French Guiana Road Passability and Surface Change (2020-2024)

HeiGIT produced this dataset of AI-derived road surface, width, and passability metrics for approximately 600 km of arterial roads in French Guiana using 2020 and 2024 PlanetScope satellite imagery. It supplements OpenStreetMap (OSM) data by providing surface predictions for segments where tags were previously missing.

Use Cases

• Identifying logistical bottlenecks for heavy transport using 'Road Width Class' and 'Humanitarian Passability Index (HPI)'
• Tracking infrastructure modernization by analyzing 'Road Surface Change' between 2020 and 2024
• Filling 'surface tag' gaps in OpenStreetMap using 'DL Surface Prediction 2024' values

Strengths

• Surface model accuracy of 89.2% significantly exceeds the 64.7% accuracy of existing OSM surface tags
• Includes multi-temporal surface predictions for 2020 and 2024
• Provides standardized width classifications based on specific vehicle clearance thresholds

Limitations

• Restricted to arterial road classes, omitting local and residential networks
• Small total coverage area of approximately 600 km
• AI-derived predictions may contain errors in regions with heavy cloud cover or dense tropical vegetation

Provenance

Source

HeiGIT (Heidelberg Institute for Geoinformation Technology)

Collection Method

AI-derived predictions from PlanetScope satellite imagery integrated with OpenStreetMap data.

Time Range

2020-2024

Freshness

Last updated March 2026; contains data up to 2024.

Geography

French Guiana