Comment 6-533
The phrase "repercussions of these abrupt climate changes" assumes that the abrupt changes (in the North Atlantic) happen and spread to the rest of the globe. This is the leading hypothesis, but by no means the only one and it is certainly not proven fact. This section should be reworded to avoid the tendency to confuse observation with hypothesis.
Response: accepted - we now say "effects", and it does not say anywhere they start in the Atlantic. (bolding mine)
Comment 6-543
This should read "There is evidence", not "there is solid evidence".
Response: accepted in part - changed to "good"
Comment 6-544
The characterization of the abrupt changes as "the South Atlantic warmed when the north warmed, and vice versa" is incorrect. Although this way of describing the data is popular, it is not very accurate. At the very least, the numerous papers pointing this out should be cited. Steig and Alley, 2002; Wunsch, 2003; Huybers, 2003; Schmittner et al., 2003; Roe and Steig, 2004. Furthermore, the purported relationship between N and S can only be demonstrated for the largest events, not for the events generally.
Response: Partly accepted. The sentence referred to discusses what the data show, and perhaps the characterisation was too simple. We thus have added "with possible lag". The references proposed do not provide any data inconsistent with this characterisation, so we hope the reviewer's concern is covered in this way. Concerning the final point: the cited Landais paper demonstrates this for a few more events (not just the largest), and the new EPICA core data demonstrate this for events in general (not ready to be cited yet).
Comment 6-565
The discussion of climate models simulating abrupt events neglects to mention that in fully coupled climate models, e.g. Manabe and Stouffer, the magnitude of meltwater forcing required is many times greater than the greatest amounts estimated to have actually occurred. Most readers will not recognize that the models e.g. of Rahmstorff or Knutti are very simplified. Throughout this section, the type of model being discussed should be clearly stated. Another more general problem is lack of attention to the fact that the leading hypothesis for abrupt climate changes -- flooding of the North Atlantic -- can readily explain only the abrupt cooling events, whereas it is the abrupt warming events that dominate the records. This is a major challenge for the scientific community and should be discussed openly and clearly. Additionally, the modeling work of Chiang, Battisti et al. on the link between the ITCZ and the D-O events should be cited in this section.
Response: rejected - the reviewer is not correct. Models with ocean GCM component (including coupled GCM) and with simplified ocean models do not show that simpler models require systematically less freshwater input; also we do not cite any shutdown experiments by Rahmstorf or Knutti we do not see the need to describe their specific models here. We do cite simulations that explain warm events. Concerning the ITCZ shift, we cite several papers already, space is limited and the reviewer does not suggest a specific additional paper to cite.
Comment 6-571
The "stochastic resonance" model of Alley et al. referred to here has been shown to be statistically unsupported by the data. Alley et al. used an inappropriate white noise background as their null hypothesis, where standard procedure would be red or colored noise. Roe and Steig (2004) showed that if the more reasonable noise background estimates are used, then the stochastic resonance hypothesis fails to meet statistical
confidence. Ditlevsen (J. Climate, 2005) repeated this result, and further showed that the statistical significant of the 1500-year cycle (upon which the stochastic resonance hypothesis depends) was weak. Subsequent work on the North GRIP ice core has further shown that the 1500-year cycle is likely an artifact in the GISP2 ice core (this paper is not yet in press, to my knowledge). These papers should be cited and a more balanced discussion given, if the stochastic resonance hypothesis (which has no basis in climate dynamics) is discussed, despite being discredited.
Response: rejected. we cite Alley et al as "showing evidence" which suggests stochastic resonance "could have triggered" the events - we do not say this is proven. What Roe and Steig as well as Ditlevsen et al suggest is merely that stochastic resonance is not statistically proven - Alley would be the first to concede that, and so do we. We therefore see no problem with our wording. By the way, the stochastic resonance hypothesis does have a basis in climate dynamics; the mechanism can be shown to work in dynamical climate models (see Ganopolski, Phys.Rev. Let. Phys. Rev. Let. 88(3),038501).
Comment 6-574
The idea that "climate models tend to underestimate the size and extent of past abrupt climate changes" attributed to Alley et al. (2003) is an opinion, not a scientifically demonstrated fact. If this statement is to remain in the document, it should be balanced by the point that "Other authors argue that the magnitude and extent of past abrupt changes, as evidenced in the proxy data, is smaller than generally stated (Wunsch, QR, 2006).
Response: Noted. We do in fact provide critical and balancing laguage, when we write: "However, such a general conclusion is probably too simple, ..."
Comment 6-611
The discussion of the recent results of Overpeck et al. (2006), implying that future warming and its influence on the Greenland ice sheet can be inferred from the paleoclimate modeling results for the last interglacial, may greatly overstate the relevance of these results. The Overpeck et al. results used a state-of-the-art but nonetheless highly idealized ice sheet model, that may not represent the processes correctly. The all important basal conditions of the ice sheet, and resolution of ice stream processes, are simply not realistically simulated yet. Additionally, while Arctic warmth in summer may have been as great at the LIG as in our near future, the radiative forcing during summer was (as stated in the chapter) about 10% greate.r. The effects of CO2 from anthropogenic activities do not come anywhere near this. Without detailed energy balance modeling,
which has not been done, it is not at all clear how relevant the LIG results are to the future. While these results should certaintly be discussed, these important caveats deserve more attention. It is also critical that the issue of timescale be discussed. The Overpeck et al. results do NOT tell us how quickly the ice sheet will melt. For policy makers, this is of course the critical issue. [Note that this is all handled much better in the Summary for Policy Makers, and I recommend taking some of the language from there and using it in this chapter.]
Response: Noted in revisions
Comment 6-685
This sentence should read as follows, to accurately reflect what is shown in the literature: "The 8.2 kyr event is INTERPRETED as a brief adjustment of the Atlantic meridional overturning circulation, though direct evidence for such changes in limited due to the small magnitude of the meltwater forcing, compared with e.g. the Heinrich events (Bianchi and McCave, 1999; Risebrobakken et al., 2003; McManus et al., 2004)." It is simply inaccurate to state that the 8.2 kyr event is "recorded" as a change in meridional overturning.
Response: Noted
Comment 6-1015
I agree with the definition of Dansgaard-Oeschger (DO) events as "Abrupt warming events followed by gradual cooling." However, since the evidence for ABRUPT warming is restricted to the North Atlantic region, it should not be stated that the DO events are recorded "elsewhere". The definition should simply read .. "recorded in Greenland ice cores".
Response: Taken into consideration, the correspondence with changes outside the Atlantic should, however, be addressed
Comment 6-1021
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.
Response: Accepted
