Health ministries in low- and middle-income countries routinely plan vaccine distribution, emergency response and capital budgets using facility databases that are years out of date, incomplete, or simply wrong. A clinic built by an NGO last year, a pharmacy destroyed in a flood six months ago, a rural health post relocated by a district authority — none of these appear in the national register until a slow, expensive ground survey catches up. The gap between the official map and reality costs lives and misallocates scarce resources at scale.
Satellite-derived facility mapping closes that gap systematically. A constellation of sub-metre optical satellites acquires tasked imagery over populated areas on a regular cadence; ML pipelines trained on building morphology, roof spectral signatures and road-network proximity flag new, changed and destroyed structures for human analyst review. The output is a living geodatabase — every facility tagged with a unique ID, facility type, operational status, catchment population estimate and last-verified date. Change alerts fire automatically when imagery shows structural modification.
The operational payoff is immediate. Planners routing a vaccination campaign know which sites are functional before the cold chain leaves the capital. Emergency coordinators responding to a cyclone can pull a damage-assessed facility layer within 48 hours of the event. Donors and finance ministries can audit capital investment against satellite-confirmed construction. None of this is possible when the underlying map is a spreadsheet that nobody trusts.
Frequently asked
Why does a government need its own satellite capability to map health facilities? Can't it just buy imagery from Planet or Maxar?
Buying imagery works for a one-time project, but a sovereign registry requires continuous updates, and commercial vendors can change pricing, reprioritise tasking, or restrict access under export controls with little notice. A government-owned nanosatellite constellation — even a modest 6-to-12 satellite multispectral system — guarantees tasking priority, data sovereignty, and an unbroken archive that commercial contracts rarely provide. It also builds the domestic skilled workforce that makes all downstream health data systems possible.
What resolution do satellites actually need to identify a health facility reliably?
For urban environments, 0.5–1.0 m resolution (Planet SkySat, Maxar WorldView) is sufficient to identify rooftop health-cross markings, vehicle parks, and facility footprints. In rural settings where facilities may be small single-room structures, 3 m imagery is often borderline; 0.5 m is strongly preferred. SAR imagery at 1–3 m resolution (Capella Space, ICEYE) complements optical data in perpetually cloudy regions, and several microsatellite SAR constellations are now cost-accessible for national procurement.
How does satellite mapping connect to the WHO concept of a Master Facility List (MFL)?
The WHO MFL framework (WHO/HIS/EHT/2012.1) defines a single authoritative, government-owned register of every health facility with standardised attributes: GPS location, ownership, service level, and operational status. Satellite imagery provides the geometric foundation — verifying that a facility physically exists at a claimed coordinate, detecting new construction, and flagging closures. Without the satellite layer, MFL coordinates are self-reported and frequently wrong; the WHO estimates positional errors of several kilometres are common in paper-based submissions.
What is the role of GNSS in this application versus satellite imagery?
Satellite imagery identifies facility locations from orbit; GNSS receivers held by health workers or survey teams provide the precise ground-truth coordinate (≤1.5 m CEP with dual-frequency receivers) that is written into the MFL record. The two are complementary: imagery scales to national coverage in a single pass, while GNSS provides the legally defensible, attribute-rich ground record. An owned LEO navigation augmentation service further improves GNSS accuracy in areas with poor ionospheric correction data.
How often does facility data need to be refreshed, and what satellite revisit cadence supports that?
Health facility landscapes in fast-growing countries change meaningfully every 12–18 months through new construction, facility upgrades, and closures. A quarterly satellite revisit cadence is sufficient for change detection when using automated building-footprint algorithms applied to medium-resolution imagery (3–5 m). Daily revisit constellations like Planet's Dove fleet are more than adequate for trigger-based alerts (e.g., detecting construction starts), but quarterly change-detection runs keep compute and analyst costs proportionate.
Can satellite imagery detect functional status — whether a facility is actually open and operating?
Not directly. Imagery can confirm physical presence, roof integrity, vehicle activity, and absence of obvious destruction, but it cannot confirm whether staff are present, drugs are in stock, or services are being delivered. Functional status requires integration with routine health management information systems (HMIS), facility surveys, or mobile reporting tools. Satellite mapping establishes the spatial backbone; functional data layers sit on top of it.
What cybersecurity and data-governance obligations apply to a nationally owned facility registry?
A national MFL integrating satellite-derived coordinates constitutes critical national infrastructure. It should conform to ISO/IEC 27001 for information security management and OGC catalogue standards (OGC 07-006r1) for controlled interoperability. For conflict-affected or high-security facilities, coordinates should be access-tiered: full precision for authorised health logistics users, administratively aggregated for public portals. The ICRC's Digital Threats in Armed Conflict guidance also recommends that facility coordinate databases never be hosted on publicly reachable endpoints without strong authentication.
How do developing nations finance a satellite health-mapping programme without large upfront budgets?
Several pathways exist. The World Bank's Health Nutrition and Population portfolio has committed $340M across 34 projects (FY2020–2024) with geospatial components eligible for satellite tasking costs. Regional satellite consortia — modelled on EUMETSAT's cost-sharing framework — allow groups of smaller nations to co-fund a shared microsatellite and split data access costs. Alternatively, a government can negotiate OECD Official Development Assistance (ODA) funding for an initial commercial imagery baseline while ring-fencing domestic budget for an eventual sovereign asset, treating the commercial phase as a proof-of-concept that builds regulatory and analytic capacity.