Sea level is rising.
That is not a prediction.
It is an observation.
The more difficult question is something else:
How much was sea level supposed to rise — and by when?
For more than four decades, scientific reports, government assessments, newspapers and public figures have published numbers for future sea-level rise.
Some were projections.
Some were scenarios.
Some were upper bounds.
Some became headlines.
Some deadlines have now passed.
Others remain decades away.
And if all of those numbers are placed in the same category, almost any conclusion can be manufactured.
So this audit begins with four distinctions:
GLOBAL MEAN SEA LEVEL ≠ LOCAL SEA LEVEL
PROJECTION ≠ UNCONDITIONAL PREDICTION
HIGH-END SCENARIO ≠ MOST-LIKELY OUTCOME
2100 ≠ TODAY
With those rules established, we can ask the question that matters:
What did the predictions actually say — and what happened when their deadlines arrived?
First: Sea Level Really Is Rising
Any audit that begins by denying the observational record would fail before it starts.
The Intergovernmental Panel on Climate Change estimates that global mean sea level increased by approximately 0.20 meters — 20 centimeters — between 1901 and 2018, with an assessed range of 15 to 25 centimeters.
The rate also increased.
According to the IPCC, average global sea-level rise was approximately:
1.3 mm per year from 1901–1971
1.9 mm per year from 1971–2006
3.7 mm per year from 2006–2018.
The satellite era provides an even more direct record.
NASA reports that global mean sea level has risen by approximately 10 centimeters since 1993.
NOAA's 2025 assessment places global mean sea level at approximately 111.2 millimeters above the 1993 level.
So the starting point is clear:
SEA-LEVEL RISE IS REAL.
The audit is about the accuracy of specific forecasts, not the existence of the phenomenon.
The 1983 EPA Scenarios
Now we reach one of the most useful historical tests.
In the early 1980s, the U.S. Environmental Protection Agency examined the possible consequences of greenhouse warming for future sea levels.
The projections were striking.
An EPA retrospective records that Hoffman and colleagues developed a range of scenarios in 1983 based on high and low assumptions about major uncertainties.
Their assessment estimated that sea level was most likely to rise:
26 to 39 centimeters by 2025
and
91 to 137 centimeters by 2075.
The first deadline has now arrived.
2025.
So unlike projections for 2050 or 2100, this one can actually be examined.
What Happened by 2025?
This is where the comparison becomes more complicated than simply putting two numbers beside each other.
Modern satellite altimetry began in 1993.
By 2025, NASA's satellite record showed approximately 10 centimeters of global mean sea-level rise since 1993.
NOAA's independent 2025 assessment gives approximately 11.1 centimeters above the 1993 reference level.
But we cannot simply write:
1983 PREDICTION: 26–39 CM
versus
OBSERVED: 10–11 CM
and declare the problem solved.
Why?
Because the baselines differ.
The historical projection was not defined as “rise from 1993.”
Satellite measurements are.
A valid audit must compare like with like.
Baselines Matter
Sea-level predictions are meaningless without their baseline.
Suppose one projection says:
+30 cm by 2025 relative to 1980
while an observation says:
+10 cm since 1993.
Those are not directly comparable numbers.
Thirteen years have disappeared from the comparison.
That does not mean comparison is impossible.
It means the historical baseline has to be reconstructed carefully using tide gauges and other observational records.
The broader observational evidence still indicates that the rise realized by 2025 was below the 26–39 cm range described in the historical EPA assessment.
But the difference should not be exaggerated by pretending the satellite record begins at the same date as the projection.
This distinction is essential.
Why Were Early Projections So High?
The 1983 assessment was being produced during an era of much greater uncertainty about future greenhouse-gas concentrations, climate sensitivity, thermal expansion and ice-sheet behavior.
Researchers were exploring possible futures.
The EPA summary itself describes the Hoffman estimates as a variety of sea-level-rise scenarios based on high and low assumptions for major uncertainties.
That matters.
A scenario is not necessarily equivalent to:
“This will happen.”
But the same EPA summary also describes 26–39 cm by 2025 as the range considered “most likely” in that assessment.
So it cannot simply be dismissed as an extreme theoretical maximum either.
The correct conclusion is narrower:
The 2025 rise was below the range anticipated in that early assessment.
Prediction Failure Does Not Mean Sea-Level Rise Failed
This is another distinction that should be obvious but often disappears in public arguments.
Suppose a doctor predicts that someone's temperature will increase by four degrees.
It increases by one degree.
The predicted magnitude was wrong.
The temperature still increased.
The same logic applies here.
PREDICTED SEA-LEVEL RISE TOO HIGH ≠ NO SEA-LEVEL RISE
And the reverse is also true:
SEA LEVEL ROSE ≠ EVERY SEA-LEVEL PREDICTION WAS ACCURATE
Both propositions can coexist.
The Measurement Era Changed Everything
In 1983, scientists did not have today's global satellite-altimetry network.
That changed in the early 1990s.
TOPEX/Poseidon began the modern satellite sea-level record, followed by the Jason missions and Sentinel-6.
These satellites allow researchers to measure global sea-surface height repeatedly across most of the world's oceans.
NOAA describes the reference satellite series as beginning with TOPEX/Poseidon and continuing through Jason-1, Jason-2, Jason-3 and Sentinel-6 Michael Freilich.
NASA now maintains a continuous global record beginning in 1993.
That means modern projections can be tested against measurements far more directly than early projections could.
And the Rate Is Accelerating
This is where the audit cuts the other way.
The fact that an early projection overshot the observed 2025 level does not imply that future sea-level rise will simply continue at the twentieth-century average.
NASA reports that the annual rate of global sea-level rise has more than doubled during the satellite era.
The IPCC similarly finds acceleration in the longer observational record.
The causes are well understood at the broad physical level.
Water expands as it warms.
Mountain glaciers lose mass.
Greenland loses ice.
Antarctica contributes additional mass loss.
Water previously stored on land can also move into the oceans.
So an early forecast can overshoot one deadline while the underlying physical process continues — and accelerates.
There is no contradiction.
Local Sea Level Is Not Global Sea Level
Now consider the Maldives.
Or Miami.
Or New York.
Or Stockholm.
Or Louisiana.
A global mean does not tell us exactly what sea level is doing at any one coast.
Local relative sea level can differ because of:
land subsidence,
land uplift,
ocean circulation,
winds,
gravitational effects from changing ice masses,
sediment compaction,
and other regional factors.
NOAA distinguishes these measurements explicitly: tide gauges measure sea level locally relative to land, while satellite measurements provide the average height of the global ocean.
This is why statements such as:
“Sea level has risen X centimeters.”
need another question:
Where?
The Maldives Prediction Was a Different Kind of Claim
We already audited the famous 1988 Maldives warning separately.
That claim was not merely:
“Global mean sea level will rise.”
It was associated with the far more dramatic consequence that the Maldives could be completely covered within approximately 30 years.
That deadline passed around 2018.
The Maldives remained.
That does not demonstrate that sea level did not rise.
It demonstrates that:
SEA-LEVEL RISE ≠ COMPLETE SUBMERGENCE OF THE MALDIVES BY 2018
Those are different claims.
And they require different verdicts.
What About Today's 2050 Prediction?
Modern assessments still project substantial future sea-level rise.
The IPCC's Sixth Assessment Report projects global mean sea level by 2050, relative to 1995–2014, at approximately:
0.18 meters under SSP1-1.9
to
0.23 meters under SSP5-8.5
with likely ranges around those central values.
Notice something important.
The spread between emissions scenarios is relatively modest by 2050.
Why?
Because much near-term sea-level rise is already influenced by warming and ice loss already underway.
By 2100, however, the scenarios diverge much more strongly.
The same IPCC assessment gives likely 2100 ranges extending approximately from:
0.28–0.55 m under SSP1-1.9
to
0.63–1.01 m under SSP5-8.5
relative to 1995–2014.
Those are projections.
Their deadlines have not arrived.
We Cannot Grade 2100 in 2026
This rule sounds trivial.
Yet it eliminates a surprising amount of bad argument.
A projection for 2100 cannot be classified today as:
FAILED
or
CONFIRMED.
We can ask whether observations currently fall within the projected trajectory.
We can evaluate the assumptions.
We can compare successive generations of models.
We can examine whether observed rates are accelerating.
But the actual deadline remains in the future.
Therefore:
2100 ≠ EXPIRED
Exactly the same rule applied to our polar-bear article.
The major two-thirds-by-mid-century projection cannot be given a final verdict in 2026 because mid-century has not arrived.
An honest prediction audit cannot move the deadline simply because doing so would produce a more dramatic conclusion.
The Danger of the High-End Number
Sea-level projections often contain ranges.
Suppose an assessment gives:
0.5 to 1.0 meters
under one scenario.
A headline may become:
SEA LEVEL COULD RISE ONE METER
A political speech may become:
SCIENTISTS SAY SEA LEVEL WILL RISE ONE METER
Years later, somebody checks the observation against one meter.
The scientific statement has changed three times.
This is precisely why our audit preserves the original language.
COULD ≠ WILL
UPPER RANGE ≠ CENTRAL ESTIMATE
SCENARIO ≠ FORECAST
LOW-CONFIDENCE HIGH-IMPACT OUTCOME ≠ MOST-LIKELY OUTCOME
Those distinctions are not excuses.
They are part of the prediction.
But Scientists Also Have to Own the Numbers
There is another side to this.
It would be equally misleading to publish dramatic numerical projections, allow them to dominate public communication, and later insist that they were never predictions of any kind because they technically belonged to scenarios.
If a report assigns a date, a range and an assessed likelihood, that information is testable once the deadline arrives.
The correct response is not to erase the historical number.
It is to preserve its conditions.
Then test it.
That is what the evidence audit does.
What Actually Happened?
The observational answer is straightforward.
Global mean sea level rose.
It rose throughout the twentieth century.
The rate increased.
Satellite measurements show continued rise since 1993.
NASA reports roughly 10 centimeters since 1993, while NOAA's 2025 assessment places the annual global mean approximately 111 millimeters above the 1993 level.
The IPCC estimates approximately 20 centimeters of global mean rise between 1901 and 2018.
So any prediction of:
NO SEA-LEVEL RISE
would have failed badly.
But some early projections of the magnitude and timing also overshot what occurred by their deadlines.
Both facts belong in the record.
THE VERDICT
THE OBSERVATION: Global mean sea level is rising, and the rate has accelerated.
THE LONG-TERM RECORD: Approximately 20 cm of global mean sea-level rise occurred between 1901 and 2018, according to the IPCC.
THE SATELLITE RECORD: Global mean sea level rose approximately 10–11 cm between 1993 and 2025.
THE EXPIRED HISTORICAL CASE: A 1983 EPA assessment reported Hoffman et al.'s most-likely range of approximately 26–39 cm by 2025. The observed rise by the deadline was lower, although precise comparison requires accounting for the different historical baseline.
THE MALDIVES CASE: Sea-level rise occurred, but the widely reported complete-submergence deadline around 2018 did not.
THE MODERN 2050 AND 2100 PROJECTIONS: NOT EXPIRED.
VERDICT:
SEA-LEVEL RISE: CONFIRMED.
ACCELERATION: OBSERVED.
SOME EARLY MAGNITUDE/DEADLINE PROJECTIONS: OVERSHOT.
2050/2100 PROJECTIONS: TOO EARLY FOR A FINAL VERDICT.
And that is exactly why the question:
“Were the sea-level predictions right?”
has no honest one-word answer.
You first have to ask:
Which prediction?
Which baseline?
Which scenario?
Which location?
Which deadline?
Only then can the number be tested.
From the Book
This article is part of the evidence-audit approach used in Patrick Panteix's CLIMATE: THE PREDICTIONS THAT EXPIRED — What We Were Told. What Actually Happened. Who Profited.
The method remains simple:
Preserve the prediction.
Preserve the number.
Preserve the baseline.
Preserve the scenario.
Preserve the deadline.
Then compare it with what actually happened.
100 predictions. 100 verdicts.
SOURCES
U.S. Environmental Protection Agency — Sea Level Rise and Wetland Loss: An Overview
EPA historical review recording the Hoffman et al. (1983) scenarios, including an estimated most-likely sea-level rise of 26–39 cm by 2025 and 91–137 cm by 2075.
https://nepis.epa.gov/Exe/ZyPURL.cgi?Dockey=P1015HEJ.txt
NASA — NASA Analysis Shows La Niña Limited Sea Level Rise in 2025
NASA's satellite assessment reports approximately 10 cm of global mean sea-level rise since 1993 and explains the observed 2025 change.
https://www.nasa.gov/earth/nasa-analysis-shows-la-nina-limited-sea-level-rise-in-2025/
NOAA — BAMS State of the Climate
NOAA's 2025 assessment reports global mean sea level at approximately 111.2 mm above the 1993 reference level.
https://www.ncei.noaa.gov/bams-state-of-climate
NOAA — Laboratory for Satellite Altimetry: Sea Level Rise
Satellite-altimetry measurements and global mean sea-level time series from TOPEX/Poseidon, Jason and Sentinel missions.
https://www.star.nesdis.noaa.gov/socd/lsa/SeaLevelRise/
NASA PO.DAAC — Global Mean Sea Level Dataset
NASA satellite-based global mean sea-level dataset beginning in 1993.
https://podaac.jpl.nasa.gov/dataset/NASA_SSH_GMSL_INDICATOR
IPCC — AR6 Working Group I, Chapter 9: Ocean, Cryosphere and Sea Level Change
Assessment of observed historical sea-level rise and projections for 2050 and 2100 under different emissions scenarios.
https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_Chapter09.pdf
IPCC — AR6 Working Group I: Summary for Policymakers
Reports approximately 0.20 m of global mean sea-level rise between 1901 and 2018 and the acceleration in the rate of rise.
https://www.ipcc.ch/report/ar6/wg1/chapter/summary-for-policymakers/
NOAA Ocean Service — Is Sea Level Rising?
Explanation of the difference between local tide-gauge measurements and satellite measurements of global mean sea level.
https://oceanservice.noaa.gov/facts/sealevel.html
