Spatial and temporal assessment of sustainable development indicators for the China-Pakistan transportation corridor

ABSTRACT

The China-Pakistan Transportation Corridor (CPTC) has long faced great challenges in natural disasters and sustainable development. This study develops a localised indicator evaluation system for the CPTC sustainable development. It contains 27 Tier I and 8 Tier II indicators, covering 11 sustainable development goals. Based on this evaluation system, a single-goal, multidimensional, and comprehensive evaluation of the CPTC sustainable development level is conducted for 2015, 2017 and 2020. The results show that, in terms of single goals, Xinjiang has the highest sustainable development level. Khyber Pakhtunkhwa has the poorest level of sustainable development. The Islamabad Capital Territory is better developed for all goals, especially in SDG5, SDG8 and SDG9. Besides these three goals, SDG6, and SDG11 are more prominent in Punjab, whereas SDG2 and SDG3 performed worst. From a multidimensional evaluation, the CPTC sustainable development level has been steadily increasing. However, the social and environmental dimensions of Khyber Pakhtunkhwa have experienced a ‘regression’. The comprehensive evaluation results show that the level of sustainable development in all provinces of the CPTC increase, except for Khyber Pakhtunkhwa. This study identifies CPTC sustainability issues and proposes targeted recommendations for sustainable development.

KEYWORDS:

• Sustainability indicator evaluation system
• sustainability evaluation
• China-Pakistan transportation corridor
• localisation

1. Introduction

Since the twenty-first century, with rapid industrialisation and the expansion of the scope of human activities(Yang et al. 2022; Yu and Luo 2019), the economy has developed rapidly, which has led to an imbalance between society, economy and environment. To address this issue, in September 2015, the United Nations released ‘Transforming Our World: the 2030 Agenda for Sustainable Development’ based on the Millennium Development Plan. The agenda proposed 17 Sustainable Development Goals (SDGs) with 169 targets and 231 indicators (Ashizawa, Sudo, and Yamamoto 2022). The UN Global Indicators Framework for Sustainable Development (GIFSD) offers a framework for evaluating sustainable development (Diaz-Sarachaga, Jato-Espino, and Castro-Fresno 2018; United Nations 2015). Many countries monitor changes in national SDGs through voluntary national review reports submitted to the UN Political Forum on Sustainable Development (Allen et al. 2019). However, issues such as inadequate evaluation systems and lack of data on evaluation indicators, seriously affect evaluations across different regions (Guo et al. 2021, 2022).

The China-Pakistan Transportation Corridor (CPTC), situated at the confluence of the Himalayas, Karakorum, and the Hindu Kush Mountains in the high Asian region, faces significant challenges in green development. These challenges stem from the region's underdeveloped economy, complex climate (Shelton and Dixon 2023), and frequent natural disasters (Ali et al. 2019; Chang et al. 2021). With the further implementation of the Belt and Road Initiative (BRI) in recent years, what changes have occurred in the CPTC regarding the economy, environment, agriculture, urbanisation (Jiang et al. 2022), disaster prevention, and mitigation? How well does the CPTC implement SDGs? Will the 17 SDGs be achieved as expected? These issues are urgent concerns worldwide.

To address the challenges of regional evaluation, the current focus of SDGs evaluation is transitioning from top-level theoretical research to specific quantitative research that presents a variety of evaluation methods. Several studies have started from the single goal of SDGs (Boeren 2019; Campbell et al. 2018; Ferguson and Roofe 2020; Sadoff, Borgomeo, and Uhlenbrook 2020) and evaluated them from different dimensions and spatial scales. However, since the achievement of SDGs depends on natural and social systems characterised by complexity, diversity, dynamics, and interconnection (Wu et al. 2022), the adoption of an approach based on single-goal evaluation cuts off their holistic and systematic nature, lacks consideration of the relationships among indicators, and fails to form a comprehensive understanding of regional sustainable development level. Comprehensive evaluation is currently a hot topic in the evaluation of sustainable development. Allen et al. (2020) combined GIFSD to form a new dashboard evaluation method for comprehensively evaluating the progress of 144 indicators corresponding to 86 development sub-goals in Australia. McArthur and Rasmussen (2019) proposed a normalisation method to diagnose the inherent diversity of indicators and evaluate the human impact of failing to meet the SDGs.

However, there are some significant problems with the current evaluation of sustainable development. First, special regions were not evaluated. Most of the current SDGs have been assessed at both national and global levels, and there is a lack of evaluation studies for extreme environmental and transboundary regions such as the CPTC. Because these areas span different countries, it is more difficult to evaluate sustainability using uniform indicators. Second, there is a lack of research that aims to capture the spatial and temporal dynamics of the SDG indicators (Giupponi and Gain 2017). Sustainable development evaluations considering different spatial (Kussul et al. 2020) and temporal dimensions can provide a more comprehensive understanding of the sustainable development process in the study area and better decision support for local development.

To address the aforementioned issues and further understand the sustainable development process of the CPTC since the initiation of the SDGs, this study uses data from 2015 (the start year of SDG implementation), 2020 (the current status with ready data support and the end year for some indicators such as target 3.6: By 2020, halve the number of global deaths and injuries from road traffic accidents), and 2017 (the middle year between 2015 and 2020 to gain a better grasp on SDG implementation trends). The CPTC is selected as the research area. We apply statistical and spatial data to formulate a localised sustainable development indicator system for the CPTC. Based on this evaluation system, a single-goal, multidimensional and comprehensive evaluations of the CPTC sustainable development level are conducted in these specified years. First of all, we combined multiple data sources, including big earth data, to standardise the inconsistent scale of data into unified sustainable development indicators. This addressed the CPTC issue of 8 Tier II (no data). On the other hand, we discovered the lagging development in the central part of the CPTC and imbalanced social, economic and environmental development issues. Consequently, a serial of targeted sustainable development suggestions for CPTC were proposed.

2. Materials and methods

2.1. Study area

The study focuses on administrative and geographical units along the China-Pakistan Transportation Corridor (CPTC)'s Karakorum Highway (KKH). The study area encompasses a buffer zone extending 100 km on either side of the KKH, covering a vast area of 281,340,000 km². The geographical coordinates range from 40°46′13.75″N to 32°59′20.79″N, and from 71°14′9.81″E to 78°36′37.1″E (Figure 1). The highway straddles China and Pakistan, starting from Kashgar City in the Xinjiang Uyghur Autonomous Region of China in the north, passing through the Khunjerab boarder port at the China-Pakistan border crossing, and ending in Takot in Pakistan in the south. The total length of the highway is 1224 km, accounting for one-third of the total length of highways in China and two-thirds in Pakistan. This transportation corridor offers great convenience for transportation and trade between China and Pakistan as well as with neighbouring countries such as Iran and Afghanistan.

Figure 1. Overview of the study area.

Note: XJ, Xinjiang Uyghur Autonomous Region; FCT, Islamabad Capital Territory; NWFP, Khyber Pakhtunkhwa; AK, Azad Kashmir; P, Punjab; NA, Gilgit-Baltistan; FATA, Federally Administered Tribal Area.

CPTC is one of the most treacherous and undulating areas in the world. Crossing three major extreme high mountain ranges, namely the Karakorum Mountains, the Hindu Kush Mountains and the western end of the Himalayas, the altitude is generally higher than 4000 m, and it is characterised by alpine climate. Precipitation and temperature are greatly influenced by latitude, with an annual average temperature of 16°C. Predominantly, this area is arid or semi-arid, with an average annual precipitation that less than 250 mm. Winter in this climate zone is long and cold, often windy and snowy. Summer is cool but short, the terrain is precarious, and prone to geological disasters such as landslides, mudslides and glacial lake outbursts.

2.2. Data source and pre-processing

2.2.1. Data source

The data used in this study were divided into two main parts: (1) statistical data with spatial attributes, and (2) remote sensing monitoring data, including land cover, night-time lights, and other geodata. These data sources are listed in Tables 1 and 2, respectively.

Table 1. Geospatial and statistics data used for the study.

Data	Source	Usage
Road vector data (OSM 2023)	OpenStreetMap (https://www.openstreetmap.org)	Density of urban roads
Water system vector data (OSM 2023)	OpenStreetMap (https://www.openstreetmap.org)	Density of water system network
Human-caused disaster data (SATP 2023)	https://www.satp.org/datasheet-terrorist-attack/official-data/pakistan	Human-caused disaster data events
Natural Disaster Data (EM-DAT 2020)	EM-DAT (https://www.emdat.be/)	Population death rate due to natural disasters
Population Distribution Data (WorldPop 2020)	world pop (https://www.worldpop.org)	Arable land per capita
Gross Domestic Product (GDP) (Kummu, Taka, and Guillaume 2020) (The World Bank 2023)	DRYAD (https://datadryad.org/stash) World Bank (https://data.worldbank.org.cn/indicator)	GDP per capita
Pakistan Statistics Data (The World Bank 2023)	World Data Bank official website (https://databank.worldbank.org/)	All statistical indicators
Xinjiang Statistical Data (NBS 2020)	Xinjiang Statistical Yearbook	All statistical indicators

Table 2. Remote sensing data sources used in the study.

Data	Source	Spatial resolution	Times	Usage
GLC_FCS30 (Zhang et al. 2022)	https://data.casearth.cn	30 m	2020	Extract land cover data
Landsat8 OLI (USGS 2020)	Google Earth Engine (GEE)	30 m	2013, 2015, 2017, 2020	Extract land cover data and NDVI, MNDWI, NDBI
NPP–VIIRS interannual nighttime light remote sensing dataset (NOAA 2020)	National Oceanic and Atmospheric Administration (NOAA) (https://ngdc.noaa.gov)	500 m	2015, 2017, 2020	VANUI

2.2.2. Data pre-processing

2.2.2.1. Human activity intensity of the VANUI value construction

Humans are the main factors in sustainable development and the most subjective and dynamic factor in the pursuit of the SDGs. NPP/VIIRS nighttime light data, Normalised Difference Vegetation Index (NDVI), Moderate Normal Different Water Index (MNDWI), and Normalised Difference Built-Up Index (NDBI) data were used to obtain the nighttime light adjustment index Variable of Anthropogenic Nighttime Urban Intensity (VANUI), which was used to characterise the intensity of human activities (Chu et al. 2020).

VANUI was derived from adjusting the DN value of the nighttime light data (Zhang, Schaaf, and Seto 2013). This adjustment was performed by applying the indices such as NDVI, MNDWI, and NDBI collectively to establish an adjustment coefficient for DN:$(1-\mathrm{NDVI}-\mathrm{MNDWI}+\mathrm{NDBI})$. It served the purpose of mitigating the impact of vegetation, water bodies, and other factors on the NPP/VIIRS nighttime light data. The human-activity intensity adjustment index VANUI was obtained as follows: (1) $\mathrm{VANUI}=(1-\mathrm{NDVI}-\mathrm{MNDWI}+\mathrm{NDBI})\times D{N}_{\mathrm{nor}}$(1) where$\mathrm{VANUI}$ denotes Human activity intensity,$D{N}_{\mathrm{nor}}$ denotes the normalised nighttime light data, $D{N}_{\mathrm{nor}}=(DN-D{N}_{\min })/(D{N}_{\max }-D{N}_{\min })$,$D{N}_{\max }$, $D{N}_{\min }$ denote the maximum $\mathrm{DN}$ value and minimum $\mathrm{DN}$ value in the nighttime light data .

Figure 2. Spatialised geographic data for 2015.

2.2.2.2. Geographic data spatialisation

Compared with statistical indicators, geospatial indicators can provide a full coverage and more detailed insights at specific geographical locations. Therefore, this study used land cover, nighttime light, and other geospatial data to spatially grid indicators such as population, GDP, road density, and water system density along the CPTC (Figure 2). There were three main steps for obtaining geospatial gridded data: extracting and calculating land cover data, analysing vector data changes, and clipping existed geospatial data. Specifically, extracting and calculating land cover data involved acquiring the required land cover based on specific classifications, and reclassifying them to form corresponding geospatial indicator data. Vector data change calculation involved setting up establishing a grid based on acquired vector data (e.g. roads and water systems), and performing analyses to obtain corresponding geospatial indicator data. Clipping existed geospatial data was the last step of the data set ready for evaluation.

Figure 3. Technological roadmap.

2.2.2.3. Xinjiang and Pakistan statistical data spatialisation

The statistical data for each province of Pakistan in 2015, 2017 and 2020 were screened from the World Data Bank. The statistics of Xinjiang were obtained from the China Xinjiang Statistical Yearbook. All the selected screened statistics were localised against the SDG indicator framework and assigned to each provincial administrative unit in the study area to spatially represent the regional statistics.

2.3. Methods

Quantitative research, approached from various evaluative perspectives and directions, can push a comprehensive understanding of the CPTC sustainable development level. The methodology of this study involved two main steps: section 2.3.1 Indicators system construction, including (1) formation of localised sustainable development indicators based on the characteristics of the CPTC. (2) Harmonisation all indicators so that they can be aggregated to form sustainable development indicators system; section 2.3.2 Sustainable development evaluation, including (1) The benchmarking of selected indicators against SDGs for single goal evaluation of sustainable development. (2) Evaluation conducted across social, economic, and environmental dimensions, based on established goals. And finally, (3) a comprehensive evaluation carried out by considering the scores and weights of each indicator. The specific methodology is illustrated in Figure 3.

2.3.1. Indicators system construction

2.3.1.1. Indicators localisation

Localisation emphasises the need for foreign elements to be adapted to the new environments in order to meet specific local requirements (Kramer and MacKinnon 1993). The process of localising sustainable development indicators (Patole 2018) requires the screening and improvement of indicators based on the natural terrain, population characteristics, and socioeconomic factors. This leads to the indicators that are unique to the local context. The CPTC, which spans China and Pakistan, is marked by significant variations in terrain, a small population size, and frequent natural disasters. Consequently, the localisation of indicators is essential to establish a sustainable development indicator system for CPTC.

This study aims to construct a localised sustainable development evaluation system for CPTC. The selection of indicators adhered to the principles of objectivity, accessibility, and rationality, in accordance with the UN SDGs evaluation requirements. Specific localisation methods included the following aspects: (1) The CPTC is characterised by large terrain undulations, uneven population distribution, and difficulties in acquiring statistical data, so spatialised indicators (such as land restoration and degradation) that can more easily reflect the actual state of sustainable development are mainly selected; (2) The CPTC is sparsely inhabited and concentrated, with an aggregated distribution of various land-use types, so remote sensing data (such as land-use data) are more capable of highlight the sustainable development level of the CPTC; (3) The CPTC is prone to natural disasters due to its topography and climate, so localised indicators for disasters are also formed; (4) The CPTC has undergone significant changes in recent years as a result of the BRI, so more resource environment and socio-economic indicators are selected to analyse the their changes.

2.3.1.2. Indicators harmonisation

Due to the diverse nature and origin of localised indicators, it is necessary to harmonise and aggregate them for better assessing sustainable development. This study mainly completed the harmonisation and aggregation of indicators through the following three steps. Firstly, we improved most indicators by converting them into dimensionless data such as percentages or ratios. Secondly, we had standardised all indicators (0–100) to make them dimensionless and facilitate their comparison. Specific standardisation methods were as in Equations (2) and (3), and the positive and negative directions of the specific indicators were shown in the properties of Table 3. Lastly, we spatialised all indicator values, including geospatial and statistical data, by gridding them and resampling them into uniform grid data with a spatial resolution of 100 m, making it easier to aggregate and calculate. (2) $\mathrm{Positive}\mathrm{indicator}\mathrm{score}:X_i=100\times {\displaystyle \frac{(x_i-x_{\min })}{(x_{\max }-x_{\min })}}$(2) (3) $\mathrm{Negative}\mathrm{indicator}\mathrm{score}:X_i=100\times {\displaystyle \frac{x_{\max }-x_i}{(x_{\max }-x_{\min })}}$(3) where$X_i$ denotes the score of each indicator, $x_i$ denotes the raw data of the indicator, $x_{\max }$ and $x_{\min }$ are the maximum and minimum values of the relative levels in the data obtained for the indicator, respectively. Specific scores for each sustainability indicator were obtained by normalisation.

Table 3. Sustainable development indicator system for CPTC.

System level	Goal level	Indicator level	Indicator Descriptions	Calculation methods	Tier Class	Properties	Weights
Economic Development	GOAL 1: No Poverty	Per capita disposable income of residents (CNY)	Disposable income/population	B	Tier I	+	0.010227
	GOAL 2: Zero Hunger	Arable land per capita (m^2)	Area of arable land/Population	B	Tier II	+	0.017017
		Heavy or moderate food insecurity rate (%)	Thresholds for heavy or moderate insecurity according to international standards, population with food insecurity/Population	A	Tier I	−	0.016333
		Per capita unit production of food crops (kg)	Crop production/Population	B	Tier II	+	0.012082
	GOAL 8: Decent Work and Economic Growth	GDP growth rate (%)	(GDP this year − GDP last year)/GDP last year	A	Tier I	+	0.05899
		GDP per capita (CNY)	GDP/Population	A	Tier I	+	0.019511
		Long-term unemployment rate for men and women (%)	Number of long-term unemployed individuals of a specific gender/Total number of individuals in the labour force of that gender	A	Tier II	−	0.023085
		Average annual salary for male and female staff (CNY)	Total annual salary for a specific gender/Total number of individuals of that gender	B	Tier II	+	0.102901
	GOAL 9: Industry, Innovation and Infrastructure	Proportion of households using the network (%)	Number of households using the network/Total number of households	B	Tier I	+	0.046719
Social Equity and Security	GOAL 3: Good Health and Well-being	Number of hospitals (Count)	The number of medium and large hospitals in the region	B	Tier I	+	0.010421
		Under 1 year old mortality rate (%)	Number of deaths under 1 year old/Number of live births	B	Tier I	−	0.011018
		Number of traffic accidents (Count)	Number of traffic accidents per year in the region	A	Tier I	−	0.011883
		COVID-19Mortality (%)	Number of COVID-19 deaths/Number of COVID-19 cases	B	Tier I	−	0.012353
		Proportion of population with access to vaccines (%)	Number of individuals with access to vaccines/Total population	A	Tier I	+	0.054866
	GOAL 4: Quality Education	Percentage of financial expenditure on education (%)	Total expenditure on education/Total government expenditure	B	Tier I	+	0.02336
		Literacy rate over 10 years old (%)	Number of literate individuals over 10 years old/Total population over 10 years old	B	Tier II	+	0.038014
	GOAL 5: Gender Equality	Education ratio of male to female (%)	Number of males enrolled in education/Number of females enrolled in education	A	Tier I	−	0.014101
		Male-to-female population ratio (%)	Number of males population/Number of females population	B	Tier II	−	0.026927
		Proportion of males to females using cell phones (%)	Number of males using cell phones/Number of females using cell phones	A	Tier II	−	0.010318
	GOAL 16: Peace and Justice Strong Institutions	Human-caused disaster data events (Count)	Number of disasters caused by non-natural factors such as terrorist attacks	B	Tier I	−	0.009437
	GOAL 11: Sustainable Cities and Communities	Power lighting ratio (%)	Power consumed by lighting/Total power consumed	B	Tier I	+	0.010672
		Proportion of homeless population (%)	Number of homeless individuals/Total population	B	Tier I	−	0.011941
		Density of urban roads (km/km^2)	Total length of all roads/Area of study area	B	Tier I	+	0.050361
		Garbage collection system (Count)	Calculation of the number of garbage collection points in the region	B	Tier II	+	0.014514
	GOAL 8: Decent Work and Economic Growth	Labour force participation rate (%)	Number of people in the labour force/Total working-age population	B	Tier I	+	0.011505
	GOAL 1: No Poverty	Population death rate due to natural disasters (%)	Number of deaths due to natural disasters/Population at risk	B	Tier I	−	0.011198
		Proportion of the population living in poverty (%)	Number of people in poverty/Total population	A	Tier I	−	0.00952
Ecological Quality	GOAL 15: Life on Land	Land degradation rate (%)	Land degradation area/Total land area	A	Tier I	−	0.034013
		Forest cover (km^2)	Extraction of forest cover areas in the region using land use data	A	Tier I	+	0.075414
		Area share of protected areas (%)	Total Area of Protected Areas/Total Land Area	A	Tier I	+	0.056393
		Land utilisation rate (%)	Actual land area utilised/Total land area	B	Tier I	+	0.018718
		Land restoration rate (%)	Land restoration area/Total land area	B	Tier I	+	0.051322
	GOAL 6: Clean Water and Sanitation	Proportion of ecosystem changes related to water (%)	Area of water-related ecosystem change/Total ecosystem area	A	Tier I	+	0.034013
		Density of water system network (km/km^2)	Total length of water system network/Area of study area	B	Tier I	+	0.065871
		Rate of drinking improved water (%)	Number of people drinking improved water/Total population	A	Tier I	+	0.014982

2.3.2. Sustainable development evaluation

Accurate and quantitative evaluations of sustainable development can serve as a foundational reference for future regional development decisions. A singular approach to such assessment does not yield a holistic understanding of CPTC (Hutton and Chase 2016). This research examined the sustainable development of CPTC across three key dimensions: single goal evaluation, socio-economic-environmental evaluation, and comprehensive evaluation. The objective is to provide a comprehensive insight into CPTC's sustainable development trajectory, while also pinpointing areas requiring improvement and identifying potential shortcomings.

2.3.2.1. Single goal evaluation

The scores of each SDG were further calculated based on the standardised score of each indicator obtained in the previous calculation. Considering that each indicator is equally important for its corresponding goal, an equal-weighted average was calculated for each indicator, which was expressed as the ratio of the sum of the scores of each indicator to the number of indicators: (4) $Y_h={\displaystyle \frac{{\sum }_1^n{X}_i}{n}}$(4) where $Y_h$ denotes the score of $h$-th goal, $X_i$ denotes the score of each indicator corresponding to indicator$i$,$n$ denotes the number of indicators corresponding to the goal.

2.3.2.2. Socio-economic-environmental multidimensional evaluation

The equal-weighted average method was used to calculate the development of the Socio-economic-environmental multidimensional. Because the principle that equal importance should be given to the economic, social, and environmental dimensions for each sustainable development goal. The overall performance of each province was calculated across these dimensions as follows: (5) $Z_k={\displaystyle \frac{{\sum }_1^n{Y}_h}{n}(k=1,2,3)}$(5) where $Z_k(k=1,2,3)$ denote the dimensions of economy, society, and environment respectively, $Y_h$ denotes the goal score corresponding to goal$h$,$n$ is the number of goals contributing to the dimension.

2.3.2.3. Comprehensive evaluation

The entropy method was used to determine the weight of each indicator at the comprehensive sustainable development level. Currently, a large number of sustainable development evaluations used principal component analysis (Dos Santos et al. 2019; Kaymaz, Birinci, and Kızılkan 2022) to determine weights. However, this method is highly subjective, whereas the entropy method is a more objective method for determining weights through the degree of variability of statistical data. The information carried by the entropy value was used to calculate the weight of each sustainable development indicator, which was then used for comprehensive sustainable development evaluation.

Assuming that there are $n$ indicators for an objective in the sustainable development system of the study area, the information entropy value $E_j$ for the $j$ indicator is calculated as: (6) $E_j=-{\displaystyle \frac{1}{\mathrm{lnm}}\sum _{i=1}^m{p}_{\mathrm{ij}}\ln p_{\mathrm{ij}}}$(6) where m denotes the total number of levels for the $j$-th indicator, $p_{\mathrm{ij}}$ is the proportion of the j-th indicator's $i$-th level to the total number of observations. The information entropy value reflects the degree of uncertainty or randomness in the distribution of the indicator levels. Higher entropy values indicate greater uncertainty, whereas lower value indicates greater predictability. The information utility values $d_j$ is: (7) $d_j=1-E_j$(7) the larger the information utility value, the more important the indicator, and the greater the influence on the evaluation results. Thus, the weight of the $j$-th indicator can be obtained using the following equation: (8) $W_j={\displaystyle \frac{d_j}{{\sum }_{j=1}^n{d}_j}}$(8) The standardised indicator data were multiplied and summed with the weights calculated using the entropy method to obtain the comprehensive score of sustainable development indicators for the CPTC. (9) $S=\sum _1^m(X_i\times w_i)$(9) where $S$ is the final calculated composite sustainability score; $X_i$ is the score for the $i$-th indicator,$w_i$ is the calculated weight for each objective; $m$ is the total number of indicators.

3. Results

3.1. Sustainable development indicators system

We modified and screened the UN SDG Indicators framework to form a CPTC localised indicator system (Table 3). The indicators selection criteria mainly include rationality (whether the selected indicators are suitable for CPTC region), objectivity (whether the acquired data can objectively reflect the corresponding indicators), and accessibility (whether the selected indicators can be obtained through statistical data and geospatial data).

The system consists of three levels: (1) Indicator level, consisting of a total of 35 indicators, with 27 Tier I indicators (with available methodology and data) and 8 Tier II indicators (with available methodology but no data); (2) Goal level, including SDG1, SDG2, SDG3, SDG4, SDG5, SDG6, SDG8, SDG9, SDG11, and SDG15; (3) System level, which evaluates the sustainable development of CPTC from social, economic and environmental dimensions. In Table 3, ‘A’ demonstrates that the calculation method for the United Nations sustainable development indicators is directly applied to obtain the results; ‘B’ denotes a new calculation method suitable for CPTC, based on the United Nations sustainable development indicator calculation method, with refinements and adjustments to the indicators. Specifically, we converted indicators with dimensions into ratios or percentages without dimensions to allow for a cross-regional comparison of indicators.

3.2. Sustainable development evaluation

3.2.1. Single goal evaluation

To visualise the sustainable development level of each province in the CPTC, we use a traffic light dashboard proposed by Sachs et al. (2022). The seven provinces are ranked 1–7 in terms of their SDG scores, and are organised into four gradients: the first gradient is for the provinces ranked 1–2, indicating the highest sustainable development level of CPTC, denoted by ‘’; The second gradient is provinces ranked 3–4, indicating a higher sustainable development level, denoted by ‘’; provinces ranked 5–6, indicating an average sustainable development level, are indicated by ‘’; and province ranked 7, indicating a poorest sustainable development level, is indicated by ‘’. ‘’ indicate that no data are available for this corresponding SDG. A final visualisation map of each sustainable development goal situation for each province is derived (Figure 4).

Figure 4. CPTC sustainable development goals traffic light dashboard.

The dashboard and single goal scores reveal that Xinjiang Uygur Autonomous Region of the CPTC is outstanding in sustainable development. The region has high, rapid progress in all of the SDGs, with seven ranking in the top positions. Khyber Pakhtunkhwa Province has more room for improvement, especially in SDG1 (No Poverty), SDG8 (Economic Development), and SDG11 (Sustainable Cities). It can also be deduced that Khyber Pakhtunkhwa province is economically backward and has a slow urbanisation process in CPTC. As the capital of Pakistan, Islamabad Capital Territory has superior and accelerated progress towards various sustainable development goals compared to other provinces, especially SDG5 (Gender Equality), SDG8 (Economic Growth), and SDG9 (Sustainable Industrialisation) are in the first rank; The data deficit in Azad Kashmir has been substantially reduced, placing it in the middle of the pack overall, with balanced development across all objectives; In FATA, the SDG2 (Zero Hunger) and SDG16 (Strong Institutions for Peace and Justice) are developing at a faster pace compared to the SDGs in other areas, while all others are developing steadily; In Gilgit-Baltistan, SDG16 and SDG5 are developing at a higher level compared to the SDGs in other areas, while SDG4 is developing poorly, and education needs further improvement; In Punjab, SDG5 (Gender Equality), SDG6 (Clean Water), SDG8 (Economic Development), SDG9 (Sustainable Industrialisation) and SDG11 (Sustainable Cities) are in the first rank and perform well compared to the SDGs in other areas, while in SDG2 (Zero Hunger) and SDG3 (Healthy Life) have the poorest performance. It also shows that although the province has a high level of economic development, there are many poor people and a large gap between the rich and the poor. Gilgit-Baltistan and FATA stand in the middle of sustainable development.

3.2.2. Socio-economic-environmental multidimensional evaluation

Owing to the numerous development goals, it is not easy to compare the shortcomings of the development of each region’s development using only quantitative research on each development goal. At the same time, it is difficult to form a comprehensive understanding of CPTC sustainable development. Economic, social, environmental analyses can qualitatively describe the development status of each CPTC dimension. A comprehensive sustainable development evaluation of CPTC is conducted to understand its overall development. The sustainability scores calculated for each dimension in 2015, 2017 and 2020 are listed in Table 4, the spatial distribution of the sustainability level score for the socio-economic-environmental dimensions of the CPTC is shown in Figure 5.

Figure 5. Spatial distribution of multidimensional sustainability scores.

Table 4. socio-economic-environmental dimension sustainability score.

Year	Dimension	Xinjiang Uygur Autonomous Region	Islamabad Capital Region	Khyber Pakhtunkhwa	Azad Kashmir	Punjab	Gilgit-Baltistan	Federally Administered Tribal Areas
2015	Economy	61.6709	26.2704	17.8203	17.6571	22.1441	16.7186	21.2462
	Social	45.2088	51.7896	66.5704	32.0352	67.1167	51.3457	36.5067
	Environment	30.8343	21.8502	43.8656	37.2855	42.0708	29.6228	17.751
2017	Economy	61.9533	36.2808	18.5582	32.0724	23.2687	22.3529	31.3721
	Social	60.8656	54.8593	34.4730	46.06	68.7017	52.8786	41.9091
	Environment	59.9625	34.2648	41.8372	39.7458	50.2555	34.9836	28.9088
2020	Economy	68.1485	40.1653	22.9985	39.9811	34.4022	30.3301	36.3408
	Social	64.1856	55.1441	29.7088	54.2855	53.6689	53.4539	56.7540
	Environment	63.3272	46.3657	34.2195	44.7123	54.5377	41.4118	30.6170

On the economic front, the overall level of economic sustainable development has increased from 2015 to 2020. However, the rate of development in 2017–2020 was slower than 2015–2017 due to the impact of the COVID-19 epidemic (Nundy et al. 2021), especially in the capital regions of Azad Kashmir and Islamabad. On a social level, the overall level of the region is increasing, especially in the northern and southern regions; however, but the rate of development has slowed significantly. Individual provinces, such as Khyber Pakhtunkhwa, have experienced a severe regression in terms of social sustainable development. This may be related to the impact of COVID-19 (Fenner and Cernev 2021; Nundy et al. 2021; Wang and Huang 2021), the frequency of natural disasters (Rahman et al. 2021; Waqas et al. 2021) and land degradation (Ahmad et al. 2021; Zhao et al. 2023) in the province. The most rapid development among the three dimensions was in the environmental dimension, which improved significantly in all provinces except Khyber Pakhtunkhwa.

Overall, the social, economic and environmental dimensions of sustainable development in the CPTC have been steadily increasing from 2015 to 2020 (Figure 5), while only Khyber Pakhtunkhwa province is showing ‘regression’, which has negative environmental and social growth. The reason may be related to economic development that brings social imbalances and environmental damage (e.g. natural disasters and land degradation) and COVID-19. The Xinjiang Uyghur Autonomous Region has developed rapidly in the past five years with balanced and rapid development in all three dimensions, ranking first in terms of economic, environmental and social aspects. It has become the most prominent region for sustainable development in the CPTC. Punjab has performed well in economic and environmental aspects but has also experienced negative social development. Therefore, Punjab should pay more attention to social issues such as gender equality, health care, and education in its subsequent development to ensure a balanced approach to the 2030 Sustainable Development Goals; Islamabad Capital Territory has experienced rapid economic and environmental development in the past five years, and has achieved a balanced development in all three aspects of social, economic, and environmental development; FATA’s sustainable development level is low in 2015, but it has developed rapidly in recent years and is now basically on par with the sustainable development levels of Pakistani provinces; Gilgit-Baltistan is the lowest economic level in the district in 2015, but after five years of development, it has caught up with Khyber Pakhtunkhwa; Azad Kashmir has improved in social, economic and environmental dimensions, with significant progress being made in social sustainability as the fastest-growing aspect.

3.2.3. Comprehensive evaluation

The overall sustainable development of the CPTC is ‘high on both ends and low in the middle’. The northern Xinjiang Uyghur Autonomous Region has the highest level of sustainable development, while the southern of Punjab Province has a relatively high level of sustainable development. In contrast, the central part of the CPTC, including the Gilgit-Baltistan and Azad Kashmir, are located in areas with high levels of environmental development but a low level of economic sustainability, resulting in low levels of overall sustainable development. Subsequently, the overall sustainability scores for each province of CPTC were calculated, as shown in Figure 6.

Figure 6. Spatial distribution of the integrated sustainability score of the CPTC.

From 2015 to 2017, the overall of sustainable development level of the entire CPTC, except for the Khyber Pakhtunkhwa, has been increasing. It also fully reflects the advantages of the BRI, driving the sustainable development of the whole CPTC (Cao 2019; Guo et al. 2018). The Xinjiang Uyghur Autonomous Region has consistently ranked first among the entire CPTC and has the fastest improvement rate of sustainable development. However, compared with 2015–2017, the development rate has slowed down, which is associated with the COVID-19 in 2019 (Health 2020; Nundy et al. 2021; Wang and Huang 2021). In addition, Islamabad Capital Territory is experiencing rapid sustainable development. Other provinces have shown a slow increase in sustainability levels. The overall development level of the Khyber Pakhtunkhwa province is declining, and this decline is more serious. Although the economic sustainability score is increasing, the overall level is declining because of the significant decline in the social and environmental scores. This warns us that economic development should be accompanied by social and environmental factors to achieve sustainable development.

4. Discussion

4.1. Performance of VANUI

Land use data is closely related to the intensity of human activities, as indicated by the VANUI. The modified nighttime light data can effectively capture and characterise the intensity of human activities. We find that there is a strong correlation between VANUI and different types of land use, with buildings exhibiting the highest correlation, followed by farmland, water/grassland, forest/bushland, and snow/ice, which is consistent with Zhang, Schaaf, and Seto (2013). These results show that the land use image can be used as an effective reference to verify the accuracy of VANUI (Figure 7).

Figure 7. Comparison of human activity intensity (VANUI) with land use data. (a), (b), and (c) are the 2020 VANUI; (d) and (e) are the corresponding land use for (b) and (c).

4.2. Sustainable development evaluation

(1)	Evaluation method

The selection of a suitable localised indicator framework is a prerequisite for sustainable development evaluation. The Sustainable Development Solution Network (SDSN) proposed by Lynch and Sachs (2021.) provides a practical method and basis for assessing various national levels around the world of sustainable development. In this study, we developed a CPTC regional-level localised sustainable development evaluation system based on SDSN. In this assessment system, the selection of the number of indicators and the localisation of indicators are grounded in and integrated with the region. We spatialise all evaluation indicators (including geographical and statistical indicators), to generate spatial data with consistent grids. This approach allows for a comprehensive, multidimensional grid-based evaluation using gridded spatial information that can cover and reflect actual geospatial differences across the entire region, providing a more detailed representation of the level of sustainable development. For example, Figure 8(b) better reflects the detailed population distribution in CPTC compared with Figure 8(a). It is clear from Figure 8(b) that the population in the N.W.F.P. province is mainly concentrated in the southern region, which cannot be captured by the administrative division population distribution data. In addition, we reference the evaluation method of Sachs et al., but our method not only includes single-goal and comprehensive sustainable development evaluation, but also evaluates the sustainable development goals in three dimensions of social, economic, and environmental aspects, comprehensively grasping the level of sustainable development of CPTC from different perspectives. Huan et al. (2021) developed a composite SDG index to assess the comprehensive performance of achieving SDGs. However, their evaluation results were also only presented at the regional level for each country and did not include a detailed grid-based evaluation. In addition, Shen et al. (2020) and Vanham et al. (2018) had conducted assessments of water resource sustainability based on individual targets of SDG6. Although these assessments provided more specific insights, they tended to fragment the overall concept of sustainable development. In contrast, our study evaluates the sustainability of the research area from multiple dimensions, providing a more comprehensive understanding of its shortcomings in terms of sustainable development.

Figure 8. Comparison of population distribution between (a) administrative division data and (b) gridded data.

Additionally, the evaluation data (including geospatial data and statistical data) we used are all open-sourced data from various official websites, which enhances the reproducibility and the extensibility of our research. And we change sustainable development indicators into ratio form, standardisation, gridding, aggregate all sustainable development indicators together, which also provides a solution for the aggregation of indicators with different sources and natures. Moreover, this study presents a reference case for sustainable development evaluation in regions that lack data and span multiple administrative units.

(2)	Evaluation result

By the single goal, multidimensional and comprehensive evaluations of CPTC, we find that the sustainable development level in CPTC hugely has improved from 2015 to 2020. However, the development rate has significantly slowed, and the spatial level of sustainable development shows a pattern of ‘high at both ends and low in the middle’. The sustainable development level in terms of society, economy and environment has been improved in all areas of CPTC except for Khyber Pakhtunkhwa, which is consistent with the findings of other relevant studies in the region. In economic development, according to WEO statistics (Outlook 2023), the national per capita income of Pakistan was INR 29,176.78 in 2015 and INR 34,586.67 in 2020, which also showed that the Pakistani economy has been growing steadily for the five years. However, Ashraf (2023) suggests that the economic growth rate has declined, which is consistent our findings on the economic sustainability. In terms of environmental sustainability, Ashraf further proves the increase in CO₂ emissions in the CPTC, but this does not contradict the increased level of environmental sustainability in our study, because a single CO₂ emission can’t be used as a criterion for good or bad environmental sustainability. Demena and Afesorgbor (2020) also prove that foreign direct investment, such as BRI, enhances environmental sustainability, which confirms the results of our study on environmental aspects. In terms of social justice, Pakistan has seen an increase in women's employment over the past several years due to higher education levels (Abbass et al. 2022) and a deeper concept of gender equality (Majid and Siegmann 2021), which reflects a significant improvement in the social aspects of Pakistan (Kanwal, Chong, and Pitafi 2018), which is consistent with our conclusion that CPTC's level of social sustainability has increased.

4.3. Analysis of driving factors

The change in the spatial and temporal pattern of the sustainable development level of the CPTC as a major transportation corridor of the China-Pakistan Economic Corridor has been influenced by various factors such as policy implementation, geographical conditions, climate, and social environment.

1. BRI has promoted for sustainable development. From 2015 to 2020, the level of sustainable development of the CPTC increased steadily, which is inextricably linked to the BRI (Cao 2019). The Chinese government's BRI and the UN SDGs have the same roots and are mutually reinforced. The BRI is centred on corridor construction, which has led to the development of transportation, energy, and infrastructure in the CPTC (Ali 2018).It directly driven the development of SDG7, SDG8, and SDG9, providing more employment opportunities and economic development, and indirectly driven the development of SDG2, SDG4, and SDG6 within the region.

2. The sustainable development rate has slowed due to the COVID-19 pandemic. Our study found a slowdown in 2017–2020 compared with 2015–2017, which is related to the global outbreak of COVID-19 at the end of 2019 (Nundy et al. 2021). The pandemic increased the number of deaths and had a significant direct impact on SDG3, SDG4, and SDG16, as well as on SDG2, SDG7, and SDG8, posing an indirect threat. This has led to a slowdown in the CPTC sustainable development.

3. Geographic conditions affect the spatial distribution of sustainable development. The CPTC crosses the Karakorum Mountains, the Hindu Kush and the western end of the Himalayas, with large topographical fluctuations, showing a topographic feature of ‘high in the middle and low at the ends’. Consequently, the social and economic aspects of sustainable development within the CPTC also show a pattern of ‘high at both ends and low in the middle’, with Gilgit-Baltistan and Azad Kashmir in the middle having a low level of sustainable development, while Xinjiang in the north and Islamabad in the south have a high level of sustainable development in comparison.

4. The reasons for the ‘regression’ of sustainable development in Khyber Pakhtunkhwa. From 2015 to 2020, the Khyber Pakhtunkhwa Province is adversely affected by accelerated urbanisation and overgrazing, leading to land degradation, loss of agricultural land, and a significant decrease in grassland and arable land (Ahmad et al. 2021) (Figure 2). These factors have contributed to the deterioration of the ecosystem. In addition, the region is vulnerable to natural disasters such as floods (Waqas et al. 2021) (Figure 8(a)), cyclones, and droughts (Rahman et al. 2021) due to its topography and other geographic factors. These natural disasters have further hindered sustainable development in the region (Hamidi, Zeng, and Khan 2020). The frequency of natural disasters in the CPTC is shown in Figure 9(b), based on data from EM-DAT. Furthermore, the lack of educational resources (Khan 2021), social security issues (Din et al. 2022), and low levels of healthcare in Khyber Pakhtunkhwa (Israr et al. 2020) have negatively impacted the social aspects of development in the province.

Figure 9. Natural disasters in CPTC (a is Flood hazard of Khyber Pakhtunkhwa province in 2016 (PDMA, 2016) Image source: https://floodlist.com/asia/pakistan-least-46-killed-floods-khyber-pakhtunkhwa; b is Number of Natural Disasters in the CPTC 2015-2020).

4.4. Development suggestions

1. Targeted remediation of shortcomings of sustainable development. The central region of the CPTC needs targeted interventions, especially SDG4 (Quality Education) and SDG8 (Decent Work and Economic Growth), which are shortcomings of the sustainable development goals of the region. Targeted enhancement of SDG4 and SDG8 sustainable development levels to narrow the gap between them and the respective sustainable development goals. Simultaneously, the coordinated development of each region should be achieved, bridging the gap between the central northern and southern regions in terms of development goals. However, these interventions should not be performed at the expense of environmental goals. There is a pressing need to balance economic development with social equity, security, and ecological quality to achieve smooth and comprehensive sustainable development of society, economy, and environment.

2. Optimising the strategies for economic development. CPTC has witnessed significant land degradation and a marked expansion in cultivated land, leading to poor environmental indicators. For instance, Khyber Pakhtunkhwa witnessed an increase in sustainable economic development from 2015 to 2020, but a decline in environmental sustainability. Therefore, we recommend undertaking an industrial transformation and promotion of green and sustainable industries. Reducing the development of heavy industry and agriculture, combining the characteristics of its own environmental and geographical advantages, developing tourism and high-tech industries, and enhancing the economy while improving the level of sustainable development from environmental and social aspects.

3. Improve the ability to prevent and mitigate disasters. The CPTC is very vulnerable to natural disasters (Figure 9), such as floods, landslides, and mudslides due to its undulating terrain and complex topography, which are important factors that limit its development. The disaster prevention and mitigation capacity of the region must be improved to enhance sustainable development.

4.5. Future work direction

In this study, we use spatialised and statistical indicators to construct the CPTC's sustainable development indicator system, and conduct single-goal evaluation, multidimensional evaluation, and comprehensive evaluation of the CPTC are carried out for the SDGs proposed by the United Nations. Follow-up research can be improved and developed in the following aspects: (1) Due to the special geographical location of the study area, data acquisition is relatively limited. In future research, we can combine various data, such as field surveys, to strengthen the collection and integration of big earth data, enrich and expand the number of evaluation indicators, and evaluate them in more detail. (2) SDGs are not completely independent from each other, and there are synergies or trade-offs among them. Actions taken for one goal may enhance (McCollum et al. 2018) or offset exist between another goal, resulting in an imbalance in the SDGs (Mulligan et al. 2020). For example, using coal to improve energy access (SDG7) accelerates climate change (SDG13) and damages human health through air pollution (SDG3). In future research, synergy and trade-off analyse among indicators can be strengthened by considering the interrelationships among the SDGs in an integrated manner to contribute more deeply to the overall evaluation. (3) Finally, all the 17 SDGs are accessed from a more comprehensive perspective. That means more related indicators at the goal and indicator levels can be involved to support a comprehensive and localised assessment of sustainable development in this region in the future.

5. Conclusion

Oriented to the gap of sustainable development evaluation in high Asian area, this study develops a localised indicator evaluation system for the sustainable development of CPTC. Based on this evaluation system, a single-goal, multidimensional and comprehensive evaluation of the CPTC sustainable development level in 2015, 2017 and 2020 is conducted. The sustainable development level of the CPTC greatly improve from 2015 to 2020, and the development rate from 2015 to 2017 is higher than that from 2017 to 2020. This reflects the advantages of the BRI, which has benefited the entire CPTC. From a spatial perspective, the sustainable development levels of the CPTC show a pattern of ‘high at the ends and low in the middle’, with Xinjiang in the north and Islamabad in the south having higher sustainable development levels, while Gilgit-Baltistan and Azad Kashmir in the centre having lower sustainable development levels and Khyber Pakhtunkhwa province typifies a ‘regression’ in sustainable development levels. Despite the economic scores improved considerably, the social and environmental scores decrease significantly, leading to lower overall scores. This is a warning to the government to intervene in economic development and focus on the social and environmental aspects to achieve sustainable development. Targeted interventions are needed in the South to address these gaps, especially in socioeconomic-related SDGs. This study not only fills in the gaps of 8 Tier II (no data) in CPTC, but also identifies existing sustainability issues and offers targeted recommendations for sustainable development.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Publicly available datasets were analysed in this study, specific data sources can be found in Tables 1 and 2.

Additional information

Funding

This work was supported by the National Key R&D Program of China (2022YFF0711600); the Construction Project of the China Knowledge Centre for Engineering Sciences and Technology (CKCEST-2023-1-5); the Natural Science Foundation of Jiangsu province named as ‘Hyperspectral inversion of photosynthetic and physiological indicators for Suaeda salsa based on UAV in coastal wetland under water and salt stress’ (BK20221397).