Introduction

Urban human settlements are complex systems that incorporate various factors including social policies, economic development, population, and resources. Human-centred urban settlements are the foundation for their survival and development (Li et al., 2010). Rapid urbanisation is an important opportunity for urban development; however, it has also created significant resistance to sustainable urban development, with problems such as the urban heat island effect, air pollution, and low efficiency of construction land—all of which have a negative impact on the quality of urban human settlements (Li et al., 2023; Liu et al., 2022; Neff et al., 2022; Ren et al., 2023; Xiao et al., 2022; Yao et al., 2022). Cities play a vital role in achieving the United Nations Sustainable Development Goals (SDGs) as their key spatial carriers (M. Chen et al., 2022). The rapid expansion of urbanisation is not synchronous with the development of human settlements, and the quality of human settlements has gradually become an important measure of whether a city is moving towards sustainable development (Li, Zhu et al., 2023). Therefore, with the combined background of growing urbanisation and the SDGs, the study of urban human settlements has become subject of considerable importance to the academic community.

Since Wu Liangyong proposed the concept of ‘Human Settlements Science’, there has been significant development in both theoretical and practical research (Wu, 2001, 2011). The study of human settlements is an open, interdisciplinary system that includes geography, ecology, social sciences, and economics. Research data on urban human settlements have gradually transitioned from initial questionnaire surveys and statistical data to remote sensing data and network big data, resulting in a transformation from traditional statistical data to multisource big data and compensating for the limitations of poor accuracy in traditional statistical data (Yang, You et al., 2020). Research methodologies have shifted from initial quantitative geographic models to new econometric models and big data analysis techniques (X. Li et al., 2022). For example, regarding the spatial differentiation of human settlements, GeoDetector, geographically weighted regression analysis, and space syntax models have emerged as primary methods. Research on human settlements involves theoretical underpinnings, spatiotemporal evolution characteristics, and relationships with the population and the economy (Guan et al., 2022a; Li et al., 2015; Tang et al., 2017; Xue et al., 2021; Yu et al., 2023). Regarding future trends in human settlements, data are gradually becoming multi-sourced, and the scale is gradually evolving from macro and medium-sized scales, such as global, regional, urban agglomerations, and cities, to micro scales, such as buildings and grids (Chen et al., 2017; Yang and Zhang, 2016). Human settlements directly impacted living standards and regional coordinated development, which are critical for increasing urban competitiveness and sustainable development.

Human settlements suitability (HEI) is a critical component of human settlements and encompasses both natural and humanistic elements. Natural elements are the foundation that influences HEI, whereas humanistic elements are the adaptation and change of the substrate by humans (Guan et al., 2022b; Liu et al., 2023). Currently, the study of HEI is primarily based on a natural perspective, using Feng Zhiming’s comprehensive evaluation framework of HEI, superimposing natural elements such as topography, climate, hydrology, vegetation, and the selection of watersheds and urban agglomerations as the study area to obtain the results of the evaluation of natural HEI (Feng et al., 2008; W. Li et al., 2022). Some researchers have added elements that influence HEI, enriching their theoretical bases and research frameworks. For example, Luo et al. (2021) investigated the HEI in Chongqing’s main urban area from the perspective of urban ventilation, which is critical for urban planning and environmental management. Traditional studies on humanistic HEI are primarily based on statistical data, which do not show spatial differentiation characteristics within statistical units. Geospatial big data, such as networks and remote sensing images, have become an inevitable trend in conducting HEI research, compensating for poor accuracy and providing effective data support for revealing differences in HEI within administrative units (Kang et al., 2021; Long et al., 2018; Ta et al., 2016). To assess the development status of HEI, the combined impact of natural and humanistic elements should be considered, which will not only facilitate the formulation of effective urban development policies and planning recommendations by governmental departments but will also contribute to decision-making support for improving the human settlements quality.

Grid-scale HEI research focuses mostly on natural components such as climate, topography, vegetation, and hydrology, with less emphasis on humanistic factors. Further, insufficient attention has been paid to the characteristics of HEI at different urban scales and climate zones and their spatial disparities along urban–rural gradients (URGs). Provincial capital cities are typically the region’s political and administrative centres, as well as key drivers of regional economic development, with well-developed infrastructure. Consequently, this study fully exploits the benefits of integrating data from multiple sources, proposes a framework for assessing HEI based on integrated natural and humanistic elements, reveals the spatial pattern of HEI in China’s provincial capital cities, investigates the differences in HEI at different urban scales and climate zones, focuses on the analysis of HEI along the URGs, and summarises the intrinsic mechanism of the differences in HEI. This will provide strong decision-making assistance for urban spatial planning, ecological civilisation formation, and sustainable urban development.

Materials and methodology

Study area

The study area comprises 31 provincial capital cities in China (Fig. 1), including 4 municipalities directly under the central government and 27 provincial capitals and autonomous region capitals, belonging to five urban scales and five climate zones (tropical monsoon, subtropical monsoon, temperate monsoon, plateau mountain, and temperate continental climate zones). Provincial capital cities are characterised by urbanisation level, high population density, and economic development. Exploring the HEIs of provincial capital cities can provide scientific references for urban planning and governance, which is critical to the SDGs 11: Sustainable cities and communities.

Fig. 1: Geographical location of the study area.
figure 1

There are five types of urban scales, including super city (6), megacity (10), type I large city (9), type II large city (5), and medium-sized city (1).

Full size image

Selection of elements

Natural elements are the primary factors affecting human settlements, and humanistic elements are important for sustainable urban development. To comprehensively evaluate the HEI in provincial capital cities in China, a framework of evaluation elements, including natural and humanistic elements, was constructed.

Selection of natural elements: (1) Terrain (DEM): Terrain conditions are important for characterising regional topography and geomorphic features on a macro scale, as they influence air circulation, transportation, and building conditions. (2) Vegetation (Enhanced Vegetation Index, EVI): Vegetation is an essential indicator of ecosystem quality and has been frequently employed in research on global climate change and agricultural monitoring and assessment (Banerjee et al., 2023). (3) Land surface temperature (LST): The spatial distribution of LST represents the spatial heterogeneity of the energy balance and surface processes and is useful for understanding the regional energy balance and climate change (Yang, Wang et al., 2020). (4) Precipitation (PRE): The amount of precipitation can accurately characterise the region’s basic meteorological conditions and indirectly reflect the spatial characteristics of regional water resources. (5) Frequency of geological disasters (FGD): Geological disasters have an obvious impact on transportation, harm the ecological environment, restrict the availability of land resources, and threaten the safety of people’s lives and property (Uehara et al., 2022).

Selection of humanistic elements: (1) Economic development level (NTL): Nighttime light data reflect the intensity of human activity, economic development, and urbanisation level, and studies have demonstrated a strong relationship with socioeconomic variables, such as Gross Domestic Product (GDP), population, poverty, and carbon emissions (J. Chen, M. Gao et al., 2022; Yang et al., 2022). This study uses nighttime light data to represent the spatial pattern of economic development in provincial capital cities. (2) Built-up area distance (BA_ED): Based on nighttime light data, the threshold method was used to determine the range of built-up areas, and the Euclidean distance algorithm was used to calculate the distance from each grid to the built-up area of the capital city. (3) PM2.5: PM2.5, which can reduce urban air quality and impair the normal life and productivity of urban residents (Yue et al., 2021). (4) Road density (RD): The integrity of urban road networks can significantly increase the accessibility of urban transportation and shorten commuting times, thereby facilitating productive activity. This study used RD to assess the status of urban transportation networks. (5) Point-of-interest (POI) kernel density: POI is a fundamental data point based on location services that are commonly used in urban research, such as the identification of urban functional zones (Luo et al., 2023; Y. Chen and J. Yang et al., 2022). POI kernel density indicates the distribution of different POI types inside the city, revealing the heterogeneity of the spatial structure and industrial organisation (Xue et al., 2020; Yu and Xiao et al., 2023).

Data and processing

The data materials used in this study include remote sensing images (NTL, LST, EVI, and PRE), POI, geological disasters, and urban road network data, which are detailed in Table 1.

Table 1 Data sources and descriptions.
Full size table

Methodology

The framework of this study consists of three parts: (1) Constructing a basic database of natural elements (DEM, LST, EVI, PRE, and FGD) and humanistic elements (NTL, POI kernel density, RD, BA_ED, and PM2.5) for HEI using the batch processing method of remote sensing data after pre-processing the data with projection transformation and cropping. (2) Using the ArcGIS fishnet tool to create a grid of 1000 m * 1000 m as the basic unit, we calculated the correlation between each element in the grid and the population based on the random forest method and assigned weights. The spatial patterns of the HEI were statistically measured using natural elements, humanistic elements, and weights. (3) Using GIS spatial analysis methods, we identified variances in HEI across different urban scales and climate zones, with a focus on studying the patterns of HEI along the URGs.

Kernel density estimation (KDE)

POI data contain spatial geographic information on urban infrastructure, and spatial distribution patterns and clustering trends are the primary means of assessing the urban public infrastructure level. Data on geological disasters in China, including avalanches, ground collapse, landslides, and mudslides, can be used to measure the frequency of geological catastrophes. After the above data were screened and cleaned, an appropriate density index was created using the KDE method, and density index for each grid was partitioned and tallied.

KDE is the most intuitive method for measuring the degree of agglomeration and was used to estimate the smoothed empirical probability density function. It is widely used to delineate urban functional zones and POI recommendations (Meng et al., 2024; Zhang et al., 2021). The formula is as follows:

$${\rm{f}}\left({\rm{s}}\right)=\mathop{\sum }\limits_{i=1}^{n}\frac{1}{{h}^{2}}\Phi \left(\frac{s-{c}_{i}}{h}\right)$$
(1)

In the formula, \({\rm{f}}\left({\rm{s}}\right)\) is the KDE function located at position \({\rm{s}}\); \(h\) is the attenuation value (bandwidth); \({c}_{i}\) is the position of the \({{\rm{i}}}^{{th}}\) POI; \({\rm{n}}\) is the number of POI points whose path distance from position \({\rm{s}}\) is not higher than \(h\); and \(\Phi\) is the predetermined kernel function.

RD

The progressive improvement of urban transportation networks serves to improve living standards and promote regional economic development (Chen et al., 2020; Zarrinpanjeh et al., 2022), hence it can be considered a factor influencing HEI. In this study, RD was used to characterise the convenience of an urban transportation network, which is defined as the ratio of the total length of roads within a certain range to the area of the range. The intersection was obtained using open street map road vector data and a 1000 m * 1000 m grid, followed by the resampling of partition statistics to generate grid-based RD. The formula used is as follows:

$${{RD}}_{i}=\frac{\sum {L}_{i}}{{A}_{i}}$$
(2)

where \({{RD}}_{i}\) is the density of the road network, \({L}_{i}\) is the length of the road in the grid (km), and \({A}_{i}\) is the grid area.

HEI model

Determining the weights of elements is an important step in the scientific and rational assessment of HEI. Given the elements’ different scales and cross-sectional comparability, they were standardised using the extreme-difference standardisation approach. To determine the impact of various elements on HEI, a correlation analysis between each element and the population density was performed using the random forest method. The indicator weight of a single element was calculated based on the percentage of the correlation coefficient between a single factor and population density to the overall correlation coefficient. The formula for the HEI (ranged 0–100) is as follows:

$$\begin{array}{l}{\rm{HEI}}={\rm{a}}* {\rm{NDEM}}+{\rm{b}}* {\rm{NLST}}+{\rm{c}}* {\rm{NEVI}}+{\rm{d}}* {\rm{NPRE}}+{\rm{e}}* {\rm{NFGD}}+{\rm{f}}\\\qquad\qquad\qquad\qquad* {\rm{NNTL}}+{\rm{g}}* {\rm{NPOI}}+{\rm{h}}* {\rm{NRD}}+{\rm{i}}* {\rm{NBA}}\_{\rm{ED}}+{\rm{j}}* {\rm{NPM}}2.5\end{array}$$
(3)

in which \({\rm{NDEM}}\), \({\rm{NLST}}\), \({\rm{NEVI}}\), \({\rm{NPRE}}\), \({\rm{NFGD}}\), \({\rm{NNTL}}\), \({\rm{NPOI}}\), \({\rm{NRD}}\), \({\rm{NBA\_ED}}\), and \({\rm{NPM}}2.5\) represent standardised DEM, LST, EVI, PRE, FGD, NTL, POI kernel density, RD, BA_ED, and PM2.5 data, respectively; \({\rm{a}}\), \({\rm{b}}\), \({\rm{c}}\), \({\rm{d}}\), \({\rm{e}}\), \({\rm{f}}\), \({\rm{g}}\), \({\rm{h}}\), \(i\), \({\rm{j}}\) are the corresponding weights of each element. Based on the natural breakpoint method, HEI is divided into five zones: highly suitable (>26.41), comparatively suitable (17.63–26.41), generally suitable (11.25–17.63), critically suitable (6.46–11.25), and unsuitable zones (<6.46).

Results

Spatial patterns of HEI

The HEI of China’s provincial capital cities ranges from 0 to 51.54, with an overall gradual increase from northwest to southeast (Fig. 2). The high-value areas of HEI all appear in built-up areas and account for a relatively small proportion; the medium-value areas gradually extend outward with the high-value areas as the centre, which are potential areas for improvement of HEI; low-value areas are generally found in unsuitable areas, such as lakes, rivers, and forests. Public and transportation facilities in built-up areas were far better than those in surrounding areas, and the suitability of humanistic attributes was higher, resulting in a higher HEI in built-up areas than in surrounding areas. Concurrently, HEI is closely related to the city’s transportation development, and HEI shows the spatial distribution characteristic of ‘strip type’, which is specifically manifested in the trend of decreasing along the city’s main transportation routes to the surroundings. Further, coastal cities have geographical location advantages, better living conditions, better public infrastructure levels, and an oceanic climate that makes the human body more comfortable. The HEI in coastal cities is generally higher than those in inland cities, which is consistent with previous research (Fang et al., 2022).

Fig. 2: Spatial distribution characteristics of HEI in 2020.
figure 2

There are five categories: highly suitable (>26.41), comparatively suitable (17.63–26.41), generally suitable (11.25–17.63), critically suitable (6.46–11.25), and unsuitable zones (<6.46).

Full size image

Spatial differentiation characteristics of HEI at different urban scales and climate zones

According to the spatial pattern of the HEI, spatial differences were primarily concentrated in the urban built-up areas. Therefore, this study focused on the built-up areas of provincial capital cities and investigated the differences in HEI across urban scales and climate zones, with an emphasis on assessing HEI patterns along the URGs. Based on nighttime light data, the threshold approach was used to obtain the built-up area ranges of the provincial capital cities, which were then integrated with land-use data to verify their accuracy. To analyse HEI patterns along the URGs, the first step was to reveal the difference in the mean HEI within the built-up areas (the average value of all pixels within the provincial capital cities’ built-up areas), as well as to clarify the difference in the proportion of HEI categories, which includes three perspectives: each city, different urban scales, and different climate zones.

Spatial differentiation characteristics of HEI in each city

Overall, the mean HEI value inside the built-up areas was 8.81, indicating a generally suitable zone (Fig. 3a). Haikou, Guangzhou, and Changsha are the top three cities for average HEI in built-up areas, and the bottom three are Yinchuan, Tianjin and Urumqi. Sixteen provincial capitals exceeded the average HEI for all cities, while 15 did not. As shown in Fig. 3b, all cities had the highest percentage of unsuitable zones, followed by critically suitable zones. The percentage of highly suitable zones was the lowest. Cities such as Changsha, Haikou, and Shijiazhuang had a comparatively high percentage of generally suitable zones, which resulted in high mean HEI. Only six provincial capital cities (Haikou, Changsha, Shijiazhuang, Zhengzhou, Guangzhou, and Hangzhou) had a proportion above the generally suitable zone level, which was greater than 30%. This indicates that the HEI level in China’s provincial capital cities is relatively low and requires further improvement.

Fig. 3: Spatial differentiation characteristics of HEI at different urban scales and climate zones.
figure 3

Average HEI within urban built-up areas: a every city, c different urban scales, d different climate zones; Classification percentage in HEI within urban built-up areas: b every city, e different urban scales, f different climate zones.

Full size image

Spatial characteristics of HEI at different urban scales

In terms of the HEI in built-up areas of various urban scales, the pattern was that the HEI steadily increased as the urban scales expanded (Fig. 3c). Specifically, the average HEI in the built-up areas of megacity was the highest (9.21), followed by super city, type I large city, type II large city, and Medium-sized city. As shown in Fig. 3e, regardless of the urban scales, the percentage of HEI categories was in the following descending order: unsuitable, critically suitable, generally suitable, comparably suitable, and highly suitable zones. Regarding the percentage above the generally suitable zone, megacity > type II large city > super city > type I large city > Medium-sized city, with each accounting for less than 30%.

Spatial characteristics of HEI at different climate zones

For different climate zones, the difference in HEI inside built-up areas was significantly greater than that at the urban scales (Fig. 3d), as indicated by the following order of climates: tropical monsoon (10.81), subtropical monsoon (9.06), temperate monsoon (8.74), plateau mountain (8.27), and temperate continental (6.77). Built-up areas in tropical monsoon climate zones were the most suitable for human habitation, while those in temperate continental climate regions were the least suitable. This discrepancy can be attributed to the warm and humid tropical monsoon climate, which is conducive to human comfort, whereas the temperate continental climate is cold and dry, making it unsuitable for human activity. Figure 3f shows the percentages of HEI categories in various climate zones. In the unsuitable zone, temperate continental climate had the largest percentage (62.23%), while tropical monsoon climate had the lowest percentage (33.33%). The percentage of critically suitable zones with a subtropical monsoon climate and a plateau mountain climate was greater than 30%, which is high. Only the tropical monsoon climate accounts for more than 40% of at or above the generally suitable zone, while the subtropical monsoon, temperate monsoon, and plateau mountain climates account for more than 20% of at or above the generally suitable zone. Because of the high percentage of unsuitable zones, the percentage of temperate continental climates over the generally suitable zone level was only 15.03%, suggesting that HEI’s overall performance is poor.

Spatial patterns of HEI along the URGs

We concentrated on the centroid of urban built-up areas and establish 20 1-km buffer zones to quantitatively assess HEI discrepancies along the URGs. Overall, all provincial capital cities showed a decreasing tendency in HEI as the distance from the city centre increased; however, the degree of decline varied. (Fig. 4). As the distance from the city centre increased by 1 km, the HEI in Beijing, Shanghai, Tianjin, Guangzhou, Chongqing, and Chengdu decreased by an average of 1.05, 1.39, 0.36, 0.33, 0.67, and 1.57, respectively. Notably, in Tianjin and Guangzhou, the HEI of Tianjin first fluctuated and rose to 7 km, and then gradually decreased to level off at 14 km until stabilising, whereas the HEI of Guangzhou fluctuated more but generally showed a decreasing trend. The HEIs in Changsha, Hangzhou, Kunming, and Wuhan first increased slowly and then declined progressively until stabilisation. The HEIs of Harbin, Jinan, Nanjing, Shenyang, Xi’an, and Zhengzhou showed a considerable reduction, followed by a levelling-off trend, reducing on average by 0.90, 0.90, 0.99, 1.33, 1.05, and 1.19, respectively, with a 1 km increase in distance from the city centre. As the distance from the city centre increased by 1 km, the HEIs of Changchun, Fuzhou, Guiyang, Hefei, Nanchang, Nanning, Shijiazhuang, Taiyuan, and Urumqi gradually decreased and finally levelled off. As the distance from the city centre increased by 1 km, the HEIs of Changchun, Fuzhou, Guiyang, Hefei, Nanchang, Nanning, Shijiazhuang, Taiyuan, and Urumqi gradually declined and ultimately flattened. The HEIs of Haikou, Hohhot, Lanzhou, Xining, and Yinchuan decreased on average by 0.71, 0.89, 0.37, 0.52, and 0.36, respectively, with an increase of 1 km in distance from the city centre, and the magnitude of their decrease was significantly smaller than that of megacity. For Lhasa, its HEI decreased by 0.38 on average as the distance from the city centre increased by 1 km, and the downward tendency slowed as it reached 9 km, gradually returning to the equilibrium state. Differences in HEI along the URGs reflect disparities in the infrastructure level and economic development between urban and rural areas, to which the government, urban planners, and other sectors of society should pay more attention.

Fig. 4: HEI along the URGs.
figure 4

The Y-axis represents the average HEI, and the X-axis represents the URGs; and the blue solid line is the difference curve of HEI along the URGs, and the red dotted line is its fitting curve.

Full size image

As illustrated in Fig. 5a, megacity had the greatest variation in HEI between urban scales along the URGs, followed by super city, Type I large city, Type II large city, and Medium-sized city. This indicates that the HEI in megacity centres is significantly higher than those in peripheral areas. This is closely related to the focus of megacity planning and governance, economic development, and infrastructure construction being more tilted towards the city centres, which may also lead to the unbalanced development of megacity human settlements in the inner-city space. The average difference in the HEI along the URGs was closely related to the climate zones (Fig. 5b), as indicated in the following order: temperate monsoon > subtropical monsoon > tropical monsoon > temperate continental > plateau mountain. Temperate, subtropical, and tropical monsoon climate zones are more suitable for human life and development because of factors such as urbanisation, and the average difference in HEI along the URGs is quite substantial. The temperate continental climate zone is primarily located in northwest China, with a dry environment and poor overall climatic conditions, leading to a relatively low average HEI difference along the URGs. Owing to the high altitude and unique geographic characteristics of the plateau mountain climate zone, the HEI difference between the urban core area and the outer locations was slight.

Fig. 5: Average differences in HEI along the URGs.
figure 5

a different urban scales and b different climate zones.

Full size image

Discussion

The mechanism in spatial patterns of HEI along the URGs

The spatial differences in HEI are primarily concentrated in urban built-up areas, with urban scales and climate zones playing key roles. Based on the variations in HEI across urban scales and climate zones, this study focused on investigating the mechanisms of HEI along the URGs.

The impact of urban land-use types on HEI

There is a strong correlation between the different urban land-use types and HEI. Reasonable urban land-use planning and layouts are critical components for improving HEI and ensuring sustainable urban development (Yu, Yung et al., 2023). The HEI varies significantly between land-use types (Fig. 6). Business office land is usually located in city centres or core business districts, which have a superior geographic position and a high infrastructure level, leading to the highest level of HEI. The HEI of medical and health land, transportation station land, commercial service land, sports and cultural land, residential land, administrative office land, and educational research land decreased sequentially, which usually considers land types, transportation accessibility, and the contribution of different land types to the HEI (Dorosan et al., 2024). Parks and green space land are typically distributed throughout cities or suburbs, resulting in relatively low transportation accessibility. Additionally, parks and greens pace land are surrounded by insufficient support facilities, resulting in low HEI. The poor HEI of airport facility land is closely related to airport facility land typically being located in peripheral urban areas, which are frequently associated with traffic congestion and land development policy restrictions (Da Silva et al., 2020). Industrial land has the lowest HEI of any land type for the following reasons: First, when other land-use types are converted into industrial land, the first issue is air pollution, which affects not only industrial land but also the surrounding area. Second, the continuous development of industry attracts several people to work, yet its own products require strong logistics and transportation capacities, which may cause traffic congestion and safety issues. Finally, because industrial land is mostly used for production, the land-use type is very simple and lacks infrastructure, thus lowering the HEI.

Fig. 6: Average HEI for different urban land-use types.
figure 6

There are differences in the average HEI based on different land use types, which increases from top to bottom.

Full size image

Consistency analysis of economic, population, and HEI

HEI is the foundation of regional development, with a profound interactive impact on the geographical layout of population and economic development. First, the GDP decomposition method based on nighttime light data derived the spatial pattern of GDP in provincial capital cities (Wang and Sun, 2022). Second, based on the WorldPop dataset as the spatial distribution of the population, the random forest method was used to undertake a correlation analysis of the economy, population, and HEI. The HEI was consistent with the spatial pattern of the economy and population, with a decreasing trend from the city centre to the outside. This finding suggests that economically developed and densely inhabited locations tend to have higher HEI. As shown in Fig. 7a, b, there is a significant positive correlation between the HEI and economy and population, with correlation coefficients of 0.567 and 0.603, indicating that a developed economic level and gradual population agglomeration have contributed to the enhancement of the HEI, which is in accordance with previous studies (W. Chen et al., 2022). The strong positive correlation between economy, population, and HEI does not imply that economic development and population concentration will always promote HEI, which is influenced by a combination of factors including geographic location, climatic conditions, and ecological environment quality.

Fig. 7: Consistency analysis of HEI with economy (GDP) and population (POP).
figure 7

a Spearman correlation between HEI and GDP, b Spearman correlation between HEI and POP, and c cumulative percentage curves of HEI with GDP and POP.

Full size image

The cumulative percentage curve depicts the degree of difference between population, economy, and HEI; this study shows the cumulative percentage curve of HEI (X-axis) versus population and economy (Y-axis, Fig. 7c). These findings revealed that the population was concentrated in locations with generally suitable and comparatively suitable zones, whereas areas with higher economic development were primarily dispersed in areas with generally suitable zones. The cumulative percentage curves for population, economy, and HEI all have an “S” shape, but the trends are not entirely consistent. When the HEI ranges from 3 to 40, the cumulative percentage of GDP exceeds the cumulative percentage of the population, indicating that economic development is faster than population concentration in this range. The HEI exhibits a bidirectional relationship with economic development. On the one hand, stronger economic development implies better infrastructure and innovation capabilities, which contribute to HEI improvement. On the other hand, suitable human settlements are more likely to attract capable labour resources, thereby driving economic development even further. The HEI has a direct effect on the spatial pattern of population, and a better HEI may lead to a gradual agglomeration of the population, resulting in the steady growth of urban scales. Economy and population have significant impacts on HEI (Jia and Gu, 2017; Yang et al., 2018). Population agglomeration supports economic development and HEI. However, after the population exceeds a certain threshold and the economy is completely developed, transitional resource development and progressive degradation in the ecological environment quality occur, resulting in a decrease in the HEI (Xu et al., 2022).

Recommendations from multidimensional perspectives

Provincial capital cities, which serve as the driving forces of regional economic development, have well-developed industrial structures and robust urban competition. The study uncovered that the HEI of some cities increased and then decreased, indicating that the HEI of the city centre is not the highest value for the entire city but rather higher within a few kilometres of the city centre, reflecting the urbanisation problems of high population density and traffic congestion in the core of the built-up area. Further, previous research on HEI has paid little attention to the differences in HEI between different urban scales and climate zones as well as their characteristics along the URGs. Therefore, this study investigated the spatial differences in HEI from a multidimensional perspective and proposes recommendations and regulatory directions for HEI based on research perspectives (urban scales, climate zones, and URGs) with the goal of providing effective management and decision-making references for the sustainable development of provincial capital cities. The specific recommendations are as follows:

Different urban scales: Super city and megacity is essential locations for gathering population and economic growth, as well as areas with lower ecological environment quality. Consequently, these communities should establish a green economic development model that harmonises the economy, population, and ecological environment to mitigate environmental damage (Hao et al., 2023; Huang and He, 2023; Li and Hou et al., 2023; Nie et al., 2023). Simultaneously, efforts should be undertaken to promote weight loss and physical fitness in super city and megacity as well as to maintain population levels under reasonable control. Type I large city and Type II large city face challenges, such as unequal regional development and inadequate infrastructure (Q. Zhang et al., 2022; Wan et al., 2023). Therefore, it is necessary to improve the multidimensional and three-dimensional coordination mechanism of urban infrastructure to ensure long-term sustainable development. Medium-sized city (Lhasa) is among the world’s highest-altitude cities with limited urbanisation and economic development. In the subsequent urban development process, Lhasa should prioritise ecological environmental protection based on its natural characteristics, encourage population agglomeration and economic development, and generally improve its HEI (Chen et al., 2018).

Different climate zones: Regarding the tropical monsoon climate zone (Haikou), economic development has been relatively rapid in recent years, and future development should focus on environmental protection while creating a landscape pattern appropriate for the local climate zones to enhance the HEI (H. Zhang et al., 2022). Cities in the subtropical monsoon climate zone are concentrated in large areas south of the Qinling-Huai River and east of the Tibetan Plateau. They should maintain the momentum of economic development and the radiation effect on surrounding cities, achieving harmonious development between the economy and the natural environment (Ni et al., 2023; Y. Zhang et al., 2023). Temperate monsoon climate cities typically have good infrastructure; however, they also need to increase green coverage based on their climate characteristics, alleviate summer urban heat island effects (Cai et al., 2023), improve air quality, and use new energy to replace traditional coal for heating to alleviate environmental pollution in urban winters (Chen et al., 2023; Han et al., 2023; Wang et al., 2023). Lhasa and Xining belong to the plateau mountainous climate zone, and urban planning and administration must fully consider the climate background and terrain conditions with a focus on environmental protection. Simultaneously, Lhasa and Xining should develop a comprehensive infrastructure system for healthcare, culture, sports, and other areas with an emphasis on increasing the HEI of human factors. Northwest China has a temperate continental climate, and drought and water scarcity are major impediments to the transformation and development of agriculture and rural revitalisation in the region; thus, the primary goal of enhancing HEI is to focus on the management and use of water resources, as well as to achieve a sustainable supply of water resources from multiple sources (Rowe et al., 2017).

Along the URGs: The spatial pattern of economic development and public infrastructure levels in provincial capital cities is imbalanced, mainly reflecting the disparity between urban and rural areas (Zamir and Wang, 2023). To alleviate this imbalance, the following recommendations are proposed: (1) Urban areas: increase the greening areas, improve air quality, and create a liveable natural environment background (Wu et al., 2023); the environmental monitoring and assessment system should also be perfected to protect the safety and health of the urban environment; build an efficient, convenient, and excellent public transportation network, and encourage inhabitants to use the public transportation to alleviate traffic congestion, air pollution, and other urban issues (Alvarado-Molina et al., 2023; Harleman et al., 2023); establish an outstanding land management system, and reasonably optimise urban population distribution and landscape patterns (Zeng and Zong, 2023). (2) Transitional zones between urban and rural areas: enhancing the infrastructure level is the first task in this region to improve HEI through environmental education and publicity to raise residents’ awareness of environmental protection (B. Zhang et al., 2023). It will be necessary to consider the protection of the ecological environment and economic and social development, while simultaneously laying out the buildings and infrastructure in the transitional areas to improve the level of regional sustainable development. (3) Rural areas: the overall level of HEI in rural areas is not high, and there is an obvious gap with urban areas, which are mostly represented by poor infrastructure in rural areas and uneven economic development (Hu and Wang, 2020; Wang and Zhu, 2023). In accordance with its own ecological environment quality and level of economic development, China should promote the key tasks of gradually improving rural infrastructure in a coordinated manner and take the path of rural revitalisation, industrial integration, and development to enhance its core competitiveness. It should also increase policy support, promote the construction of information technology for the management of rural human settlements, and comprehensively enhance the level of appropriateness of HEI.

Limitations

This study comprehensively assessed the spatial pattern of HEI and investigated the differences in HEI between different urban scales, climate zones, and URGs perspectives, which can serve as a reference for the adjustment and optimisation of territorial spatial development patterns. However, owing to the limitations of data acquisition, this study only examined the current state of HEI in Chinese provincial capital cities in 2020, ignoring the evolution of a long-term series. In a follow-up study, the comprehensive impact of natural and humanistic factors on HEI should be fully considered, the differences in HEI under different preference perspectives in a long-term series should be investigated (Wang et al., 2017), and more targeted recommendations and strategies for the enhancement of human settlements should be put forward.

Conclusion

China’s provincial capital cities are currently undergoing rapid urbanisation and industrialisation, resulting in environmental challenges that must be addressed. Enhancing the human settlements quality has become a primary goal of contemporary urban development, planning, and governance. In view of this, this study used multisource data and combined natural and humanistic elements to comprehensively assess the current state of HEI in China’s provincial capital cities in 2020, analysed the differences in HEI across different urban scales and climate zones, and focused on revealing the spatial patterns of HEI along the URGs. The main conclusions are as follows:

  1. (1)

    Spatial pattern of HEI. The HEI in provincial capital cities in China ranged from 0 to 51.54, with a gradually increasing tendency from northwest to southeast, with obvious coastal directionality and decreasing spatial distribution characteristics along the city’s main transportation routes to the surrounding areas.

  2. (2)

    HEI varies significantly in built-up areas with different urban scales and climate zones. The average HEI in the built-up areas of provincial capital cities reached 8.81, which is generally suitable zone. The HEI essentially followed the trend of gradually increasing with the expansion of urban scales; the differences in HEI in built-up areas with different climate zones were significantly larger than the urban scales. The percentages of the HEI categories, in descending order, were unsuitable zones, critically suitable zones, generically suitable zones, comparatively suitable zones, and highly suitable zones.

  3. (3)

    HEI along the URGs. HEI within the built-up area of the provincial capital city has a declining tendency with increasing distance from the urban centre, and HEI variation along the URGs is strongly correlated with urban scales and climate zones. Further, the primary focus of this study is the mechanism of HEI along the URGs within the built-up areas of provincial capital cities. HEI varies significantly among various urban land-use types and has a significant impact on how the population and economy interact spatially.

Developing appropriate policies and strategies to improve the HEI of China’s provincial capital cities is of great significance for enhancing comprehensive competitiveness and sustainable development capacity as well as for the optimisation of the territorial spatial layout. Therefore, it is necessary to reveal the spatial heterogeneity of HEI based on multidimensional perspectives of various urban scales, climate zones, and URGs. In the future, a more comprehensive, objective, and scientific evaluation framework should be constructed by taking multidimensional factors into full consideration, and evaluation studies on HEI at different spatial scales should be conducted to provide powerful scientific references for the realisation of the SDGs 11: Sustainable cities and communities.