Niue Road Surface and Passability Data with 2020-2024 AI Predictions

A source of AI-derived road surface, width, and passability metrics for arterial roads in Niue, generated by HeiGIT using PlanetScope satellite imagery from 2020 and 2024. It covers road segments classified in OpenStreetMap as motorway, trunk, primary, and secondary, including their link classes. The data includes a Humanitarian Passability Index (HPI) to assist in logistical planning and infrastructure monitoring.

Use Cases

• Logistics planning for heavy vehicles using the Road Width Class and Humanitarian Passability Index (HPI)
• Infrastructure development monitoring by identifying unpaved-to-paved transitions via the Road Surface Change indicator
• OpenStreetMap data enrichment by filling missing surface tags with DL Surface Prediction values

Strengths

• AI surface model achieves 89.2% accuracy compared to 64.7% for standard OSM tags
• Multi-temporal coverage (2020 and 2024) allows for road surface change detection
• Standardized width classification based on specific vehicle clearance (e.g., <3.5m, 3.5-5.5m, >5.5m)

Limitations

• Small geographic sample size restricted to the island of Niue
• Excludes non-arterial road networks which may be critical for local access
• AI predictions may be affected by environmental factors in satellite imagery like cloud cover or shadows

Provenance

Source

Heidelberg Institute for Geoinformation Technology (HeiGIT)

Collection Method

Deep learning predictions derived from PlanetScope satellite imagery integrated with OpenStreetMap road geometries

Time Range

2020 - 2024

Freshness

The dataset uses imagery from 2020 and 2024, with the last metadata update in March 2026.

Geography

Niue