Uncovering urban morphology and environmental interactions of small towns using self organizing maps in Qinba Mountains of Southern Shaanxi China – Nature

Executive Summary

This report presents a comprehensive analysis of the interplay between environmental factors and urban morphology in 358 small towns within the Qinba Mountains of Southern Shaanxi, China. The study provides a quantitative framework to support the achievement of the Sustainable Development Goals (SDGs), particularly SDG 11 (Sustainable Cities and Communities) and SDG 15 (Life on Land). Utilizing Earth observation data and advanced spatial analysis, including a U-Net3+ deep learning model, the boundaries of 123,992 buildings were precisely extracted. A Self-Organizing Map (SOM) algorithm identified six distinct environmental typologies and eight urban morphological clusters, revealing significant heterogeneity in both natural conditions and development patterns. Correlation analysis demonstrated that topographical constraints (slope) and ecological health (NDVI) are the most significant factors shaping urban form. The findings underscore the adaptive strategies employed by small towns in response to environmental limitations, with compact and moderate-sized developments emerging as the most prevalent and sustainable forms. This research offers critical insights for formulating evidence-based, place-specific urban planning policies that balance development with ecological preservation, thereby advancing the agenda for resilient and sustainable communities in sensitive mountainous ecosystems.

Introduction: Aligning Urban Morphology with Sustainable Development Goals

The Role of Small Towns in Achieving SDG 11

Small towns are indispensable for achieving balanced regional development and serve as critical nodes for strengthening urban-rural linkages, a key target of SDG 11.a. Unlike large metropolises facing acute pressure on land and resources, small towns offer a model for more sustainable, compact, and resilient communities (SDG 11.3). In the context of China, these towns are pivotal for coordinating development between urban and rural areas, promoting diversified local economies, and easing ecological pressures. This study focuses on small towns in the ecologically sensitive Qinba Mountains, examining how their physical form, or morphology, adapts to environmental conditions. Understanding these adaptations is essential for creating inclusive, safe, and sustainable human settlements in challenging terrains.

Ecological Imperatives in Mountainous Regions: Connecting to SDG 15 and SDG 6

The Qinba mountainous region is not only a focus for development but also a crucial ecological security barrier, directly contributing to SDG 15 (Life on Land), particularly target 15.4 concerning the conservation of mountain ecosystems. As the primary water source for the national South-to-North Water Transfer Project, the sustainable management of this area is also vital for SDG 6 (Clean Water and Sanitation). Urban development in this region must navigate the inherent conflict between economic growth and the preservation of ecological integrity. This study addresses this challenge by analyzing how environmental factors shape urban morphology, providing a basis for planning that respects ecological limits and protects biodiversity while fostering community well-being.

Methodological Framework for Sustainable Urban Analysis

Data Acquisition and Technological Innovation for SDG 9

In line with SDG 9 (Industry, Innovation, and Infrastructure), this study leverages advanced technology and high-resolution Earth observation data to build a robust analytical framework. Data was sourced from multiple public datasets to create a comprehensive environmental profile for each town.

• Environmental Data: Included Digital Elevation Models (DEM), slope, precipitation, land cover (GlobeLand30), Normalized Difference Vegetation Index (NDVI), soil type (HWSD), and river networks (OpenStreetMap).
• Urban Morphology Data: High-resolution satellite imagery from Google Earth was used to train a U-Net3+ deep learning model for automated building footprint extraction, a significant innovation for data-scarce regions.

Analytical Approach: Integrating Machine Learning for Evidence-Based Planning

A multi-stage analytical process was employed to quantify the relationship between environmental conditions and urban form, providing a scientific basis for sustainable planning.

1. Building Footprint Extraction: A U-Net3+ model identified 123,992 building footprints, providing the foundational data for morphological analysis.
2. Boundary Delineation: A concave hull algorithm precisely defined the boundaries of 358 small towns, capturing their true spatial extent.
3. Clustering Analysis: The Self-Organizing Map (SOM) algorithm, an unsupervised neural network, was used to cluster towns based on environmental and morphological characteristics, identifying distinct typologies.
4. Spatial Pattern Analysis: The Minimum Spanning Tree (MST) method was applied to analyze the spatial distribution and connectivity of towns within each cluster, revealing how they organize across the landscape.

Analysis of Environmental and Morphological Patterns for SDG Implementation

Environmental Clustering: Identifying Typologies for Targeted Interventions (SDG 11, SDG 15)

The SOM analysis identified six distinct environmental clusters, highlighting the region’s ecological diversity. This classification allows for the development of targeted policies that align with local conditions, a core principle of sustainable development.

• Cluster 4 (26.3% of towns): Characterized by low slope and elevation, representing areas most favorable for development but with lower vegetation (NDVI), indicating a need to integrate green infrastructure to meet SDG 11.7 (access to green and public spaces).
• Cluster 2 (16.2% of towns): Exhibited strong ecological characteristics with the highest NDVI and high elevation, suggesting these areas are critical for conservation efforts under SDG 15.
• Cluster 6 (Fewer towns): Adapted to challenging terrain with the highest slope values, demonstrating resilience but requiring careful planning to mitigate risks from natural disasters (SDG 11.5).

Urban Morphology Clustering: Assessing Progress Towards Sustainable Urban Forms (SDG 11.3)

Eight distinct urban morphological clusters were identified, ranging from compact to expansive forms. The prevalence of certain types indicates adaptive strategies that align with sustainable urbanism.

• Cluster 3 (145 towns): The most common type, characterized by compact development with small areas and high roundness. This form is highly efficient in its use of land, directly supporting SDG 11.3 (inclusive and sustainable urbanization).
• Cluster 1 (16 towns): Displayed elongated, linear forms, reflecting adaptation to valley corridors. Planning for these towns must focus on maintaining ecological connectivity along these corridors.
• Cluster 8 (4 towns): Represented the largest and most expansive urban forms, located in environmentally favorable areas. These settlements require careful management to prevent unsustainable sprawl.

Spatial Distribution and Connectivity: Insights from MST Analysis

The MST analysis revealed how different urban and environmental types are organized spatially. The dominant pattern for compact towns (Cluster 3) was a dendritic, tree-like structure following valley systems. This demonstrates an organic adaptation to topography that maintains regional connectivity, a crucial element for strengthening urban-rural partnerships as envisioned in SDG 11.a. In contrast, towns in more favorable environments (Cluster 4) showed a more concentrated spatial pattern, while those in challenging terrains (Cluster 2) were more dispersed, reflecting a direct response to environmental constraints.

Correlation and Implications for Sustainable Urban Planning

Key Environmental Drivers of Urban Form

Correlation analysis confirmed that environmental factors are significant drivers of urban morphology, providing quantitative evidence for planning decisions.

• Slope: Showed the strongest negative correlation with urban size (Area: r = -0.39), confirming that steep terrain is a primary constraint on urban expansion. This underscores the need to direct development towards flatter areas to minimize land degradation (SDG 15.3).
• NDVI: Exhibited a strong negative correlation with building area (r = -0.50), highlighting the direct trade-off between urban development and the preservation of green spaces. This finding is critical for balancing SDG 11 with SDG 15.
• Land Use: Artificial surfaces were positively correlated with larger urban forms (r = 0.42), indicating that once development begins, it tends to expand. This reinforces the importance of establishing clear urban growth boundaries.

Integrated Analysis: Linking Environmental Conditions to Morphological Outcomes

A heatmap analysis revealed strong associations between environmental conditions and resulting urban forms. For instance, in high-slope areas (Environmental Cluster 6), 60% of towns adopted a compact morphology (Morphological Cluster 3). This demonstrates that in extreme environmental conditions, urban form options are highly constrained, forcing development into more resource-efficient patterns. Conversely, areas with balanced environmental conditions supported a wider variety of urban forms. This evidence strongly supports the need for context-specific planning that adapts development strategies to local environmental carrying capacity.

Discussion: Policy Recommendations for Sustainable Mountain Communities

Advancing SDG 11 through Context-Specific Planning

The findings provide a basis for differentiated planning strategies tailored to the identified urban morphologies, ensuring that interventions are effective and sustainable.

1. For Compact Forms (Cluster 3): Policies should focus on enhancing public space quality and accessibility to ensure high density does not compromise livability, contributing to SDG 11.7.
2. For Linear Forms (Cluster 1): Planning should strengthen development along existing transportation corridors while establishing green belts to prevent lateral expansion and protect adjacent ecosystems (SDG 15.1).
3. For Large-Scale Developments (Cluster 8): Strategies must prioritize optimizing the ecological footprint through green infrastructure, mixed-use zoning, and sustainable transport to mitigate environmental impacts.

Balancing Development and Conservation (SDG 11 & SDG 15)

Sustainable development in mountainous regions requires a delicate balance between meeting human needs and preserving ecological integrity. This study’s results lead to several key recommendations:

• Prioritize Adaptive Compact Development: Encourage urban forms that efficiently utilize limited buildable land while respecting topographical constraints.
• Integrate Environmental Sensitivity: Development plans must be based on a thorough assessment of environmental factors, particularly slope and vegetation cover, to minimize negative impacts.
• Adopt a Network-Based Approach: Focus on developing integrated systems of small towns along natural corridors to enhance regional connectivity and resilience, rather than pursuing isolated projects.
• Match Development Scale to Carrying Capacity: Restrict large-scale urban projects to areas with favorable environmental conditions to avoid overwhelming sensitive ecosystems.

Conclusion: A Framework for Resilient and Sustainable Small Towns

This study successfully developed and applied an innovative framework integrating Earth observation, deep learning, and spatial analysis to uncover the complex relationships between urban morphology and environmental factors in the small towns of the Qinba Mountains. The research provides robust, quantitative evidence demonstrating that small towns have developed diverse and sophisticated adaptation strategies in response to their local environmental contexts. The predominance of compact and moderate-sized development forms highlights a pathway toward sustainable urbanism that is naturally suited to mountain environments. The findings offer invaluable insights for policymakers and urban planners, providing a scientific basis for creating targeted strategies that promote sustainable development. By understanding how environmental constraints shape urban form, planning can more effectively guide future growth to build resilient communities, protect unique ecological characteristics, and contribute meaningfully to the global Sustainable Development Goals.

Analysis of Sustainable Development Goals in the Article

1. Which SDGs are addressed or connected to the issues highlighted in the article?

1. SDG 11: Sustainable Cities and Communities

   The article is fundamentally centered on sustainable urban development. It analyzes urban morphology and environmental interactions to provide “valuable insights for sustainable urban planning in mountainous regions.” The study explores how small towns can develop in a way that is both resilient and environmentally conscious, addressing issues of land use, urban expansion, and the unique challenges of settlements in difficult terrains. It also highlights the role of small towns as “crucial nodes to link urban and rural areas,” which is a key aspect of integrated regional planning under SDG 11.

2. SDG 15: Life on Land

   The research is set in the Qinba mountainous region, described as a “crucial ecological security barrier” and an ecologically sensitive area. The analysis extensively covers terrestrial ecosystem factors, including soil type, slope, land use (forests, cropland), and vegetation coverage (NDVI). By examining the trade-offs between urban development and ecological preservation (e.g., the negative correlation between NDVI and building area), the study directly addresses the sustainable management of mountain ecosystems, the protection of biodiversity, and the prevention of land degradation issues like “soil erosion.”

3. SDG 6: Clean Water and Sanitation

   The article explicitly states that the study area is the “primary water source for the national South-to-North Water Transfer Project.” This highlights the critical importance of the region’s water-related ecosystems. The analysis includes “distance to rivers” as a key environmental factor, demonstrating the close relationship between settlement patterns and water resources. The findings contribute to understanding how to manage urban development in a way that protects these vital freshwater sources and their surrounding ecosystems.

4. SDG 9: Industry, Innovation, and Infrastructure

   The study’s methodology relies heavily on technological innovation. It employs “Earth observation data,” “U-Net3+ deep learning model,” and “advanced spatial analysis methods” like Self-Organizing Maps (SOM) to create a “quantitative framework for analyzing environmental-morphological relationships.” This application of advanced technology and data-driven research to solve complex urban planning challenges aligns with the goal of fostering innovation for sustainable and resilient infrastructure and development.

5. SDG 13: Climate Action

   The article notes that the study area “often suffers from floods, landslides, and other natural disasters.” While not the primary focus, the analysis of topographical factors like slope and elevation, combined with the goal of promoting sustainable planning, implicitly addresses the need to build resilience and adaptive capacity to climate-related hazards. Sustainable urban planning in such a vulnerable mountainous region is crucial for mitigating the risks associated with extreme weather events, which are expected to increase with climate change.

2. What specific targets under those SDGs can be identified based on the article’s content?

• SDG 11: Sustainable Cities and Communities

  • Target 11.3: By 2030, enhance inclusive and sustainable urbanization and capacity for participatory, integrated and sustainable human settlement planning and management in all countries.
    
    Explanation: The entire study is designed to support this target by providing a “quantitative framework” and “valuable insights for sustainable urban planning in mountainous regions.” It analyzes urban forms and environmental constraints to guide better planning decisions.
  • Target 11.4: Strengthen efforts to protect and safeguard the world’s cultural and natural heritage.
    
    Explanation: The article identifies the study area as a “crucial ecological security barrier,” which qualifies as significant natural heritage. The research aims to balance development with the preservation of this ecological integrity.
  • Target 11.5: By 2030, significantly reduce the number of deaths and the number of people affected… by disasters, including water-related disasters…
    
    Explanation: The study area is described as being prone to “floods, landslides, and other natural disasters.” By analyzing how environmental factors like slope and river proximity influence settlement patterns, the study provides a basis for planning safer, more resilient towns.
  • Target 11.a: Support positive economic, social and environmental links between urban, peri-urban and rural areas by strengthening national and regional development planning.
    
    Explanation: The article explicitly states that small towns are “crucial nodes to link urban and rural areas” and that they “coordinate development processes effectively between urban and rural areas.” The analysis supports better regional planning by understanding the role and form of these intermediary settlements.

• SDG 15: Life on Land

  • Target 15.1: By 2020, ensure the conservation, restoration and sustainable use of terrestrial and inland freshwater ecosystems and their services, in particular forests, wetlands, mountains and drylands…
    
    Explanation: The study focuses on a mountainous region, analyzing land cover data that includes “forest, grassland, shrubland, wetland, water body” to understand how urban development impacts these ecosystems.
  • Target 15.3: By 2030, combat desertification, restore degraded land and soil…
    
    Explanation: The article mentions that “mountain towns face some natural and ecological issues, including soil erosion.” The analysis of soil types and slope is directly relevant to identifying areas vulnerable to land degradation and informing restorative planning.
  • Target 15.4: By 2030, ensure the conservation of mountain ecosystems, including their biodiversity, in order to enhance their capacity to provide benefits that are essential for sustainable development.
    
    Explanation: This is a core theme of the article. The entire research is dedicated to understanding and promoting sustainable development in the “Qinba mountainous region” to balance “critical ecological conservation with the socioeconomic development needs.”

• SDG 6: Clean Water and Sanitation

  • Target 6.6: By 2020, protect and restore water-related ecosystems, including mountains, forests, wetlands, rivers…
    
    Explanation: The study area is identified as the “primary water source for the national South-to-North Water Transfer Project.” The analysis of river proximity and land use in this critical watershed directly contributes to understanding the pressures on these water-related ecosystems and how to protect them.

3. Are there any indicators mentioned or implied in the article that can be used to measure progress towards the identified targets?

• For Target 11.3 (Sustainable Urbanization)

  • Indicator 11.3.1: Ratio of land consumption rate to population growth rate.
    
    Explanation: The article does not measure population growth, but it provides extensive data on land consumption. The analysis of “artificial surfaces,” the extraction of “123,992 building footprints,” and the calculation of morphological indicators like “area,” “total building area,” and “fractal dimension” are direct measures of urban land use and expansion, which form the numerator of this indicator.

• For Target 15.1 (Conservation of Terrestrial Ecosystems)

  • Indicator 15.1.1: Forest area as a proportion of total land area.
    
    Explanation: The study uses “land cover data… which includes various land cover types such as cropland, forest, grassland, shrubland.” This data can be directly used to calculate the proportion of forest area, providing a baseline for monitoring changes over time.

• For Target 15.4 (Conservation of Mountain Ecosystems)

  • Indicator 15.4.2: Mountain Green Cover Index (MGCI).
    
    Explanation: The article explicitly measures and analyzes the “Normalized Difference Vegetation Index (NDVI),” which is a primary component of the MGCI. The study’s detailed spatial analysis of NDVI values across the mountainous region provides the exact type of data needed to calculate and monitor this indicator.

• For Target 6.6 (Protection of Water-Related Ecosystems)

  • Indicator 6.6.1: Change in the extent of water-related ecosystems over time.
    
    Explanation: The study uses land cover data that identifies “water body” areas and analyzes the proximity of towns to rivers. This provides a spatial baseline of the extent of these ecosystems, against which future changes due to urban development or restoration efforts can be measured.

4. Create a table with three columns titled ‘SDGs, Targets and Indicators” to present the findings from analyzing the article.

Source: nature.com