Geospatial Analysis of Access from Buildings to Health Facilities in Bayelsa, Nigeria

Ensuring timely access to health care in river-delta settings is challenging. This study analyzed geographic accessibility for Bayelsa State, Nigeria at the building level by harmonizing open data in Google Earth Engine to 100 m resolution and estimating least-cost travel time with WHO AccessMod. The cost surface combined Copernicus land cover, OSM roads and waterways, and mode-specific speeds for walking, motorized travel, and boats. AccessMod times were attached to Google Open Buildings footprints, yielding a per-building dataset. We summarized travel time by LGA; screened 5-km hexagons with an impedance ratio (minutes/km) to distinguish near-but-slow barriers from true remoteness; and identified frontier settlement clusters using DBSCAN for TT > 60 min. LGA medians 1–3 min, with Yenegoa, Kolokuma/Opokuma, and Ogbia achieving ≈universal ≤60 min coverage. Long-tail delays align with riverine belts, especially Brass and Ekeremor (e.g., P95 ≈ 65 min). Hex-grid classification across 175,352 buildings shows well-connected 65.7% and typical 29.3%, while actionable problem categories are limited but important: barrier hotspots 1.8%, heterogeneous pockets 1.1%, and remote 0.8% (with 1.4% “No estimate”). Frontier clusters concentrate on the Brass peninsula, the western Ekeremor coast, and scattered southern creek belts (e.g., Odioma-Diema, Cape Formosa, Iduwini wards; median TT typically 62–138 min). Findings indicate that Bayelsa’s equity gap is driven less by system-wide distance than by localized impediments. Micro-infrastructure (footbridges, jetties, short feeder links) can resolve near-but-slow hotspots, while outreach/transport support or strategic siting is warranted for truly remote clusters. The reproducible, open-data workflow generalizes to other deltaic contexts.