Spatial Nonlinear Effects of Street Vitality Constrained by Construction Intensity and Functional Diversity - A Case Study from the Streets of Shenzhen

Jilong Li, Niuniu Kong, Shiping Lin, Jie Zeng, Yilin Ke, Jiacheng ChenISPRS Int. J. Geo Inf 2024 · DOI

TL;DR

This study investigates how the "built environment" of city streets, specifically construction intensity and functional diversity, influences street vitality. Using a novel spatial autoregressive threshold model and big data from Shenzhen, the research reveals that these factors have a significant, but nonlinear, impact on street vitality, with distinct thresholds beyond which their positive effects diminish. This work offers crucial insights for urban planners aiming to enhance vibrant urban spaces.

What Problem Does This Paper Address?

Urban vitality, particularly the vibrancy of streets, is a key indicator of a city's health and economic prosperity. While previous research has explored the relationship between the built environment and street vitality, many studies have overlooked the crucial aspect of spatial spillover effects – how the vitality of one street influences its neighbors. Furthermore, the nonlinear nature of these relationships, where effects change dramatically beyond certain thresholds, has also been underexplored. This paper aims to bridge this gap by examining how construction intensity and functional diversity, two key components of the built environment, nonlinearly affect street vitality, considering both their direct impact and their spatial influence on surrounding streets.

Methods

The core of this research lies in the development and application of a spatial autoregressive threshold model. This model uniquely combines two established techniques:

  • Panel Threshold Model (PTM): Used to identify and quantify the existence of thresholds in the relationship between explanatory variables (construction intensity, functional diversity) and street vitality. This allows researchers to pinpoint specific values where the impact of these factors changes.
  • Spatial Panel Durbin Model (SPDM): Incorporated to account for spatial spillover effects, meaning it considers how the vitality of neighboring streets influences the street being analyzed.

By integrating these, the model can simultaneously analyze the nonlinear (threshold) effects and the spatial (spillover) effects of the built environment on street vitality. The study also constructs spatial weight matrices at different distances (500m, 1000m, 1500m) to capture how street vitality's spatial influence varies with scale, reflecting the concept of "living circles."

Datasets

The study utilizes a comprehensive set of multisource big data from Shenzhen, China:

  • Baidu Building Data: Information on building footprints.
  • Baidu Huiyan LBS Data: Location-based service data to estimate population density and thus street vitality.
  • Amap POI Data: Points of Interest data, categorized into 16 types, used to measure functional diversity and construction intensity.
  • Road Network Data: From OpenStreetMap, used to define street segments.
  • Street View Images: Processed using deep learning for semantic segmentation to extract features like sky view factor and green view index.

Key Insights and Ideas

  • Nonlinear Relationships are Key: The study strongly emphasizes that the impact of construction intensity and functional diversity on street vitality isn't linear. There are specific "tipping points" or thresholds.
  • Thresholds Matter: Beyond a certain level of construction intensity or functional diversity, the positive impact on street vitality doesn't just plateau; it can diminish. This is a critical insight for urban planning, suggesting that "more is not always better."
  • Spatial Spillover is Significant: Streets don't exist in isolation. The vitality of a street is influenced by its neighbors, and this influence weakens with distance, following a "distance decay" pattern.
  • Accessibility is Paramount in Dense Cities: For a megacity like Shenzhen, public transportation and public facilities are the most significant drivers of street vitality, highlighting the importance of efficient urban infrastructure for resident well-being and economic activity.
  • Greenery's Dual Role: While green spaces (GVI) generally have a positive association with street vitality under functional diversity constraints, they can have a negative impact under high construction intensity. This suggests a complex interplay between different built environment factors.

Key Results

  • Construction Intensity Threshold: A single, consistent threshold of 0.1466 was identified for construction intensity across all spatial scales (500m, 1000m, 1500m). Streets below this are "low-intensity," and above are "high-intensity."
  • Functional Diversity Thresholds: Functional diversity exhibits two thresholds, which vary slightly with spatial scale:
  • 500m: 0.6832 and 2.2065
    • 1000m: 0.6832 and 1.4325
    • 1500m: 0.6832 and 1.2724
  • Spatial Autocorrelation: Street vitality shows significant positive spatial autocorrelation, meaning vibrant streets tend to be near other vibrant streets. This clustering effect weakens as the spatial scale (distance) increases.
  • Impact of Variables:
    • Public transit facilities and public facilities and services consistently show the strongest positive impact on street vitality across all scales.
  • Construction intensity and functional diversity have significant positive impacts, but their effects are modulated by the identified thresholds.
  • Sky view factor (SVF) positively correlates with street vitality, while green view index (GVI) shows a negative correlation under construction intensity constraints but a positive one under functional diversity constraints.

Limitations and Future Work

The authors acknowledge several limitations:

  • LBS Data Representativeness: Baidu Huiyan LBS data, while extensive, may not fully capture the population distribution due to users with privacy settings, no smartphones, or those not using Baidu apps.
  • Data Granularity: Despite processing, some data coarseness and discrepancies may remain.
  • Complexity of Street Vitality: The selected control variables might not fully explain the multifaceted nature of street vitality.
  • Street Classification Simplification: Real-world streets are complex, and the study's classification might oversimplify their intricate nature.

Future research directions suggested include:

  • More Detailed Indices: Exploring more nuanced measures of functional diversity.
  • Broader Spatial Scales: Investigating the impacts of factors on larger spatial scales to reach more comprehensive conclusions.
  • Varying Threshold Intervals: Further reducing the interval between distance thresholds to potentially reveal more detailed spatial effects.
  • Cross-City Validation: Examining whether similar spatial effects occur in different types of cities and regional contexts.
  • In-depth Exploration of Spatial Effects: Deeper investigation into spatial agglomeration, spillover, heterogeneity, and scale effects.

Summary updated

BibTeX
@article{li2024spatial,
  author = {Jilong Li and Niuniu Kong and Shiping Lin and Jie Zeng and Yilin Ke and Jiacheng Chen},
  title = {{Spatial Nonlinear Effects of Street Vitality Constrained by Construction Intensity and Functional Diversity - A Case Study from the Streets of Shenzhen}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2024},
  doi = {10.3390/ijgi13070238},
}
Figure