Tuvalu's Department of Lands and Survey built a machine-learning-assisted digital twin of Funafuti under the Pacific Community's Digital Earth Pacific programme, detecting coastline change and 200+ solar installations from drone imagery — at no cost to government.
1000 images
Geotagged drone images captured (early 2024)
200 installations
Solar-panel installations identified via ML feature extraction
40 points
Ground control points captured
1 week
Orthophoto delivery time
Details
Maturity
Pilot
Promoter
Tuvalu Department of Lands and Survey (with Pacific Community's Digital Earth Pacific programme)
Period
2023–2024
Keywords
geospatial AI, digital twin, climate adaptation, land administration
Context
In September 2023, Tuvalu's Department of Lands and Survey, working with the Office of the Prime Minister, joined the Pacific Community's Digital Earth Pacific programme and partnered with nonprofit PLACE and Esri to build a national digital twin of the 26 km² archipelago.
Objectives
The programme aims to map Tuvalu's coastline and built environment to support climate adaptation and land administration, with the wider Digital Earth Pacific programme explicitly designed for replication across 27 Pacific countries and territories.
Activities
In early 2024, eVTOL drone flights over Funafuti captured over 1,000 geotagged images and nearly 40 ground control points across two atolls. Machine-learning feature extraction identified more than 200 solar-panel installations on existing housing stock, and change-detection comparing 2019 and 2024 imagery flagged considerable coastline change at the southern tip of the atoll. A true orthophoto was delivered within one week of data collection. Privacy safeguards included masking faces and vehicle plates in street-level imagery.
Results
The data-collection phase was provided at no cost to the government through PLACE's partnership model. Fiji's Ministry of Lands is already an active user of the wider Digital Earth Pacific platform.
Conclusions
No source publishes cost-benefit, accuracy or adoption figures specific to Tuvalu, and coverage describes the current scope as limited to Funafuti and select atolls, with nationwide completion framed as a future goal rather than an accomplished fact.
Implementation
Indicative cost
Low (< €50k)
Time to results
Short (< 1 year)
Staffing & skills
Tuvalu Department of Lands and Survey (lead agency), Office of the Prime Minister of Tuvalu, PLACE (nonprofit partner) and Esri (technology partner), Pacific Community's Digital Earth Pacific programme
Conditions for success
Data-collection phase delivered at no cost to government through PLACE's partnership model
Shared Digital Earth Pacific platform designed for replication across 27 Pacific countries and territories
Privacy safeguards including masking of faces and vehicle plates in street-level imagery
Commonly funded by
Philanthropic / foundation funding
Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.
Do you run this practice?
Claim it —
verified implementers get a public contact pathway and can propose corrections.
Data sources
Where this practice's information was retrieved from, and when.
Mozambique's INGD and INAM are piloting machine-learning forecasts atop a multi-hazard early-warning system that helped cut Cyclone Freddy's 2023 death …