Historical satellite analytics
Earthlytix turns satellite archives, from 1960s reconnaissance film to today's constellations, into georeferenced, analysis-ready maps. We measure how cities grew, forests fell and coastlines moved.
Explore
Each map pairs declassified film from the Cold War with today's high-resolution satellite imagery. Drag the handle to compare, then pan and zoom anywhere in the scene.
Historical imagery: U.S. Geological Survey declassified archive, shown at reduced resolution. Top-of-page preview: contains modified Copernicus Sentinel data 2026.
The archive
Most satellite analytics begin with recent imagery. We go back to the first satellite photographs of the Earth, then harmonise every era that followed so you can compare like with like.
Declassified Cold-War missions such as CORONA, GAMBIT and HEXAGON captured detailed imagery decades before commercial satellites existed.
The longest continuous multispectral archive. It is the backbone for land-cover trends across half a century.
Sub-metre commercial imagery for individual buildings, roads, plots and infrastructure.
Frequent revisits for current conditions, ongoing monitoring and near-real-time alerts.
How it works
Historical imagery is hard to use. It is distorted, uncalibrated and unreferenced. Our pipeline handles that so you get clean answers, not raw pixels.
We search every relevant archive for the best scenes covering your area and dates.
Film is scanned, corrected for camera and terrain distortion, then co-registered to modern imagery.
Machine-learning models map land cover consistently across sensors and decades.
Change is mapped and measured: hectares cleared, area built, shoreline moved, rate per decade.
Solutions
We map built-up extent for every available year, from 1960s film to today. Reports cover growth rates, the direction of sprawl and the farmland or green space each city absorbed.
Consistent forest and non-forest maps across decades quantify clearing, degradation and regrowth. They also give you the historical baseline needed for EUDR and other supply-chain due diligence.
Comparing any two dates reveals new roads, construction, mining, reservoirs and land clearing. Each change is classified by type and ranked by size.
We classify vegetation, built-up land, water and more, harmonised across sensors. That way a 1966 map and a 2025 map can be compared directly.
Solutions imagery: U.S. Geological Survey declassified archive and modified Copernicus Sentinel data 2025–2026. Shanghai change map: automated classification compared with © ESA WorldCover project 2021, contains modified Copernicus Sentinel data (2021) processed by the ESA WorldCover consortium.
Who it's for
Reconstruct urban growth to plan housing, infrastructure and green space.
Historical forest baselines for EUDR, zero-deforestation commitments and sustainability claims.
Shoreline retreat, land reclamation and mangrove and wetland change since the 1960s.
Site history for acquisitions, insurance, permitting and contamination risk.
Long-term environmental baselines for studies, monitoring programmes and impact reporting.
Early film shows landscapes and sites that later development has erased.
Start a project
Send us an area of interest and a time period. We'll check archive coverage and suggest the right approach.