Comparative Spatial Justice and Urban Heat Resilience: Rotterdam vs. Guangzhou

This repository contains the computational pipeline and analytical workflows for evaluating the relationship between urban green infrastructure allocation, socioeconomic vulnerability, and Land Surface Temperature (LST).

Repository Architecture

To replicate the study, your local directory must be structured as follows. Create a /data folder in your root directory.

├── code/
│   ├── 01_SJI_complete_rotterdam.R   # Full-resolution SJI for Rotterdam
│   ├── 02_SJI_comparative_baseline.R # Simplified, harmonized SJI for both cities
│   └── 03_network_thermal_analysis.R # Network routing, hex-aggregation, and UHI plotting
├── data/                             # User-created local data directory
│   ├── rotterdam/                    # Place downloaded Rotterdam files here
│   └── guangzhou/                    # Place downloaded Guangzhou files here
└── README.md                         # Project documentation

Data Requirements & Acquisition

Because raw data are not hosted in this repository, you must download the source files directly from the providers below and place them into their respective subfolders within /data.

1. Rotterdam Data

• Socioeconomic Indicators: Download the neighborhood data from the CBS StatLine Portaal.

• Physical Livability: Download the open dataset from the Leefbaarometer 2024 Open Data Portal.

• Building Footprints (BAG): Use the PDOK Services Plugin in QGIS to download building functions and layers for the Rotterdam study area.

• Network & Green Infrastructure: Use the QuickOSM Plugin in QGIS to download keys highway and green areas for Rotterdam.

2. Guangzhou Data

• Demographic Density: Download the gridded population raster from the WorldPop Open Spatial Repository.

• Network & Green Infrastructure: Use the QuickOSM Plugin in QGIS to extract building footprints, residential land use, pedestrian paths, and green spaces from OpenStreetMap.

3. Land Surface Temperature Data

• Create an account in USGS Earth Explorer (https://earthexplorer.usgs.gov/)

• Set an area around Rotterdam/Guangzhou with small percentage of cloud coverage (e.g., < 40%)

• Set the search criteria to Landsat 8/9 Collection 2 Level-2 Surface Temperature satellite imagery

• Download the ST_B10 file from the Surface Temperature data package

Execution Workflow

All scripts are written in R and utilize relative paths Ensure your working directory is set to the project root before executing.

1. Pre-processing: Download and crop all spatial data layers (including LST rasters) to your respective city administrative boundaries.

2. Run Spatial Justice Index:

• To analyze Rotterdam with its full suite of local indicators, execute code/01_SJI_complete_rotterdam.R.

• To run the comparative model across both cities, execute code/02_SJI_comparative_baseline.R.

3. Run Spatial Routing and Thermal Overlays: Execute code/03_network_thermal_analysis.R. This script processes the pedestrian networks, calculates distance thresholds, aggregates data into spatial hexagons, and generates final correlation plots against the LST layers.