“Sinking cities” are not merely a dramatic metaphor: for many coastal metropolises, the combined effects of sea-level rise and land subsidence are causing visible land loss, more frequent flooding and damage to infrastructure. Some cities—Jakarta, Venice, Bangkok, New Orleans, Miami and Shanghai—are now facing rates that make adaptation extremely costly; elsewhere, the processes are slow but inexorable. Solutions range from barriers and pumps to delta reforestation and managed retreat—but none is simple or inexpensive. We have synthesised studies, recent data and practical examples to provide a clear overview of the problem and the available options.
For further context, see The first liveable “underground” cities: a futuristic solution to overpopulation and climate change, an article that explores an aspect directly related to this topic.
1. What are “sinking cities”? Definitions and nuances
In practice, a “sinking city” refers to a combination of two processes:
- Sea-level rise (SLR)—a global phenomenon driven by the thermal expansion of ocean water and the melting of glaciers and ice sheets (eustasy).
- Vertical land subsidence—the actual lowering of the ground surface relative to a fixed datum, caused by local factors including aquifer extraction, hydrocarbon extraction, sediment compaction, consolidation of organic soils (peat), mining, urban compaction or permafrost thaw.
Importantly, the actual risk in a coastal city is the product of the two processes (SLR + subsidence). In many cases—for example, Venice—local subsidence, rather than rising seas alone, has explained half of the “relative sea-level rise”.
2. The main causes: brief, but technically accurate
2.1 Sea-level rise: global forces
- Thermal expansion of the oceans and melting glaciers and ice sheets increase the volume of water on a planetary scale (IPCC).
- An important signal: century-scale projections (RCP/SSP scenarios) indicate that rises of hundreds of centimetres may be possible by 2100–2300, depending on the emissions pathway.
2.2 Human-caused subsidence: local and often rapid
- Groundwater extraction: when people pump water from aquifers, soil particles can compact irreversibly and the surface settles (for example, Jakarta and Bangkok). This is one of the most common causes of urban subsidence.
- Hydrocarbon extraction and mining: voids left by extraction can cause differential subsidence.
- Sediment compaction and peat oxidation: in deltas and wetlands, drainage for agriculture or urbanisation dries and oxidises peat, which shrinks in volume (for example, parts of the Mississippi Delta).
- Consolidation due to the weight of buildings: large buildings on soft soils cause slow, cumulative settlement.
- Permafrost thaw: in Arctic areas, thaw removes support from the ground, leading to subsidence and instability.
2.3 Dangerous interactions
- When a city combines intensive water extraction with fragile geomorphology, such as a delta, subsidence can reach tens of millimetres per year—a pace that becomes catastrophic over decades. The World Economic Forum and other analyses estimate that a significant share of the global population—tens of percent in certain countries—will be affected by subsidence by 2040.
3. Detailed case studies
We selected cities that illustrate different types of risk: delta megacities (Jakarta), heritage built on piles (Venice), delta conditions and water extraction (Bangkok), metropolises with fragile infrastructure and land loss (New Orleans), and North American and Asian cities facing a mix of risks (Miami and Shanghai).
3.1 Jakarta, Indonesia: the classic example of a lethal combination—rapid subsidence plus SLR
Jakarta is perhaps the most frequently cited example: entire neighbourhoods have already sunk below sea level because of extensive groundwater extraction for industry and urban use, combined with reduced sediment supply from human-altered deltas. Reported rates vary by area, from a few mm per year to 10–11 cm per year in some sectors; recent studies show very large spatial variations and locations with extreme subsidence. A 2024–2025 study confirms sinking on the order of tens of mm per year in parts of northern Jakarta. In response, the Indonesian government decided to move the capital to Borneo (Nusantara)—a decision motivated in part by Jakarta’s flood and subsidence risk.
Practical consequences: the drainage network becomes ineffective, pumps and barriers cost billions, properties become vulnerable, and tidal flooding (tide plus storm surge) becomes frequent—some areas are already below sea level for much of the time.
3.2 Venice, Italy: heritage and acqua alta—how much is caused by the sea and how much by subsidence?
Venice has a historic mix of factors: the city was built on wooden piles in estuarine sediments; alongside sea-level rise, there is a significant subsidence component resulting from sediment compaction and historic hydrogeological changes. Scientific literature shows that subsidence contributed significantly to relative sea-level rise over the past century. In recent years, the frequency of acqua alta episodes has increased, while the MOSE flood-defence infrastructure—mobile barriers—has entered operation to limit extreme flooding, although the system remains controversial on ecological, political and cost grounds.
Lessons: major works such as barriers and pumps can buy time, but may alter the lagoon’s sediment dynamics and have side effects, such as changing sand deposition and the emergence of new islands. Costs and maintenance also remain long-term challenges.
3.3 Bangkok, Thailand: groundwater extraction and a compact delta
The Bangkok–Chao Phraya area has experienced significant subsidence since the 1970s, caused primarily by over-pumping of aquifers for drinking water and industry. Journals and studies from 2022–2024 report cumulative settlement of decimetres in some areas and continue to warn of flood vulnerability; recent studies use SAR radar data to map subsidence in real time. Policy strategies have included restrictions on private wells, an emphasis on shifting to alternative sources, and modern pumping systems.
3.4 New Orleans, United States: coast, wetland loss and continuing subsidence
The New Orleans metropolitan area is affected by a series of processes: rapid land loss in the Mississippi Delta through erosion and marsh destruction, sediment consolidation, industrial extraction and historic water management, including levees that block the natural deposition of sediment. Recent 2025 studies using satellite radar show that certain areas are sinking at rates on the order of tens of mm per year—up to nearly 47 mm per year in some sections—while major wetland-restoration initiatives, such as sediment-diversion projects, face political and financial obstacles. The cancellation of some major projects has underscored the conflict between local economic interests and the need for long-term resilience.
3.5 Miami, United States: rising levels, coastal soils and vulnerable infrastructure
Miami and South Florida have highly porous coastal soils that allow saltwater to rise through drainage systems, combined with sea-level rise and severe high tides, known as king tides. Local projections from Miami-Dade County show increases of tens of centimetres by 2050, and adaptation planning is under way, including resilience strategies, pumps and raised roads. However, costs and hydrogeological limitations make permanent solutions difficult.
3.6 Shanghai and the Shanghai Plain: focused on managing underground resources
Shanghai historically experienced subsidence caused by water extraction during the 20th century, but rates have fallen following strict regulation and aquifer-recharge policies. Recent studies show that areas at greater risk are concentrated in densely urbanised districts, although the picture is complex because of underground infrastructure and soil composition. Continued monitoring and regulation of extraction have a visible effect.
4. Impacts: infrastructure, the economy and people
4.1 Critical infrastructure
- Roads, bridges, tunnels, pipelines and pumping stations all suffer from differential subsidence, in which some sections “sink” faster than others. This increases maintenance costs and shortens the lifespan of investments. Barriers and dikes must also be designed to accommodate vertical movement. Studies show that hundreds or thousands of buildings could enter the “risk zone” in the coming decades.
4.2 Socioeconomic impacts
- Property losses, falling real-estate values, internal migration—with the urban poor becoming the most affected—rising insurance costs, tourism losses, for example in Venice, and hardship for vulnerable communities, such as fishers and delta communities. Managed retreat, or planned relocation, raises ethical dilemmas: who receives the money, who pays and how can social cohesion be preserved?
4.3 Public health and the environment
- Saltwater contamination of freshwater aquifers, a greater risk of vector-borne diseases due to habitat changes, and loss of coastal ecosystems that buffer storms, such as mangroves and marshes. Wetland restoration brings co-benefits, including biodiversity and carbon sequestration, but requires time and space.
5. Responses and adaptation strategies
Strategies can be grouped into hard engineering, nature-based solutions, urban operations and planning, and socioeconomic governance (social resilience).
5.1 Barriers, pumps and movable gates
- Examples include MOSE in Venice, the Thames Barrier in London and works in New Orleans. These can reduce immediate risk, but involve enormous costs, ecological effects and a need for permanent maintenance. MOSE has demonstrated its effectiveness at certain times, but also its consequences for lagoon hydrodynamics and the controversy surrounding its costs and benefits.
5.2 Restoring wetlands, mangroves and deltas
- Restoring marshes and sediment movement can raise land in deltas over decades, but requires changes to river-management policies, such as reconnecting rivers with floodplains. The Netherlands’ “Room for the River” is an example of reshaping space to give the river water and room, rather than constructing massive dikes solely to confine its course. This type of intervention reduces risk and creates multifunctional green spaces.
5.3 Aquifer regulation and water policy
- Reducing illegal pumping, charging for water extraction and artificially recharging aquifers can all reduce subsidence rates in many cities that have relied on groundwater. Bangkok and Shanghai offer examples of how regulation can temper the phenomenon, although such measures bring social and economic costs, including relocating industries and changing water sources.
5.4 Planned relocation
- Redistributing people and infrastructure away from the most vulnerable areas, including coastal reforestation to stabilise the shoreline, is politically painful but realistic when the cost of protection exceeds the benefits. Practical examples are rare and political: there are local pilots and buyout programmes in flood-prone areas of the United States. The economic study of relocating Indonesia’s capital towards Nusantara is relevant: sometimes moving administrative functions is seen as the only long-term solution.
5.5 Combined and adaptive solutions
- The idea is not to choose a single solution that “fixes everything”, but a sequence of measures that allows change: for example, temporary protection first, through barriers and pumping, followed by ecological restoration and, if necessary, retreat. Each step has a trigger threshold based on continuous monitoring. Phased planning has become the norm in cities that cannot afford permanent protection.
6. How much does it cost?
- Costs vary dramatically: large barrier projects, such as MOSE and US systems protecting major cities, cost billions of dollars. Restoration interventions may cost a fraction of this amount, but take time—decades—to deliver benefits. Social costs, including housing loss and migration, are difficult to quantify but significant. Economic analyses by organisations such as Oxford Economics point to enormous spending associated with moving a capital or making the necessary investments in major metropolises.
7. Climate justice and equity
- Vulnerability is not evenly distributed: vulnerable groups, including poor people, migrants and Indigenous communities, suffer disproportionately. Decisions about allocating funds—such as which neighbourhood is to be protected—have moral implications: “who stays and who leaves?” Examples from New Orleans, where levels of protection differ between neighbourhoods, show that inequalities are reproduced in adaptation plans as well.
8. Projections and scenarios: short term, 2030–2050, and long term, 2100
- By 2050, many cities will experience more frequent coastal flooding and “nuisance flooding”, or tidal flooding that disrupts urban life, and the need for adaptation will become urgent. NOAA and local authorities, including Miami-Dade, project rises of tens of centimetres by 2050; combined with local subsidence, this could mean metres of relative effect.
- By 2100, depending on the emissions pathway, some low-lying areas could become effectively uninhabitable without permanent protection. Costly measures, such as barriers, may keep some areas functioning, but only at high prices and with continuous maintenance.
9. Practical recommendations
- Continuous monitoring: install GNSS stations, SAR radar imaging and hydrological networks to detect subsidence early. Recent observations in New Orleans demonstrate the crucial role of radar satellites.
- Strict regulation of water extraction: taxes, limits and alternative supply sources. Bangkok and Jakarta offer practical lessons.
- A mix of infrastructure measures: combine hard and nature-based solutions, including marsh and mangrove restoration. “Room for the River”-type programmes can be replicated where possible.
- Fiscal planning and dedicated funds: governments and international donors should create adaptation-finance mechanisms, including relocation and green-reconstruction funds.
- Social justice policies: community consultation, compensation programmes and social insurance for those who are relocated.
- Research and development: invest in new solutions, including aquifer recharge, flexible infrastructure materials and efficient pumping systems.
There is no one-size-fits-all solution
The “sinking cities” phenomenon varies at the local scale: some cases are dominated by local water extraction, such as Jakarta and Bangkok; others by combined historic and eustatic processes, such as Venice; while elsewhere land loss is accelerated by ecosystem loss, as in New Orleans. An effective response will be hybrid: technical measures, ecological restoration and, where unavoidable, planned relocation and social policies to mitigate injustice. Monitoring, long-term governance and smart financing remain essential.
