Svalbard and Jan Mayen Islands: Road Surface Data

OpenStreetMap road data for Svalbard and Jan Mayen Islands is enhanced with deep learning predictions for surface type. The dataset includes AI-derived classifications for paved and unpaved roads, combining OSM tags with predictions from Mapillary street-level imagery. It is produced by the Heidelberg Institute for Geoinformation Technology (HeiGIT) and was last updated in March 2026.

Use Cases

• Transportation planning based on road surface type and highway classification attributes.
• Infrastructure analysis using combined OSM and deep learning surface classifications.
• Climate emissions assessment leveraging the distinction between paved and unpaved road networks.
• GIS applications utilizing geospatial road data with urban/rural classification attributes.

Strengths

• Integrates multiple data sources: OpenStreetMap, Mapillary imagery, and GHSU/AFRICAPOLIS urban layers.
• Provides AI-derived surface classifications (paved/unpaved) with confidence metrics like 'pred_class' and 'n_of_predictions_used'.
• Includes a companion dataset derived from PlanetScope satellite imagery for arterial roads.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count and total road length statistics are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

Heidelberg Institute for Geoinformation Technology (HeiGIT), based on OpenStreetMap (OSM).

Collection Method

Surface classification derived using a hybrid deep learning approach on Mapillary street-level imagery, combined with OSM tags and urban layers.

Freshness

Last updated 2026-03-03 09:11:35.549765; freshness should be verified.

Geography

Svalbard and Jan Mayen Islands.