Global Mean Sea Level Determination: An Ocean Of Uncertainty

What follows is a good overview explaining why the measurement of Global Mean Sea Level (GMSL) is fraught with much uncertainty and so subject to substantial error. Michael Limburg of the European Institute for Climate and Energy tells us why.

Tide gauge system (Credit: University of Colorado)


On Determining Global Mean Sea Level

by Dipl.-Ing. Michael Limburg

An exact determination of GMSL is a very difficult if not a fundamentally impossible task. Even more difficult is determining sea level rise (or drop) over time. Different authors using the same datasets arrive at completely different results. It’s little wonder sea level expert W. Siefert in Hamburg recently said in an interview: “When examined closely, sea level is being exposed more and more as a pure mathematical prop, inadequate and, foremost, not very meaningful. Especially when it is to be used as a sole standard of measure, or used to derive horror scenarios…(1)“

This is also confirmed by researchers like Douglas [Douglas, 1994], who illustrated in great detail why e.g. Barnett (1984), Emery and Aubry (1991) Pirazzoli (1993) concluded:

…the determination of a single sea-level curve of global applicability is an illusory task.”

Douglas hoped that improved research instruments would bring better and more reliable results in the years ahead. And with the possibilities presented by satellite altimetry, these results may now be at hand. But so far mostly controversial results have been produced and no really reliable findings have been gained.

And later, in spite of the many new instruments and techniques that can now be used to find the much sought signal of global warming on GMSL, they say…

…these tools seem to have raised more questions than they have answered.”

Obtainable accuracy

It must also be allowed to say that a claimed measurement accuracy of a few tenths of a millimetre per year for the (only measurable) Relative Sea Level (RSL) and the GMSL are not possible with the available historical data. Only the newest satellite altimetry tools may allow this in principle. Therefore using the unit of measure “mm” is grossly misleading. The IPCC,  many experts, publicly informed media, and laymen are hence falsely claiming an accuracy that simply cannot be reached. These figures are computed values only.

In reality sea level changes can be measured accurately only to centimetres, and often only to within several centimetres – and very often not even that. Munk [Munk, 2003] (3) confirms this writing that…:

“…the jury is still out on the interpretation of the tide gauge records.”

For sea level and for trends from global historical temperature data, the old saying of Carl Friedrich Gauss (1777-1855) – known as a great inventor of many basic statistical principles and algorithm – remains true:

Nothing shows a lack of mathematical understanding more than an exaggeratedly accurate calculation.”

Therefore the only thing certain is that the statistical construct of GMSL over the last 120 years indicates a rise of between 10 and 20 cm/century. The error lies in the scale of the calculated value, and likely may be even higher. While Mörner expects a mean rise of 10 cm/century, the IPCC (AR4) sees approx. 19 cm/century. Today IPCC experts see an increase in the rate of rise over the last 20 years, while others explicitly exclude such a rate increase.

Why all the uncetrtainty?

It is due to built-in systematc errors. Many of these errors are widely unknown in their historical size, appearance and direction. They are included in the data and they involve an array of factors such as: dtermination of sea level measurement reference points, datasets of various lengths, contaminated datasets, rapid shifting of tectonic plates and their vertical components, barometric pressure, density of water, etc, etc.

A close analysis of all these error factors indicate that the errors are of a systematic nature and, because they are mostly subtle, they cannot be determined on the scale of the sought local sea level rise. Therefore they have to be indicated using error bars in accordance with good scientific practice. But this has very rarely been done. Credible figures regarding the attainable accuracy are as a rule the exception, e.g. Mörner +10 ± 10 cm by the year 2100 (or + 5 ± 15 cm) [Mörner, 2004] (2). Anything else has to be taken with much caution.

Assigning the causes of sea level rise

Detailed attempts to determine GMSLR are made by authors of the IPCC and others by breaking it down into various components. But one has to keep in mind that that this approach is prone to failure. Cazenave et al [Cazenave, 2004] is quoted on this:

…for the past 50 years, sea-level trends caused by change in ocean heat storage also show high regional variability,” ´


..has led to questions about whether the rate of 20th-century sea-level rise, based on poorly distributed historical tide gauges, is really representative of the true global mean.”

The estimates for eustatic and steric components cannot be brought in agreement with the observed data. The movement of single tectonic plates with speeds in the range of more than 15 cm/year, the vertical components therein, which can decisively impact the volume of the ocean above it, is certainly a cause of the observed changes in RSL and thus GMSL. But recording these changes and quantitively attributing them to a source has been impossible up to now.
Also a potential temperature-dependency is not detectable over the last 1000 years, as clearly shown by Storch et. al [Storch, 2008] (5). In their model that looks back, they did not find any correlation between sea level trend and temperature.

For all measurement locations, sea level changes can be more easily attributed to natural changes (glacial isostatic adjustment GIA / post glacial rebound PGR or other tectonic shifts) or, similar to the UHI for temperature, attributed indeed to man-made social-economic factors, e.g. urban growth and thus sinking. A greenhouse effect is not necessarily needed to explain it. This means future projections of the GMSL are purely speculative because of the great lack of understanding of the involved processes and the lack of data.

Nothing makes this more explicit than the wide range of estimates among IPCC lead authors (e.g. Rahmstorf) and other specialists: Jevreva, Mörner or Singer. Rahmstorf [Rahmstorf, 2007a](4) believes a maximum of 140 cm is possible by the end of the century, James Hansen estimates up to 600 cm under certain conditions, the IPCC shows estimates between 14 to 59 cm (final), Singer only 18-20 cm, and Mörner [Mörner, 2004] a mere 10 cm. This might be the reason why the IPCC authors are unusually cautious ( see for reference AR4 of WG I Observations: Oceanic Climate Change and Sea Level on page 410 Chapter 5.5.2.):

…there is an increasing opinion that the best estimate lies closer to 2 mm/yr than to 1 mm/yr…”


When basing conclusions solely on the ever-escalating opinions of a few scientists – some of them well known for blowing the horn of alarmism – political leaders should not decide on extremely costly measures to curb completely doubtful global sea leve rise, which is a mere statistical construct from the very beginning.

The only rational conclusion one can draw is: All global mean sea level rise claims with an accuracy of better than ± 10 cm/century have to be taken with great caution. Breaking down the rise into components and attributing a respective rise to each, especially to temperature rise, is not possible with today’s level of knowledge. Each assignment of factors is simply too speculative.

Michael Limburg, Germany
Vice President of the European Institute for Climate and Energy (EIKE)

Overview of sources:
1 Source Emder Zeitung, Sonntagsblatt, 26 February 1995. Prof. Dr. Winfried Siefert, sea level expert, worked years at the Hamburg authorities for economics and transportation, power and port building.
2. Mörner, N-A (2004) Estimating future sea level changes from past records. Global and Planetary Change
40:49 – 54.
3. Munk (2003) Uncertainties about Temporal Changes in Ocean Properties, their Effects on Sea Level, and
What they Imply about Global Warming. Science Volume 6, Number 28:
4. Rahmstorf, S a (2007) A semi-empirical approach to projecting future sea-level rise. Science 315:368–
5. Storch, HvZ, Eduardo González-Rouco, Jesús F. (2008), Relationship between global mean sea-level and
global mean temperature in a climate simulation of the past millennium. Ocean Dynamics


11 responses to “Global Mean Sea Level Determination: An Ocean Of Uncertainty”

  1. Juraj V.

    Can anyone show a photograph of any coast mark or harbor “then and now”, showing that the sea level visibly increased, even those claimed 20cm? My favourite one is Pointe-du Hoc in Normandy, even the tide is to be considered.



    1. Ed Caryl
    2. DirkH

      You have to take into account whether the particular spot subsides or is in a glacial rebound, which makes the comparison even more complicated. As far as i know, tectonic plates can “hinge”; in other words, when they rebound on one side they may subside on the opposite side.

  2. MostlyHarmless

    I’d agree that “the determination of a single sea-level curve of global applicability is an illusory task”. I’ve been doing an analysis of sea-levels around Australia, and it’s clear that any attempt to produce such a curve for that continent would be “an illusory task”, and even if it were done, would be meaningless; it would be quite likely not to represent the actual change anywhere around the coast. I say change rather than rise, because my results so far show not a rise, but a fall , or at worst stasis, since 1999 at all sites.

    Quoting figures in mm per year is largely meaningless, as the data shows large rises and falls over a few years, with long relatively stable periods between. By choosing start and end dates, it’s possible to produce almost any figure (within reason) for any site. For example, the CSIRO quotes a figure of 7.7 mm/yr for Darwin, between 1990 and 2008. This period includes the sharp rise of about 10 cm over a few years in the late 90s, which accounts for just about ALL the rise from 1966 to present. A more realistic average over the period from 1966 produces a much lower figure, as might be expected, but of course that doesn’t fit their agenda.

  3. Bob in Castlemaine

    Thanks Pierre for a thought provoking article.
    Here is a piece on sea level from University of Puerto Rico (1) it includes the usual genuflection acknowledging the AGW orthodoxy:
    Recent studies have highlighted carbon dioxide and other greenhouse gases as central to our understanding of global climate change. …etc, etc…
    But nonetheless it includes some interesting discussion on some of the factors you address.

    MostlyHarmless #5 and #6, it’s great to see someone is setting out to challenge the baseless alarmism concerning catastrophic sea level rise promoted by CSIRO and Australian federal, state and local governments.
    Australian coastal local governments now have legislation restricting coastal building based on irrational predictions of sea level rise, the impact is illustrated by these recent articles, (2) (3). The sad fact is that this sort of madness is affecting peoples lives by destroying the value of their properties based on shonky ideologically blinkered “science”. High time these meddling fools let people run their own lives, and make their own real world risk assessments.




    1. MostlyHarmless

      Thanks for that first link – it’s a good summary. There’s a slightly longer series for Stony Point than shown on the WUWT post, which shows that sea level has DROPPED since the 1980s.

      Luckily the BOM hasn’t found a valid reason for “homogenising” gauge data yet. I’m being unfair to them, they do a great job in making data easily accessible, even providing a helpline and response form for assistance if needed. Your mild rant reflects my own thinking on this subject. If I can speed the arrow on its way to the heart of this sort of alarmism, I’ll have done my bit.

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this. More information at our Data Privacy Policy