Climatic Research Unit Documents: IPCC Working Group Fourth Assessment Expert Review


Intergovernmental Panel on Climate Change

  • IPCC Working Group I Fourth Assessment Report
  • Expert and Government Review Comments on the Second-Order Draft
  • Confidential
  • June 15, 2006


6-42, p. 12

I have four chief concerns with this chapter. First, there are numerous important references left out, and an over-emphasis on papers by the authors themselves, which do not accurately reflect the communities’ view. In general, the certainty with which this chapter presents our understanding of abrupt climate change is overstated. There is confusion between hypothesis and evidence throughout the chapter, and a great deal of confusion on the difference between an abrupt “climate change” and possible, hypothetical cuases of such climate changes (e.g. Heinrich events). Second, the use of the terms “very likely”, “likely”, etc. are not in conformance with the rest of the IPCC document — some things that are virtually certain are listed as “likely” and mere hypotheses, largely untested, are listed as “very likely”. This carelessness does not add credibility to this chapter. Third, extensive reference is made to a very few recent papers that have not yet been thoroughly considered by the scientific community, and whose relevance to future climate is, in my judgement, greatly overstated. Finally, the choice of words to define — or not define — in the Glossary is strange. A definition (and a very poor one) of Heinrich events is given, but there is no definition for “Holocene”, even though that term is used throughout the text. I would additionally note that overall, the chapter does a fine job at dealing with the “Hockey Stick” controversy, but a very poor job dealing with abrupt climate change and its possible relevance to the future. There are numerous glaring omissions of citations — notably no mention is made of the work by Wunsch, Seager and Battisti, challenging the standard “Broecker-type” hypothesis for abrupt climate change. [Eric Steig (Reviewer‘s comment ID #: 252-1)]

6-56, p. 16

This whoile section consists of a succession of extravagant extreme claims of almost complete knowledge of paleoclimate processes which is not supported by credible literature. [VINCENT GRAY (Reviewer‘s comment ID #: 88-731)]

6-81, p. 19

Replace “very likely” by “generally expected” [VINCENT GRAY (Reviewer‘s comment ID #: 88-674)]

6-85, p. 19

I have looked through the draft chapter 6 and find it an impressive document. However, bullet 4 on page 6.2, starting “global mean cooling and warming…..” strikes me as incorrect and misleading. Whereas the mean rate of temperature change over the Pleistocene may have been 10 times slower than that projected for the next century, there is clear evidence that for specific major climatic transitions, global (or at least hemispheric) temperature changes in the past have been at least as rapid as those projected by climate model simulations and incorporated in the last IPCC report. The most obvious case in point is the global warming at the start of the Holocene, ca. 11.5 ka BP. Russell Coope, more than 20 years ago, showed from beetles that UK temperatures rose faster than could be dated within the errors of 14C dating. Subsequently this was confirmed by Greenland ice cores based on layer counting (full glacial to interglacial in less than 100 years), and by the Cariacos basin marine record. I have worked on varved lake records from both the tropics (Roberts et al Nature 1993 366, 146-148) and the Mediterranean (Roberts et al The Holocene, 2001, 11, 719-734) where this climate transition was accomplished in substantially less than a century. In short, several independent lines of evidence show that the climate system has been capable of flipping from one meta-stable state to another, very different one over timescales that could be experienced by a single human lifetime. This is not an unimportant conclusion in terms of the potential for non-linear responses of future climate to GHG forcing. I also looked for supporting argument for bullet 4 later in chapter 6, but found nothing of substance. In short, this particular bullet seems in need of critical reassessment before the definitive version of the next IPCC report emerges, or simpler still – just cut it. [C Neil Roberts (Reviewer‘s comment ID #: 216-1)]

6-113, p. 22

Attribution of abrupt climate change only to changes in Atlantic Ocean circulation ignores other explanations including possible nonlinear responses of tropical Pacific variability to radiative forcing directly overhead (Clement et al. 1997; Cane and Clement 1999; Mann et al. 2005). These two leading theories may be partly reconciled by emerging evidence that big changes in the Atlantic can modulate ENSO frequencies, (see recent paper by Dong et al. 2006. Geophysical Research Letters), possibly at multiple time scales. Note that allusion is already made to the dynamic ocean thermostat theory on another of the major findings (page 6-3, lines 39-41). [Govt. of United States of America (Reviewer‘s comment ID #: 2023-363)]

6-203, p. 33

likely as not? UGH! YOU ARE GOING TO GET KILLED OVER THIS TERM! How about stating it something like – IT CANNOT BE DEMONSTRATED WITH ANY DEGREE OF CONFIDENCE THAT PAST WARM PERIODS WERE COMPARABLE TO OR GREATER THAN THE LATE TWENTIETH CENTURY. I don’t iinsist on this phrasing, but save yourself a lot of grief and choose something – else those 11 letters are going to get you into trouble! [Thomas Crowley (Reviewer‘s comment ID #: 51-20)]

6-480, p. 60

I have a standard gripe that the claims of uncertainties in the Crowley 1995 estimate reflect the fact that people have not read the details of the paper – because in fact an uncertainty analysis was conducted, which is why there is a large spread of values. and I find it notable that the 600-1000 Mt values cited in the ipcc report fig very snugly in the range I state. I furthermore point out in my paper that hte C13 data arenot bulletproof, for glacial stage 6 has very different C13 changes despite the fact that boundary conditiosn were virtually identical to stage 2. [Thomas Crowley (Reviewer‘s comment ID #: 51-28)]

6-596, p. 76

This sentence is misleading: because I used published relative sea level data from coral terraces in the calibration step of my method, my curve cannot be interpreted as ice equivalent sea levels (see Waelbroeck et al. (2002), section 4, § 2) and has no such value. Only complete earth models like those developed by Peltier or by Lambeck and co-authors can yield estimates of the global impact of ice sheets build up and melting. There is an ongoing debate on the total ice-volume equivalent sea level depression that prevailed at the LGM: Peltier’s ICE-5G model yields an estimate of 118.5 m at 21 cal. Ky BP, whereas Lambeck’s model yields an estimate of about 140 m at 21 cal. Ky BP (Lambeck and Chappell, 2001; Lambeck et al., 2002a; Lambeck, 2004). Additional reference: Lambeck, K., 2004. Sea-level change through the last glacial cycle: geophysical, glaciological and palaeogeographic consequences. C. R. Geoscience 336, 677-689. [Claire Waelbroeck (Reviewer‘s comment ID #: 279-8)]

6-696, pp. 92-3

Fig 6.10. I here repeat a point made in my comments on the FOD. It is statistically invalid and visually misleading to overlay the black instrumental line on this diagram. The coloured graph lines show proxy records that end at 1980. If you want a line that continues up to more recent years that then you must use the proxy records that continue past 1980, not switch to a different type of series. There are up to date proxy records available, but as I’m sure the authors of this chapter are aware, they depart from the surface instrumental record, many of them declining after 1980. By failing to show this, and including the surface temperature data in black, it constitutes a misrepresentation, since the black line is an invalid forward extrapolation of the proxy data. If the reason for not showing the updated proxies is that they are not considered to be good representatives of temperature anymore, then by what right does the Figure insinuate that they were good proxies 8-10 centuries ago? It is no defence to claim that MBH99 established a statistically skillful relationship between the proxy network and the instrumental data, since that claim has been refuted, as discussed above. McIntyre and McKitrick (2005a,d) showed that the pre-1450 RE statistic was incorrectly benchmarked, yielding a spurious inference, and the r2 stat calculated by MB&H themselves, which showed the lack of skill, was simply not reported. The failure of the r2 and CE stats is confirmed by Wahl and Ammann. The squared correlation between the MBH long proxies and the instrumental record is nearly zero (MM05a,c). The mean correlation between the long NOAMER proxies and gridcell temperatures in the MBH98 data set (which dominate the pre-AD1450 portion) is -0.08 (McIntyre and McKitrick 2005c), and the RE significance benchmark is above the MBH98 RE score, using all available implementation of the Mann code (McIntyre and McKitrick 2005d). The surface instrumental record cannot be used as a statistically valid extrapolation for the proxies after 1980. [Ross McKitrick (Reviewer‘s comment ID #: 174-35)]

6-1167, p. 117

You allude to the fact that these reconstructions are “not entirely independent inasmuch as there are some predicotrs that are common”. This is a very misleading description. For the medieval period, there is massive overlap in all the cited studies. The six series of Briffa (2000) together with bristlecones/foxtails are used in only slightly varying combinations in all of the cited studies. If there are problems with only a few canonical series (as arguably has already been demonstrated with the birstlecones/foxtails) then the entire corpus of studies may fall. Problems can be observed elsewhere e.g. the Yamal series and the Polar Urals Update have very different properties with the Yamal series being a big contributor to HS-ness while the Polar URals series has a strong MWP. The Polar Urals Update correlates better to gridcell temperature than the Yamal series and one cannot help but suspect that the decision to use the Yamal series in all studies except Esper has been done with one eye on the MWP-modern relationship. [Stephen McIntyre (Reviewer‘s comment ID #: 309-62)]

6-1021, p. 166

This is a very poor definition of Heinrich event, confusing fact with hypothesis. Stating that Heinrich events are “indivative of cold periods” is both unimportant and potentially misleading. Furthermore, there is room for debate about how many Heinrich events there are. By some measures, there are only four, by others there are 8 or more. The definition should simply read as follows: Heinrich event: An interval of rapid flow of icebergs from the margins of ice sheets into the North deposition of sediment eroded from the land. [Eric Steig (Reviewer‘s comment ID #: 252-26)]

Share this: