Réunion Road Surface and Passability Data for 1,300 km of Arterial Roads

HeiGIT produced this dataset covering approximately 1,300 km of arterial roads in Réunion using 2020 and 2024 PlanetScope satellite imagery. It combines OpenStreetMap attributes with AI-derived predictions for road surface, width, and humanitarian accessibility to fill gaps in existing map data.

Use Cases

• Logistics planning using 'Road Width Class' to determine if segments support two-lane heavy truck flow
• Infrastructure monitoring via 'Road Surface Change' to identify transitions from unpaved to paved roads between 2020 and 2024
• Emergency response routing based on 'Humanitarian Passability Index (HPI)' scores to assess accessibility risks

Strengths

• 89.2% surface prediction accuracy compared to 64.7% for standard OSM tags
• Multi-temporal 2020-2024 surface change detection
• Standardized Humanitarian Passability Index (HPI) for logistical planning

Limitations

• AI-derived predictions may contain errors inherent to satellite imagery interpretation
• Limited to high-capacity arterial road classes only

Provenance

Source

HeiGIT (Heidelberg Institute for Geoinformation Technology)

Collection Method

Deep learning analysis of PlanetScope satellite imagery integrated with OpenStreetMap road segment data.

Time Range

2020 - 2024

Freshness

Last updated March 2026; utilizes satellite imagery from 2020 and 2024.

Geography

Réunion