Comparative analysis of spatial patterns and driving factors of historic districts in the Chinese four ancient capitals

Abstract

The four ancient capitals of China—Beijing, Nanjing, Xi’an, and Luoyang—are rich in historical and cultural heritage. However, rapid urbanization poses significant challenges to conserving historic districts. This study analyzes their spatial patterns and driving factors across natural, economic, cultural, and ecological dimensions. Using single-factor analysis, hot and cold spot analysis, and optimal parameter probes, findings reveal that 63% of historic districts in Beijing, Nanjing, and Xi’an are concentrated in older urban cores, while peripheral areas exhibit fragmented, dispersed patterns. Luoyang with fewer historic districts, concentrated in specific areas. This study highlights driving factors influencing the spatial distribution of historic districts, historic sites, and cultural blocks. Economic factors predominantly shape spatial patterns, followed by ecological and natural influences, while cultural factors are particularly significant for cultural blocks. This study recommends strengthening regional protection, enhancing peripheral cultural conservation, integrating ecological restoration, and promoting multi-sector collaboration for sustainable historic district preservation.

Introduction

The four ancient capitals of China—Beijing, Nanjing, Xi’an, and Luoyang—symbolize the nation’s rich historical and cultural legacy, having witnessed the flourishing development of Chinese civilization¹. Urban culture, as a synthesis of material and spiritual elements throughout urban development, encompasses architectural styles, folklore traditions, and spiritual beliefs. Various historical periods and regional contexts have shaped distinctive urban cultures, reflecting their time’s social conditions and incorporating the heritage and evolution of earlier cultures. As a vital component of cultural heritage, historic districts preserve relatively intact architectural forms and traditional lifestyles, serving as both an epitome and a spiritual symbol of a city’s historical trajectory^2,3. Additionally, these districts retain essential urban memories, reflecting the unique characteristics of specific historical periods and embodying significant artistic, cultural, and economic values^4,5. However, with the accelerating pace of urbanization, historic districts face dual protection and development pressures. As areas bearing substantial cultural value and urban memory, many of these districts are in decline or at risk of disappearing. This weakens the transmission of social experiences and erodes the social cohesion of cities, potentially transforming them into fragmented spaces devoid of shared values^6,7,8,9. Thus, it is crucial to thoroughly investigate the spatial patterns of historic districts and their driving factors. Such an exploration provides a foundation for developing scientifically grounded conservation and development strategies and balancing urban cultural heritage’s sustainable preservation with modern urbanization.

Historic districts encompass not only streets, buildings, and landscapes at a material level but also embody a wealth of intangible cultural heritage (ICH), including historical memories and the spirit of place, which reflect values, esthetic sensibilities, lifestyles, and folk arts^10,11,12. As visual representations of a city’s historical evolution, these districts preserve non-renewable historical information and possess significant historical and cultural value^13,14. Their spatial composition includes natural landforms, architectural relics, street networks, stone carvings, historical legends, folklore, and festive rituals. These elements, shaped over time, collectively contribute to the unique local culture and identity¹⁵. In contemporary cultural development, artistic creations within historic districts, such as paintings, music, and handicrafts, serve as vehicles for cultural transmission and reflect significant artistic value^16,17.

In recent years, research on the spatial patterns of historic districts has deepened, particularly in the key areas of urban morphology, street networks, and building distribution. Scholars have extensively explored these dimensions using various methods^18,19,20. Lyu et al.⁵ analyzed the Yushan Historic District in China using spatial syntax, revealing the profound impact of historical evolution and urban development on its spatial structure. Similarly, Xu et al.²¹ studied the Baihuazhou Historic District in Jinan through spatial syntax, visual mapping, and survey analysis, discovering that its disordered architectural fabric and lack of spatial coherence resulted in poor accessibility, ultimately hindering its functionality and cultural transmission. Wu et al. examined the spatial distribution characteristics of hutongs in historic districts, identifying multiple clusters through sample square analysis and kernel density methods. Their findings illustrated the social and cultural clustering unique to hutongs and the complex interplay of social interactions and spatial organization²². Lotfata & Lotfata highlighted the detrimental impact of modern physical changes on the spatial structure of historic districts, emphasizing the urgency of preserving cultural heritage amidst modernization²³. Although these studies reveal the spatial characteristics and evolutionary trends of individual cities or specific historic districts, they are limited in providing systematic comparative analyses across multiple cities. Furthermore, advances in GIS and remote sensing technologies have furnished more precise tools for studying the spatial patterns of historic districts, enabling researchers to more clearly delineate their spatial structures and changes, thus offering a scientific foundation for the preservation and development of historical and cultural heritage.

In the analysis of the spatial patterns of historic districts, commonly used methods include spatial syntax^5,24,25,26, typological approaches^27,28, and kernel density analysis^22,29,30,31. While these methods accurately reveal spatial structures and building distributions, they also exhibit certain limitations. Spatial syntax excels at quantifying urban spatial structures, particularly the accessibility and connectivity of street networks, but it often overlooks non-physical factors, such as economic, social, and cultural influences³². Typological methods, on the other hand, summarize planar and formal characteristics through shape syntax, revealing the correlation between spatial morphology and cultural features, yet their quantitative analysis capacity remains limited. Kernel density analysis effectively identifies spatial hotspots and quantifies architectural agglomeration effects, reflecting socio-cultural phenomena. However, its statistical approach is mainly descriptive, providing insufficient insight into the underlying driving factors. In the study of the drivers of historic districts, qualitative analyses have predominantly been emphasized, with quantitative research relatively underdeveloped, leading to a lack of in-depth understanding of the driving mechanisms. Nonetheless, systematic research on drivers is essential for comprehending the spatial distribution of historic districts. Hu & Gong proposed a framework consisting of five drivers: livability, social, cultural, natural, and economic, which provides a valuable reference for assessing the sustainable redevelopment of historic districts in the United States¹⁰. Zhu et al. found that the spatial distribution of ancient sites in Shaanxi is predominantly concentrated within 10 kilometers of rivers, underscoring the critical role of water resources in shaping the formation and development of historic districts³³. Li³⁴ suggested that social development, natural environment, geographic elements, and economic and political factors collectively influence the spatial layout of ancient sites in China. Zhang et al.³⁵ in their study of the Changsha Historic District, further argued that optimizing land use patterns not only aids in preserving historic sites but also enhances accessibility, thereby promoting the sustainable development of historic districts. Overall, the driving factors of historic districts can be categorized into four main dimensions: natural environment, economy, history and culture, and ecology. A comprehensive exploration of these specific influencing factors is crucial for understanding the challenges faced by historic districts amid modernization. It provides a theoretical foundation for formulating effective conservation and development strategies. While previous research has yielded significant insights into historic districts’ spatial patterns and driving factors, most studies have been limited to single cities or specific historic districts, lacking systematic cross-city comparisons. Moreover, existing research on driving factors has predominantly focused on traditional villages, historical monuments, and industrial heritage, leaving the exploration of historic district drivers relatively underdeveloped. This study addresses this gap by comparing historic districts’ spatial patterns and driving factors in China’s four major ancient capitals.

This study integrates the driving force perspective of cultural ecology theory with the spatial construction perspective of spatial production theory to systematically analyze the spatial patterns and evolutionary mechanisms of historic districts (Fig. 1). Cultural ecology theory emphasizes how external factors—natural, ecological, economic, and cultural—jointly influence the formation, development, and structural transformation of these districts^36,37,38. According to this theory, the spatial configuration of historic districts is shaped not only by the natural environment and ecological carrying capacity but also by economic vitality, social interactions, and long-term cultural accumulation. As such, it provides a robust framework for explaining the spatial distribution, stability, and evolutionary trajectories of historic districts under various natural and economic conditions^37,38. However, a notable limitation of cultural ecology theory is its primary focus on external influences, which results in less attention being paid to the roles of policy interventions, capital flows, and planning mechanisms in spatial production. This gap restricts its capacity to fully explain the spatial dynamics of historic districts amid ongoing socio-economic transformations.

Fig. 1

Theoretical framework diagram.

In contrast, spatial production theory analyzes the spatial patterns of historic districts through a social construction lens, asserting that spatial practice, representations of space, and representational spaces together form the essence of spatial production^39,40,41. This theory emphasizes that space is a continuous process of formation, reproduction, and reconfiguration, with government planning, market forces, and social dynamics playing key roles in shaping the spatial form of historic districts. While spatial production theory provides valuable insights into how policy regulations, market mechanisms, and social interactions influence the spatial evolution of historic districts, its primary limitation is its relatively limited focus on ecological and natural factors in shaping spatial patterns.

By integrating cultural ecology theory and spatial production theory, this study establishes a complementary analytical framework that allows for a more comprehensive examination of both external drivers and internal spatial constructs affecting historic districts. Cultural ecology theory explains how natural, ecological, economic, and cultural factors influence the spatial form of these districts, while spatial production theory enhances this perspective by highlighting the role of policy, planning, and capital in shaping and reproducing space. Building on this integrated framework, the study not only identifies the spatial distribution patterns of historic districts under different environmental conditions but also explores how policy interventions, capital investment, and social interactions drive spatial transformations through spatial reproduction. By adopting this analytical approach, the study constructs a more holistic model of spatial evolution in historic districts, offering systematic theoretical support for their sustainable conservation and spatial planning within the broader context of historical and cultural heritage preservation.

The theory of cultural ecology was first proposed by Julian Steward in 1955 in his book The Theory of Cultural Change⁴². The theory has culture at its center and views nature, the economy, and society as its ecological environment³⁶, emphasizing the interdependence and interaction between culture and the ecological environment, which form the cultural ecosystem’s theoretical basis^43,44. In recent years, cultural ecology theory has been widely used in traditional village conservation⁴⁵, cultural landscape⁴⁶, and historic districts. Ling et al.⁴⁶ and Li⁴⁷ have applied cultural ecology theory to study red cultural areas, albeit with different focal points. Ling et al. examine the development of red tourism, analyzing how cultural ecological changes influence the integration of tourism resources⁴⁶. Li investigates the role of social, economic, and ecological factors in revitalizing red cultural resources, with particular attention to their implications for landscape design⁴⁷. Similarly, Huang et al.³⁷, Ha et al.³⁸, and Chen et al.⁴⁵ employ cultural ecology theory to explore the evolution of spatial structures and the influence of natural, ecological, and social environments on historical and cultural settlements. Huang et al.³⁷ emphasize architectural spatial adaptation, Chen⁴⁵ investigates the evolutionary mechanisms of spatial patterns, and Ha et al.³⁸ construct a broader cultural ecosystem framework. However, none of these studies conduct cross-regional comparisons, making generalizing common evolutionary patterns across different cultural spaces difficult. Additionally, several scholars, including Chen & Zhang⁴⁸, Huang⁴⁹, and Ling & Li⁵⁰, have applied cultural ecology theory to the conservation and renewal of historic districts, albeit with varying research emphases and methodologies. Chen & Zhang⁴⁸ and Huang⁴⁹ focus on the impacts of economic activities, social structures, and the physical environment on cultural preservation, proposing strategies for conserving and renewing cultural resources. However, these studies rely primarily on qualitative analyses and lack spatial quantitative measurements, limiting their capacity to identify the driving factors behind historic district transformations and their interrelationships. In contrast, Ling & Li⁵⁰ explore the coupled relationship between street space and social behaviors, examining how neighborhood environments shape residents’ behavioral patterns. Nonetheless, Ling’s study does not delve deeply into historic districts’ spatial patterns or driving mechanisms. Although these studies collectively adopt cultural ecology theory to analyze the conservation and renewal of historic districts, they exhibit several methodological limitations, including a reliance on single-case studies, a lack of cross-regional comparisons, and insufficient spatial quantitative analysis. To address these gaps, this research systematically compares the spatial patterns and driving mechanisms of different historic districts using GIS, hotspot analysis, and optimal parameter geodetector quantitative methods. By integrating spatial measurement techniques and cross-regional analysis, this study seeks to overcome the methodological shortcomings of previous research and contribute to a more comprehensive understanding of the spatial evolution of historic districts.

The spatial production theory proposed by Henri Lefebvre posits that space is not merely a product of natural geography and economic development but is actively constructed through the interplay of social, political, and cultural forces. This theory has been extensively applied to research on historic district renewal, spatial reproduction, and urban spatial evolution in recent years. Scholars such as Zhang & Ren⁵¹ and Yao et al.⁵² have explored how capital, policy, and social structures shape spatial patterns within the framework of spatial production theory. Specifically, Zhang & Ren investigate the transformation of industrial heritage into consumer spaces, analyzing how commercial development reconfigures social space and marginalizes industrial culture⁵¹. Meanwhile, Yao et al.⁵² examine spatial production processes in historic districts, focusing on how capital influences spatial reproduction and residents’ identity by interacting with physical, mental, and social spaces. In the study of spatial evolution of historic districts, Degen et al.⁵³, Zhang et al.⁵⁴, Cheng et al.⁵⁵, and Hu⁵⁶ all adopt spatial production theories, but with different research perspectives and focuses. Degen et al.⁵³, Zhang et al.⁵⁴, and Hu⁵⁶ examine the interaction mechanisms between capital, tourism, and historic districts, consistently identifying capital, power, and social actors as key drivers of spatial development. However, their analytical approaches differ. Degen et al.⁵³ adopt a postmodern consumerist perspective to compare the evolution of tourism spaces on Pingjiang Road and Shantang Street in Suzhou, revealing distinct modes through which consumerism influences historical spaces. Zhang et al.⁵⁴ investigate the reproduction of consumer spaces in Beijing’s historic districts, exploring how capital shapes spatial patterns and impacts residents’ spatial rights. Hu⁵⁶ examines the Jiangnan Canal historical and cultural district, focusing on how cultural tourism patterns influence spatial structures and tourists’ perceptions. In contrast, Cheng et al.⁵⁵ take a different approach by analyzing the spatial production process in the Xishi Temple neighborhood in Beijing, emphasizing the role of culture in shaping the spatial structure of historic neighborhoods through representational space. Unlike other studies that highlight capital as the dominant force in spatial evolution, Cheng et al.⁵⁵ argue that the influence of capital on spatial transformation is neither a straightforward boom nor bust. Instead, they underscore the moderating role of culture in spatial reproduction, thereby highlighting the complexity and plurality of spatial evolution in historic districts. While spatial production theory has contributed to a deeper understanding of how capital, power, and social relations reshape historic districts, existing studies have primarily focused on the social construction of space, often overlooking the role of policy and planning in shaping spatial patterns through institutional design and intervention. This gap highlights the need for further research on how governance frameworks, regulatory policies, and urban planning strategies interact with economic and social forces to influence the spatial evolution of historic districts.

From the perspective of driving forces in cultural ecology theory and spatial construction in spatial production theory, this research establishes a comprehensive analytical framework to systematically examine the influence of natural environment, economy, history, culture, ecology, and planning policies on the spatial distribution of historic districts and their driving mechanisms. By utilizing optimal parameter-based Geodetector methods, the study uncovers the underlying mechanisms driving these factors and elucidates their interrelationships. Additionally, the research provides an in-depth examination of how multiple factors interact to shape the evolution of spatial patterns in historic districts, thus offering a scientific basis for integrated planning and cultural heritage protection. The findings contribute to the sustainable development of historic districts by offering theoretical insights that can guide urban planners and policymakers in the practical preservation and rational utilization of historic and cultural spaces within contemporary urban contexts.

Hypothesis 1: The spatial patterns of historic districts, historic sites, and historical and cultural blocks in China’s four ancient capitals are shaped by the interplay of natural, economic, cultural, and ecological factors, exhibiting significant variations across different regions and developmental stages. Economic density fosters the concentration of historic districts in urban cores. In contrast, peripheral areas tend to develop a more dispersed and fragmented pattern due to ecological constraints and land use transformations. While cultural factors define the intrinsic value of historic districts, their spatial distribution is regulated mainly by economic growth and ecological suitability, resulting in dynamic spatial evolution.

Hypothesis 2: Ecological suitability plays a crucial role in the spatial stability and sustainability of historic districts, historic sites, and historical and cultural blocks, demonstrating dynamic variations across different scales. Areas with favorable ecological conditions exhibit better continuity and stability in historical spatial patterns. In contrast, regions with fragile ecosystems or significant environmental shifts are more susceptible to spatial fragmentation and functional decline. The influence of ecological suitability on different historical areas is scale-dependent, varying across specific regions and development stages and forming a complex interactive relationship with economic and cultural factors.

Methods

Study areas

This study selects four of China’s major ancient capitals—Beijing, Nanjing, Xi’an, and Luoyang—as the study area (Fig. 2), primarily based on their distinct advantages in historical significance, diversity of district types, and regional differences. First, these cities hold significant positions in Chinese history, boasting rich historical and cultural resources and profound historical heritage. Beijing, the ancient capital of five dynasties, has over 3000 years of urban history and has served as the capital for >860 years. It is also the current capital of the People’s Republic of China¹. The city’s urban layout is characterized by a symmetrical structure centered around its central axis, with hutongs and siheyuan forming the unique style of its old city. Nanjing, the ancient capital of the Six Dynasties and the Republic of China, has more than 2400 years of history. The city is encircled by mountains and water, where traditional districts are integrated with the natural environment^57,58. Xi’an, the ancient capital of 13 dynasties, has a 3100-year history and served as the capital for over 1100 years. As the starting point of the ancient Silk Road, the city is renowned for its city walls and bell and drum towers, which are organized within a distinctive grid-like street layout^59,60,61. Luoyang, the ancient capital of nine dynasties, boasts over 4000 years of urban history and more than 1500 years as a capital. Its unique historical and cultural identity is evident in its urban layout, which is centered around the Luo River and a street and alley system that mirrors the river’s flow¹.

Fig. 2

Study areas.

The historical migration of ancient Chinese capitals reveals a west-to-east and north-to-south trajectory, mainly following the Xi’an–Luoyang–Kaifeng–Beijing axis. From the Xia to the Northern Song Dynasty, capitals were primarily situated in the Central Plains, with Xi’an, Luoyang, and Kaifeng as focal points. In later periods, from the Southern Song to the Qing Dynasty, capital cities moved toward coastal regions, reflecting the southward shift of economic centers. This trend followed a path from Hangzhou to Nanjing and ultimately to Beijing, marking a significant transition in China’s political geography. Xi’an and Luoyang were instrumental in shaping early Chinese political history. At the same time, Nanjing emerged as a southern economic and cultural hub during the Northern and Southern Dynasties and the early Ming Dynasty. Beijing solidified its role as the political center of China from the Yuan Dynasty onward, maintaining this status through the Ming and Qing Dynasties.

Secondly, the diversity of historic district types in the four ancient capitals offers a multidimensional analytical perspective. These cities encompass many historic district types, including palaces, city walls, commercial areas, religious quarters, and traditional residential neighborhoods. These districts reflect the characteristics of urban planning and social functional differentiation across different historical periods, providing a rich dataset for studying the spatial patterns of historic districts and their driving factors.

Finally, regional differences. These four cities are geographically located in distinct cultural and environmental zones, each reflecting unique lifestyles and urban forms. Beijing is characterized by flat terrain in the northern part of the North China Plain. Nanjing benefits from a varied geographical setting with mountains and waterways in the lower Yangtze River region. Xi’an, located in the Guanzhong Plain, is naturally fortified by surrounding mountains, making it a strategic defense location. In the Central Plains, Luoyang historically served as a crucial transportation and military hub. Analyzing the spatial patterns and driving factors of these historic districts provides critical insights into the historical evolution and cultural characteristics of these ancient capitals, offering valuable references for the preservation and development of their historic districts.

The concepts of historic district, historic site, and historic and cultural block are closely related, complementary, and interdependent. A historic district is an area that preserves a certain number of historic buildings or structures, with its overall appearance well-maintained. On the other hand, a historic site refers to an area rich in historical relics where the cultural landscape is harmoniously integrated with the natural environment. Such sites authentically represent a particular historical period’s traditional style and regional or national characteristics while preserving a unique cultural environment. Meanwhile, as defined by relevant government regulations, a historical and cultural block is an area where historical buildings and cultural relics are concentrated. These blocks reproduce traditional patterns and historical features in a more comprehensive and complete form. These three concepts often intersect and overlap in research. For instance, some scholars consider historic and cultural blocks a specific type of historic site⁶². At the same time, other studies tend to group historic and cultural blocks and historic sites under the broader category of historic districts. For example, Wu & Lin examine the Beijing Guozijian historical and cultural block as a historic district⁶³. Article 28 of the Regulations for the Protection of Xi’an Famous Historical and Cultural Cities, issued on 5 April 2017, explicitly designates blocks within the ancient city walls, such as Beiyuanmen, Sanshuxue Street, Zhubashi, Defu Lane, and Xiangzimiao Temple, as historic districts. In protecting and inheriting historical and cultural heritage, historic districts, historic sites, and historical and cultural blocks are spatially interconnected and intertwined, collectively forming the city’s historical heritage and cultural landscape. To systematically explore changes in spatial patterns and the influence of driving factors, this study subdivides its research objects into three categories: historic districts, historic sites, and historical and cultural blocks. This classification enables a more precise analysis of the similarities and differences among the three spatial types and uncovers the differential impacts of driving factors at various spatial levels.

Data sources

This study utilizes a diverse range of multidimensional datasets, including statistical, geospatial, economic, historical, cultural, natural, and ecological data, to systematically analyze the spatial patterns and driving factors of historic districts in four ancient capital cities. The primary data sources on historic and cultural sites were drawn from several key planning documents, including the Design Guidelines for the Protection and Renewal of Beijing Historic and Cultural Districts, the Protection Plan of Xi’an Historical and Cultural Cities (2020–2035), the Nanjing Conservation Plan of Historic City (2010–2020), and the Luoyang Historical and Cultural City Protection Regulations. Geospatial data, such as latitude and longitude, were obtained from Baidu Maps and processed using ArcGIS to create point distribution maps of the historic districts. The administrative boundary data for the study area were sourced from the 2022 China’s Administrative District Vector Boundary Data provided by the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences (https://www.resdc.cn).

In the analysis of driving factors, the data were categorized into four main groups: natural, economic, historical and cultural, and ecological (Table 1). The natural environmental factors include the Normalized Difference Vegetation Index (NDVI, X7), Annual Precipitation (X8), Annual Mean Temperature (X9), Distance to River (X12), and Digital Elevation Model (DEM, X15), which capture the natural geographic characteristics of each city. Economic factors encompass GDP Density (X4), Population Density (X5), Night Light Intensity (X6), Distance to Prefecture-Level City (X13), and Distance to Roads (X14), reflecting the economic development and infrastructure of the regions studied. Cultural factors include the Number of A-Class Scenic Spots (X1), ICH Sites (X2), and Cultural Relics under National Protection (X3), which assess the cultural value and conservation status of the historic districts. Lastly, ecological factors such as the Land Use Diversity Index (X10) and the Percentage of Urban Land Use (X11) measure the ecological aspects of land use and urbanization levels within the cities.

Table 1 Data sources and attributes

Analysis of variance

To analyze the spatial differences between the four ancient capitals—Beijing, Xi’an, Nanjing, and Luoyang—and their drivers, we used analysis of variance (ANOVA). We collected data on historic districts and normalized them. ANOVA was then used to test the hypothesis that the means of these variables were significantly different in the four groups. The F-statistic was calculated to compare differences in means between groups and differences in means within groups. The results from the ANOVA and post-hoc tests were interpreted to understand the spatial differences and driving factors for each historic district. The analysis involved examining how historical, cultural, and socio-economic factors contributed to the unique spatial patterns observed in each capital. By identifying significant differences, we could draw meaningful conclusions about the underlying drivers shaping the historic districts.

Hotspot analysis

To identify and analyze spatial clusters of significant activity within the historic districts of Beijing, Xi’an, Nanjing, and Luoyang, we employed hotspot analysis using the Getis-Ord (${G}_{i}^{* }$) statistic. This method detects areas with high concentrations (hotspots) or low concentrations (coldspots) of specific attributes, such as historical sites, population densities, or economic activities. The ${G}_{i}^{* }$ statistic is calculated as follows:

$${G}_{i}^{* }=\frac{\mathop{\sum }\nolimits_{j=1}^{n}{w}_{{ij}}{x}_{j}-\bar{X}\mathop{\sum }\nolimits_{j=1}^{n}{w}_{{ij}}}{S\sqrt{\frac{n\mathop{\sum }\nolimits_{j=1}^{n}{w}_{{ij}}^{2}-{\left(\mathop{\sum }\nolimits_{j=1}^{n}{w}_{{ij}}\right)}^{2}}{n-1}}}$$

(1)

where x_j is the attribute value at location j, w_ij is the spatial weight between locations i and j, $\bar{X}$ is the mean of the attribute values, and S is the standard deviation of the attribute values. Areas with high ${G}_{i}^{* }$ values indicate statistically significant hotspots, while low ${G}_{i}^{* }$ values indicate coldspots. This analysis allowed us to visualize and interpret the spatial distribution of key factors influencing the historic districts, providing a deeper understanding of the spatial patterns and their underlying drivers.

Optimal parameter geodetector

To investigate the driving factors of spatial patterns in the historic districts of Beijing, Xi’an, Nanjing, and Luoyang, we utilized the Optimal Parameter Geodetector. This method assesses the influence of various factors on spatial heterogeneity by quantifying the relationship between the spatial distribution of a dependent variable and multiple independent variables. The core of this method is the power of determinant (PD) value, calculated as:

$${PD}=1-\frac{\mathop{\sum }\limits_{h=1}^{L}{N}_{h}{\sigma }_{h}^{2}}{{N\sigma }^{2}}$$

(2)

where L represents the number of strata (or classes), ${N}_{h}$ is the number of units in stratum h, ${\sigma }_{h}^{2}$ is the variance of the dependent variable in stratum h, N is the total number of units, and ${\sigma }^{2}$ is the variance of the dependent variable in the entire region. A higher PD value indicates a stronger explanatory power of the independent variable on the spatial distribution of the dependent variable. This method allows for the identification of optimal parameters that best explain the spatial patterns and helps in understanding the complex interactions between various driving factors and the observed spatial heterogeneity in the historic districts.

Results

Comparative analysis of spatial patterns of historic districts

The study reveals notable variations in the spatial distribution of historic districts across the four cities (Fig. 3A). In Beijing, 33 of the 43 historic districts are concentrated in the Dongcheng and Xicheng districts, which represent the historical and cultural core of the city. These areas house numerous historical relics, reflecting Beijing’s rich cultural heritage and its long-standing role as a center of power. Although geographically distant, the peripheral districts of Mentougou, Miyun, and Yanqing each contain two historic districts, suggesting that these areas still hold some historical significance. In contrast, Shijingshan, Haidian, Shunyi, and Fengtai districts each have only one historic district, while Chaoyang, Changping, Daxing, and Tongzhou have none, underscoring their secondary historical importance.

Fig. 3: Spatial distribution of historic districts, historic sites, and historical and cultural blocks.

A Spatial distribution of historic districts. B Spatial distribution of historic sites. C Spatial distribution of historical and cultural blocks.

In Nanjing, 39 of the 77 historic districts are located within the old city, with a significant concentration in Qinhuai District, which reflects its deep historical and cultural roots. Gulou, Xuanwu, and Qixia districts also contain a number of historic districts, though with less density. The peripheral areas of Jianye, Jiangning, and Liuhe show a more dispersed distribution, indicating their relatively marginal role in Nanjing’s historical development.

In Xi’an, 32 of the 45 historic districts are located in the old city, with the majority concentrated in Beilin, Lianhu, and Xincheng districts, which form the cultural core of the ancient capital. Fewer historic districts are found in the peripheral districts of Baqiao and Gaoling, where historic resources are more scattered, reflecting weaker preservation efforts in these areas.

Luoyang, in contrast, exhibits a much sparser distribution, with only two historic districts in total. One of these is located in the old city, indicating a limited concentration of historical resources. This sparse distribution can be attributed to Luoyang’s unique historical trajectory, which has hindered its ability to widely preserve its cultural heritage.

The spatial distribution of historic sites also demonstrates differences across the cities (Fig. 3B). In Nanjing, the highest concentration of historic sites is in Qinhuai District, followed by Gulou and Qixia Districts. Xuanwu, Yuhuatai, Pukou, and Liuhe districts each have between four and seven historic sites, with a more scattered distribution. Jianye and Jiangning districts have only two and one historic site, respectively, while Lishui and Gaochun districts have none, highlighting the uneven distribution of historic sites in Nanjing. In Xi’an, historic sites are mainly concentrated in Beilin, Lianhu, and Xincheng districts, followed by Yanta District. In contrast, Baqiao District has only two historic sites, while Gaoling, Shuangyi, Lintong, Chang’an, and Weiyang districts each have one. Yanliang, Zhouzhi, and Lantian districts have no historic sites.

Finally, the distribution of historic and cultural blocks reveals additional spatial patterns (Fig. 3C). In Beijing and Luoyang, the distribution of historic and cultural blocks mirrors that of the historic districts, indicating a strong concentration of historic resources. In Nanjing, these blocks are mainly concentrated in Qinhuai District, with two blocks each in Xuanwu and Gaochun Districts and one in Gulou District. Other districts have no historic and cultural blocks. In Xi’an, the historic and cultural blocks are predominantly located in Lianhu, Xincheng, and Beilin districts, with only one block per district, and none in the remaining districts.

An ANOVA revealed significant differences in the spatial distribution of historic sites, historic districts, and moderate differences in historic and cultural blocks across the four ancient capitals of China (Table 2). Specifically, the ANOVA results for historic sites produced a p value of 0.003 (≤0.05), indicating a highly significant difference in their distribution. Similarly, the p value for historic districts was 0.006 (≤0.05), showing a significant difference in distribution patterns. In contrast, the ANOVA results for historic and cultural blocks yielded a p value of 0.245 (>0.05), indicating no statistically significant difference. However, the Eta-square (η²) value was 0.11, suggesting that 11.0% of the variance in the data can be attributed to differences between the groups. Additionally, Cohen’s f value of 0.352 indicates a medium effect size, highlighting a moderate degree of difference in the spatial distribution of historic and cultural blocks.

Table 2 A one-way analysis of variance

The spatial distribution of historic districts reveals clear patterns of hotspot and cold spot areas (Fig. 4A). In Nanjing, 99% of the hotspot areas are concentrated in Gulou, Xuanwu, Qinhuai, Jianye, and Jiangning districts, representing the city’s highest concentration of historic and cultural resources. Additionally, 95% of the hotspot areas cover Qixia and Yuhuatai districts in Nanjing, as well as the Dongcheng district in Beijing, and the Lianhu and Beilin districts in Xi’an. This highlights the significant role these districts play in the preservation and distribution of historic districts. Furthermore, 90% of the hotspots are located in Beijing’s Xicheng District and Xi’an’s Xincheng District, underscoring the cultural importance of these areas. In contrast, 90% of the coldspots are concentrated in the Luolong District of Luoyang, reflecting a sparse distribution of historic districts and a more scattered presence of cultural resources.

Fig. 4: Analysis of cold hotspot in historic districts, historic sites, and historical and cultural blocks.

A Cold and hotspot analysis of historic districts. B Cold and hotspot analysis of historic sites. C Cold and hotspot analysis of historical and cultural blocks.

For the distribution of historic sites (Fig. 4B), 99% of the hotspot areas are concentrated in Nanjing’s Qixia, Pukou, Gulou, Xuanwu, Qinhuai, Jianye, and Jiangning districts, as well as Xi’an’s Lianhu, Xincheng, and Beilin districts. These areas form the core of historic and cultural resources in both cities, with a dense distribution of historic sites. In addition, 95% of the hotspot districts include Xi’an’s Yanta District, further emphasizing its significance. By contrast, 90% of the hotspot districts are located in Nanjing’s Liuhe District, which, though on the city’s outer periphery, still holds a notable concentration of historical sites.

In terms of historic and cultural blocks (Fig. 4C), 99% of the hotspots are concentrated in Beijing’s Chaoyang, Xicheng, Dongcheng, and Fengtai districts, which form the most densely populated areas of the city’s historic and cultural resources. Additionally, 95% of the hotspots are located in Haidian District, indicating its importance in the distribution of historic and cultural blocks. Although it is not as densely populated as the central districts, it remains culturally significant.

Optimal parameter selection

Before applying the optimal parameter geographic detector method to analyze the driving factors influencing the spatial patterns of historic districts, it is essential to discretize continuous factors to obtain optimal parameters. Figure 5 illustrates the grading standard, while Fig. 6 presents the interval numbers and corresponding q values for various classification methods. In Fig. 6, the horizontal axis represents the number of intervals, and the vertical axis represents the q values of the independent variables. The q values differ depending on both the classification method and the number of intervals. Specifically, for the same number of intervals, q values vary across different classification methods, and for the same classification method, q values change with varying interval numbers. Therefore, the combination of classification method and interval number significantly influences the q values. The optimal parameters for each factor are determined by the classification method and the interval number that corresponds to the maximum q value. For example, the q value for driving factors was maximized when using the quantile classification method with 5 intervals.

Fig. 5: Grading standard.

X1: A-class scenic spots; X2: Intangible cultural heritage; X3: Cultural relics under national protection; X4: GDP density; X5: Population density; X6: Nighttime light intensity; X7: NDVI; X8: Precipitation; X9: Temperature; X10: Land use diversity; X11: Percentage of urban land use; X12: Distance to river; X13: Distance to prefecture-level city; X14: Distance to roads; X15: DEM.

Fig. 6: Different classification methods and q values of the number of intervals.

X1: A-class scenic spots; X2: Intangible cultural heritage; X3: Cultural relics under national protection; X4: GDP density; X5: Population density; X6: Nighttime light intensity; X7: NDVI; X8: Precipitation; X9: Temperature; X10: Land use diversity; X11: Percentage of urban land use; X12: Distance to river; X13: Distance to prefecture-level city; X14: Distance to roads; X15: DEM.

Detection of driving factors in the historic districts

This study identifies the key factors influencing the spatial distribution of historic districts (Fig. 7A), historic sites (Fig. 7B), and historical and cultural blocks (Fig. 7C), revealing the differentiated roles of economic, natural, cultural, and ecological factors in shaping various types of historic spaces. Overall, the spatial patterns of these three types of historic areas are primarily driven by economic factors. In contrast, other factors exert varying degrees of influence depending on the specific type of historic space. Economic factors emerge as the dominant drivers across all three categories, with GDP density, Population density, and nighttime light intensity ranking among the top determinants. This suggests that urban economic vitality, infrastructure development, and population concentration play a decisive role in shaping the spatial distribution of historic areas. However, the influence of natural, cultural, and ecological factors varies significantly depending on the type of historic space. The spatial pattern of historic districts is predominantly shaped by ecological factors, with urban land use percentage (q value 55.29) and land use diversity (q value 41.76) ranking highly among the driving factors. This indicates that land use structure and urban planning are critical in defining the spatial organization of historic districts, influencing their multifunctionality and potential for sustainable development. In contrast, the spatial distribution of historic sites is primarily governed by natural factors, particularly temperature (q value 46.04) and precipitation (q value 34.22). These findings highlight the long-term impact of climatic conditions on the weathering of building materials, structural stability, and heritage conservation, suggesting that the spatial configuration of historic sites is closely linked to geographic adaptability and the preservation mechanisms of the natural environment. Meanwhile, the spatial pattern of historical and cultural blocks is predominantly driven by cultural factors, with ICH (q value 90.21) and national key cultural relic protection units (q value 63.35) ranking as the most influential determinants. This underscores the role of cultural policies, heritage tourism, and conservation initiatives in shaping the spatial distribution of historical and cultural blocks, positioning them as vital hubs for cultural resource clustering and the development of the cultural industry. Beyond economic factors, ecological variables also play a crucial role in shaping the spatial pattern of historic districts. Land use percentage (q value 55.29) and land use diversity (q value 41.76) significantly influence the spatial organization of these neighborhoods. The proportion of urban land use determines the spatial distribution of different land use types, particularly the allocation of cultural and historical land, which is critical for maintaining the spatial coherence of historic districts. Moreover, diverse land use patterns enhance the multifunctionality of these districts, integrating cultural, residential, commercial, and public service attributes. This integration, in turn, fosters the sustainable development of historic districts.

Fig. 7: Analysis of driving factors.

A Driving factors of historic districts. B Driving factors of historic sites. C Driving factors of historical and cultural blocks.

In the analysis of historic districts, the factors are ranked as follows: X5 > X4 > X6 > X11 > X13 > X7 > X10 > X9 > X8 > X3 > X2 > X12 > X15 > X14 > X1. Population density (q value 79.13, p < 0.001), GDP density (q value 71.95), and nighttime light intensity (q value 56.57%) are the most significant factors influencing the spatial distribution of historic districts. High population density areas typically serve as social and cultural hubs of the city, where abundant human resources and strong consumer demand drive the preservation and revitalization of historic districts. Similarly, areas with high GDP density tend to possess greater resource availability and economic vitality, providing essential funding for neighborhood renovation and functional optimization. Nighttime light intensity, as a key indicator of economic activity, further underscores the central role of urban vitality in shaping the spatial structure of historic districts.

Although natural factors like NDVI (q value 46.93%), temperature (q value 36.95%), and distance to the river (q value 35.53%) have weaker explanatory power, they still affect spatial distribution to some extent. Cultural factors exhibit limited explanatory power, with Cultural Relics under National Protection (q value 32.92%) and ICH (q value 25.04%) showing relatively minor effects. A-class scenic spots (q value 10.66%) have the least impact. Overall, economic factors, particularly population density, GDP density, and night light intensity, dominate the spatial distribution of historic districts. Ecological and natural factors are secondary, while Cultural factors have the least influence. These findings suggest that future urban planning should focus on balancing economic growth with the protection of cultural heritage.

In the analysis of historic sites, the factors are ranked as follows: X4 > X9 > X5 > X6 > X8 > X13 > X11 > X7 > X15 > X3 > X12 > X1 > X10 > X2 > X14. Economic factors are the dominant determinants of the spatial distribution of historic sites. Among them, GDP density (q value 61.64, p < 0.001), population density (q value 45.88), and nighttime light intensity (q value 38.82) further suggest that areas with high GDP density usually have resource advantages and active economic activities, which provide the necessary financial support for site maintenance and development. Densely populated areas are more likely to form clusters of historic sites, which may result from concentrated cultural demands and frequent social activities. Natural factors also play an important role in the distribution of sites. Among natural factors, temperature (q value 46.04) is a key variable significantly influencing the spatial patterns of historic sites. Rising temperatures and increased sunlight exposure accelerate the weathering of building materials, mainly stone and other fragile materials, making historic buildings more vulnerable to damage. Additionally, variations in temperature and relative humidity directly affect construction materials’ physical and chemical properties, exacerbating weathering and pollution. Precipitation (q value 34.22) further underscores the combined impact of natural factors on the spatial distribution of historic sites. Rainfall, particularly acidic rain, accelerates the chemical weathering of stone, leading to surface erosion and salt crystallization, exacerbating the deterioration of historic structures. Extreme weather events, such as heavy rain or drought, pose additional threats to cultural heritage, increasing the complexity of preservation efforts.

Ecological factors, such as the percentage of urban land use (q value 32.93%) and land use diversity (q value 12.77%), have a more limited explanatory power but still exert some influence. Cultural factors play a relatively weak role, with Cultural Relics under National Protection (q value 23.35%) having some effect on spatial distribution, while A-class scenic spots (q value 16.15%) and ICH (q value 9.74%) have minimal impact. Other variables, such as DEM (q value 26.55%), distance to river (q value 18.06%), and distance to road (q value 8.19%), show weak correlations, indicating a limited influence on the spatial distribution of historic sites. Overall, economic factors, particularly GDP density, population density, and night light intensity, are the primary drivers of historic site distribution. Natural factors, including temperature and precipitation, also have a notable impact, while ecological and historical-cultural factors play a lesser role.

For the spatial distribution of historic and cultural blocks, the factors are ranked as follows: X4 > X6 > X2 > X3 > X10 > X5 > X11 > X13 > X7 > X1 > X14 > X8 > X15 > X12 > X9. Specifically, GDP density (q value 92.95%, p < 0.001) is the most decisive factor influencing the spatial distribution of historic and cultural blocks. Night light intensity (q value 90.98%) also shows a strong correlation, highlighting the significant impact of economic activity. Historical and cultural factors, such as ICH (q value 90.21%) and Cultural Relics under National Protection (q value 63.35%), play a more prominent role here than in historic districts or sites. ICH, including traditional handicrafts, folk activities, and folk arts, is a vital carrier of the cultural essence of historic districts. It reflects a region’s unique historical and cultural values and acts as an important resource for cultural innovation and renewal. Cultural relics under national protection, as tangible carriers, directly represent the district’s historical value and spatial core position. These protected sites not only signify the cultural importance of the district but also influence its spatial patterns through their distribution characteristics. A-class scenic spots (q value 20.26%), while still important, have a relatively limited influence.

Ecological factors, including land use diversity (q value 61.96%) and the percentage of urban land use (q value 35.68%), indicate that land use structure significantly impacts the spatial distribution of cultural blocks. Natural factors, such as NDVI (q value 20.54%), precipitation (q value 16.41%), DEM (q value 15.56%), and distance to the river (q value 14.11%), have a minor influence, with temperature (q value 13.97%) having the least effect.

The interaction between multiple variables significantly enhances the explanatory power of spatial distribution patterns in historic districts (Fig. 8A). Based on q value analysis, the combined effects of several variables stand out. The q values for X5 ∩ X4, X5 ∩ X3, X6 ∩ X4, X6 ∩ X5, X8 ∩ X4, X8 ∩ X5, X9 ∩ X5, X10 ∩ X5, X11 ∩ X8, X12 ∩ X5, X13 ∩ X5, X14 ∩ X5, and X15 ∩ X5 are all above 80%. Notably, DEM and Population density (X15 ∩ X5), Precipitation and Population density (X8 ∩ X5), and Distance to road and Population density (X14 ∩ X5) rank the highest, with q values of 87.05%, 86.48%, and 84.25%, respectively. These results suggest that the interaction between population density and various physical and geographical factors plays a crucial role in shaping the spatial distribution of historic districts. In particular, the combination of population density and road distance has a particularly strong influence.

Fig. 8: Interactions between multiple variables.

A Variable interactions in historic districts. B Variable interactions in historic sites. C Variable interactions in historical and cultural blocks.

Furthermore, the interaction between GDP density (X4) and population density (X5) yields q values above 60%, demonstrating the robust impact of economic development and population agglomeration on the spatial patterns of historic districts. In contrast, the interaction between ICH (X2) and A-class scenic spots (X1) has a relatively weak explanatory power, with a q value of only 16.21%, indicating that these cultural resources have a limited influence on the spatial distribution of historic districts. Overall, the analysis reveals that economic factors, especially population density, significantly impact spatial distribution patterns, particularly when interacting with natural and geographical variables.

For historical sites, the interaction between multiple variables also significantly increases the explanatory power of spatial distribution (Fig. 8B). Q values for combinations such as X1 ∩ X5, X2 ∩ X5, X3 ∩ X5, X4 ∩ X5, X2 ∩ X6, X4 ∩ X6, X4 ∩ X8, X5 ∩ X8, X4 ∩ X13, and X5 ∩ X15 all exceed 70.00%. The interaction between population density (X5) and DEM (X15) ranks highest, with a q value of 81.58%, followed by GDP density (X4) and population density (q value 76.86%), and population density combined with precipitation (X5 ∩ X8) (q value 76.24%). These findings emphasize that population density, particularly when combined with physical factors like DEM and precipitation, significantly influences the spatial distribution of historical sites.

Weaker interactions, such as Distance to road and Cultural Relics under National Protection (X14 ∩ X3), Land use diversity and GDP density (X10 ∩ X4), and ICH and A-class scenic spots (X2 ∩ X1), have q values of 12.10%, 14.91%, and 16.9%, respectively, indicating their more limited effects on the spatial distribution of historical sites.

In the analysis of historical and cultural blocks, the interaction between variables also demonstrates strong explanatory power for spatial distribution (Fig. 8C). The q values for Land use diversity and Temperature (X10 ∩ X9), Distance to road and Precipitation (X14 ∩ X8), and DEM and Land use diversity (X15 ∩ X10) are particularly high, at 98.20%, 97.45%, and 97.14%, respectively. These results suggest that the interaction between ecological factors plays a significant role in determining the spatial patterns of historical and cultural blocks.

Moreover, the interaction between ICH (X2) and GDP density (X4) also shows strong q values, exceeding 0.9000, highlighting the dual role of cultural resources and economic development in the preservation and development of historical and cultural blocks. The q values for Cultural Relics under National Protection (X3) and Land use diversity (X10) are above 0.6000, further emphasizing their influence on the spatial distribution of these blocks.

Discussion

To verify the credibility of this research, this study referenced other related studies and compared their findings with its results. Chahardowli & Sajadzadeh identified economic factors as the most important drivers for developing the historical core and urban renewal in Hamadan city⁶⁴. Similarly, Wright & Eppink found that economic value is a standard driver of heritage value, with heritage sites in areas of higher population density being more valuable⁶⁵. These findings align with this study’s conclusion that economic factors significantly influence the spatial distribution of historic districts. Wang et al.⁶⁶ demonstrated that the spatial distribution of ICH in the Yellow River Basin is primarily driven by socio-economic factors, with natural environmental factors playing a minor role. This is consistent with this study’s conclusion that economic factors, such as GDP and population, dominate spatial distribution. However, the studies diverge in the role of natural factors. While this study found that natural factors such as land use diversity and NDVI have some influence on the spatial distribution of historic districts, these factors are less significant for ICH. This difference stems from the distinct dependencies of the two: ICH relies more on the cultural heritage environment, whereas the spatial distribution of historic districts is more closely linked to ecosystem structures. Zhang et al.⁶⁷ showed that the spatial distribution of industrial heritage in China is influenced by various factors, with GDP per capita having the highest q value among economic factors and elevation and average precipitation being the dominant natural factors. Conversely, the distance to city centers and rivers had the least influence. These findings are partially consistent with this study, as GDP density is also highly significant for the spatial distribution of historic districts. At the same time, the influence of distance to city centers and rivers is minimal. However, a key difference lies in the role of DEM. In the study of industrial heritage, DEM was a significant driver, whereas its influence on historic districts was minimal. This difference may be attributed to the functional attributes and formation mechanisms of the two types of heritage. Natural conditions often constrain industrial heritage sites, such as the need to avoid flood risks or access energy, which are influenced by topographic elevation. Therefore, elevation plays a critical role in the distribution of industrial heritage, unlike in historic districts.

This study provides an in-depth analysis of the spatial distribution of historic districts, historic sites, and historical and cultural blocks in China’s four ancient capitals—Beijing, Nanjing, Xi’an, and Luoyang. The findings reveal both significant concentrations and marked differences in the spatial patterns of these historic districts. The high concentration of historic districts in the old urban cores of Beijing, Nanjing, and Xi’an reflects the shifting political, economic, and cultural centers of these capitals throughout history^68,69,70,71. Historically, these cores were centered around clusters of significant buildings such as palaces, city walls, and major temples, forming concentrated areas of historical significance. However, with modern urbanization, the distribution of historic districts has become increasingly uneven. Historic resources in the old city centers are better preserved, while those in peripheral areas are more scattered, failing to form the same cohesive cultural aggregation seen in the urban cores^17,72. In some cases, these peripheral areas are at risk of losing their cultural heritage altogether.

Luoyang stands out as an exception due to its relatively sparse distribution of historic resources, which are concentrated in a few specific areas. This pattern reflects Luoyang’s unique historical development. As the capital of multiple dynasties, Luoyang has experienced numerous relocations and reconstructions, preventing the formation of highly concentrated historic districts like those in Beijing, Nanjing, and Xi’an. This distinct historical and cultural trajectory sets Luoyang apart from the other ancient capitals, highlighting the city’s divergent path in historical preservation⁷³. Overall, these differences in spatial distribution reflect each city’s unique historical evolution and illustrate regional variations in how historic districts are conserved amidst modern urbanization.

The study also underscores the central role of economic factors in shaping the spatial distribution of historic districts, historic sites, and cultural blocks⁷⁴. Economic activity, particularly in urban centers, has contributed to both the preservation and development of historical heritage. However, it has also introduced the risk of over-commercialization^75,76. The rising value of land in these areas has placed development pressure on historic districts, leading to a gradual weakening of their historical and cultural functions as they are increasingly repurposed for tourism and commercial ventures^77,78,79. Additionally, while cultural factors may appear to have a more limited impact in shaping spatial distribution, they are critical for sustaining the vitality of historic districts, particularly where strong cultural conservation policies are in place. Districts that retain traditional community structures and cultural activities demonstrate greater cultural continuity and social cohesion. By contrast, those that have been over-commercialized or overly modernized often lose their original historical atmosphere and cultural vibrancy.

Although natural environmental factors exert an indirect influence, they play an essential role in the spatial distribution and preservation of historic districts. Temperature and precipitation are critical drivers affecting historic sites, directly threatening building materials’ physical and chemical stability. Rising temperatures and increased sunlight accelerate weathering processes, especially for fragile materials such as stone and wood, making them more vulnerable to damage. Similarly, rainfall and acidic precipitation intensify the chemical weathering of stone, causing surface erosion and salt crystallization, significantly compromising historic buildings’ structural integrity⁸⁰. Extreme climatic events, such as heavy rainfall and droughts, cause more direct damage to cultural heritage. These events amplify building materials’ deterioration and disrupt historic districts’ ecological balance, diminishing their social functions and cultural appeal. For instance, in Xi’an, the semi-humid continental monsoon climate increases the difficulty of maintaining historic structures. Repeated freeze-thaw cycles lead to micro-cracks in stone walls, while high summer temperatures contribute to the deformation and degradation of wooden structures.

Additionally, the prolonged environmental pressures from climate change have caused the gradual disappearance of cultural features reliant on natural landscapes, significantly impacting these historic areas’ landscape character and cultural values. To address these challenges, enhancing vegetation cover to improve NDVI values offers a practical approach to ecological restoration. Vegetation serves as a natural buffer, mitigating the adverse effects of extreme climate conditions on building materials, such as reducing the erosive impact of heat and acidic rain. Moreover, it enhances the ecological quality of the area and strengthens its overall resilience, supporting the long-term preservation and sustainability of historic districts. These findings suggest that future conservation policies should prioritize not only the physical preservation of historic districts but also the maintenance of socio-cultural elements to ensure the long-term vitality of these areas. Specifically, efforts should focus on promoting social and cultural heritage in tandem with economic development to avoid eroding the historical identity of these districts.

In summary, this study highlights a close correlation between the uneven distribution of historic districts and socio-economic inequality. Economically developed urban cores tend to have better access to cultural preservation resources, while economically disadvantaged peripheral areas often suffer from a lack of policy support and financial investment. This leads to unequal outcomes in the preservation of historic and cultural resources. As a result, urban planning must pay greater attention to socio-economic equity to ensure that the benefits of historic district preservation are more evenly distributed across different social groups. Ultimately, the preservation of historic districts should not solely serve the goals of tourism and economic growth; it must also focus on safeguarding social and cultural heritage and enhancing the well-being of local communities.

The study found that the dominance of economic factors in the distribution of historic districts is closely related to the historical orientation of China’s national policies. In the early years of reform and opening up, China’s level of urbanization was low, and the government prioritized the expansion of cities and the development of new districts. In contrast, preserving old urban areas and historic districts was relatively neglected. In the 1990s, with the increasing tension on land resources and the problem of urban development coming to the forefront, the renovation of old cities became a key priority, and the need to preserve historic districts was progressively put on the agenda. Over time, international urban regeneration theories were introduced into China and gradually developed in localized practice, promoting a systematic process of preservation and regeneration of historic districts. Between 1949 and 1976, the preservation of historic districts was severely constrained by political turmoil and economic prioritization policies. Despite a small number of preservation measures, a large number of historic buildings were destroyed during this period⁵⁴. After the reform and opening up, with the approval of the Instructions on the protection of China’s historical and cultural cities in 1982 and the promulgation of the Cultural Relics Protection Law of the People’s Republic of China, the protection of historic districts has gradually moved towards a legalized track^81,82, however, pragmatism-orientated urbanization still poses a persistent threat. Entering the 21st century, policies and regulations have become more refined, and conservation efforts are gradually moving towards holistic and progressive organic renewal, reflecting China’s in-depth exploration of the conservation and sustainable use of historical and cultural resources^75,83.

Currently, several significant models are gradually taking shape in the practice of historic neighborhood conservation and renovation, including the government-led model^82,84, commercial development^75,85, dynamic renewal model^4,14, and the ecological districts model¹⁰. The government-led model tends to neglect the transmission of historical and cultural values despite improving land utilization. While promoting economic development, the commercial development model often leads to over-commercialization, causing historic districts to lose their original cultural atmosphere. Although the dynamic renewal model enhances the vitality and functional renewal of the blocks, it fails to effectively maintain the social and cultural vitality of the historic districts due to over-emphasis on development and modernization, resulting in the gradual weakening of the historical and cultural connotations. The ecological district model, on the other hand, advocates the preservation of the living functions of the aborigines, which is more sustainable in theory but more challenging to operate in practice. In practice, while these models promote economic vitality, they also inevitably bring the challenges of cultural connotation loss and social structure change. In terms of the evolutionary path of historic districts, alterations have gone through different stages, from commercial property development to leisure and entertainment consumption, and then to residential regeneration. Although these paths have enhanced economic vitality to a certain extent, they have also brought about problems such as weakening traditional cultural connotations and reducing community identity. Lack of policy planning and laws and regulations, neglect of ICH, homogenization brought about by commercialization, and weakening of traditional culture are all challenges that need to be addressed in the conversion of historic districts.

Comprehensive improvements and innovations are needed at the policy, legal, and social perception levels to achieve the sustainable development of historic districts. Firstly, policies to protect historic districts need to pay more attention to the integrity and sustainability of cultural heritage, avoiding over-commercialization and the loss of cultural connotations while maintaining economic vitality. Secondly, urban planning should pay more attention to the long-term impact of ecological factors to ensure that the sustainable development of historic districts and environmental protection go hand in hand. Finally, at the social level, public awareness and the enhancement of social and cultural values are crucial, and only with broad social participation can the cultural heritage of historic districts be genuinely protected and passed on.

To achieve comprehensive conservation and sustainable development of historic districts, policymakers should implement more detailed measures across several key areas:

Research shows that economic factors dominate the distribution of historic districts, with economically developed regions typically possessing more excellent preservation resources, while less developed areas suffer from weaker conservation capacity. This unequal distribution of resources places historical and cultural assets in economically disadvantaged regions at a higher risk of loss. For cities like Luoyang, where historical resources are relatively scarce, it is crucial to enhance the protection and utilization of less developed areas based on regional economic conditions and the distribution characteristics of historic districts. Adjusting resource allocation mechanisms to direct limited preservation funds and policy support to weaker areas can ensure more balanced protection of historical and cultural resources across urban regions. Additionally, providing these areas with technical support and introducing digital recording tools and monitoring systems can enhance their conservation capabilities and ensure the scientific and practical implementation of preservation measures.

Studies indicate an increasing imbalance in the distribution of historic districts, with historical resources in old city centers receiving better protection while those in peripheral areas are more dispersed. To address this, the scope of conservation must be expanded to include peripheral areas, enhancing their historical and cultural preservation through infrastructure development, promoting cultural activities, and optimizing landscape design. Improving the transportation network and guiding systems in peripheral areas can enhance accessibility and usability, encouraging greater public participation. Furthermore, leveraging ICH, traditional crafts, and local festivals to develop cultural and creative industries can transform peripheral districts into cultural consumption hubs. Lastly, creating an interactive cultural network between core city areas and peripheral neighborhoods, supported by coherent cultural tourism routes and resource integration, can promote core-periphery linkage development, strengthening the city’s overall cultural continuity and historical memory.

Natural factors such as temperature, precipitation, and extreme weather events pose significant challenges to historic districts’ spatial distribution and conservation. Integrating ecological restoration and green infrastructure into historic district planning can increase green space coverage and improve environmental quality. For instance, monitoring vegetation cover using the NDVI can enhance climate adaptability, while environmentally friendly materials and technologies can be employed to restore historic buildings. Establishing a dynamic environmental monitoring system can enable continuous management of the natural environment in landmark districts, improving their resilience to climate change and ensuring sustainable preservation.

The conservation of historic districts is highly complex and influenced by economic, natural, ecological, and cultural factors. This complexity necessitates interdisciplinary and cross-sector collaboration in conservation efforts. An effective multi-sectoral collaboration mechanism is urgently needed to achieve this goal. Governments should coordinate relevant sectors to jointly formulate policies integrating cultural heritage preservation, ecological protection, and economic development. At the same time, private enterprises, academic institutions, and non-governmental organizations should be actively involved in conserving and reusing historic districts, providing diversified and innovative solutions for cultural heritage preservation and development.

While this study offers an in-depth analysis of the spatial distribution and driving factors of historic districts in China’s four ancient capitals and provides valuable policy recommendations, it is not without limitations. Firstly, the study relies predominantly on quantitative analytical methods. Although these methods reveal significant spatial patterns and drivers, they may not fully capture the complexity and diversity of cultural heritage conservation. A more nuanced understanding of these factors could be gained through the incorporation of qualitative methods. Secondly, the geographical scope is confined to the four ancient capitals, which limits the generalizability of the findings. Future research could broaden the study to include other regions, which would allow for a more comprehensive understanding of historic district conservation across different cultural and geographic contexts. Expanding the geographic scope would provide insights into the diversity and complexity of conservation challenges in various settings. Additionally, while this study considered economic, natural, environmental, historical, cultural, and ecological factors, it did not fully explore the interactions between these variables. Moreover, the failure of this study to conduct longitudinal historical comparisons is mainly restricted by the limitations of access to historical data, which makes it difficult to systematically construct comparisons of spatial evolution in different historical periods. Future studies should further integrate historical maps, archival records, and urban planning data, and adopt longitudinal analysis methods in order to reveal the dynamic change patterns of historic districts and provide a more forward-looking scientific basis for the conservation and planning of historic districts.

Future research should focus on how these drivers intersect and how policies can effectively balance them to achieve sustainable conservation and development of historic districts. Understanding these interactions will be critical for developing holistic conservation strategies. Future research should also adopt an interdisciplinary approach, integrating insights from social sciences, environmental studies, and urban planning to formulate more comprehensive strategies for conserving historic districts. With advancements in technology, tools such as big data, artificial intelligence, and remote sensing can further enhance the precision of spatial analyses and offer a more dynamic understanding of changing trends in historic districts. Moreover, future studies should pay particular attention to issues of social equity. Efforts to protect historical and cultural heritage should ensure that the benefits are distributed across a broad range of social groups, including marginalized communities. Addressing social equity in heritage conservation will be essential to achieving inclusive and sustainable outcomes.

This study systematically examines the spatial patterns and driving factors of historic districts, sites, and cultural blocks in China’s four major ancient capitals—Beijing, Nanjing, Xi’an, and Luoyang—using multidimensional data and advanced spatial analysis methods. The findings reveal a strong spatial concentration of historic districts in urban cores, with notable regional differences. Beijing, Nanjing, and Xi’an exhibit clustering in old city centers, while Luoyang’s historic areas are more scattered, reflecting unique spatial evolution patterns. Economic factors, particularly GDP density, population density, and nighttime light intensity, are identified as primary drivers of historic area distribution, emphasizing the role of economic vitality and social aggregation in heritage preservation and renewal. Ecological and natural factors, including land use diversity and NDVI, play secondary but significant roles, enhancing spatial efficiency and ecological adaptability. Historical and cultural factors, while shaping the core value of cultural blocks, are less influential on the distribution of historic districts and sites, serving more as inherited resources. The study offers theoretical support for conservation and planning, advocating for strategies that enhance cultural resource protection in peripheral areas, integrate ecological considerations, and foster multi-stakeholder collaboration. Future research should further explore the dynamic interplay of economic, natural, cultural, and environmental factors to achieve comprehensive heritage conservation and sustainable development.