Coastal planning decisions in the UAE are often made on data that is already out of date by the time it reaches a planner’s desk. The satellite archives exist. The environmental datasets exist. What is still rare is turning that record into a forward-looking signal instead of a historical one. That gap is the starting point for a new proof of concept Sparkgeo is building with Buro Happold and SOUR.
Together, they will develop proof-of-concept pilots (PoCs) across maritime and infrastructure, insurance and risk, the built environment, public safety, and geospatial intelligence through Dubai’s Geospatial Accelerator, an initiative from DIFC Innovation Hub and the Mohammed Bin Rashid Space Centre (MBRSC).
The Geospatial Accelerator attracted 73 applications from 24 countries, 45 of which progressed to interview, and led to the selection of ten scale-ups including:
- AtlasAI
- FlyPix AI
- Geospatial Insight
- GHGSat
- Latitudo40
- Little Place Labs
- SatSure
- Sparkgeo
- Stellaria
- V-Nova
It is the first programme launched under the Space for Earth Labs platform, pairing SpaceTech scale-ups with UAE public and private sector organisations to build and test proof-of-concept projects.
Sparkgeo is the technical partner on Buro Happold’s project, working alongside their built-environment team to apply geospatial intelligence, EO data and AI-enabled analytics to a specific question: how will the UAE coastline change over the next 25 to 50 years, and what does that mean for the habitats along it.
“A warning is only as good as the action it triggers. When warnings speak the language, landscape, and lived reality of the people receiving them, they can improve quality of life and save lives.”
What the proof of concept tests
Underpinning the work is Prescient, Sparkgeo’s geospatial data platform. Prescient ingests spatial data, including Earth Observation archives, in this case, a 12,000-image catalog of Landsat imagery alongside habitat maps and ancillary datasets like bathymetry and coastal surveys, and organizes them against open standards (STAC and OGC API) so the data is structured the same way regardless of source. That consistency is what makes it possible to run analysis without a separate data wrangling step first.
For this project, that means a multi-year shoreline change detection pipeline: classifying erosion, accretion and habitat extent across the historical satellite record, ranking erosion hotspots by rate, then projecting the trend forward rather than treating today’s conditions as a fixed baseline. The output is a practical planning signal, not a full wave-climate model: where erosion is accelerating, which habitats are most exposed, and how risks compound against the development already in place.
“Resilient coastlines depend on resilient ecosystems. Our aim is to better understand how habitats may evolve in the future so that today’s planning decisions work with nature, rather than against it.”
For Sparkgeo, the appeal of the partnership is straightforward. Most coastal planning still runs on data that’s already out of date by the time it’s used. This project is about closing that gap, giving the built environment sector a predictive signal instead of a historical record.
What’s next
Participating teams will develop and test their proof-of-concept work before presenting the outcomes at the Dubai AI Festival next week. For Sparkgeo, the project sits alongside a broader focus on building data pipelines, such as applying Earth Observation data to the problems that matter most in conservation and infrastructure planning: giving the people managing coastlines, habitats and assets a clearer view of what is coming, not just what already happened.