Which satellite data types are actually used to identify poor households?

Practitioners combine multiple inputs: daytime multispectral imagery (roof material, building density, road access), nighttime light intensity from VIIRS or DMSP sensors, SAR backscatter for building volume and flood exposure, and radio frequency signals (mobile network activity from operators like Spire or HawkEye 360 acting as proxies for phone ownership). No single layer is sufficient; ensemble models trained against Living Standards Measurement Study (LSMS) survey data tie them together.

Does a government need its own satellite to do this, or can it just buy commercial imagery?

Buying commercial imagery from Planet or Maxar is faster to start, but it creates dependency: pricing can double, access can be restricted during crises, and raw data rarely leaves the vendor's platform, which prevents independent auditing of targeting decisions. A sovereign microsatellite constellation in LEO—even at 3–5 m resolution—gives the state persistent, unencumbered access to the imagery and removes a single point of failure in programme delivery.

How accurate is satellite-based targeting compared to conventional proxy means testing?

A 2022 study supported by the World Bank found satellite-derived consumption estimates reduced inclusion and exclusion errors by roughly 22 percentage points compared to self-reported household surveys alone, particularly in rural areas where enumerator access is limited. Accuracy degrades significantly below the village level, so satellite data is best used to rank geographic units rather than select specific households.

What resolution of satellite imagery is needed?

For building-level asset proxies (roof type, structure count) you need sub-5-metre optical or sub-1-metre for urban areas. For agro-ecological vulnerability and vegetation stress (NDVI, NDWI) 10–30 m resolution (Sentinel-2 class) is sufficient. A national programme can therefore mix a sovereign microsatellite constellation for medium-resolution wide-area coverage with selective tasking of commercial high-resolution assets for validation strips.

How does this interact with national data protection law?

Satellite-derived poverty indices are statistical products attached to geographic units (village polygons, grid cells), not individuals—so they sit below the threshold of personal data in most jurisdictions. The risk arises when those indices are joined to beneficiary registries. That linkage should occur only within a national social registry system governed by domestic legislation, aligned with ISO/IEC 27001 for system security and ideally audited by the national statistics office.

Can this work in conflict zones or refugee settings?

Yes, and it may be the only scalable option where census infrastructure has collapsed. UNHCR has piloted satellite-assisted targeting for cash assistance in Syria, South Sudan, and the Sahel, using building damage indices from SAR and displacement tracking. The key constraint is not imagery availability but the absence of a beneficiary registry to attach the geographic scores to—UNHCR's PRIMES system partially fills that gap.

What is the typical revisit frequency needed for a cash transfer targeting update cycle?

Targeting lists for most national cash transfer programmes are updated annually or biannually, which means near-daily revisit is not operationally necessary. A sovereign constellation achieving 5–7 day revisit at 3–5 m resolution is more than adequate. More frequent revisit adds value for shock-responsive programmes that need to adjust rosters after floods, droughts, or displacement events within weeks rather than months.

How much does it cost to build a national satellite-based poverty analytics capability?

A minimally viable sovereign capacity—two to four microsatellites with ground processing, an Earth Observation data platform, and three years of model training and calibration surveys—is typically costed in the $40–120 million range depending on launch vehicle, orbit insertion, and ground segment complexity. This is typically offset within two to three programme cycles by reduced leakage in a national cash transfer programme; the World Bank estimates leakage in poorly targeted programmes can exceed 30% of total transfer value.

13.6.4Live

Cash Transfer Programme Targeting

Woman at a mobile money agent receiving a cash payment in a rural African village — A hand calculating finances on a smartphone illustrates how satellite poverty mapping directs cash transfers to those most in need.

Satellite Poverty Mapping to Cash Transfer Targeting

1
Observation
Multispectral satellites capture settlement, cropland density
2
Data Downlink
Raw imagery relayed to regional ground stations
3
Poverty Analytics
ML models score deprivation index per household cluster
4
Transfer Targeting
Verified beneficiary lists trigger mobile cash payments

Satellite imagery feeds deprivation models that automatically generate verified recipient lists for cash transfer programmes.

Page in the build queue

This application has its place in the atlas. The full page — short and long descriptions, what-matters bullets, sovereignty rationale, and the eight-field architecture block — is being written next.

Maturity reading: Operational systems exist with paying customers. Phase 1 built §4.1 (Maritime Intelligence) end-to-end as the depth-and-format reference. Section 4 is the next section to graduate from scaffold to full content; the rest follow on the roadmap.

Related applications

13.6.3 · Previous

Subnational Poverty Indices

All in poverty analytics

13.6.5 · Next

Service Access Inequality Mapping

Bring this into your sovereign programme.

All solutions Speak with engineering