If Thwaites Glacier in West Antarctica were to collapse entirely, it would raise global sea levels by roughly two feet on its own, and its loss could destabilize neighboring glaciers enough to eventually contribute several more feet. That alone would reshape coastlines worldwide, but the consequences extend well beyond flooding. The massive influx of cold, fresh meltwater would slow deep-ocean circulation, shift weather patterns in the Southern Hemisphere, and alter marine ecosystems from Antarctica to the tropics. The glacier is already retreating, and the physical processes driving that retreat look increasingly difficult to reverse.
Why Scientists Call It the Doomsday Glacier
Thwaites Glacier sits in the Amundsen Sea sector of West Antarctica and drains an area roughly the size of Great Britain. It earned its nickname not because its collapse is imminent in any Hollywood sense, but because the geometry of the bedrock beneath it creates a self-reinforcing retreat. The ground under Thwaites slopes downward as you move inland, meaning that as the glacier’s grounding line (the point where ice lifts off the seafloor and starts to float) retreats, it exposes ever-thicker ice to warm ocean water. Thicker ice at a deeper grounding line means faster melting and faster flow, which pushes the grounding line farther inland, which exposes still-thicker ice. This feedback loop is what glaciologists call marine ice sheet instability, and modeling shows that losing ice-shelf buttressing can trigger it, leading to around 0.6 meters of sea-level contribution within a century just from the West Antarctic Ice Sheet’s interior basins being opened up.1The Cryosphere. Stabilizing effect of mélange buttressing on the marine ice-cliff instability of the West Antarctic Ice Sheet
There is a related and more dramatic possibility called marine ice-cliff instability. If retreating ice exposes cliffs taller than about a kilometer at the ocean boundary, those cliffs may not be able to support their own weight. Modeling work shows that such cliffs can fail through a combination of viscous deformation, shear-band formation, and brittle cracking, with surface crevasses propagating through the full thickness of the ice and causing chunks to calve off by forward rotation.2PubMed Central. Marine ice-cliff instability modeling shows mixed-mode ice-cliff failure and yields calving rate parameterization West Antarctica’s basins are particularly vulnerable because they contain over-deepenings greater than a kilometer in depth. If cliff collapse were to begin in earnest, it could drive catastrophic retreat on timescales of decades to centuries.3PubMed. Transition to marine ice cliff instability controlled by ice thickness gradients and velocity
What Is Happening Right Now
Thwaites is not melting primarily from the surface the way you might picture an ice cube on a countertop. The main driver is warm, salty Circumpolar Deep Water that intrudes across the continental shelf and reaches the underside of the glacier’s floating ice shelves. This water is warm only by Antarctic standards, just a couple of degrees above freezing, but that is enough to erode the ice from below at rates far faster than snowfall can replenish it.4Journal of Physical Oceanography. Inflow of Warm Circumpolar Deep Water in the Central Amundsen Shelf Sensors placed beneath the floating ice via hot-water drilling have confirmed that this submarine melting is the central process driving the glacier’s retreat.5EGUsphere. Thwaites Eastern Ice Shelf Cavity Observations Reveal Multi-year Sea Ice Dynamics and Deep-Water Warming in Pine Island Bay, West Antarctica
Beneath the grounded portion of Thwaites, subglacial water channels also play a role. Research modeling the hydrology underneath the glacier found that channelized drainage amplifies submarine melting at the terminus and ice shelf, while simultaneously increasing friction farther upstream. The net effect is complicated: the channels speed up melting at the front while somewhat slowing flow deeper inland, but the melting at the front is winning the tug-of-war, and grounding-line retreat accelerates under present-day conditions.6The Cryosphere. Persistent, extensive channelized drainage modeled beneath Thwaites Glacier, West Antarctica7Journal of Glaciology. A fast and unified subglacial hydrological model applied to Thwaites Glacier, Antarctica
Recent work has also identified a tipping-point behavior in ocean water intrusion at the grounding zone. A small increase in ocean temperature can push the system across a threshold where the intrusion of warm water beneath grounded ice shifts from bounded (limited penetration) to effectively unbounded (runaway widening). The difference between a stable and unstable outcome in one model came down to an ocean warming of just a fraction of a degree, with dramatic implications for how quickly the grounded ice could thin.8Nature Geoscience. Tipping point in ice-sheet grounding-zone melting due to ocean water intrusion
The Ice Shelf That Acts as a Cork
Much of the near-term concern centers on the Thwaites Eastern Ice Shelf, the remaining large floating extension of the glacier. This shelf is pinned against a submarine ridge about 40 kilometers offshore. That contact acts like a doorstop: it provides “backstress” that slows the flow of ice from the interior toward the ocean. Satellite data show that this pinning point has been shrinking rapidly, with grounded ice elevations continuing to drop. If current thinning rates persist, the shelf could unpin from the seafloor within less than a decade, and ice discharge along a 45-kilometer stretch of the grounding line could increase by up to 10 percent as a result.9The Cryosphere. Weakening of the pinning point buttressing Thwaites Glacier, West Antarctica
The shelf has already been fragmenting. Over the past several years it has experienced brittle failure along a major shear zone, the formation of tensile cracks across its main body, and the release of tabular icebergs from both flanks. This damage originated when the shelf accelerated between 2002 and 2006, seeding cracks that basal melting and positive feedback between damage and strain rates then widened. Ice-sheet modeling suggests that further damage or complete unpinning of the shelf may not immediately trigger dramatically increased ice loss from the wider West Antarctic Ice Sheet, but the calving response remains highly uncertain.10The Cryosphere. Rapid fragmentation of Thwaites Eastern Ice Shelf That uncertainty is the core of the problem: once the cork is gone, we do not have confident predictions of how quickly the bottle empties.
How Sea Levels Would Actually Change
If you picture meltwater from Antarctica spreading evenly across the world’s oceans, you get a neat but wrong answer. Sea-level rise from ice-sheet collapse is profoundly uneven, a pattern scientists call a sea-level fingerprint. When a massive ice sheet shrinks, its gravitational pull on surrounding ocean water weakens, so water close to Antarctica actually drops. Meanwhile, water redistributes toward the far field, and the loss of ice mass shifts Earth’s rotation axis, creating additional asymmetries. A 2009 study found that a West Antarctic collapse would produce sea-level rise along the U.S. coastline about 5 to 10 percent higher than the global average, with peak values in the North Pacific reaching about 20 percent above the average. The same study showed the gravitational and rotational effects would push rises two to three times higher than previously estimated for U.S. coastal sites specifically.11PubMed. The sea-level fingerprint of West Antarctic collapse
Further modeling found that the specific geometry of where the ice melts within Antarctica matters too. When melting is concentrated near the ice-sheet margins rather than distributed uniformly, the excited polar motion roughly doubles, amplifying the rotational feedback signature by about 10 percent of the global average value. That rotational effect increases sea-level rise over North America and the southern Indian Ocean while reducing it over parts of Asia and southern South America.12Geophysical Journal International. On the robustness of predictions of sea level fingerprints In practical terms, this means places like New York, Miami, and the Pacific coast of North America face disproportionately large increases compared to the global mean.
Even modest additional sea-level rise has outsized effects when combined with storm surges and tides. Under current projections, extreme sea levels that historically struck a given coast once per century could become annual events at roughly 60 to 82 percent of the world’s tide gauges by 2100, depending on the emissions pathway.13Weather and Climate Extremes. Storm surges and extreme sea levels: Review, establishment of model intercomparison and coordination of surge climate projection efforts (SurgeMIP) Adding several feet from a West Antarctic collapse on top of that baseline would push those numbers even further.
Disrupting the Ocean’s Deep Circulation
The consequences of Thwaites melting are not limited to higher water. A massive release of cold, fresh meltwater into the Southern Ocean would dilute the dense, salty water that normally sinks near Antarctica to form Antarctic Bottom Water, the cold, oxygen-rich current that fills the deepest layers of every ocean basin. Modeling shows that enhanced meltwater fluxes could cause a near-complete shutdown of Antarctic Bottom Water formation within about 50 years under mid- to high-range warming scenarios. After 200 years, up to half of the original volume of this abyssal water mass could disappear as a result of warming from below in the absence of fresh cold-water input from the surface.14Journal of Climate. Projected Slowdown of Antarctic Bottom Water Formation in Response to Amplified Meltwater Contributions
This matters because Antarctic Bottom Water carries heat, carbon dioxide, and nutrients from the surface to the abyss. Without it, the deep ocean warms, its capacity to absorb carbon shrinks, and nutrient cycles that support fisheries at lower latitudes are disrupted. Earlier modeling work confirmed the general pattern: Antarctic meltwater reduces surface salinity, suppresses deep convection, and cools surface waters in the Southern Ocean.15Annals of Glaciology. Effect of meltwater input from the Antarctic ice sheet on the thermohaline circulation Observational studies have already detected signals consistent with this: meltwater and changing winds together have contributed to Southern Ocean surface cooling and periods of Antarctic sea-ice expansion over recent decades, in the opposite direction from what a simply warming world might suggest.16Geophysical Research Letters. Winds and Meltwater Together Lead to Southern Ocean Surface Cooling and Sea Ice Expansion
Ecological Ripple Effects
More meltwater pouring off Antarctica does not just change temperature and salinity; it changes ocean chemistry. Glacial meltwater delivers iron-rich particles to the surface Southern Ocean, and iron is the nutrient that limits phytoplankton growth across vast stretches of the region. Recent research found that these particles are rich in iron(II), a more biologically available form than the iron(III) typically found in seawater, and that organic carbon coating on the particles slows their oxidation, keeping them useful to phytoplankton for longer. Accelerating melt could therefore boost biological productivity in nearby waters.17PubMed Central. Antarctic glaciers export carbon-stabilised iron(II)-rich particles to the surface Southern Ocean
That sounds like a silver lining, and in the very near term it might be: more phytoplankton means more food for krill, which feeds whales, seals, and penguins. But the picture gets complicated quickly. If deep-ocean circulation slows as described above, the upwelling of nutrients from the abyss weakens, and phytoplankton productivity in the broader Southern Ocean could eventually decline even as iron-fertilized blooms spike near glacier outlets. Meanwhile, freshening of surface waters changes stratification, affects where sea ice forms, and alters habitat for species adapted to specific salinity and temperature ranges. The net ecological outcome depends on which effect dominates where, and that is genuinely hard to predict.
This Has Happened Before
The West Antarctic Ice Sheet is not a permanent feature. During the Last Interglacial period, roughly 129,000 to 116,000 years ago, polar temperatures were warmer than today and global sea levels stood about 6 to 9 meters higher than present. Greenland’s ice sheet, mountain glaciers, and ocean thermal expansion can account for only about 3 meters of that rise, which means Antarctica had to contribute the rest. Evidence from a blue-ice record in the Weddell Sea Embayment shows substantial ice loss in that region during the Last Interglacial, most likely driven by warming of Southern Ocean waters and associated destabilization of subglacial gas hydrates. Ice-sheet modeling suggests that millennial-scale ocean warming could have triggered a multi-meter rise in global sea levels from Antarctica alone.18PubMed Central. Early Last Interglacial ocean warming drove substantial ice mass loss from Antarctica
A recent critical review of all available evidence from that period, including far-field sea-level records, ice cores, glacial geology, and genomic data from organisms whose distributions would have been affected by open seaways, suggests that the Amundsen Sea sector (where Thwaites sits) was the most likely epicenter of collapse. The review identifies two plausible configurations of how much ice remained, corresponding to maximum Antarctic contributions of either 4 or 6 meters of sea-level rise. Genomic evidence from marine organisms indicates that seaways opened between the Weddell, Amundsen, and Ross Seas, reinforcing the picture of a dramatically diminished ice sheet.19Quaternary Science Reviews. Past interglacial West Antarctic Ice Sheet collapse: a critical review of evidence and approaches The precedent is clear: under conditions only modestly warmer than today, the West Antarctic Ice Sheet can and has lost most of its mass.
How Fast Could It Happen
This is the question with the widest uncertainty. Simulations of Thwaites Glacier specifically suggest that the onset of rapid collapse, defined as contributing more than one millimeter per year of sea-level rise, falls somewhere in the range of 200 to 900 years from now.20PubMed. Marine ice sheet collapse potentially under way for the Thwaites Glacier Basin, West Antarctica That is a large spread, and it reflects genuine uncertainty about ocean warming rates, ice dynamics, and whether feedback mechanisms like cliff collapse will kick in. The lower end of that range, a couple of centuries, is fast by geological standards but still spans multiple human generations. The upper end buys more time, but it does not change the eventual outcome if warming continues.
Satellite monitoring is tightening the uncertainty. Radar interferometry from missions like COSMO-SkyMed now provides grounding-line positions for dozens of Antarctic glaciers, including Thwaites, Pine Island, Totten, and Moscow University glaciers, allowing researchers to measure retreat rates over time rather than relying on snapshots.21PubMed Central. Antarctic grounding line delineation from the Italian Space Agency COSMO-SkyMed DInSAR data Each new dataset narrows the models’ range. For now, though, the honest answer is that we know the direction (retreat is under way and accelerating) but cannot pin down the speed with confidence.
The Economic Scale of the Problem
Adapting coastlines to rising seas is expensive under any scenario, and underestimating the upper end of possible sea-level rise makes the price tag worse. A 2023 analysis of global coastal adaptation costs found that failing to account for the full range of sea-level and socioeconomic uncertainty leads to large underestimates. Under a high-emissions pathway, the 95th-percentile cost of adaptation was underestimated by about $3 trillion in net present value, a gap of roughly 46 percent.22PubMed Central. Sea Level and Socioeconomic Uncertainty Drives High‐End Coastal Adaptation Costs Even under lower-emissions scenarios the underestimate ran to hundreds of billions. These numbers cover seawalls, managed retreat, building elevation, and similar measures; they do not include the economic value of lost land, displaced communities, or disrupted industries. The costs of inaction would be far larger.
The question of who pays for climate-related losses is unresolved in international law. Frameworks like the Paris Agreement acknowledge the concept of loss and damage, and principles like common but differentiated responsibilities suggest that nations with larger historical emissions and more resources should bear a greater share. But no enforceable regime currently exists to assign costs.23Indonesian Journal of Environmental Law and Sustainable Development. Who Pays for Climate Loss and Damage? State Obligations Under International Law Explained For low-lying island nations and coastal developing countries, the gap between what the law recognizes and what it can enforce is existential.
Could We Stop It With Engineering
A handful of researchers have explored the idea of building underwater curtains or artificial barriers on the seafloor to block warm Circumpolar Deep Water from reaching glacier grounding lines. The concept targets the root cause: if you can keep warm water away from the ice, you slow basal melting. Modeling suggests the approach could meaningfully reduce melt rates in the Amundsen Sea embayment. But the researchers themselves are cautious, noting that any construction in the region would require detailed environmental impact assessments under the Antarctic Treaty System and the Madrid Protocol, and that the potential effects on marine life need thorough study before anything could be attempted.24PubMed Central. The potential for stabilizing Amundsen Sea glaciers via underwater curtains
Beyond the technical hurdles, the governance challenges are steep. Antarctica is managed under an international treaty system that prohibits military activity and mineral extraction and requires environmental protection. Deploying large-scale infrastructure to alter ocean circulation around the continent would be a form of glacial geoengineering with significant and unforeseeable technical and environmental risks, raising questions that the existing treaty framework was never designed to answer.25International Affairs. ‘Ice sheet conservation’ and international discord: governing (potential) glacial geoengineering in Antarctica Getting dozens of treaty nations to agree on an intervention with uncertain consequences, in one of the most remote and ecologically sensitive places on Earth, would be a political challenge at least as large as the engineering one. For now, the most realistic lever remains reducing the greenhouse-gas emissions that are warming the ocean in the first place.