Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.

Settings

Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup

Settings


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe


Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...



Username
Password
Keep me logged in
New? Register here
Forgot your password?

Latest Posts

Archives

Misinterpreting a retraction of rising sea level predictions

Posted on 25 February 2010 by John Cook

A new skeptic argument has emerged that upon close inspection, is a polar opposite to the scientific reality. This week, scientists who published a 2009 paper on sea level rise retracted their prediction due to errors in their methodology. This has led some to claim sea levels are no longer predicted to rise. This interpretation was helped no doubt by the unfortunate Guardian headline "Climate scientists withdraw journal claims of rising sea levels". However, when you read the article and peruse the peer-reviewed science on future sea level, you learn that the opposite is the case.

The IPCC 4th Assessment Report predicted sea level will rise between 18 to 59 cm by the year 2100. Many consider this a conservative estimate as observed sea level rise is tracking at the top range of IPCC estimates (Rahmstorf 2007, Allison 2009). However, a study led by Mark Siddall examined how sea levels have changed over the past 22,000 years in response to temperature change (Siddall 2009). This enabled them to predict how sea level would respond to future warming, estimating sea level rise between 7 to 82 cm by the year 2100. Siddall's paper concluded that this increased confidence in the IPCC projections.

However, a later study using similar methods to Siddall 2009 came to dramatically different results, estimating sea level rise of 75 to 190 cm by 2100 (Vermeer & Rahmstorf 2009). Why the discrepancy? Judging by the acknowledgement in Siddall's retraction, one speculates that Vermeer and Rahmstorf discovered flaws in Siddall's methodology and notified the authors. Siddall saw that the errors undermined their results and retracted their paper. So we have two papers using similar methods - one predicting low sea level rise, the other predicting high sea level rise. The low sea level rise is found to be in error. While some are spinning this result to imply no sea level rise, in actuality it increases our confidence in high sea level rise.

Vermeer's results are confirmed by another study that approach the sea level question from a different angle, examining the dynamics of calving glaciers (Pfeffer 2008). The conclusion was a predicted sea level rise of 80 cm to 2 metres by 2100. Further evidence of the ice sheets' high sensitivity to warmer temperature comes from paleoclimate studies of the last interglacial period 125,000 years ago. At that time, global temperatures were around 2 degrees warmer than now. This is the amount of warming expected for some of the lower emission scenarios. At that time of the last interglacial, sea levels were at least 6 metres higher than present levels. So while we expect sea levels to rise up to 2 metres by 2100, they will continue to rise afterwards to at least 6 metres.

Future sea level rise will be one of the most serious impacts of global warming on humanity, with much of the world's population clustered around coastlines. That my daughter will see sea level rise of 1 to 2 metres in her lifetime is for me, an unhappy prospect. This scientific reality is a stark contrast to the 'Now You Can Forget About Those Rising Seas' attitude. Despite the serious picture painted by the peer-reviewed science, these kinds of misinterpretations turn the climate debate in an almost farcical direction. One could blame the Guardian for a carelessly worded headline. More blame should be apportioned to those who pontificate from their soapboxes without bothering to acquaint themselves with the science. That skeptics allow themselves to be tossed and turned by media headlines is the very antithesis of genuine skepticism.

0 0

Bookmark and Share Printable Version  |  Link to this page | Repost this Article Repost This

Comments

1  2  3  Next

Comments 1 to 50 out of 107:

  1. John
    This effect is what I wwas attempting to point out in the Volcanos thread. Not only are the El Ninos more pronounced but the La Ninas as well.
    0 0
  2. Is that a peer-reviewed journal? doesn't seem to me.

    Anyway, it's interesting to see that removing ENSO reveals that there was no significant warming trend from 1935 to 1985. In fact, there only seems to be a warming episode in the 1990's. What's the peer-reviewed explanation? Did CO2 suddenly become a powerful GHG in 1990?

    Also why did the authors not use the satellite data? Isn't that cherry picking?

    I'm just teasing. If you pretend to use "peer-reviewed" evidence to counter claims, you can do better than that! Maybe find papers in actual scientific journals?
    0 0
    Response: The Bulletin of the Australian Meteorological and Oceanographic Society is a peer reviewed journal. Don't forget that 1991 featured a huge volcanic eruption that lowered global temperatures significantly - their analysis finds the global cooling would've been even worse if there weren't El Nino conditions at the time. So global temperatures were recovering from the eruption throughout the 90's.
  3. John
    I don't know what the relationship is yet but mentally put a vertical by at 1918, 1938, 1958, 1978 and 1998 in figure 3 to contrast the graph against the PDO.
    0 0
  4. John
    Sorry,That was supposed to be 1938, 1968 and 1998 for PDO. I don't know how I did that. Sun on the brain I guess.
    0 0
  5. Good post and good website. It makes sense to remove the "el Nino of the century" anomaly in discerning the trend. Then there's the 11-year solar cycle, which we are now at the minimum on. For a variety of reasons, the "global warming stopped in 1998" is one of the more ridiculous arguments from the contrarian groups, although it's one that is easy to communicate to the layperson, so I understand why it's used.
    0 0
  6. NewYorkJ

    No! There is no 11-year solar signature!!! Hey, that's a contrarian argument too! Don't you know that the Sun has no influence on climate?

    As for Pinatubo, I was just saying that there was no trend up until 1990, so that's before Pinatubo. Furthermore, its effect lasted for about 2-3 years, not a whole decade.

    I don't see the point in trying to debunk something that does not need to be debunked. If temperatures have been stable since 1998, then so be it! Why twist the data this way or that way, use a running average, remove ENSO, remove Pinatubo, add this or remove that, and say, yes, there is a trend, the trend is still there, don't you see it?! It is as simplistic an argument as saying there is a downtrend since the last 7 years. A trend is a trend only up until it's no longer a trend. It could just be part of a long cycle, as far as I'm concerned.

    Anyone looking at those graphs objectively would say: there was warming up until 1940, then not much, then warming from 1990-2000, then not much again. Does that disprove AGW? No. Does the existence of a trend prove AGW? No again.
    0 0
  7. John
    You might be interested in this relevant article: Stay cool about short-term climate forecasts posted Thursday, May 01, 2008.
    0 0
  8. Very funny how so many attribute Global Warming trends to human activity, yet when defending periods of no warming, or cooling, they claim natural causes.
    0 0
    Response: The cooling period between 1940 to 1970 was due in some part to human activity as sulfate pollution increased atmospheric albedo, preventing as much sunlight from reaching the earth's surface.
  9. BTN,

    Well, there IS an asymmetry in the situation:
    - the expectation of GW fits into the framework that is CONSISTENT with the atmospheric physics that we know about.

    - the claim that GW is not happening poses problems for this framework. Specifically, what happened to the greenhouse effect? And if the answer is, There IS not greenhouse effect, then the problem becomes, Then why is the Earth's surface so warm, compared to what it "should" be?

    Since the expectation of GW is the result of the framework theory, it has to be DISPROVEN to be invalidated. The GW-denying theory is NOT consistent with the rest of atmospheric physics, so if you want to make that claim, you do have a higher standard of evidence to meet.
    0 0
  10. nealjking
    That is not how science works. GW is a hypothesis, albiet a fair one. The problem is that it has been accepted as a working hypothesis but has not been proven. Climate change, on the other hand, is obvious.
    Most of the natural forcing are proven, albeit not all.
    Science works by proving your hypothesis valid, which may or may not prove an alternate hypothesis invalid (two different functions can have similar results).
    Yes I have a higher standard of evidence to meet, but I do not deny GW, rather I am very skeptical of CO2 as the cause of GW or that AGW is actually an important factor in climate change. There have been very good peer reviewed papers indicating natural causes that fit the picture much better than CO2 induced AGW. The jury is still out on this. I am not posting links here because they do not relate directly to this issue (Did global warming stop in 1998?) but if you read through the rest of this site you will find plenty of links to some very interesting papers (both from John and in the comments). But "framework theory, it has to be DISPROVEN to be invalidated." is a false premise as it too is a hypothesis, albeit a good one.
    0 0
  11. Is everyone forgotting the famous quote of Kevin Ternberth from NCAR?

    http://www.npr.org/templates/story/story.php?storyId=88520025

    "...But if the aquatic robots are actually telling the right story, that raises a new question: Where is the extra heat all going?

    Kevin Trenberth at the National Center for Atmospheric Research says it's probably going back out into space. The Earth has a number of natural thermostats, including clouds, which can either trap heat and turn up the temperature, or reflect sunlight and help cool the planet.

    That can't be directly measured at the moment, however.

    "Unfortunately, we don't have adequate tracking of clouds to determine exactly what role they've been playing during this period," Trenberth says."

    Hello? Can you say, "Heat exchange system?"
    0 0
  12. Actually, if you look at a chart showing the last 5 million years you can easily see the slow increase in temperature. The most notable part is that while the highs are marginal (small slope) the lows are drastic (large slope). This can not be seen on the 450 thousand year graph at all (like the one on wikipedia). Can you say "cherry picking"?
    0 0
  13. John PS
    Can you say "cherry picking"? is in reference to Wikipedia, not your article.
    0 0
  14. John said>> They find the linear trend over 1998 to 2007 is a warming trend in all three data-sets.

    Taken alone, none of the three data sets in your Figure 2 above show any credible warming trend. In this data, the trend lines appear to trend very slightly upward only because of the skewing effect of the two lowest points in 1999 & 2000.
    0 0
  15. Re #12:

    Quietman, the last 5 million years has shown a slow decrease in temperature. That can be seen here, for example:

    Lisiecki, L. E., and M. E. Raymo (2005), A Pliocene-Pleistocene stack of 57 globally distributed benthic δ18O records, Paleoceanography, 20, PA1003.

    or:

    http://en.wikipedia.org/wiki/Geologic_temperature_record


    what data are you loking at?
    0 0
  16. Re #14 and more generally.

    There seems to be a bit of nonsense over this. The gloabl temperature of 1998 was lifted by around 0.2 oC above the trend by the strongest El Nino of the 20th century.

    The temperature of 2005 was statistically indistinguishable from 1998. However this was reached without the "aid" of the massive El Nino warming (e.g. http://data.giss.nasa.gov/gistemp/2005/)

    So one could say that "global warming stopped in 2005". But why bother?!


    Since 2005 the solar cycle has been in its waning phase (we're pretty much smack at the bottom now)...we've had a La Nina suppressing temperatures during the early months of this year.

    As John Cook illustrates internal variations and extrinsinc factors introduce "noise" onto the long term trend. So it's pretty unremarkable that the temperature goes up and down a bit as it rises under the influence of enhanced greenhouse forcing.

    Probably the next record warm year will occur during the next significant El Nino or two....
    0 0
  17. The wording I used is misleading and I did not realize how it would be taken. The high points decline slightly but the lower end is constantly getting warmer, ie. instead of saying it's getting warmer I should have said its not getting as cold. Just follow the trend line of the coldest points to see what I meant.
    0 0
  18. chris
    forget it, the damn graph was drawn backwards.
    0 0
  19. Re #16,

    Chris,

    Despite a very verbose reply in refering to #14, you totally ignored the point of my post, which was in direct response to John's claim that;

    "They (Fawcett and Jones) find the linear trend over 1998 to 2007 is a warming trend in all three data-sets."

    My point was that, based on the data sets presented in figure 2 above, the linear trends only appear to trend very slightly upward because of the skewing effect of the two lowest points in 1999 & 2000.

    There is therefore no credible warming trend in the data presented at all.
    0 0
  20. No, I addressed your point in my response in post #14, Healthy Skeptic. To be specific:

    The linear warming trend in the Fawcett/Jones data is exactly that. A linear warming trend (or, to be precise, three linear warming trends, two strongish, one weak), is simply a mathematical fact.

    You consider that the linear trend is "skewed" "by the effect of the two lowest points in 1999 and 2000. But those two points are representative of the global temperature anomaly around that time. The point that "skews" the progression of the global temperature is 1998 which was lifted around 0.2 oC above the long term trend by the strongest El Nino of the 20th century.

    So there is a very clear warming trend in the data. It's mathematically precise in relation to linear regression of an 11 year running average of the raw data, and is more apparent in the analysis presented in which the effects of internal variations (due to ENSO) are removed.

    The point that Jones and Fawcett are making is that there is no scientific basis for proposing that there hasn't been any greenhouse-induced warming since 1998.

    Of course one can argue endlessly over temperature variations during very short periods, and the fact that the temperature anomaly has been steadyish for the last couple of years is cat-nip for those who want to progress spurious "arguments". Fawcett and Jones are just pointing out (yet again!) that those arguments don't have much basis in fact.

    Of course we'd like to wait a few years to see how things progress. In the meantime the most reliable means of adressing the temperature anomaly trend is to consider substantial running averages in which internal variations are roughly averaged out....
    0 0
  21. Chris,

    LOL! Talk about creative interpretation of the data!

    Unless there is a clear and continuous upwards trend in a set of data, applying a linear trend to it means absolutely nothing. This sort of misinterpretation is a trap you young players. There is no continuous upwards trend from 2002 to 2007 (which represents 60% of the data). A linear trend of this region is dead flat!

    If 1998 was such an "anomalous year" why do the values from 2002 to 2007 statistically differ very little from the 1998 value, and how does this fact support a "warming" trend?
    0 0
  22. re #21

    Unless there is a clear and continuous upwards trend in a set of data, applying a linear trend to it means absolutely nothing


    Not really HS. A linear trend is a linear trend. The whole point of determining a linear trend over a significant time period (7 or 10 or 11 or whatever years), is to establish significant progression of a variable in the context of stochastic variability ("noise").

    If 1998 was such an "anomalous year" why do the values from 2002 to 2007 statistically differ very little from the 1998 value, and how does this fact support a "warming" trend?


    That's the point, HS. 1998 was anomalously warm; the strong El Nino raised the global temperature by around 0.2 oC above the trend. Now (2002 through 2007) the global temperature has reached the anomalously warm temperature of 1998 without the warming "pulse" of a strong El Nino...

    ..that's how the warmth of 2002-2007 supports a warming trend. It's about 0.2 oC warmer in this period than the equivalent period 10 years previously.
    0 0
  23. Like I said...talk about creative interpretation of the data!
    0 0
  24. Have a look at this:

    http://www.friendsofscience.org/assets/documents/FOS%20Essay/GlobalTroposphereTemperaturesAverage.jpg

    which shows a linearised temp trend downwards since 2002, an actual decrease of 0.2C in 6 years. Yes, 1998 was an anomaly, but the cooling trend is continuing.
    0 0
  25. Thanks Mizimi,

    I can't wait for Chris' creative interpretation of this new data.

    When reading the protestations of fervent AGW proponents I am constantly reminded of creationist leader Henry Morris who decreed that any scientific evidence that did not support a 'young earth' was to be "explained away".
    0 0
  26. re # 2
    "The Society keeps members informed on current news and activities through the regular distribution of the bi-monthly Bulletin of the Australian Meteorological and Oceanographic Society."

    That does not sound like a peer review journal to me.
    Sounds more like an academy mouthpiece. However, I do not know anything about the journal, this is just what i picked up at their website.


    re # 9
    "Since the expectation of GW is the result of the framework theory, it has to be DISPROVEN to be invalidated."

    that is not how science works
    You generate a hypothesis
    you find supporting data
    if over the years ANY data is found that disproves the hypothesis. The hypothesis is wrong.

    or at least that is the scientific metyhod i have been used to in my line of work.
    Karl popper agrees.
    0 0
  27. oh
    and yes
    any hypothesis must be falsifiable
    that it means that there must be a clear testable means ( now or in the future) wherby you say that if X is true the hypothesis must be false.

    so how do we falsify AGW. what X must be true for AGW to be falsifiable.
    0 0
  28. In a way, this paper --> investigation by another group --> questions raised --> voluntary retraction is exactly the sort of way in which science is supposed to work. However, it's going to be tough convincing some of the non-scientists of this, since they don't know just how well scientists police themselves.
    0 0
  29. "Skeptics" (we'll call them that for now) confuse the facts supporting anthropogenic climate change (ACC), and the effects of those changes.

    Estimations of changes to sea levels falls into the category of effects of ACC.

    Whether or not the models to project the effects of ACC were off a bit still does not change the facts supporting ACC.
    0 0
  30. The reaction to this one really has been in the up is down category.

    Many or most scientists who work in glaciology, and many who study sea level, consider the IPCC's projections to be only a lower bound on future sea level rise. That is, actual sea level rise is almost certain to exceed the IPCC projections. Why? Because the IPCC assumed that glacier flow would not speed up as temperatures warm. Their reason for excluding a contribution from enhanced glacier flow in the future is that the uncertainties in future projections of it were large based on the literature as of the IPCC cutoff date. In the Third report a few years ago, they had tried to include this contribution, but this time they did not.

    There are two problems with the IPCC's decision. First, a slowdown in glacier flow with warming temperature is very, very unikely. What we have seen so far is that glacier flow is speeding up with increasing temperature. It is just hard to predict how much more speedup there will be in the future. Second, the major contribution to sea level from Greenland and Antarctica does not come from meltwater running off, but from glaciers flowing faster and dumping ice into the ocean, where it melts. So the bottom line is that the IPCC projection excluded a component that is highly uncertain, but almost certainly positive. Sea level rise will almost certainly be faster than the IPCC projection, but there is disagreement on how much faster.

    That's the context of the Siddall paper. They got a relatively low projected future rate, which would have been a fairly big deal given that a lot of other work in the last few years has pointed to much higher rates. John is absolutely right -- the authors retracted their paper's conclusions of low future sea level rise, because when they fixed their error, they get a much higher future prediction.

    They explained the error in the retraction letter. Their basic method was sound. But the time step they used in the integration of their model was too long to be used for 20th century sea level rise and 21st century predictions. Effectively, by using a time step that was appropriate only for reconstruction over a much longer time scale, they smoothed out the recent and projected rise, getting a lowball estimate.
    0 0
  31. Nice work John. And once again more references than I can look at.
    0 0
  32. From the abstract:

    "First, we tested the sensitivity of our results to the length of the time step used in the integration of the model for the period of deglaciation, which we found to be robust. However, we overlooked that the simulations of the twentieth and twenty-first centuries are sensitive to this time step, which led to an overestimation of the sea-level response to warming in the simulations for these centuries."

    Aren't they being a bit more specific than saying that their methods are wrong and their conclusions are simply invalid? Are they not saying that their methodology led them to overestimate sea level rise?
    0 0
  33. An interesting peer reviewed article published in Geophysical Research Letters is :"Recent global sea level acceleration started over 200 years ago?", Jevrejeva, Moore, Grinsted, and Woodworth. GRL 2008.

    http://www.pol.ac.uk/psmsl/author_archive/jevrejeva_etal_1700/2008GL033611.pdf

    Only 4 pages long and a reasonably easy read, but you can also just skip down to Figure 3, which extracts from the sea level instrumental record a basic oscillatory trend of 60 year period which is superimposed on an accelerating trend starting in the late 1700's.
    0 0
  34. Global sea level linked to global temperature
    http://www.pnas.org/content/early/2009/12/04/0907765106.full.pdf

    Vermeer & Rahmstorf 2009 write:
    "When applied to observed data of sea level and temperature for 1880-2000, and taking into account known anthropogenic hydrologic contributions to sea level, the correlation is >0.99, explaining 98% of the variance"

    Unfortunately it does not make sense. Since observed sea level data are taken from IPCC AR4 WG1 where twentieth century sea level rise is estimated to be 14±10 cm (note the wide error margin), a 0.99 correlation during calibration phase is meaningless for a model with built-in exponential behavior. The 190 cm figure is baseless, Vermeer & Rahmstorf 2009 should be retracted as well.

    Even more important. Contrary to your statement "The low sea level rise is found to be in error. While some are spinning this result to imply no sea level rise, in actuality it increases our confidence in high sea level rise" Siddall's retraction letter says: "we overlooked that the simulations of the twentieth and twenty-first centuries are sensitive to this time step, which led to an overestimation of the sea-level response to warming in the simulations for these centuries"

    http://www.nature.com/ngeo/journal/vaop/ncurrent/full/ngeo780.html

    Overestimation. The plain meaning is that the 82 cm figure is too large, even for the worst model scenario.

    We have two statements:

    1. Sea level rise during 21st century is 75-190 cm
    2. 7-82 cm is an overestimation

    They are extremely hard to be seen as strengthening each other.
    0 0
  35. Re # 7: "The 190 cm figure is baseless, Vermeer & Rahmstorf 2009 should be retracted as well."

    If you are so confident in your science and analysis (doubtful if you have done any) then please submit a comment/rebuttal to PNAS requesting a retraction of Vermeer and Rahmstorf, or even better, submit a paper describing your own data, methodology and analysis and conclusions. If you can convince the reviewers and scientific community that your analysis and projections are both superior and reliable, then I will accept your lower estimate until such time as newer evidence suggests otherwise.
    0 0
  36. This is the amount of warming expected for some of the lower emission scenarios. At that time of the last interglacial, sea levels were at least 6 metres higher than present levels. So while we expect sea levels to rise up to 2 metres by 2100, they will continue to rise afterwards to at least 6 metres.


    It's likely more than 6 meters. Sea level changes due to ice melt are very slow to materialize, and the last interglacial was relatively short-lived.
    0 0
    Response: Good point. Kopp found it very likely (95% probability) that in the last interglacial, sea levels were at least 6.6 metres higher than today. It's likely (67% probability) that sea levels exceeded 8 metres (Kopp 2009). Note that temperatures during the last interglacial were around 2 degrees warmer than now. This is the amount of warming expected for some of the lower emission scenarios. This is important to realise when you compare current emissions to the IPCC scenarios. I will be posting about that later today (if time permits).
  37. #7 Berényi Péter, it helps to read the Siddall paper, and also Vermeer and Rahmstorf's post at RealClimate last August, which appears to be where they spotted the error (especially check the update just before the footnotes):

    http://www.realclimate.org/index.php/archives/2009/08/ups-and-downs-of-sea-level-projections/

    In the paper, they modeled sea level as being a function of a*dT+b (they used a different notation; this matches Vermeer and Rahmstorf in RealClimate and is simpler. dT is change in temperature). The a and b are determined empirically by fitting the data. Their error caused them to get incorrect estimates for both a and b. Their retraction mentions only the effect on a. If you correct the error in their code, then their model no longer fits 20th century sea level rise (it underpredicts it), so its prediction for the 21st century obviously can't be trusted. If you increased b so that their model fits 20th century rise, then you would get a larger 21st century prediction. This might mean that you can't model both post-LGM and 20th century sea level change with the same a and b. (They mention a second error, and I don't know what effect that has, but it really doesn't matter because the entire result needs to be thrown away and redone).

    Or better yet, revert back to Rahmstorf's 2007 paper or to Grinsted et al. (2009), both of which lack errors, and fit the historical data. These papers predict higher future rates, by the way. I have not looked at Vermeer and Rahmstorf (2009), just downloaded it but don't have time to read it now.

    By the way, your statement 2 is based on your own misinterpretation (jumping to conclusions). So your final argument is meaningless.
    0 0
  38. Berényi Péter sez:

    Siddall's retraction letter says: "we overlooked that the simulations of the twentieth and twenty-first centuries are sensitive to this time step, which led to an overestimation of the sea-level response to warming in the simulations for these centuries


    He's assuming that "response" means "sea level rise". That's not the fact.

    From Rahmstorf's original criticism at Real Climate, which eventually led to the correspondance which led to Siddal retracting the paper:


    Siddall et al. in contrast find a time scale of 2900 years, but introduce a non-linearity in the equilibrium response of sea level to temperature (see their curve in Fig. 1 and footnote 3 below): it flattens off strongly for warm temperatures.


    The response which is overestimated is this flattening of sea level rise, not sea level rise itself.

    Remove that overestimated flattening and the sea level rise estimates ... rise!

    Here's more of Rahmstorf's RC note on this:

    The reason for both the long time scale and the shape of their equilibrium curve is that this curve is dominated by ice volume changes. The flattening at the warm end is because sea level has little scope to rise much further once the Earth has run out of ice. However, their model is constructed so that this equilibrium curve determines the rate of sea level rise right from the beginning of melting, when the shortage of ice arising later should not play a role yet. Hence, we consider this nonlinearity, which is partly responsible for the lower future projections compared to R07, physically unrealistic. In contrast, there are some good reasons for the assumption of linearity (see below).


    Note that the retraction thanks Vermeer and Rahmstorf for bringiing the overestimation (of flattening) to their attention, so Siddal at least is convinced they're right.

    Berényi Péter, why did you assume "response" meant "sea level rise"? It's a reasonable guess, but you don't have to guess, you know, you can always read instead.
    0 0
  39. In the paper, they modeled sea level as being a function of a*dT+b (they used a different notation; this matches Vermeer and Rahmstorf in RealClimate and is simpler. dT is change in temperature).


    That sounds way too simplistic. I haven't taken the time to read the relevant papers, though.

    A model of SLR should consider ice melt and thermal expansion separately. Thermal expansion is non-linear. I was looking at density of water given its temperature and salinity, and a quadratic equation fits it quite well. It's complicated to do an ocean estimate because the temperature of ocean water varies with depth, and modeling this seems non-trivial by itself.

    Ice melt looks even more complicated. You can probably come up with a model for 'equilibrium' ice volume. But then the speed of ice melt is tricky. This would have to depend on the difference between the current ice volume and the equilibrium ice volume, times the surface area of heat transfer.
    0 0
  40. #12 joseph, I did simplify it considerably (maybe too much). Siddall et al. defined the equilibrium sea level in terms of the inverse hyperbolic sine of a quantity that was linear in temperature. The modeled sea level then follows a differential equation of which the equilibrium sea level as a function of time is the inhomogeneous part.

    dhogaza's explanation (#11) is better than mine was.
    0 0
  41. #8 Albatross at 06:51 AM on 25 February, 2010:
    "If you can convince the reviewers and scientific community that your analysis and projections are both superior and reliable, then I will accept your lower estimate until such time as newer evidence suggests otherwise"

    Man, try to think for yourself. It is not so difficult. Of course I don't have any estimate of my own. I just put two estimates side by side, one from Vermeer & Rahmstorf, from Siddall's retraction letter the other one. I care neither for reviewers nor the scientific community, but I do care for logic & truth.

    Appealing to authority is NOT the way science is done. Not even scientific authority is competent in this respect. As long as you are able to understand stuff, you are self sufficient.
    0 0
  42. Let me start off on my first post here and say I am not a scientist and I admit that reading through these posts makes my brain hurt sometimes. I am trying to learn though.

    I do have a question for Berényi Péter...in your self sufficiency of logic and truth suppose an error creeps in to your work which results in your conclusion being in error. You review your work following the same line of thought and as the process is unchanged then the error remains. In your eyes then your result is "Truth" but ultimately it is false. That is the benefit and need of reviewers and the community at large. The more input, the more eyes on a problem the more self correcting for errors.
    0 0
  43. shdwsnlite, you sure sound like a scientist! I am one, and I agree with you completely.
    0 0
  44. #10 Jeff Freymueller at 07:39 AM on 25 February, 2010:
    "Vermeer and Rahmstorf's post at RealClimate last August, which appears to be where they spotted the error"

    OK. In that post they say:
    "To constrain the value of a – which dominates the 21st Century projections — one needs to look at the “new rise”. How much has sea level rise accelerated over the 20th Century, in response to rising temperatures? That determines how much it will accelerate in future when warming continues"
    http://www.realclimate.org/index.php/archives/2009/08/ups-and-downs-of-sea-level-projections/

    Sounds reasonable. Let's see.

    PSMSL (Permanent Service for Mean Sea Level) tide gauge data from Proudman Oceanographic Laboratory
    http://www.pol.ac.uk/psmsl/datainfo/
    http://www.pol.ac.uk/psmsl/psmsl_individual_stations.html

    New York tide gauge
    http://www.pol.ac.uk/psmsl/pubi/rlr.annual.plots/960121.gif
    http://www.pol.ac.uk/psmsl/pubi/rlr.annual.data/960121.rlrdata

    No 20th century acceleration at New York tide gauge. Not a bit. Some deceleration, if anything.

    If global sea level rise is accelerating nevertheless, we can conclude that New York is accelerating upwards on a slightly faster rate. Odd enough.

    There are quite some tide gauges in PSMSL with long records. Would anyone set about computing acceleration for each one? It is not difficult, just time consuming. BTW, acceleration should be more precise than average rate of rise, since crustal slabs are not particularly brisk.
    0 0
  45. #15 shdwsnlite at 09:26 AM on 25 February, 2010:
    "The more input, the more eyes on a problem the more self correcting for errors"

    Agreed. This is what I am trying to do here. Unfortunately most guys can only quote authoritative sources instead of having their eyes on the problem itself.

    I am prone to err as anyone, correction is welcome. However, in this topic I have not seen a valid one yet.
    0 0
  46. #18 Berényi Péter,
    You're claiming you can eye-ball a single tide guage and determine if the global sea-level rise is accelerating or decelerating? Or even a few individual tide guages? That's amazing!

    In #7 you wrote:
    "The 190 cm figure is baseless, Vermeer & Rahmstorf 2009 should be retracted as well."

    I'm sure it will, if (and only if) someone comes along to point out an actual flaw in their methodology or conclusions. But eye-balling individual tide guages doesn't accomplish that.
    0 0
  47. @ 18 Berényi Péter..-- you posted:
    New York tide gauge
    http://www.pol.ac.uk/psmsl/pubi/rlr.annual.plots/960121.gif
    http://www.pol.ac.uk/psmsl/pubi/rlr.annual.data/960121.rlrdata

    No 20th century acceleration at New York tide gauge. Not a bit. Some deceleration, if anything.

    How did you reach that conclusion? What was the process?
    0 0
  48. #18 Berényi Péter, their model sea level includes a response time in the differential equation that they solve. This deals with the fact that the response to a step increase in temperature would not be a step increase in sea level, but instead sea level would start to rise for a while, then level off after some time. It looks to me like they estimated this constant based on the fit to the last 22,000 years of post-LGM sea level rise (the details are in the supplementary material, not the main paper).

    This means that if there was no change in temperature starting in 1900, for example, you would still expect sea level to rise for a while because of past temperature changes, but the rate of rise would go down with time. So the 'new rise' is the response to the temperature changes of the 20th century. I find Rahmstorf's approach easier to understand, but the way, and it is probably more useful to discuss that approach. (Another time, at least for me, I need to log off here and get some work done).

    As for acceleration of sea level rise, you can see from Figure 3 of Vermeer and Rahmstorf that the rate of global sea level rise has increased with time (and this rate is faster than the average over the last few thousand years). Whether or not you can detect an acceleration in a single tide gauge record is really beside the point, as I said a few days ago.
    0 0
  49. #21 shdwsnlite at 10:41 AM on 25 February, 2010:
    "What was the process?"

    Fitted least squares parabola to data (960121.rlrdata), coefficient of x^2 negative.
    0 0
  50. Berényi Péter writes: I am prone to err as anyone, correction is welcome. However, in this topic I have not seen a valid one yet.

    You must have missed dhogaza's comment here, which did a nice job of correcting your misunderstanding about the meaning of the expression "an overestimation of the sea-level response". What was overestimated was not the rate of rise, but the time scale of the leveling off in rise once available land ice has nearly all melted. That's something that won't happen until nearly all the ice is gone, so the implication of this is clearly that Siddall's paper underestimated 21st century SLR.
    0 0

1  2  3  Next

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.



The Consensus Project Website

TEXTBOOK

THE ESCALATOR

(free to republish)

THE DEBUNKING HANDBOOK

BOOK NOW AVAILABLE

The Scientific Guide to
Global Warming Skepticism

Smartphone Apps

iPhone
Android
Nokia

© Copyright 2014 John Cook
Home | Links | Translations | About Us | Contact Us