General
“New research models indicate a 60% probability of Thwaites Glacier fully detaching from the Mouginot Ridge before 2100." (Post caption variant: "The study identifies a 60% chance of Thwaites detaching from the Mouginot Ridge before 2100, which would trigger a self-reinforcing retreat. Even under zero-melt scenarios, the glacier continues…”
Plain restatementA recently published ice-sheet modeling study estimates roughly a 60 percent probability that Thwaites Glacier's grounding line retreats past (becomes unpinned from) a seafloor feature called the Mouginot Ridge by the year 2100.
This claim checks out in substance. A real peer-reviewed study published in Nature Communications on 12 October 2026, led by Mattia Poinelli with colleagues at UC Irvine, NASA JPL and other institutions, does find roughly a 60 percent probability that Thwaites Glacier becomes unpinned from a seafloor feature called the Mouginot Ridge before 2100. The important missing detail is that this number is conditional. It applies to the subset of model runs that assume a 4 to 6 kilometre wide grounding zone, while runs using a 2 to 3 kilometre zone give about 45 percent and runs with no grounding zone give zero. The projections also hold ocean conditions at their 1996 to 2023 average, so they are a baseline rather than a worst case. The phrase "fully detaching" is looser than the study's term "unpinning," and the study does not predict that the glacier collapses by 2100. One further issue is that the post credits the same study with the finding that Thwaites keeps losing ice for 150 years even with no ocean melting, but that result comes from a separate British Antarctic Survey paper in a different journal.
[drifted from the evidence:] New research models indicate a [drifted from the evidence:] 60% probability of Thwaites Glacier fully detaching from the Mouginot Ridge before 2100." (Post caption variant: "The study [drifted from the evidence:] identifies a 60% [drifted from the evidence:] chance of Thwaites [drifted from the evidence:] detaching from the Mouginot Ridge [drifted from the evidence:] before 2100, which would trigger a self-reinforcing retreat. Even under zero-melt scenarios, the [drifted from the evidence:] glacier continues losing ice for 150 years...")
A [added by the neutral restatement:] recently published ice-sheet modeling study [added by the neutral restatement:] estimates roughly a 60 [added by the neutral restatement:] percent probability that Thwaites [added by the neutral restatement:] Glacier's grounding line retreats past (becomes unpinned from) [added by the neutral restatement:] a seafloor feature called the Mouginot Ridge [added by the neutral restatement:] by the [added by the neutral restatement:] year 2100.
Red-tinted words in the claim drifted from the evidence. Green-tinted words are what a neutral restatement needs.
The trace / claim to source
- The study exists, is peer reviewed, and was published in Nature Communications on 12 October 2026. The institutions and lead author named in the post are correct.
- The 60 percent figure is real and appears both in the paper's results and in the lead author's quoted statements.
- "Mouginot Ridge" is a genuine named bathymetric feature, not an invented term.
- "Before 2100" matches the study's timeframe.
- The claim that unpinning would trigger self-reinforcing retreat is supported: the bed inland is predominantly retrograde, so retreat continues to the next high point.
- The post's 30 mm / 1.2 inch by 2100 figure and the "double previous estimates even without additional warming" framing match the paper's abstract.
- The post's "150 years of continued loss under zero-melt" statement corresponds to a real finding, though from a different study. The BAS paper found Thwaites continues to lose ice at a decreasing rate for 150 years despite the absence of ocean melting, demonstrating that Thwaites' present-day mass loss trend cannot be reversed by ocean cooling alone over policy-relevant centennial timescales.
- Omitted qualifier (the main problem): the 60 percent is not the study's overall probability. It is the probability within the subset of simulations using a 4–6 km grounding zone. The paper's own best-fit calibration range was 2–5 km, and 2–3 km runs give roughly 45 percent, while 0–1 km runs give zero. Presenting a single unconditional "60% probability" removes the conditionality that the paper itself foregrounds. This same simplification originated in the UC Irvine press release, so the post inherited rather than invented it.
- Omitted qualifier (scenario): the projections assume ocean and climate held at the 1996–2023 average. The post does gesture at this with "even without additional warming," but does not state that the figure is a constrained baseline rather than a full-range forecast.
- Wording escalation: the source term is "unpinning" or "grounding-line retreat beyond Mouginot Ridge." The claim's "fully detaching" is vaguer and, read quickly, can suggest the glacier itself breaking away or collapsing. The study does not project collapse by 2100.
- Source conflation: the caption says "the study" for both the 60 percent figure and the 150-year zero-melt finding. These come from two separate papers by two separate teams in two different journals (Poinelli et al. in Nature Communications; Williams et al. of the British Antarctic Survey in Geophysical Research Letters). The post's hashtags list British Antarctic Survey, but the text merges the two studies into one.
- Model output framed as near-certainty: "making future sea level rise unavoidable" is close to the BAS authors' own language about committed loss, but the post's overall framing presents modeled probabilities with more finality than the papers do.
- Secondary figure without clear sourcing: the post's "complete collapse could push oceans more than two feet higher, potentially leading to a rise of around 10 feet globally" blends two different quantities. Coverage of the BAS study puts total Thwaites collapse at an estimated 65cm to global sea level rise, roughly two feet. The ~10 foot figure refers to knock-on destabilization of the wider West Antarctic Ice Sheet, a separate and longer-timescale proposition, and the post does not make that distinction.
- I retrieved the Nature Communications text through search extracts rather than the complete article, so I could not read the full Figure 5b cumulative probability curve or the exact sentence giving the 60 percent value in the paper's own words. The 60 percent is directly attributed to the lead author in the institutional release and is consistent with the paper's stated 45 percent value for the 2–3 km case.
- Whether the paper assigns any weighting across grounding-zone widths (which would yield a single integrated probability) could not be determined from the extracts available.
- The post's "4 percent of worldwide sea level rise" figure for Thwaites and the "2028 instrument deployment" plan were not independently verified in this investigation.
The study is real, recent, and peer reviewed. The paper reports that grounding-zone processes govern the glacier's detachment from stabilizing seafloor ridges, enabling unpinning as early as 2038 and producing up to 30 ± 6 mm of global mean sea-level rise by 2100, approximately double previous projections even without additional climate warming. On the specific 60 percent figure, the paper's own results section states the probability is conditional on how wide the modeled grounding zone is. Simulations that omit a grounding zone or employ a narrow 1 km length, regardless of the imposed melt rate, do not retreat beyond Mouginot Ridge by 2100, while simulations with a 2–3 km grounding zone predict a ~45% probability of unpinning from the Mouginot Ridge by the end of the century. The lead author stated the higher figure explicitly as conditional: computer simulations that factor in a 4–6-kilometer grounding zone predict a 60% probability of unpinning the Mouginot Ridge by the end of the century. On what unpinning would mean, the paper describes a self-reinforcing geometry: once unpinned, the glacier continuously retreats across predominantly retrograde seafloor toward the next major topographic high, South Ridge, and Mouginot Ridge therefore exerts a stabilizing influence on the present-day position of Thwaites Glacier, with the likelihood of detachment being controlled by the length of the grounding zone. The Mouginot Ridge is a real, formally named feature. It was named in the 2024 PNAS paper as a bedrock ridge at 900 m depth, roughly 5 km long along flow, 9 km wide across flow, and 200 m high. That paper described it as a temporary barrier to glacier retreat whose prograde bed slope protects the glacier from marine ice sheet instability.
Complete reasoning
The reply is formatted for pasting into the thread where the claim is circulating.
Ask this case
Answers come only from the case file above; nothing is added.
Is it true that there's a 60% chance Thwaites Glacier detaches before 2100?
A real, peer-reviewed study does report a 60 percent figure, but it only applies to model runs that assume a wider 4 to 6 kilometre grounding zone. Runs with a narrower 2 to 3 kilometre zone show about 45 percent, and runs with no grounding zone show zero, so 60 percent is not the study's single overall probability.
What does 'detaching from the Mouginot Ridge' actually mean?
The study uses the term 'unpinning,' meaning the glacier's grounding line retreats past this seafloor ridge. This is different from the glacier fully breaking away or collapsing, which the study does not project happening by 2100.
Does this mean Thwaites will keep losing ice no matter what?
The claim that Thwaites keeps losing ice for 150 years even with no ocean melting comes from a separate study by the British Antarctic Survey, not from the Nature Communications paper that produced the 60 percent figure. The post incorrectly presents both findings as coming from the same study.
Is the Mouginot Ridge a real feature, and does this reflect worst-case warming?
Yes, Mouginot Ridge is a genuine named bedrock feature described in a 2024 PNAS paper as a stabilizing barrier to glacier retreat. The 60 percent projection assumes ocean and climate conditions held at their 1996 to 2023 average, so it represents a baseline scenario rather than a worst-case warming forecast.
What about the claim that collapse could raise sea levels by 10 feet?
The investigation found this figure blends two separate things: estimates of roughly two feet of sea level rise from Thwaites itself, and a much larger, longer-timescale figure tied to broader West Antarctic Ice Sheet destabilization. The post does not distinguish between these two separate claims.