Science1 publisher2 min readPublished
Osaka rises fastest along its fault lines as its drained aquifers refill
Osaka's ground is rising about 4 millimetres a year on average, and up to 12 in places, after Japan's early-1960s curbs on groundwater pumping. At that average pace, regaining the 2 metres or more some districts sank would take at least 500 years.
The Scientist · Science desk

What happened
- Between the 1920s and 1960s, heavy pumping lowered Osaka's groundwater by up to 30 metres, and parts of the city sank more than 2 metres.
- The team tracked water levels in 44 monitoring wells from 1985, some 500 metres deep, and measured ground height with satellite radar tied to GNSS ground stations.
- Groundwater has been rising by up to a metre a year, and the land rose fastest where water levels climbed fastest.
- Uplift rates changed sharply within 100 metres, in narrow corridors that the researchers say line up almost perfectly with known tectonic faults.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- precedent Osaka gives other sinking cities a measured case that strict pumping limits, held for decades, can turn subsidence into uplift.
- constraint Uplift measured in millimetres cannot offset fast sinking soon: Jakarta's losses of more than 10 centimetres a year run 25 times Osaka's average rise.
- decision Planners counting on recovering aquifers to cut flood risk would need fault maps first, because the gain in height arrives unevenly along them.
Osaka works as a test case because its intervention has a date. Japan brought in strict groundwater regulations in the early 1960s, and levels beneath the city have recovered steadily since [3]. That gave the team a before and after inside one city. Their question was whether a recovering aquifer lifts the ground back evenly, like a balloon reinflating, or in a more complicated way [14]. The well records start in 1985 [4], more than two decades after the rules took effect [3], so the observations begin partway through the recovery.
The link between water and land is spatial. Places where water levels climbed fastest also rose fastest [6]. That is a correlation, and the simplest reading is that recovery runs a well-understood process backwards: many cities are held up in part by the water in the aquifers beneath them, and when pumping drains that water, the ground above sinks [15].
The faults were the surprise. "These corridors mapped almost perfectly onto known tectonic fault lines," the researchers wrote [8]. The alignment is an observation. The dam behaviour is the team's interpretation of it: faults, usually thought of in terms of earthquakes, act here like dams, which they picture as curtains hanging vertically underground [9].
The rates are small against the losses. Some districts sank more than 2 metres during the pumping decades [2]. At the average uplift of about 4 millimetres a year, getting that back would take at least 500 years; at the fastest local rate of 12 millimetres, roughly 170 [1]. Both figures assume today's rates hold for centuries.
The research team, writing on phys.org, frames the stakes as coastal. Sinking land and rising seas add together, so a subsiding coastal city meets higher relative sea levels sooner [13]. Uplift pulls the other way. The researchers do not set Osaka's rise against a local sea-level rate, so the evidence does not show how much flood risk the recovery has removed.
Other cities are still sinking. If subsidence in Tianjin, home to 15 million people, continues unabated, 15% of its population will be underwater by 2120, according to the team [11]. Around 80% of New Zealand's urban coastline is sinking, and groundwater may be a contributing factor in Christchurch and Wellington [12].
Better groundwater management is already the standard prescription for slowing subsidence [17]. I think Osaka supports going further and treating aquifer recovery as a flood-planning tool, on conditions the record makes plain. The refill came from restricting pumping. The restriction has held since the early 1960s [3]. And the city's faults were already mapped when the uplift corridors appeared [8].
What to watch
- The study's quantitative figures for how closely Osaka's uplift corridors track its faults, and how much uplift each metre of water rise produces.
- Whether the fault-controlled pattern appears in other cities where aquifers are recovering, such as the New Zealand sites where groundwater may drive sinking.
- A comparison of Osaka's uplift against its local rate of sea-level rise, which would show the net change in flood exposure.