Urban Ecological Quality Assessment Based on Google Earth Engine and Driving Factors Analysis: A Case Study of Wuhan City, China

Weiwei Zhang, Wanqian Zhang, Jianwan Ji, Chao ChenSustainability 2024 · DOI

TL;DR

This study leverages Google Earth Engine (GEE) and the Remote Sensing Ecological Index (RSEI) to assess urban ecological quality (EQ) in Wuhan City over two decades. It reveals a trend of initial decline followed by improvement, driven primarily by vegetation cover and dryness, offering valuable insights for sustainable urban planning.

What Problem Does This Paper Address?

Rapid urbanization poses significant challenges to urban ecological quality, leading to habitat fragmentation, reduced biodiversity, and degraded ecosystem services. Effectively monitoring and understanding the drivers of these changes are crucial for sustainable urban development. Traditional methods for ecological assessment often rely on difficult-to-obtain and outdated data. This paper addresses the need for a more efficient, dynamic, and comprehensive approach to assess urban EQ and identify its influencing factors, particularly in rapidly developing cities like Wuhan.

Methods

The study employs a multi-pronged approach:

  • Remote Sensing Ecological Index (RSEI): This index integrates four key indicators: greenness (NDVI), wetness (WET), heat (LST), and dryness (NDBSI). These are derived from satellite imagery.
  • Google Earth Engine (GEE): This cloud-based platform is used for efficient batch processing of large-scale remote sensing datasets, enabling rapid image acquisition, preprocessing, and index calculation. This overcomes limitations of traditional desktop-based analysis.
  • Principal Component Analysis (PCA): PCA is used to construct the RSEI, combining the four indicators into a single index and determining their relative contributions. The first principal component, explaining a significant portion of the variance, is used as the RSEI.
  • Optimal Parameter-Based Geographical Detector (OPGD) Model: This advanced spatial analysis tool is used to identify and quantify the influence of various natural and socio-economic factors (elevation, slope, aspect, population density, GDP, NDVI, WET, NDBSI, LST) on the RSEI. The OPGD model is novel in its ability to optimize spatial scale and discretization, providing more robust results than traditional geographical detectors.

Datasets

The study utilizes the following datasets:

  • Landsat 8 OLI remote sensing imagery: Acquired from the GEE platform, with a 30m spatial resolution, covering the period from 2001 to 2021. Images with less than 10% cloud cover were selected.
  • Digital Elevation Model (DEM): Sourced from the Geospatial Data Cloud.
  • Population Density and Gross Domestic Product (GDP) data: Obtained from the Resource Environment Science and Data Center.
  • Slope and Aspect data: Derived from detailed elevation data.

Key Insights and Ideas

  • GEE as a Game-Changer: The study highlights GEE's transformative potential for large-scale ecological assessments, streamlining complex workflows and improving efficiency compared to traditional methods.
  • RSEI's Comprehensive Nature: The RSEI effectively captures the multifaceted nature of urban ecological quality by integrating diverse indicators.
  • The "Greenness-Dryness" Dichotomy: The findings strongly suggest that healthy vegetation (greenness) is a primary driver of good ecological quality, while dryness (indicating arid conditions or lack of vegetation) is a significant detractor.
  • Urbanization's Double-Edged Sword: While rapid urbanization initially led to ecological degradation in Wuhan, the study shows that targeted environmental policies and management efforts can reverse this trend.
  • Spatial Heterogeneity Matters: The OPGD model's ability to identify optimal spatial scales is crucial, as the influence of driving factors can vary significantly depending on the resolution of analysis.

Key Results

  • Spatiotemporal Trend: Wuhan's ecological quality showed an initial decline from 2001 to 2006, followed by a gradual improvement, particularly after 2016. The proportion of "poor" EQ areas decreased, while "good" and "better" areas increased.
  • Driving Factor Ranking: The most significant drivers of EQ in Wuhan were identified as NDVI (greenness) and NDBSI (dryness). LST (heat) and WET (wetness) also played important roles. Slope and aspect had minimal impact.
  • Temperature and Precipitation: Annual average temperature and precipitation showed a non-significant impact on Wuhan's EQ during the study period.
  • Spatial Distribution: Poorer EQ was concentrated in core urban areas, while suburban and rural fringes generally exhibited better conditions, reflecting the impact of industrial development and population density.

Limitations and Future Work

  • Data Limitations: While the study used a comprehensive set of indicators, the authors acknowledge that future research could incorporate a wider range of ecological, social, and economic indicators, such as biodiversity and ecosystem services, for a more holistic assessment.
  • Single City Focus: The findings are specific to Wuhan. Comparative studies across different cities and regions are needed to identify broader patterns and context-specific factors.
  • Climate Change Impact: While temperature and precipitation were analyzed, the study suggests further investigation into the long-term effects of climate change on Wuhan's EQ.
  • Policy and Management: The study emphasizes the need for continued and enhanced comprehensive management strategies to sustain and further improve Wuhan's EQ, suggesting a need for multi-dimensional and integrative approaches to urban ecological management.

Summary updated

Abstract

Ecological quality is a critical factor affecting the livability of urban areas. Remote sensing technology enables the rapid assessment of ecological quality (EQ), providing scientific theoretical support for the maintenance and management of urban ecology. This paper evaluates and analyzes the EQ and its driving factors in the city of Wuhan using remote sensing data from five periods: 2001, 2006, 2011, 2016, and 2021, supported by the Google Earth Engine (GEE) platform. By employing principal component analysis, a Remote Sensing Ecological Index (RSEI) was constructed to assess the spatiotemporal differences of EQ in Wuhan City. Furthermore, the study utilized the optimal parameter-based geographical detector model to analyze the influence of factors such as elevation, slope, aspect, population density, greenness, wetness, dryness, and heat on the RSEI value in 2021 and further explored the impact of changes in precipitation and temperature on the EQ in Wuhan. The results indicate that (1) principal component analysis shows that greenness and wetness positively affect Wuhan’s EQ, while dryness and heat have negative impacts; (2) spatiotemporal analysis reveals that from 2001 to 2021, the EQ in Wuhan showed a trend of initial decline followed by improvement, with the classification grades evolving from poor and average to good and better; (3) the analysis of driving factors shows that all nine indicators have a certain impact on the EQ in Wuhan, with the influence ranking as NDVI > NDBSI > LST > WET > elevation > population density > GDP > slope > aspect; (4) the annual average temperature and precipitation in Wuhan have a non-significant impact on the EQ. The EQ in Wuhan has improved in recent years, but comprehensive management still requires enhancement.
BibTeX
@inproceedings{zhang2024urban,
  author = {Weiwei Zhang and Wanqian Zhang and Jianwan Ji and Chao Chen and Jianwan Ji and Chao Chen},
  title = {{Urban Ecological Quality Assessment Based on Google Earth Engine and Driving Factors Analysis: A Case Study of Wuhan City, China}},
  booktitle = {Sustainability},
  year = {2024},
  doi = {10.3390/su16093598},
}
Figure