Most climate research teams point their satellites at rising oceans. Leonard Ohenhen and colleagues pointed theirs at the ground - and found something that should make 150 million people on Java Island very uncomfortable.

Published in Science Advances, their study used Interferometric Synthetic Aperture Radar (InSAR) - basically, bouncing radar signals off the Earth from space and measuring how the ground shifts between satellite passes - to build the most detailed subsidence map of Java ever created. The verdict? Multiple coastal cities are sinking at 1 to 15 centimeters per year. That's not a typo. Parts of Jakarta, Semarang, Pekalongan, and Demak are dropping by more than a meter per decade (Ohenhen et al., 2026).

The Sea Isn't the Villain Here (Well, Not the Main One)

Here's the thing everyone gets wrong about coastal flooding: we blame the ocean. Global sea level rise gets all the headlines, the protest signs, the documentary narration. But for Java's northern coast, the land is sinking so fast that it accounts for the dominant share of relative sea level rise - that's the measurement that actually matters to the person whose living room is filling with water.

The team constructed "virtual tide gauges" every 5 kilometers along the northern coastline - a clever workaround for regions that lack ground-based monitoring infrastructure. Their projection: contemporary subsidence will dominate relative sea level budgets over the next 25 years along more than 75% of the coast. In some areas, land sinking contributes up to 85% of the total relative sea level change. The ocean is rising, sure, but the ground is dropping out from underneath people way faster.

Why Is Java Sinking? (Spoiler: We Did This)

Using machine learning spatiotemporal clustering - think of it as teaching an algorithm to find patterns in when and where the ground moves - the researchers traced the dominant cause to resource extraction. Groundwater pumping is the big one. When you suck water out of underground aquifers, the layers of sediment above compact like a sponge being squeezed dry. Multiply that by millions of wells serving homes, factories, and farms across the most densely populated island on Earth, and you've got yourself a slow-motion disaster.

This isn't news for Jakarta specifically. The city is so famous for sinking that Indonesia is literally building a new capital - Nusantara, on the island of Borneo - partly to escape the problem. North Jakarta has dropped roughly 2.5 meters over the past decade, and about 40% of the city now sits below sea level. But what this study reveals is that the problem extends far beyond Jakarta. It's an island-wide crisis affecting cities that haven't gotten nearly as much attention.

The Same Team Has Been Catching Sinking Cities Worldwide

Ohenhen and co-author Manoochehr Shirzaei have been on a global tour of subsidence mapping. In 2024, they published a Nature study showing that 24 of 32 major US coastal cities are sinking at rates exceeding 2 mm per year (Ohenhen et al., 2024, Nature). A separate European assessment found nearly half the continent's coastal floodplains are subsiding (Thiéblemont et al., 2024). The Java study completes a trifecta that drives home a global pattern: we keep framing coastal risk as a climate problem when it's often a groundwater management problem.

As Shirzaei put it: "We often frame sea-level rise hazards as a climate-driven process, but in many of the world's most vulnerable regions, human-induced sinking land is the dominant driver."

Why This Actually Matters (Beyond the Obvious)

The practical upshot is surprisingly hopeful, if anyone acts on it. Climate-driven sea level rise requires global coordination and decades of carbon reduction. Land subsidence caused by groundwater extraction can be addressed locally, through regulation, by individual governments, starting now. Cities like Tokyo and Shanghai dramatically slowed their subsidence by restricting groundwater pumping decades ago.

The virtual tide gauge methodology is also a template. Hundreds of coastal regions worldwide lack the monitoring infrastructure to understand their actual flood risk. Satellite-based approaches like this one can fill those gaps - giving policymakers hard numbers instead of hand-waving about "rising seas." If you're into visualizing complex spatial data like this, tools like mapb2.io offer mind-mapping approaches that can help make sense of layered geographic and temporal datasets.

Java, home to roughly 2% of the world's population, is running out of time to integrate subsidence data into adaptation planning. The ground isn't waiting for a committee to finish its report.

References

1. Ohenhen, L. O., Shirzaei, M., Kumar, P., et al. (2026). Land subsidence on Java Island and its contributions to relative sea level change. Science Advances. DOI: 10.1126/sciadv.aec0172

2. Ohenhen, L. O., Shirzaei, M., et al. (2024). Disappearing cities on US coasts. Nature, 627, 108-115. DOI: 10.1038/s41586-024-07038-3

3. Thiéblemont, R., et al. (2024). Assessing Current Coastal Subsidence at Continental Scale: Insights From Europe. Earth's Future. DOI: 10.1029/2024EF004523

4. Minderhoud, P., Shirzaei, M., & Teatini, P. (2025). From InSAR-Derived Subsidence to Relative Sea-Level Rise - A Call for Rigor. Earth's Future. DOI: 10.1029/2024EF005539

5. Ohenhen, L. O., Shirzaei, M., et al. (2025). Land subsidence risk to infrastructure in US metropolises. Nature Cities. DOI: 10.1038/s44284-025-00240-y

Disclaimer: This blog post is a simplified summary of published research for educational purposes. The accompanying illustration is artistic and does not depict actual model architectures, data, or experimental results. Always refer to the original paper for technical details.