#12 - On Page #7 it says:
“Several studies that use semi-empirical relationships between sea level and climate have predicted up to 1.4 meters (55 inches) of sea-level rise by AD 2100 when ice sheet contributions are included.”
NC SLR Assessment Report Critique - Page 11
The term “semi-empirical” can mean a lot of things, but usually it is code for “selected real world data was massaged by a computer program and plugged into a model that the author made up because it produced results that he had hoped to show.”
A careful reader of the NC Sea Level Assessment Report will come to the inescapable conclusion that Rahmstorf’s 2007 paper is the key pillar supporting the authors’ opinion that the IPCC’s sea-level projections are too conservative. Rahmstorf 2007 is referred to some nine times in this report, using such terms as “robust” and that “Rahmstorf’s ‘method’ for projecting future SLR has been adopted by several states and municipalities. The method has produced highly accurate hindcast results ...”.
Since this report relies so heavily on Rahmstorf’s 2007 paper (see
<<http://tinyurl.com/3bhuzd>> and <<http://tinyurl.com/456arxg>>) it is appropriate to investigate the credibility of his methodology and conclusions:
A) One expert wrote: “Stefan Rahmstorf is regarded as being outside the mainstream of current thinking, to put it charitably. Here is a rebuttal to his sea level rise estimates written by four prominent oceanographers: <<http://tinyurl.com/67qbv2y>.” Their conclusion is that Rahmstorf’s work is “simplistic". In an interview with London’s Sunday Times, one of the authors, Dr Simon Holgate, said: “Rahmstorf's real skill seems to be in publishing extreme papers just before big conferences like Copenhagen, when they are guaranteed attention.”
B) There is a very detailed critique of Rahmstorf’s 2007 paper at this website: <<http://tinyurl.com/4z3wxjo>>. This analysis concludes that:
1) Sea level rise rate vs. temperature is displayed in a way that erroneously implies that it is well fit to a line, as expressed in his equation.
2) The assumption that the time required to arrive at the new equilibrium is "on the order or millennia" is not borne out by the data.
3) Rahmstorf extrapolates out more than five times the measured temperature domain.
C) In a highly unusual move, Dr. Eduardo Zorita publicly called for Rahmstorf to be barred from the IPCC process <<http://tinyurl.com/4w382sn>>. Dr. Zorita is a leading Paleoclimatologist (headed the Department of Paleoclimate at the GKSS Research Center) and physicist who has written numerous scientific papers on climate related matters (<<http://tinyurl.com/48f5l87>>).
D) Dr. Roger Pielke, Sr. (<<http://tinyurl.com/6cbj26>>: a well respected meteorologist) wrote that there was “Blatant Cherry Picking By Stefan Rahmstorf And Colleagues In Science Magazine”. This article is very similar to the 2007 Rahmstorf report cited by the NC Sea Level Assessment, so the criticism is pertinent here. (See <<http://tinyurl.com/5stq9ap>>.)
NC SLR Assessment Report Critique - Page 12
E) “The studies (2007, etc.), led by Stefan Rahmstorf, ... have caused growing concern among other experts. They say his methods are flawed and that the real increase in sea levels by 2100 is likely to be far lower than he predicts. Jason Lowe, a leading Met Office climate researcher, said: ‘We think such a big rise by 2100 is actually incredibly unlikely. The mathematical approach used to calculate the rise is completely unsatisfactory.’” (See <<http://tinyurl.com/y9hh6lv>>.)
F) Since the 2007 Rahmstorf paper is rather technical, here is a technical expert analyzing it, and concluding that some of its key technical claims aren’t what they are asserted to be: “At the end of the day, the secret of Rahm-smoothing is that it’s a triangular filter with linear padding. All the high-falutin’ talk about ‘embedding dimension” and “nonlinear … lines’ is simply fluff. All the claims about doing something ‘new’ are untrue, as are Rahmstorf’s claims that he did not use ‘padding’. Rahmstorf’s shift from M=11 to M=15 is merely a shift from one triangular filter to a wider triangular filter – it is not unreasonable to speculate on the motive for the shift, given that there was a material change in the rhetorical appearance of the smoothed series.” (See <<http://tinyurl.com/34qa2zf>>.)
G) Rahmstorf subsequently publicly acknowledged a significant error in his 2007 paper. [Note that it had been signed off by his “peers,” and none were the wiser.] “In hindsight, the averaging period of 11 years that we used in the 2007 Science paper was too short to determine a robust climate trend. The 2-sigma error of an 11-year trend is about +/- 0.2 ºC, i.e. as large as the trend itself. Therefore, an 11-year trend is still strongly affected by interannual variability (i.e. weather)” (<<http://tinyurl.com/6z5nmpy>>).
This shows how the curves change when he makes new assumptions
<<http://tinyurl.com/nz26s9>>.
H) Dr. David Stockwell has a lengthy critique of Rahmstorf 2007 and concludes:
“It is apparent from these discussions that Prof. Rahmstorf had little
understanding of the methodology he employed, and that the view expressed in Rahmstorf et al. (2007) that: ‘The data available for the period since 1990 raise concerns that the climate system, in particular sea level, may be responding more quickly to climate change than our current generation of models indicates.’ is based in flawed and biased research.” <<http://tinyurl.com/4662alj>> (or here). And more from Stockwell is here <<http://tinyurl.com/485yuaa>>.
I) Another analyst weighs in “The non-linear trend in Rahmstorf et al. [2007] is updated with recent global temperature data. The evidence does not support the basis for their claim that the sensitivity of the climate system has been underestimated.”<<http://tinyurl.com/4cyjmvb>>.
NC SLR Assessment Report Critique - Page 13
J) World famous meteorologist, Dr. William Gray, (Emeritus Professor of Atmospheric Science, Head of the Tropical Meteorology Project, etc.) wrote me saying: “I have interacted with Stefan Rahmstorf a little bit over the last decade. I've been to a few meetings with him and he has visited and given talks at our ATS Department in Fort Collins. He is a 'far-out' global warming modeler... I do not judge Rahmstorf's model assessments of 2100 sea levels as being objective or reliable. He is very biased in his AGW views and has grossly exaggerated the warming threat to his own betterment, in my view. I've seen his 2007 Science paper. His Figure 4 graph indicating a 60-140 cm rise in sea levels by 2100 is not at all credible. North Carolina should not use this long period forecast. It is grossly exaggerated. I would anticipate a value more like 20-30 cm (8”-12”).”
K) Experts subjected Rahmstorf’s theories to testing, and they came up short <<http://www.nipccreport.org/articles/2010/dec/1dec2010a1.html>>.
L) I received this commentary from a closely involved scientist “The question of future sea-level rise is a complex one, and one where the uncertainties are very deep indeed. Let me try to summarize the problem. Within the mainstream IPCC interpretation sea level rise would be affected by different contributions: the expansion of the water column due to rising water temperatures, melting of land ice (glaciers and polar ice sheets), the gravitational effect of the disappearance of the latter, changes in ocean circulation that do not affect global mean but would do so at regional scales.”
“The IPCC climate models cannot represent all these processes, which means that there are processes that are not included at all in the climate models. One is the dynamics of polar ice sheets, the other is the gravitational effects of these ice sheets. The dynamics of polar ice sheets under rising temperatures is largely unknown. This is why the last IPCC report bolted an 'overhead' of roughly 20 cm to the contribution of the expansion of the water column simulated by climate models. This amount is however just a guess-estimate. Some researchers, like Rahmstorf have been trying to implement ad-hoc semi-empirical methods to estimate the contribution of the polar ice sheets to future sea-level rise. Basically he set up a statistical model linking the rate of sea level rise and global temperatures. The model would be calibrated with observations and then applied to the simulated global temperature rise from climate model simulations. This type of study, though much touted in certain circles, is not part of any consensus among scientist close to the IPCC, and I would even say that Rahmstorf represents a minority view here.”
For those in the know, Rahmstorf did come out with a 2009 paper that changed some things from his 2007 version. Why didn’t the 2010 NC report use the later 2009 paper? Probably because it was even more unsupportable. See this twelve (!) part critique: “Rahmstorf (2009): Off the Mark Again (Part 12) — A Mathematical Comedy” <<http://tinyurl.com/4vyuk3b>>... And here is criticism from a scientist with statistical modeling expertise who found Rahmstorf’s “method to be unreliable.” (See: <<http://tinyurl.com/4mfhgoz>>.)
NC SLR Assessment Report Critique - Page 14
In summary, for those who still advocate Rahmstorf’s approach, I put forward this observation sent to me: “I read that there is a possibility that a tremor could knock one of the cliffs off a Canary Island, which in turn could create a tsunami that would sweep across the Atlantic and inundate the entire east coast of the United States. The plausibility of this is unknown, just like the 1 meter projection of NC sea level rise by 2100. Accordingly I would argue that North Carolinians should built a gigantic sea wall to prevent being swamped by such an event. Hey, you can’t be too careful, right?”
Both Rahmstorf 2007 and the NC sea-level report reference the Church & White 2006 study (<<http://tinyurl.com/4uuoxdv>>) to support their high sea-level rise contentions. The obvious questions is: how reliable is that report?
A) A top sea level researcher wrote me in a personal correspondence: “C&W 2006 used a different technique to effectively scale the tide gauge record by the satellite altimeter data. See the CSIRO_GMSL_figure which shows the good agreement between tide gauge and satellite data. C&W is the ONLY study to achieve this agreement, and as they acknowledge in the paper, their findings are not consistent with all the other studies. This should raise red flags.”
B) Another sea level expert wrote me: “In reading and re-reading the C&W 2006 abstract and introduction, I come away with a sense that the authors approached the subject with the intent of discovering an acceleration in the rate of sea level rise. In the 1990s, global warming aficionados paid great attention to tide gauge-based sea level data as a proxy for measuring climate. However, mathematical calculations of that data could not produce a rate of rise as significant as the computer modeled scenarios featured in the early IPCC reports. The IPCC wanted/needed confirmation of the models from various approaches to measuring climate. For the tide-gauge data to produce results similar to the IPCC models a century out (and thus confirm the models), it would be necessary to find late-period acceleration in the most recent data. The C&W 2006 paper apparently intended to produce that acceleration and confirm the works cited in the IPCC's first three assessments. The C&W paper completely falls apart on the logical fallacy contained in their conclusion and is simply not "an important confirmation of climate simulations." I can say with confidence that the Church/White paper is corrupt and deserving of ridicule. Such patently bad science disappoints me. No public policy decision should be based on that paper.”
C) A recent paper (Wada, et. al. [2010]) estimates that up to .8 mm a year of sea level rise may be attributed to pumping ground water. Church et al. have not accounted for this. (See <<http://tinyurl.com/4azbchs>>.)
D) Church & White speculate on a mathematical model — which is not accurate. The point is that minor changes in curve-fitting methods can cause large changes in projected sea-level rise. (See <<http://tinyurl.com/4sr79mj>>.)
NC SLR Assessment Report Critique - Page 15
E) Church & White’s conclusions depend a lot on satellite data. This 2010 study shows that that there are potentially many very large “errors and biases” of such information. (See <<http://tinyurl.com/4nv75m8>>.)
F) Maybe the most convincing evidence that Church & White 2006 is seriously flawed, comes from Church and White. Well aware of the criticisms to its methodology, Church & White issued corrected data in 2009. The sea level rise projections using the 2009 data are about 50% of what results from using their 2006 data.
An expert in these matters kindly plugged in the C&W 2006 data into the Rahmstorf 2007 report. He then plugged in the C&W 2009 data into the Rahmstorf 2007 report. See the next page for these two graphs. It should be abundantly clear that the projected sea level rise is considerably less in using the 2009 data. A good question would be: why didn’t the 2010 NC Sea Level Assessment Report reference and use the later 2009 C&W data?
This same expert commented: “Another important point that is revealed when the 2009 Church and White data is used is that the baseline or equilibrium temperature ‘To’ drops from -0.5 degrees to -1.0 degrees. The baseline or equilibrium temperature is the temperature, presumably in the 19th century, when the sea level was unchanging. This implies that for the Rahmstorf's model to be correct and for Church's and White's sea level data to be correct, then the equilibrium temperature must be half a degree lower than Rahmstorf calculated. This is huge.”
This is a private correspondence that I received from a qualified sea level person:
“C&W 2009 extended the data set out five more years (end of 2001 to mid 2007), and also corrected data for the previous 100 years. If Rahmstorf used C&W 2009 instead of C&W 2006, the results would have been predictions of sea level rise half as great. The C&W 2009 ‘improved’ their earlier paper and came close to removing any apparent acceleration in rate of rise. But they did something Dr. Hansen is famous for: correcting the older historical data downward to make the modern observations appear more severe.”
This is a private email that I received from a different top sea level expert:
“The difference between C&W 2006 and C&W 2009 was the addition of extra data. The bizarre aspect is that it results in the lowering of tide gauge values before 1930, and the straightening of the sea level curve. This indicates a problem with their methodology, because I have been working on the key long term tide gauge records for that period, and they do not behave that way. A consequence of the adjustment for the C&W 2009 data, is that the Rahmstorf methods now predict much lower values (the same as IPCC) because the acceleration is gone.”
NC SLR Assessment Report Critique - Page 16
NC SLR Assessment Report Critique - Page 17