Sint Maarten Road Surface, Width, and Passability AI Data (2020-2024)

HeiGIT generated this geospatial dataset for Sint Maarten (Dutch part) using AI analysis of PlanetScope satellite imagery from 2020 and 2024. It maps arterial road segments including motorways, trunks, primary, and secondary roads with attributes for surface type, width, and logistical accessibility. The dataset provides surface predictions for road segments where OpenStreetMap (OSM) tags were previously missing.

Use Cases

• Assessing logistical accessibility for disaster response using the Humanitarian Passability Index (HPI) scores
• Monitoring infrastructure development by analyzing the Road Surface Change column for new paving between 2020 and 2024
• Route planning for heavy machinery based on the Road Width Class (1-3) to identify single-lane vs two-lane flow

Strengths

• 89.2% accuracy for road surface model predictions
• Multi-temporal coverage comparing 2020 and 2024 infrastructure
• Standardized Road Width Class (1-3) based on physical vehicle clearance
• Fills data gaps for the 37.5% of arterial roads lacking OSM surface tags

Limitations

• Geographic scope is limited to the Dutch part of Sint Maarten
• AI-derived predictions may contain interpretation errors compared to ground-level surveys
• Limited to high-capacity arterial road networks

Provenance

Source

Heidelberg Institute for Geoinformation Technology (HeiGIT) using PlanetScope satellite imagery

Collection Method

Deep learning models applied to 2020 and 2024 satellite imagery integrated with OpenStreetMap road geometries

Time Range

2020 - 2024

Freshness

Contains imagery-derived data from 2020 and 2024; last updated March 2026.

Geography

Sint Maarten (Dutch part)