UGS-SDC datasets

The imagery was composited, and NDVI was calculated from the red and near-infrared bands to characterize vegetation quality within green-space polygons.</p><p dir="ltr">2. Demand layer: gridded total resident population at 100 m resolution, based on the Seventh National Population Census of China (2020) obtained from the ASPECT dataset (<a href="doi.org/10.6084/m9.figshare.27323106.v1)" target="_blank">doi.org/10.6084/m9.figshare.27323106.v1)</a> (Ju et al., 2025).</p><p dir="ltr">3.

Transport network: the multimodal road network of Fujian Province, based on an OpenStreetMap (OSM) snapshot from February 27, 2026, downloaded from Geofabrik, classified into walking, driving, and bus with mode-specific speeds (walking 5 km/h; driving 20–50 km/h;bus 15–20 km/h).</p><p dir="ltr">4. Analysis grid: the study area aggregated into a uniform 200 m hexagonal grid; each cell carries population demand and all accessibility/equality measures.</p><p dir="ltr">5.

Outputs_accessibility_equality: per-capita accessible supply capacity (PCASC) and comparative indicators (CV, Gini coefficient, Spearman's ρ) computed under five models (Buffer, Gravity, 2SFCA, Ga2SFCA, SDC) across walking, driving, bus, and total modes, using a unified 15-minute travel-time threshold.</p><p dir="ltr">6. Outputs_siting_optimization: candidate-site solutions from the maximal covering location problem (MCLP) embedded with dynamic equality metrics, together with p-median, p-center, and a manually selected alternative, and their resulting changes in low-service population.</p><p dir="ltr">7 Outputs_sensitivity_analysis: sensitivity analyses of accessibility results under different travel-time thresholds and PCASC percentiles for SDC-MCLP siting optimization.</p><p dir="ltr">All geographic layers share the coordinate reference system [CRS, EPSG:4326]; vector layers are provided in GeoJSON and ESRI Shapefile (with accompanying CSV attribute tables), rasters in GeoTIFF, and tabular results in CSV and XLSX.

The dataset is intended to be used with the open-source interactive planning-support tool UGS-SDC [github.com/yuannell/UGS-SDC]. Please cite the article above when using these data. </p>