Vulnerability Assessment and Optimization Countermeasures of the Human–Land Coupling System of the China–Mongolia–Russia Cross-Border Transportation Corridor

Abstract

In recent years, the conflicts of the human–land coupling system (HLS) in the cross-border transportation corridor areas have become increasingly severe, especially in the China–Mongolia–Russia Cross-Border Transportation Corridor (CMRTC). The vulnerability assessment of the HLS-CMRTC is the key scientific issue for regional sustainable development. Based on the nearly 20 years of a scientific expedition, we set the CMRTC as the study area, constructed the vulnerability assessment index system and quantitative model, assessed the vulnerability of the HLS-CMRTC, revealed the key influencing factors, divided vulnerability risk prevention zones, and proposed the targeted optimization countermeasures. This study found that: (1) The overall vulnerability pattern of the HLS-CMRTC showed a vulnerability level gradually increasing from south to north. (2) Permafrost instability risk, land desertification, temperature increase, and backward social development were key influences. (3) Vulnerability risk prevention zones were divided into four priority and two general zones. The targeted optimization countermeasures were proposed, such as establishing an ecological security barrier, carrying out collaborative ecological risk monitoring, and early warning. The conclusions could provide a decision-making basis for the study area to reduce the vulnerability of the HLS. They could also provide reference and scientific support for achieving sustainable development of the economy and environment in similar regions of the world.

1. Introduction

The human–land coupling system (HLS) is a complex system formed by complex interactions and feedback between human activities and the natural environment [1]. It is highly vulnerable to combined effects such as human activities, climate change, and ecological degradation. The vulnerability of the HLS could be characterized by the degree to which it suffers from these adverse effects and the generated ecological risks. Carrying out a scientific assessment of the vulnerability of the HLS and proposing targeted optimization countermeasures are important foundations for the region to achieve sustainable ecological and economic development [2,3].

In recent years, academics have become increasingly aware of the complexity and vulnerability of the HLS. The theory of vulnerability and its assessment of the HLS is increasingly becoming a popular scientific topic of research in geography and related subjects [4,5]. In terms of the vulnerability assessment model, the Vulnerability Scoping Diagram (VSD) model is a proven principle for constructing the assessment index system of the HLS in three dimensions: exposure, sensitivity, and adaptability [6]. In our research, we also used the VSD model. The term “exposure” measures the degree of outside interference and influence on the HLS, “sensitivity” measures the degree of reaction to outside interference and influence, and “adaptability” measures the ability to recover from outside interference and influence [7]. It is generally accepted that the VSD model has high compatibility and a clear assessment process and has critical application value [3]. In terms of study areas, some researchers chose different spatial scales for vulnerability assessment. Chunwei Song et al. analyzed vulnerability patterns and their drivers in Northern China over the last four decades [8]. Dong Li et al. focused on analyzing the future trends of vulnerability in Liaoning Province, China [9]. Jun, L. et al., Bo, Tang et al. and Gu, Hanlong et al. chose Zhangye City, Guangzhou City, and Shenyang City, China, as study areas, respectively [10,11,12]. And Liu, C et al. focused on the socioecological economic system at the county-level in Chongqing, China [13]. Some researchers have also chosen typical climatic and ecological zones as study areas. For example, Sietz, D et al. classified global drylands into seven typical vulnerability patterns using some indicators such as poverty, water stress, and soil degradation [14]. Zhang, Q et al. quantified the impacts of climate change on vulnerability in the Tibetan Plateau region of China [15]. Yue Chen et al. compared the similarities and differences in the vulnerability of karst and non-karst areas [16]. In addition, some researchers have chosen tropical ecosystems [17], aquatic environmental systems [18], and cropland systems [19] to examine the vulnerability patterns. However, few current studies have selected cross-border regions as the study area, which will become the focus area for research in the future.

Modern development of globalization and acceleration of economic integration are becoming rapid. In order to achieve cooperation and improvement, economic and trade activities between countries have increased significantly, and many cross-border regions are gradually breaking the national boundary barriers and forming a complete HLS. Cross-border transportation routes and other infrastructures are important links for economic integration development between different countries. So, the areas along the cross-border transportation route have become typical areas for a complete cross-border HLS [20]. However, the construction of transportation routes, rapid economic expansion, and frequent migration of people damage the ecological environment profoundly [21,22], and the vulnerability of the HLS to cross-border regions is becoming increasingly acute [23]. It is quite urgent to carry out scientific prevention of ecological risks and propose countermeasures to ensure sustainable ecological and economic development in cross-border regions. However, there is a lack of a vulnerability assessment of the HLS in cross-border regions, leading to a lack of scientific and technological support for scientific decision-making on the prevention of vulnerability risks and failure to meet the urgent needs for sustainable development of cross-border regions.

The China–Mongolia–Russia Cross-Border Transportation Corridor (CMRTC) spans three countries and forms a typical cross-border HLS. This region is the most important construction area for the economic exchanges between China, Mongolia, and Russia. In this area, the physical geography is complex and changeful, the ecological environment is fragile and sensitive, and the ecological environmental risks, such as desertification, permafrost changes, and air pollution, are serious. The conflict between rapid economic development, frequent human activities, and the sensitive ecological environment has become increasingly serious in the CMRTC recently. On the one hand, land desertification and permafrost changes not only bring risks and challenges to the construction and stable operation of regional transportation infrastructure [24,25] but also cause a decline in regional land productivity [26], affecting the development of regional agriculture and animal husbandry. On the other hand, the fast growth of regional transportation infrastructure, industry, animal husbandry, and urbanization also put pressure on the stability and sustainability of the ecological environment [21,27], such as overgrazing, needless urban expansion, land desertification, and industrial construction destroying permafrost stability. The CMRTC is an important bearing area for the economic and trade exchanges between China, Mongolia, and Russia which is vital for the three countries. It is urgent to assess the vulnerability of the HLS in this region and propose targeted optimization and control countermeasures to achieve sustainable regional economic and ecological development of the CMRTC.

In this study, based on over 20 years of a scientific expedition and research accumulated by the research team, we developed a vulnerability assessment method for the HLS-CMRTC. We constructed a vulnerability assessment index system and a quantitative model to access the vulnerability of the HLS-CMRTC, which includes the joint entropy weighting method, overlay analysis, and the natural breaks method. Based on the results of vulnerability assessment, we revealed the key influencing factors, divided vulnerability risk prevention zones, and proposed the targeted optimization countermeasures. The conclusions could provide a decision-making basis for the case study area to reduce the vulnerability of the HLS. It also could provide reference and scientific support for achieving sustainable development of the economy and environment in similar regions of the world.

This paper is organized as follows: Section 1 is the introduction, Section 2 introduces the general situation of the study area, Section 3 outlines the methodological model, Section 4 and Section 5 provide the results analysis and optimization countermeasures, and Section 6 gives a summary of conclusions.

2. Study Area

The CMRTC spans China, Mongolia, and Russia, located at 39°34′ N~58°12′ N, 95°40′ E~119°55′ E, with a total size of 976,000 km² and a total length of 1230 km of cross-border railroads. It contains the Chinese cities of Huhhot, Ulanqab, and Xilinguol League; the Mongolian provinces of Dornogovi Aimag, Dundgovi Aimag, Govisumber Aimag, Tov Aimag, Ulaanbaatar, Selenge Aimag, and Darhan-Uul Aimag; as well as the Russian republic of the Republic of Buryatia, Russia (Figure 1).

Recently, the conflict between humans and land in the CMRTC has become increasingly serious. The Baikal Lake basin in the Northeastern Republic of Buryatia experienced frequent forest fires due to high temperatures and drought [28], threatening local people’s survival. There are noticeable permafrost changes in the junction of the Republic of Buryatia and northern parts of Mongolia because of low temperatures and humid environments, which risk the stable operation of traffic facilities and transport pipelines. The climate in the southern parts of Mongolia, such as Dornogovi Aimag and Dundgovi Aimag, is warming and drying, and this trend puts the area at risk of dust storms, which has an adverse effect on the everyday life of residents. Although Huhhot, Ulanqab, and Xilinguol League in China have achieved remarkable results in ecological protection in recent years, the risk of land desertification in the areas along the railroads still should be taken seriously.

3. Methods

Based on climate data, remote sensing data, soil data, protected areas data, and expedition data, we constructed a vulnerability assessment index system, which contains 18 indicators. The entropy weighting method was used to calculate the weights of different indicators. The overlay analysis and natural breaks method were used to derive and classify the results of the vulnerability. Above all data and methods are shown in the research framework and technical methodology (Figure 2).

3.1. Data Sources

This study mainly includes meteorological data, remote sensing data, soil data, protected areas data, and expedition data obtained from the scientific expedition carried out by the research team.

The meteorological raster data of temperature and precipitation were obtained from the Climatic Research Unit (CRU) at the University of East Anglia (https://crudata.uea.ac.uk/cru/data/hrg/ (accessed on 6 July 2023)) and the resolution is 0.5°. The elevation, slope, and NDVI raster data were obtained from the National Aeronautics and Space Administration (NASA, https://ladsweb.modaps.eosdis.nasa.gov/ (accessed on 6 July 2023)), the resolution of elevation and slope data is 90 m, and that of the NDVI data is 1000 m. The vector data of the protected areas were obtained from International Union for Conservation of Nature (IUCN, https://www.iucn.org/about-iucn (accessed on 6 July 2023)). The soil organic carbon content raster data were obtained from the Harmonized World Soil Database (HWSD, https://iiasa.ac.at/models-tools-data/hwsd (accessed on 6 July 2023)), and the resolution is 1000 m. The land cover raster data were obtained from the European Space Agency (ESA, https://cds.climate.copernicus.eu/cdsapp#!/dataset/satellite-land-cover?tab=overview (accessed on 6 July 2023)), and the resolution is 300 m. Socioeconomic data and distribution of permafrost were obtained from the scientific expedition carried out by the research team.

3.2. Construction of Vulnerability Assessment Index System

Taking full account of the natural background and the mechanism of the interaction between human activities and the ecological environment in the CMRTC, we built a vulnerability assessment index system of the HLS based on the VSD model. This system was made up of “Target Layer-Factor Layer-Indicator Layer.” This index system assessed the vulnerability of the study area from exposure, sensitivity, and adaptability [6].

The term “exposure” measures the degree of outside interference and influence on the HLS. The Mongolian Plateau is a region with a large proportion of animal husbandry, and the rapid development of the animal husbandry has put pressure on the local grassland ecosystems [29]. And in recent years, due to low-temperature and humid environments, significant degradation of permafrost occurred in Northern Mongolia and the border area between Mongolia and Russia [30]. So, in the CMRTC, the sources of exposure are mainly natural disasters such as land desertification and permafrost instability, as well as pressures on the HLS from human activities such as animal husbandry development and urban expansion [27,29,31]. So, we chose land cover change and permafrost instability risk levels to characterize external ecological risks and chose the change in the proportion of the primary industry, the number of livestock, and the urbanization rate to characterize the development of the regional primary industry, the animal husbandry, and the urbanization, respectively.

The term “sensitivity” measures the degree of reaction to outside interference and influence. The sensitivity of the HLS can be reflected by factors such as topographic features, climatic situations, and soil condition of the study area [32,33,34]. So, elevation and slope were chosen to characterize topographic and geomorphic features. Changes in temperature and precipitation were used to characterize climate change. Normalized Difference Vegetation Index (NDVI) is an index that characterizes the degree of vegetation cover of the study area. Soil organic carbon content is an indicator that characterizes the quality of the soil, and sensitivity increases with decreasing soil quality.

The term “adaptability” measures the ability to recover from outside interference and influence. Good human settlements and high levels of economic development, education, and medical services can enhance the regional recovery ability. The railway and highway density were used to reflect the level of improvement of regional infrastructure. The number of ecological protection policy responses reflect the degree of government intervention in ecological risks. GDP per capita indicates the level of economic development of the region. Education level and medical service level were indicated by the number of full-time students in general education schools and the number of doctors per 10,000 people, respectively.

The vulnerability assessment index system of the HLS-CMRCT was established as in

Table 1. The vulnerability assessment index system of the human–land coupling system in the China–Mongolia–Russia Cross-Border Transportation Corridor (HLS-CMRTC).

Target Layer	Factor Layer	Indicator Layer	Positive(+)/Negative(-)
Vulnerability of Human–Land Coupling System	Exposure Index	X1: Land cover change	+
		X2: Permafrost instability risk	+
		X3: Primary industry development	+
		X4: Animal husbandry development	+
		X5: Urbanization development	+
	Sensitivity Index	X6: Elevation	+
		X7: Slope	+
		X8: Distribution of protected areas	-
		X9: Temperature increase	+
		X10: Precipitation change	-
		X11: NDVI change	-
		X12: Soil organic carbon content	-
	Adaptability Index	X13: Railway density	-
		X14: Highway density	-
		X15: Ecological protection policy	-
		X16: GDP per capita	-
		X17: Education level	-
		X18: Medical service level	-

. The “+” type of relationship in the table indicates that the vulnerability rises as the indicator rises, while “-” indicates that vulnerability rises as the indicator decreases.

The system has a total of 18 indicators. Fifteen of them are numerical data that can be standardized and overlaid directly using the raw data. The other three indicators are categorical data that need to be reclassified and assigned values, which are interpreted as follows.

Land cover change: The land cover change refers to the transformation of land cover from other land cover types to bare land. Accordingly, the most notable trend of desertification occurs when other land cover types are directly transformed to bare land and the ecosystem “forest-shrubland-sparse vegetation-grassland” transformed in order is lower.

Permafrost instability risk: There are four different permafrost forms in the research area: continuous, discontinuous, sporadic patches, and no permafrost. There were varying degrees of permafrost-type changes, and the severity of permafrost instability was divided into three categories: high, medium, and low.

Distribution of protected areas: The distance to the protected areas was chosen to characterize the degree of influence of the distribution of protected areas on sensitivity. The closer the distance from the protected areas, the higher the sensitivity level. A five-level 100 m buffer zone was constructed based on the distribution of protected areas in the study area, and the distance to protected areas was split into five levels from close to far.

3.3. Quantitative Vulnerability Assessment Model

We used the entropy weighting method to assign weights to the indicators, applied ArcGIS 10.2 to carry out overlay analysis, and adopted the natural breaks method to classify the results.

(1)

Entropy weighting method

The entropy method is a mature method of determining weights which was first introduced by Shannon [35]. The entropy weighting method is a mathematical calculation of the weights, so the calculation results are more objective and realistic. This is due to the objectivity of the method in that it has a wide scope of application and high believability. When the value of a particular indicator fluctuates significantly, the entropy value of the indicator is lower, which indicates that the indicator has more information. The following outlines the precise computation process.

First, we used the polar difference method to standardize the data of 18 indicators since the scales of each indicator vary. The precise procedure is as follows:

For a positive indicator i:

$$X_i^{\prime }=\frac{X_i-\min \left\{x_i\right\}}{\max \left\{x_i\right\}-\min \left\{x_i\right\}}$$

(1)

For a negative indicator j:

$$X_j^{\prime }=\frac{\max \left\{x_j\right\}-X_j}{\max \left\{x_j\right\}-\min \left\{x_j\right\}}$$

(2)

where $X_i^{\prime }$ and $X_j^{\prime }$ are the standardized values of the metrics, $X_i$ and $X_j$ are the original values of the metrics, $\max \left\{x_i\right\}$ and $\max \left\{x_j\right\}$ are the maximum values in the metrics, and $\min \left\{x_i\right\}$ and $\min \left\{x_j\right\}$ are the minimum values in the index. $X_i^{\prime }$ and $X_j^{\prime }\mathsf{ϵ}[0,1]$.

Second, the information entropy of indicator j was calculated:

$$E_j=-\ln {\left(n\right)}^{-1}{\sum }_{i=1}^n{p}_{ij}ln{p}_{ij}$$

(3)

where:

$$p_{ij}=\frac{X_{ij}^{\prime }}{\sum _{i=1}^n{X}_{ij}^{\prime }}$$

(4)

where $X_{ij}^{\prime }$ is the standardized value of the $i$th term of indicator $j$ and $n$ is the number of elements corresponding to indicator $j$.

Third, indicator weights were determined:

$$W_j=\frac{1-E_j}{k-\sum _{j=1}^k{E}_j}\hspace{1em}k=18$$

(5)

where $k$ is the number of indicators in the vulnerability assessment index system. In our study, $k$ is 18.

(2)

Quantitative vulnerability assessment method

The following formulas were used to quantitatively assess the exposure, sensitivity, adaptability, and vulnerability:

$$EI={\sum }_{i=1}^{n_1}{X}_i^{\prime }\times W_i$$

(6)

$$SI={\sum }_{j=1}^{n_2}{Y}_j^{\prime }\times W_j$$

(7)

$$ACI={\sum }_{k=1}^{n_3}{Z}_k^{\prime }\times W_k$$

(8)

$$VI=SI+EI+ACI$$

(9)

where $EI$ is the exposure assessment index of the study area, $n_1$ is the number of exposure indicators in the assessment index system, $X_i^{\prime }$ is the standardized value of the $i$th exposure indicator, and $W_i$ is the weight of the $i$th exposure indicator; $SI$ is the sensitivity assessment index of the study area, $n_2$ is the number of sensitivity indicators in the assessment index system, $Y_j^{\prime }$ is the standardized value of the $j$th sensitivity indicator, and $W_j$ is the weight of the $j$th sensitivity indicator; $ACI$ is the adaptation assessment index of the study area, $n_3$ is the number of adaptation indicators in the assessment index system, $Z_k^{\prime }$ is the standardized value of the $k$th adaptation indicator, and $W_k$ is the weight of the $k$th adaptation indicator. $VI$ is the study area integrated vulnerability index.

Based on the above methodology, this study utilized ArcGIS 10.2 (https://www.esri.com/zh-cn/arcgis/products/arcgis-desktop/overview (accessed on 6 July 2023)) for overlay analysis of indicators and classification of results. The Overlay Analysis tool in ArcGIS 10.2 can add the standardized values of different indicators in the same area to merge features from different indicators. This study then utilized the natural breaks s tool in ArcGIS 10.2 to categorize the sensitivity, exposure, adaptability, and vulnerability into three classes: low, medium, and high. The natural breaks method uses the mathematical method of spatial clustering to determine the breakpoints of different classes to realize optimal grading by calculating the variance of data in each class and comparing the sum of the variances [36].

4. Results

4.1. Weight of Indicators

We used the entropy weighting method to assign weights to the 18 indicators and rank them by the weights. The results of the ranking are shown in

Table 2. The weight of indicators in the vulnerability assessment index system of the HLS-CMRTC.

Indicator	X12	X6	X1	X15	X2	X11	X4	X7	X9	X18	X13	X16	X3	X17	X5	X8	X14	X10
Weight	0.17	0.14	0.11	0.1	0.08	0.08	0.06	0.05	0.04	0.04	0.02	0.02	0.02	0.02	0.02	0.02	0.01	0.01

. The weights of soil organic carbon content, elevation, land cover change, ecological protection policy, permafrost instability risk, NDVI change, animal husbandry development, and slope were higher. From this, it can be seen that the ecological environment more significantly affects the vulnerability level in the CMRTC than human activities.

4.2. Exposure Assessment

The exposure of the HLS-CMRTC exhibited a steady increase from south to north (Figure 3). The percentage of high, medium, and low exposure was 5.47%, 22.73%, and 71.80%, respectively. High-exposure regions were primarily found in the central-eastern region of Tov Aimag of Mongolia and the northeastern region of the Xilinguol League of China. Medium-exposure regions were primarily found in the majority of areas in the Republic of Buryatia, the city of Ulaanbaatar, and the western region of Dundgovi Aimag of Mongolia. The low-exposure regions were concentrated in Lake Baikal; Dundgovi Aimag and Dornogovi Aimag of Mongolia; and the southwestern areas of Huhhot, Ulanqab, and Xilinguol League in China.

The distribution of permafrost instability risk significantly impacted exposure level. The unstable area of permafrost change in the CMRTC, which indicated high risk of permafrost instability, was concentrated in the Republic of Buryatia, Tov Aimag of Mongolia, and the northeastern part of Xilinguol League, China (Figure 4a). The distribution of medium- and high-exposure areas in the CMRTC was highly consistent with the distribution of unstable of permafrost change. For example, in the Baikal Lake coastal region of the Republic of Buryatia, the freeze–thaw phenomenon of the transition from continuous permafrost to discontinuous, island, and sporadic patches occurred. The freeze–swelling phenomenon of the transition from the island and sporadic patches to continuous permafrost occurred in the central part of Tov Aimag of Mongolia and the northeastern region of the Xilinguol League, China. This result is similar to the research of Gao, H et al. [30]. Permafrost instability could lead to geological hazards, resulting in the destruction of the foundation of buildings and deformation of the roadbed, threating transportation [37]. The dramatic changes in permafrost cause higher ecological risks to the region and therefore higher exposure.

The central-eastern region of Tov Aimag and the northeastern region of the Xilinguol League were more exposed than the Republic of Buryatia. There were two main reasons. First, the number of livestock in Tov Aimag and Xilinguol League was higher and had grown significantly over the past 20 years (Figure 4b) by 2.865 and 2.864 million, respectively. It is commonly recognized that the increase in the number of livestock increased land degradation and desertification risk in Mongolia [38,39,40]. Second, Tov Aimag of Mongolia has a relatively large share of its primary industry at 65.6%, and it had only decreased by 3.3% over two decades (Figure 4c).

4.3. Sensitivity Assessment

The sensitivity of the HLS-CMRTC exhibited a steady increase from south to north (Figure 5). The percentage of high, medium, and low sensitivity was 10.23%, 47.73%, and 42.04%, respectively. High-sensitivity regions were primarily found in the northeast and northwest regions of the Republic of Buryatia, as well as the vicinity of Ulaanbaatar. Medium-sensitivity regions were primarily found in Baikal Lake basin in The Republic of Buryatia, Tov Aimag and Dundgovi Aimag in Mongolia, and the area close to Xilinguol League railroad line in Chaina. The low-exposure regions were concentrated in Lake Baikal, the border region between Selenge Aimag and the Republic of Buryatia in Mongolia, Dornogovi Aimag in Mongolia, and the majority of the three cities in China.

Elevation and NDVI change were the main influences on sensitivity level. The most sensitive parts of the Republic of Buryatia were generally located in wooded areas with undulating and variable elevations. In the northeast and northwest regions of the Republic of Buryatia, the difference between the highest and lowest elevation values can be up to 3000 m (Figure 6a). Additionally, the proportion of NDVI decrease reached 50% and 31%, respectively, in Dundgovi Aimag and Tov Aimag of Mongolia, which were medium-sensitivity areas. Areas along railroads had a significant NDVI drop over the last 20 years in the three cities in China (Figure 6b).

In particular, although temperature increase was not given much weight in the indicator system, the results of observations and field research showed that temperature increase was also one of the main influences on sensitivity level. The temperature in the Northeastern Republic of Buryatia, which was one of the highly sensitive areas, has greatly grown over the last 20 years; the temperature increase was 2.99 °C, and the largest growth was 4.08 °C (Figure 6c). Hampton, S. E. also determined that the water temperature of Lake Baikal continued to increase from 1940 to 2000, and warmer temperatures in the region could exacerbate the aridification of the Mongolian Plateau [41]. The research of Obyazov V A also proved this result [42].

4.4. Adaptability Assessment

The adaptability of the HLS-CMRTC exhibited a steady decline from south to north (Figure 7). The low-adaptability regions were primarily found in the Republic of Buryatia, Selenge Aimag, Darhan-Uul Aimag, and Govisumber Aimag in Mongolia. Medium-adaptability regions were primarily found in Ulaanbaatar City, Tov Aimag, Dundgovi Aimag, and Dornogovi Aimag. The high-adaptability regions were all found in Huhhot, Ulanqab, and Xilinguol League in China.

The degree of implementation of ecological protection policies was the main reason for adaptability level in the CMRTC. Compared to China, the implementation of ecological protection policies of the Republic of Buryatia and the seven provinces in Mongolia was woefully insufficient. There are 52 ecological protection policies at the national, provincial, and municipal levels in the three cities of China, compared to 10 in the Republic of Buryatia and 8 in Mongolia. Additionally, poor GDP per capita and low level of education and medical services were the main reasons for the low level of adaptability in Mongolian provinces and the Republic of Buryatia. The GDP per capita of the Republic of Buryatia and seven Mongolian provinces generally grew slowly. The GDP per capita of the Republic of Buryatia has even dropped by 14.4% (Figure 8a). In low-adaptability regions, the number of full-time students in general education schools, which reflected the level of education, was only expected to grow by an average of 2.9% from 2000 to 2020. Selenge Aimag even had a clear decline (Figure 8b). In Selenge Aimag, Darhan-Uul Aimag, and Govisumber Aimag of Mongolia, the number of doctors per 10,000 people, which reflected the level of medical services, increased by only 12.9% on average, which significantly lagged behind other areas (Figure 8c).

4.5. Vulnerability Assessment

Based on the assessment of exposure, sensitivity, and adaptability described above, the vulnerability level of the HLS-CRTC exhibited a general trend of gradually increasing from south to north (Figure 9). Exposure and sensitivity were the major influences on vulnerability. Permafrost instability risk, NDVI change and temperature increase, and backward social development were key influences on exposure, sensitivity, and adaptability, respectively. High-vulnerability regions accounted for 16.71% of the total, mainly concentrated in the Republic of Buryatia, Tov Aimag, and Ulaanbaatar. This figure was 50.68% in the region of medium vulnerability, which was mostly found in the Northern Republic of Buryatia; Dornogovi Aimag, and Dundgovi Aimag in Mongolia; and the Northeastern Xilinguol League in China. There were two main concentrated distribution regions of low vulnerability: Lake Baikal and most of Huhhot, Ulanqab, and Xilinguol League in China. The low vulnerability area made up 32.61% of the total area.

5. Vulnerability Risk Prevention Zones and Optimization Countermeasures for Different Risk Zones

Based on the result of quantitative assessment of vulnerability of the HLS-CMRTC, we divided vulnerability risk prevention zones into priority and general zones and proposed targeted optimization countermeasures.

5.1. Vulnerability Risk Prevention Zones

Vulnerability risk prevention zones were divided into priority and general zones based on vulnerability levels. And then vulnerability risk prevention types were divided into four types based on the main influencing factors of exposure, sensitivity, and adaptability, which were permafrost instability risk, land desertification, temperature increase, and backward social development (Figure 10).

Priority prevention zone of permafrost instability risk: The majority of the Republic of Buryatia in Russia, the central-eastern portion of Tov Aimag, the northern portion of Selenge Aimag, and the northeastern portion of the Xilinguol League in China. These regions had a great risk of permafrost instability. The degree of permafrost change was particularly strong in these regions, including frequent and strong freeze–thaw in the Republic of Buryatia and freeze–swell in Tov Aimag and Selenge Aimag in Mongolia and Xilinguol League in China.

Priority prevention zone of land desertification: The northeast and northwest regions of the Republic of Buryatia; Central Selenge Aimag, Ulaanbaatar City, southwest of Tov Aimag, and Dundgovi Aimag in Mongolia; and the Northern Xilinguol League in China, including its border region with Ulanqab City. The abovementioned areas had experienced a very significant decrease in vegetation cover in the last two decades.

Priority prevention zone of temperature increase: The northeastern region of the Republic of Buryatia, which had suffered notable temperature increases.

Priority prevention zone of backward social development: The Republic of Buryatia and the seven provinces of Mongolia need to improve the activity of environmental protection policies, economic development, medical services, and education level.

General prevention zone of permafrost instability risk: Northern Tov Aimag and Selenge Aimag of Mongolia. The main types of permafrost in these regions were discontinuous and sporadic patches, there has not been a distinct change in freeze–swell and freeze–thaw phenomena.

General prevention zone of land desertification: The southern regions of Dundgovi Aimag and Dornogovi Aimag in Mongolia; the borders of Mongolia and the northern parts of Xilinguol Meng and Ulanqab in China. The vegetation cover of the abovementioned areas was low and has exhibited a declining tendency over the last 20 years.

5.2. Optimization Countermeasures

The CMRTC spans the three countries of China, Mongolia, and Russia, which form a unique cross-border HLS. In order to prevent regional vulnerability risk, the three countries should coordinate and cooperate on the ecological preservation and build safe transportation corridors.

First, the three countries should establish a framework for the vulnerability risk prevention, exchange experience in ecological risk prevention, jointly establish an ecological security barrier, and carry out collaborative ecological risk monitoring and early warning. China should actively transfer the experiences of ecological management projects, green development models, and employing renewable energy to the Republic of Buryatia and Mongolia. In recent years, a number of eco-construction projects were initiated in response to the increasingly severe land degradation, such as returning farmland to forest, natural forest protection, Beijing–Tianjin sand control, and natural grassland restoration. These projects increased the vegetation cover distinctively and decreased the risk of land desertification. In Xilinguol League and Ulanqab, the establishment of the “Yili Dairy Industry Chain”, “Meat Sheep Green Cycle Industry Chain”, and “Beef Cattle Green Cycle Industry Chain” has achieved good social, economic, and ecological benefits [43]. China also had excellent experiences in developing wind energy. According to the National Energy Administration (NEA) of China, China provided 71.67 GW of power from wind in 2020. New energy represented by wind energy has grown to be a key contribution to China’s green development.

Second, the three countries should propose prevention countermeasures addressing their own vulnerability risk and engage in joint prevention efforts in the cross-border regions. The Republic of Buryatia is an area mainly developed for mining, forestry, and mechanical engineering. But in recent years, high-polluting sectors like electric power and non-ferrous metallurgy have undergone rapid development [44]. The Republic of Buryatia has a variety of renewable energy sources due to its abundant sunny hours and strong wind, which may be used to generate power in lieu of coals [44]. The Republic of Buryatia was also a sacred site for Russian Buddhists and offers unique natural scenery with Lake Baikal, which may support tourism growth. In Mongolia, most importantly, efforts to combat desertification should be made in places where it has already occurred, and cross-border desertification control cooperation should be expanded actively. Animal husbandry is one of the most important industries in Mongolia. Since the implementation of the private sector reform, the HLS in this country has become more vulnerable due to a great increase in the number of livestock. So, in highly vulnerable areas, overgrazing should be strictly prohibited. And the country should also adjust the livestock grazing practices, such as actively promoting the design of grazing bans, grazing rests, and zonal rotation, reasonably controlling the number of livestock. The Chinese region has a more stable HLS generally. The monitoring of possible ecological risk, particularly the desertification along train lines, should be taken seriously while retaining the present rules and regulations.

Finally, there should be targeted optimization countermeasures for different risk prevention zones.

Priority prevention zone of permafrost instability risk: First, in the border regions between the Republic of Buryatia and Mongolia, strengthening joint prevention corporation between the two countries is important. The Republic of Buryatia and Mongolia should enhance coordination in permafrost research and improve the awareness of the mechanism of permafrost monitoring and forecasting to enhance permafrost stability in the area. Second, it is advised to gradually construct long-term monitoring systems for permafrost development along significant pipelines and railroads. The specific measures are carrying out long-term observation of permafrost temperature and moisture change characteristics, land surface deformation, and engineering prevention measures to detect changes and promptly issue warnings. Third, the area of vegetation cover should be increased to amplify the impact of vegetation on soil temperature and humidity and then reduce permafrost damage. Fourth, measures such as using thermal insulation materials and geothermal energy utilization should be taken into account in the design and construction of buildings or other artificial facilities to reduce heat loss to the environment. Finally, it is also required to improve scientific research and expand researchers’ knowledge of permafrost formation, development, and damage mechanism.

Priority prevention zone of land desertification: First, for areas with severe desertification, the most important task is to prevent wind and fixed sands to establish an ecological security barrier. The main measures include covering the sand surface with dense material, planting sedge plants, and creating windbreaks on the edge of the desert. Second, the government should increase pasture regulation, strengthen the efforts of afforestation, and carry out strategies like sealing, zoning, rotational grazing, and controlling livestock numbers depending on the level of desertification to protect natural pastures. Third, soil and water conservation techniques should be enhanced to minimize soil erosion and make wise use of available water resources in terms of water extraction, water transmission, and water-saving irrigation. Fourth, in desertification areas, excessive urban expansion and human activities must be limited. Finally, social knowledge and accountability for environmental preservation should be increased.

Priority prevention zone of temperature increase: First, the monitoring and warning mechanism for forest fires must be improved. Second, it is necessary to enhance the ability to adapt to climate change and be able to take countermeasures, such as strengthening fire prevention, disaster prevention, and water resource management. Third, vegetation cover must be increased, and the ability of the region to sink and sequester carbon should be enhanced. Finally, the development of towns should be planned rationally to avoid population movement.

Priority prevention zone of backward social development: The Republic of Buryatia and Mongolia should make the most of their geopolitical advantages, actively promote ecological cooperation and experience sharing with China, learn environmental protection policies in accordance with local conditions, strengthen infrastructure construction of schools and hospitals, and enhance regional natural and social resilience to risks while achieving economic growth.

General prevention zone of land desertification and permafrost instability: It is important to monitor and predict potential permafrost instability risks. And it is required to actively plant well-adapted plants, strengthen soil and water conservation, and improve the afforestation projects in desertification areas.

6. Conclusions

This study constructed a vulnerability assessment index system and quantitative model, which were used to assess the vulnerability of the HLS-CMRTC, reveal the key influencing factors, divide vulnerability risk prevention zones, and propose the targeted optimization countermeasures.

The main conclusions were as follows:

(1)

The vulnerability level overall pattern of the HLS-CMRTC showed a gradual increase from south to north. The areas with high, medium, and low vulnerability levels accounted for 16.71%, 50.68%, and 32.62%. High-vulnerability regions were concentrated in Northern Mongolia and most regions of the Republic of Buryatia. Medium-vulnerability regions were primarily found in Southern Mongolia and the eastern part of the Republic of Buryatia, and a small area in the northeastern corner of China. Low vulnerability regions were concentrated in Lake Baikal and most regions of China.

(2)

Exposure and sensitivity were the major influences on vulnerability. Permafrost instability risk, NDVI change and temperature increase, and backward social development were key influences on exposure, sensitivity, and adaptability, respectively.

(3)

Based on the main influencing factors mentioned above and the level of exposure, sensitivity, and adaptability, vulnerability risk prevention zones were divided into four priority prevention zones (priority prevention zone of permafrost instability risk, priority prevention zone of land desertification, priority prevention zone of temperature increase, and priority prevention zone of backward social development) and two general prevention zones (general prevention zone of permafrost instability risk and general prevention zone of land desertification).

(4)

To prevent the vulnerability risk of the HLS-CMRTC, the three countries should coordinate and cooperate to establish a framework for the vulnerability risk prevention, and the three countries should also propose prevention countermeasures addressing their own vulnerability risk. There should also be targeted optimization countermeasures for different risk prevention zones.

In this study, the HLS in the cross-border region was taken as a whole system, and the vulnerability of three countries was comprehensively assessed, which extended the application of traditional vulnerability assessment methods. Furthermore, we divided vulnerability risk prevention zones into priority and general zones, and we proposed target optimization countermeasures. This work combined vulnerability assessment with regional sustainable development needs in the CMRTC so that our research results could provide a decision-making support for the case study area to reduce the vulnerability of the HLS.

There may be some uncertainties of this study’s results due to the inaccuracy of remote sensing data such as NDVI data and land cover data. For example, the NDVI data used could be influenced by clouds, solar altitude angle, and atmosphere [45,46]. The land cover data may also have some inaccuracy due to the uncertainty of satellite remote sensing data [47].

The vulnerability of the HLS is constantly changing, and the development trend prediction has important significance. Meanwhile, vulnerability is affected by various factors such as climate change and human activities. Therefore, different development scenarios need to be considered for prediction. To prevent the vulnerability risk more effectively, it is necessary to propose targeted countermeasures for various scenarios. Therefore, vulnerability prediction and scenario simulation will be our research priority in the future.