land, generally with very weak impacts on the regional net carbon emissions [27,32]. However, the water body and barren land accounted for 4% of the total area in the Beijing-Tianjin-Hebei region, so this study still considered the carbon emission coefficients of the water body and barren land, which were approximately <sup>−</sup>0.03 t/hm<sup>2</sup> and <sup>−</sup>0.05 t/hm<sup>2</sup> , respectively, according to the literature survey results [35–37]. Moreover, there are various types of built-up area, e.g., urban land, rural settlements, traffic roads, factories and mines, industrial areas, oil fields, salt fields, and quarries. The built-up area carries a large amount of the energy consumed in the production and living of human beings, and it is unfeasible to calculate the carbon emissions of the built-up area according to only the area share of built-up area [38,39]. It is necessary to estimate the carbon emissions from the built-up area based on the carbon emissions generated by the energy consumption of human beings on the built-up area [40,41], which can be estimated according to the carbon emission coefficient method of the IPCC as follows:

$$EC = \sum m\_i \times \mathcal{J}\_i \times \theta\_i \tag{2}$$

where *EC* is the carbon emission from energy consumption on the built-up area, *m<sup>i</sup>* is the consumption amount of various fossil energy sources, *β<sup>i</sup>* is the standard coal conversion coefficient of each energy resource, and *θ<sup>i</sup>* is the carbon emission coefficient of each energy resource. This study used the standard coal conversion coefficient and carbon emission coefficient of each energy resource published by the IPCC guidelines [32,42,43], which are shown in Table 1.

**Table 1.** Standard coal conversion coefficients and carbon emission coefficients of various energy sources.


#### *2.4. Decomposition Analysis of Influencing Factors of Carbon Emissions*

This study explored the influencing factors of carbon emissions with the LMDI model, which is one of the most widely used methods to explore the influencing factors of energy consumption in the field of low carbon economy due to its advantages such as high operability, full decomposition, no residuals, and unique results [36,37]. Specifically, this study analyzed the effects of different influencing factors on carbon emissions from land use change according to the Kaya identity by introducing the land use factor and establishing the formula of influencing factors of regional carbon emissions from five aspects, i.e., energy consumption structure, land output intensity, land use structure, economic growth, and population scale effect, as follows [41,42]:

$$\mathbf{C} = \sum \frac{\mathbf{C}\_i}{L\_i} \times \frac{L\_i}{L} \times \frac{L}{G} \times \frac{G}{P} \times P \tag{3}$$

where *C* is the total carbon emissions from land use change (million tons), *C<sup>i</sup>* is the carbon emission amount of the *i*th land use type (million tons), *L<sup>i</sup>* is the area of the *i*th land use type (km<sup>2</sup> ), *L* is the total land area of the study area (km<sup>2</sup> ), *G* is the GDP (10<sup>8</sup> CNY), and *P* is the regional population size (10<sup>4</sup> persons).

Then, the regional total carbon emissions can be expressed as follows [40,42]:

$$f\_{\mathbf{i}} = \frac{\mathbf{C}\_{\mathbf{i}}}{L\_{\mathbf{i}}};\ \mathbf{S}\_{\mathbf{i}} = \frac{L\_{\mathbf{i}}}{L};\ q = \frac{L}{\mathbf{G}};\ \mathbf{g} = \frac{\mathbf{G}}{\mathbf{P}}\mathbf{C} = \sum f\_{\mathbf{i}} \times \mathbf{S}\_{\mathbf{i}} \times q \times \mathbf{g} \times \mathbf{P} \tag{4}$$

where *f<sup>i</sup>* , *S<sup>i</sup>* , *q*, *g*, and *P* refer to the carbon emission intensity per unit of the *i*th land use type, the effect of the land use structure, land use intensity per unit of GDP, GDP per capita, and population size, respectively. According to this formula, the contribution value and contribution rate of each factor can be further analyzed with the LMDI model. Assuming the carbon emission in the base period and the *T*th period are *C* <sup>0</sup> and *C T* , respectively, then the carbon emission change during the study period (0–*T*) can be expressed as follows [40,43]:

$$\begin{split} \Delta \mathbf{C} &= \mathbf{C}^{T} - \mathbf{C}^{0} = \sum\_{i=1,2,...,6} f\_{i}^{T} \times s\_{i}^{T} \times q^{T} \times \mathbf{g}^{T} \times \mathbf{P}^{T} - \sum\_{i=1,2,...,6} f\_{i}^{0} \times s\_{i}^{0} \times q^{0} \times \mathbf{g}^{0} \times \mathbf{P}^{0} \\ &= \Delta \mathbf{C}\_{f\_{i}} + \Delta \mathbf{C}\_{s\_{i}} + \Delta \mathbf{C}\_{q} + \Delta \mathbf{C}\_{g} + \Delta \mathbf{C}\_{P} + \Delta \mathbf{C}\_{\text{rsd}} \end{split} \tag{5}$$

$$D = \frac{\mathbb{C}^{T}}{\mathbb{C}^{0}} = D\_{f} D\_{\text{s}} D\_{\text{g}} D\_{\text{g}} D\_{P} D\_{\text{rsd}} \tag{6}$$

where ∆*C* is the carbon emission change during the study period, ∆*C<sup>f</sup> i* , ∆*Cs<sup>i</sup>* , ∆*Cq*, ∆*Cg*, and ∆*C<sup>P</sup>* are the contribution values of *f<sup>i</sup>* , *S<sup>i</sup>* , *q*, *g*, and *P*, respectively, and ∆*Crsd* is the decomposition residual. If the obtained contribution value is >0, then the factor has a pulling effect on the carbon emissions during the study period; otherwise, the factor has a suppressing effect on carbon emissions. *D* is the carbon emission change percentage between the base period and the *T*th period; *D<sup>f</sup>* , *D<sup>s</sup>* , *Dq*, *Dg*, *Dp*, and *Drsd* are the contribution rates of *f<sup>i</sup>* , *S<sup>i</sup>* , *q*, *g*, and the residual error, respectively.

The following are the relationships in the additive decomposition mode according to the LMDI model [40,44,45]:

$$\begin{aligned} \Delta \mathbf{C}\_{fi} &= \sum\_{i} \frac{\mathbf{C}\_{i}^{T} - \mathbf{C}\_{i}^{0}}{\ln \mathbf{C}\_{i}^{T} - \ln \mathbf{C}\_{i}^{0}} \times \ln \frac{f\_{i}^{T}}{f\_{i}^{0}} \\ \Delta \mathbf{C}\_{s\_{i}} &= \sum\_{i} \frac{\mathbf{C}\_{i}^{T} - \mathbf{C}\_{i}^{0}}{\ln \mathbf{C}\_{i}^{T} - \ln \mathbf{C}\_{i}^{0}} \times \ln \frac{\mathbf{S}\_{i}^{T}}{\mathbf{S}\_{i}^{0}} \\ \Delta \mathbf{C}\_{q} &= \sum\_{i} \frac{\mathbf{C}\_{i}^{T} - \mathbf{C}\_{i}^{0}}{\ln \mathbf{C}\_{i}^{T} - \ln \mathbf{C}\_{i}^{0}} \times \ln \frac{q^{T}}{q^{0}} \\ \Delta \mathbf{C}\_{S} &= \sum\_{i} \frac{\mathbf{C}\_{i}^{T} - \mathbf{C}\_{i}^{0}}{\ln \mathbf{C}\_{i}^{T} - \ln \mathbf{C}\_{i}^{0}} \times \ln \frac{q^{T}}{8^{0}} \\ \Delta \mathbf{C}\_{P} &= \sum\_{i} \frac{\mathbf{C}\_{i}^{T} - \mathbf{C}\_{i}^{0}}{\ln \mathbf{C}\_{i}^{T} - \ln \mathbf{C}\_{i}^{0}} \times \ln \frac{p^{T}}{P^{0}} \end{aligned} \tag{7}$$

The following are the relationships in the multiplicative decomposition mode according to the LMDI model [40,45]:

$$\begin{array}{l} D\_f = \exp(W\Delta\mathbb{C}\_{f\_i}); D\_s = \exp(W\Delta\mathbb{C}\_{s\_i})\\ D\_q = \exp(W\Delta\mathbb{C}\_q); D\_\mathcal{S} = \exp(W\Delta\mathbb{C}\_{\mathcal{S}})\\ D\_P = \exp(W\Delta\mathbb{C}\_P); D\_{rsd} = 1\\ W = \frac{\ln D}{\Delta\mathbb{C}} \end{array} \tag{8}$$

#### **3. Results**

#### *3.1. Land Use Change in the Beijing-Tianjin-Hebei Region*

There was remarkable land use change in the Beijing-Tianjin-Hebei region during 1990–2020. There is mainly cropland and forest land in the Beijing-Tianjin-Hebei region, where the cropland is mainly distributed in the central and southeast parts of the study area, and the forest land as well as grassland are mainly distributed in the northeast and western parts (Figure 2). The built-up area is concentrated in the central part and the peripheral zone of the central towns in the study area. The water body is very limited in the study area, which mainly includes rivers near towns and lakes in the northwest. More specifically, cropland as the main land use type accounted for approximately 51.92% of the entire area in 1990. The forest land and grassland ranked second and third, accounting for about 20.65% and 16.49% of the entire area, respectively, while the built-up area, water body and barren land accounted for 10.94% of the entire area in total. There was a decreasing trend of cropland, forest land, grassland, and barren land from 2000–2015, while the built-up area increased significantly due to the accelerated urbanization process. In particular, the cropland decreased significantly from 2015–2020, while the built-up area continued to increase, and other land types only changed slightly.

Table 2 shows the land use transfer in the Beijing-Tianjin-Hebei region from 1990–2020. A total of 73,072 km<sup>2</sup> of land was transferred during 1990–2020, among which the transferout area of the cropland ranked first, accounting for about 91.2% of the total transfer-out area. The cropland was mainly transferred into built-up area, grassland, and forest land, with the converted areas of which reaching 6726 km<sup>2</sup> and 4705 km<sup>2</sup> and 17,621 km<sup>2</sup> , respectively. Meanwhile the transfer-in area of cropland reached 19,880 km<sup>2</sup> , which was mainly converted from grassland and built-up area, accounting for about 69.4% of the total transfer-in area of cropland. The transfer-in area of built-up area increased most significantly during the study area, reaching 20,837 km<sup>2</sup> , 84.6% of which was transferred from the cropland. By contrast, watershed and barren land only changed slightly, most

*Land* **2022**, *11*, x FOR PEER REVIEW 7 of 15

ln

*C*

*<sup>D</sup> <sup>W</sup>*

*3.1. Land Use Change in the Beijing-Tianjin-Hebei Region* 

**3. Results** 

= Δ

of which was converted to cropland, accounting for 49% and 51% of their transfer-out area, respectively. ticular, the cropland decreased significantly from 2015–2020, while the built-up area continued to increase, and other land types only changed slightly.

exp( ); exp( )

(8)

exp( ); exp( )

*<sup>i</sup> <sup>i</sup> f s*

*q g q g*

There was remarkable land use change in the Beijing-Tianjin-Hebei region during 1990–2020. There is mainly cropland and forest land in the Beijing-Tianjin-Hebei region, where the cropland is mainly distributed in the central and southeast parts of the study area, and the forest land as well as grassland are mainly distributed in the northeast and western parts (Figure 2). The built-up area is concentrated in the central part and the peripheral zone of the central towns in the study area. The water body is very limited in the study area, which mainly includes rivers near towns and lakes in the northwest. More specifically, cropland as the main land use type accounted for approximately 51.92% of the entire area in 1990. The forest land and grassland ranked second and third, accounting for about 20.65% and 16.49% of the entire area, respectively, while the built-up area, water body and barren land accounted for 10.94% of the entire area in total. There was a decreasing trend of cropland, forest land, grassland, and barren land from 2000–2015, while the built-up area increased significantly due to the accelerated urbanization process. In par-

=Δ =Δ

*D D W W C C*

*D D W W C C f s*

=Δ =Δ

exp( ); 1

=Δ =

*P rsd P*

*D D W C*

**Figure 2.** Spatial pattern of land use in the Beijing-Tianjin-Hebei region from 1990–2020. **Figure 2.** Spatial pattern of land use in the Beijing-Tianjin-Hebei region from 1990–2020.


**Table 2.** Land use transfer matrix in the Beijing-Tianjin-Hebei region during 1990–2020 (km<sup>2</sup> ).
