The IPCC And 20th Century Mean Global Sea Level Rise
On the acceleration of sea level rise, the Gilligan effect, and the garbage solution.
From Chapter 3 of the AR5 SOD:
Two out of three reconstructions of GMSL (global mean sea level) from tide gauge data extending back to 1900 or earlier indicate non-zero acceleration. Estimates range from 0.000 to 0.013 [–0.002 to 0.019] mm yr–2, so it is likely that GMSL rise has accelerated since the early 1900s.
A fourth paper on this topic was recently published, that was not referenced in Chapter 3 of the SOD:
Twentieth century global mean sea level rise: is the whole greater than the sum of the parts?
J.M. Gregory et al.
Abstract. Confidence in projections of global-mean sea-level rise (GMSLR) depends on an ability to account for GMSLR during the 20th century. There are contributions from ocean thermal expansion, mass loss from glaciers and ice sheets, groundwater extraction and reservoir impoundment. We have made progress towards solving the “enigma” of 20th-century GMSLR—that is, the observed GMSLR has been found to exceed the sum of estimated contributions, especially for the earlier decades. We propose that: thermal expansion simulated by climate models may previously have been underestimated owing to their not including volcanic forcing in their control state; the rate of glacier mass loss was larger than previously estimated, and was not smaller in the first than in the second half of the century; the Greenland ice-sheet could have made a positive contribution throughout the century; groundwater depletion and reservoir impoundment, which are of opposite sign, may have been approximately equal in magnitude. We show that it is possible to reconstruct the timeseries of GMSLR from the quantified contributions, apart from a constant residual term which is small enough to be explained as a long-term contribution from the Antarctic ice-sheet. The reconstructions account for the approximate constancy of the rate of GMSLR during the 20th century, which shows small or no acceleration, despite the increasing anthropogenic forcing. Semi-empirical methods for projecting GMSLR depend on the existence of a relationship between global climate change and the rate of GMSLR, but the implication of our closure of the budget is that such a relationship is weak or absent during the 20th century.
The paper is an early online release for the Journal of Climate: abstract [here].
So . . . the addition of one more paper to the literature without an accelerating trend (2 out of 4) implies, according to the IPCC reasoning about uncertainty, that acceleration of sea level trend is ‘about as likely as not.’ Under this line of reasoning, the addition of one paper to the literature can change the assessment from ‘likely’ to ‘about as likely as not.’ And since the Gregory et al. paper seems to be superior from a methodological perspective, a more sophisticated assessment process would weight this paper more heavily. It looks to me like the IPCC needs to change this particular conclusion (especially the confidence level) in view of this new paper by Gregory et al.
My problems with the IPCC’s characterization of and reasoning about uncertainty are described in my paper Reasoning about climate uncertainty, and this particular case is a good example of the flimsiness of many of the confidence assessments, especially for data sets that have a low maturity level.
Here is something else of interest. While Chapter 3 is on the Oceans, Chapter 13 is on Sea Level Change. Chapter 3 did not mention the new Gregory et al paper, but Chapter 13 mentions the new Gregory et al. paper multiple times. In fact, Gregory is a lead author on Ch 13, and Gregory’s coauthor Church is coordinating lead author on Ch 13. The material in Ch 13 is on sea level acceleration is much better than that in Ch 3, but observed acceleration of sea level rise seems to belong to Ch 3. Hopefully, the authors of Ch 13 can motivate or provide a more sophisticated treatment of this issue in Ch 3, with a more defensible conclusion on this for the SPM.