2 min read
sinkmap-ph: how fast Philippine cities are sinking
- civic-tech
- satellite
- case-study
The fastest-sinking ground in Metro Manila is not on the coast. It is inland, in the Bulacan and Pampanga lowland, dropping faster than the bay edge most flood maps worry about. sinkmap-ph measures that drop from free satellite radar and lays it over the floods that followed.
What it measures
Ground velocity in millimeters per year from Sentinel-1 radar, processed through HyP3 and MintPy, over 2016 to 2025. Seven cities are covered across Luzon, the Visayas, and Mindanao. Dagupan alone has lost about 35 centimeters at its worst point since 2016.
How it checks out
Three of the seven reproduce their published rates (Aslan 2024) within a factor of two:
| City | Measured, 2016 to 2025 | Published |
|---|---|---|
| Metro Manila / Bulacan | ~72 mm/yr | ~109 mm/yr |
| Cebu / Mandaue | ~10 mm/yr | 11 mm/yr |
| Iloilo | ~10 mm/yr | 9 mm/yr |
The other four (Dagupan ~20, Cavite ~6, Bacolod ~4, Tacloban ~3 mm/yr) have no published anchor to check against and are coverage-gated, so they ship as measured inventory, not validated rates.
The map
A single-file MapLibre map carries a velocity layer, a 2016-to-2025 slider that accumulates the drop so you can watch a city sink, toggleable flood extents, a building-exposure read, and a methodology page. The subsidence and the flood extents sit on the same map because they overlap in places, not because one is proven to cause the other; the coincidence-not-causation note runs throughout.
What it is not
A measurement, not a forecast. It does not predict the next flood or claim subsidence caused any specific one. The four unanchored cities carry no precision claim.
All inputs are public (Sentinel-1 via ASF HyP3, Sentinel-1 flood extents). sinkmap-ph computes statistical indicators only; specific claims require independent corroboration.