Niger: 10,700 km of AI-Derived Road Surface and Passability Data

HeiGIT generated this dataset mapping 10,700 km of arterial roads in Niger using PlanetScope satellite imagery from 2020 and 2024. It provides AI-derived surface types, width classes, and passability indices for motorway, trunk, primary, and secondary road segments.

Use Cases

• Monitoring infrastructure upgrades by tracking 'Road Surface Change' from unpaved to paved
• Assessing logistical bottlenecks using 'Road Width Class' to identify single-lane constraints
• Routing humanitarian convoys based on 'Humanitarian Passability Index (HPI)' scores

Strengths

• 89.2% model accuracy for surface detection
• Multi-temporal 2020-2024 change detection
• Standardized width classification for heavy vehicle flow

Limitations

• Excludes tertiary, residential, and unclassified road types
• Surface predictions are limited to Paved/Unpaved/Unknown categories

Provenance

Source

HeiGIT (Heidelberg Institute for Geoinformation Technology)

Collection Method

Deep learning analysis of PlanetScope satellite imagery integrated with OpenStreetMap data.

Time Range

2020-2024

Freshness

Last updated March 2026; contains imagery-based analysis from 2020 and 2024.

Geography

Niger