Guam Arterial Road Surface and Passability Data (2020-2024)

HeiGIT produced this dataset of AI-derived road surface, width, and passability metrics for approximately 300 kilometers of arterial roads in Guam using 2020 and 2024 PlanetScope imagery. It supplements OpenStreetMap data, where over 56% of segments lack surface tags, by providing predictions with 89.2% accuracy.

Use Cases

• Assessing logistical accessibility for heavy machinery using the 'Road Width Class' and 'Humanitarian Passability Index (HPI)'
• Monitoring infrastructure development by tracking 'Road Surface Change' from unpaved to paved between 2020 and 2024
• Improving navigation datasets by filling missing 'surface' tags with 'DL Surface Prediction' values

Strengths

• High surface classification accuracy of 89.2%
• Fills surface tag gaps for 56.8% of OSM road segments in the region
• Standardized Humanitarian Passability Index (HPI) for logistical planning

Limitations

• Small geographic sample size limited to Guam
• AI predictions may be affected by tropical canopy cover or cloud interference in satellite imagery
• Restricted to high-capacity transportation backbone only

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

Contains temporal comparisons between 2020 and 2024 imagery; last updated March 2026.

Geography

Guam