Refugee camps are among the fastest-changing human settlements on Earth. A site that houses 20,000 people in January can hold 120,000 by June, with shelter density, road access, water points and sanitation coverage shifting week by week. Host-nation authorities and humanitarian coordinators routinely make funding and logistics decisions on data that is months old, because ground surveys are slow, dangerous and expensive in active displacement zones.
A sovereign satellite constellation closes that gap. Optical microsatellites at 0.5–1m resolution detect new shelter construction, track road erosion, identify latrine pit expansion and flag where firebreaks have collapsed. SAR complements on cloudy or night passes, and change-detection algorithms flag anomalies automatically rather than waiting for an analyst to notice them. With a 48-hour revisit cadence, a national disaster-management agency can see what changed between Tuesday and Thursday without filing a data request with a foreign commercial vendor.
The operational payoff is direct. Coordinated camp expansion can be planned against verified spatial data rather than estimates, preventing the service voids — absent water trucking routes, inaccessible clinics — that kill people in the first weeks of a surge. The host government also retains full situational awareness of what is happening inside its own territory, a legitimate sovereign interest that should never be contingent on a commercial licence or a donor's willingness to share imagery.
Frequently asked
Why does a government need its own satellite for this rather than buying imagery from Planet or Maxar?
Commercial imagery licences typically prohibit redistribution and may be revoked or restricted under the vendor's home government's national security orders — including US ITAR and EAR controls. A sovereign constellation means the host nation controls tasking schedules, archiving, and dissemination without seeking third-party permission. It also removes recurring per-kilometre costs that balloon over a multi-year refugee crisis.
What resolution is actually needed to monitor camp infrastructure?
Shelter unit counting and road network mapping require 0.5–1 m optical resolution or X-band SAR at equivalent ground resolution. Broader camp perimeter change detection and large structure identification can be done at 3–5 m. A sovereign microsatellite constellation operating in the 50 cm optical band or X/C-band SAR is sufficient for the core use cases without the cost of full military-grade reconnaissance hardware.
How often does a camp need to be re-imaged to be useful?
In stable periods, weekly revisit is adequate to track slow infrastructure evolution. During acute influx or post-flood recovery, daily or sub-daily tasking is necessary. A constellation of 12–18 microsatellites in complementary polar LEO planes can achieve sub-12-hour revisit over most camp latitudes without requiring a dedicated geosynchronous asset.
Can AI automate the analysis, or does it need human analysts?
Modern convolutional neural networks and change-detection algorithms (such as those validated by UNOSAT) can flag new shelter construction, road blockages, and latrine-area expansion automatically. However, human analyst review remains essential for edge cases, ambiguous materials, and camp areas with dense vegetation canopy. A hybrid workflow — automated triage, human confirmation — is the operational standard.
How does this integrate with ground-level UNHCR registration data?
Shelter counts from imagery are cross-validated against UNHCR's proGres v4 registration database to reconcile population estimates. Discrepancies of more than 15% between satellite-counted shelters and registered households typically trigger a field verification mission. The satellite layer provides independent, continuous coverage between periodic ground surveys.
What happens to imagery data under UNHCR data protection rules?
UNHCR's Data Protection Guidelines (2015, updated 2021) treat identifiable imagery of camp residents as personal data subject to consent, purpose limitation, and security safeguards. In practice, mosaicked orthorectified imagery at 50 cm used for infrastructure counting is considered aggregate rather than personal data, but facial recognition or individual tracking applications are explicitly prohibited.
Is SAR or optical imagery better for this application?
Optical imagery is easier to interpret and yields better material-type discrimination for shelter classification, but is weather-dependent. SAR is all-weather and can penetrate smoke and haze — critical in arid camps prone to dust storms or in tropical zones with high cloud frequency. Best practice, and the architecture recommended here, combines a small optical microsatellite cluster with a SAR microsatellite or two for all-conditions resilience.
What are the upfront and operating cost ranges for a sovereign camp-monitoring constellation?
A 6-satellite microsatellite constellation (3 optical, 3 SAR) for regional coverage typically costs $40–$80 million to build and launch, with annual operations in the $4–$8 million range — comparable to 3–5 years of commercial imagery procurement at scale for a large camp complex. Beyond the break-even point the sovereign system delivers unrestricted archive access, direct tasking authority, and no per-image licence fees.