Colombia Road Surface and Passability Data for 35,200 km of Arterial Roads

This dataset maps approximately 35,200 km of arterial roads in Colombia, providing AI-derived surface types, widths, and passability scores produced by HeiGIT. It integrates 2020 and 2024 PlanetScope satellite imagery with OpenStreetMap data to identify road surface changes and logistical accessibility. The analysis focuses on motorway, trunk, primary, and secondary road classifications.

Use Cases

• Assessing disaster relief logistics using the Humanitarian Passability Index (HPI) to identify high-risk segments
• Monitoring national infrastructure development by filtering for segments with a positive Road Surface Change indicator
• Planning heavy vehicle transport routes by selecting segments with Road Width Class 3 (>5.5 m)

Strengths

• High classification accuracy of 89.2% for road surface predictions
• Provides surface data for 0.031 million km of roads previously missing OSM surface tags
• Includes multi-temporal change detection between 2020 and 2024

Limitations

• Excludes local and residential road classes, focusing only on the arterial backbone
• Surface and width attributes are AI-derived from satellite imagery and may contain model-specific errors
• Geographic scope is restricted to Colombia

Provenance

Source

Heidelberg Institute for Geoinformation Technology (HeiGIT)

Collection Method

Deep learning analysis of PlanetScope satellite imagery integrated with OpenStreetMap road geometries

Time Range

2020 - 2024

Freshness

Last updated March 2026 using satellite imagery from 2020 and 2024.

Geography

Colombia