From The email that pointed me at it said “Sincere apologies — I am very sorry that I did not have time to write a shorter paper” and I think he is correct to apologise. It also says “Climate sensitivity section probably belongs in the paleo paper still under review” which also seems correct – to devote 15 pages (more than 1/4 of the paper!) to saying that Climate Sensitivity is ~ 3 oC is to confess that your prose is too prolix – and he doesn’t even quote JA (which brings up another topic: he spends far too much time quoting himself. The entire thing is far too inward looking).

I did some minor due diligence before writing this, to see who else had written about it. Predictably enough Curry wrote a post – but also predictably enough she didn’t actually bother read, let alone analyse, the paper; and the long tedious discussion thread there appears to be totally worthless.

DSC_5853-phoebe-a-table Speaking of which, there is some irritating twee nonsense in Hansen’s paper about Sophie and Connor’s estimes or aerosol forcing. Apparently they are Hansens grandchildren. Why their opinion matters is a total mystery. But I don’t put my cat in scientific papers and Hansen should leave his grandchildren out, if he is writing science. If he is writing propaganda, then fine, he can be photographed kissing babies all he likes.

The next issue is the degree of heat mixing into the deep ocean. Hansen would like his model (GISS modelE-R) to over-estimate it;, for reasons that will become clear when we discuss section 7, below. Why does he believe this? First, the ocean model mixes too rapidly into the deep Southern Ocean, as judged by comparison to observed transient tracers such as chlorofluorocarbons (CFCs) (Romanou and Marshall, private communication, paper in preparation). Second, the ocean thermocline at lower latitudes is driven too deep by excessive downward transport of heat, as judged by comparison with observed ocean temperature (Levitus and Boyer, 1994). Third, the model’s second-order finite differencing scheme and parameterizations for turbulent mixing are excessively diffusive, as judged by comparison with relevant observations and LES (large eddy simulation) models (Canuto et al., 2010). Notice what *isn’t* there – comparison with other models. Like I said, this is too navel-gazing. Does Hansen never talk to other modelling groups? (well, I’m being a bit unfair; perhaps his paper is just badly edited, because later on there is some (slightly non-systematic) comparison against some other modelling groups. But that is rather post-hoc). Of his 3 reasons: the first is unpublished. The second appears to be an ad-hoc comment he has just made here, rather than what you’d expect (a paper discussing that issue). The third is possible; I haven’t read Canuto so can’t judge. But 1 out of 3 isn’t great, and his solution: for the time being, we estimate alternative climate response functions based on intuition tempered by evidence of the degree to which the model tracer transports differ from observations in the Southern Ocean and the deepening of the thermocline at lower latitudes isn’t going to play well with anyone who doesn’t already believe him. I can’t see how this could get published.

Weirdly, page 22 claims that the Hadley Centre model (which one?) has a sensitivity of 10 oC to 2*CO2, which doesn’t sound at all plausible. Implicitly, he is talking about HadCM3, in which case he is simply wrong.

Then (section 7) Hansen goes on to say, if the models mix heat down too fast, yet roughly reproduce the observed T over the last century, this must be because they have overestimated the forcing; which he proposes is because they have underestimated the aerosol cooling. In which case – ta da – we’ll be in trouble later on, because the warming will be faster than we expect. And who knows? He could be correct. Certainly there is room for much uncertainty in the aerosol forcing.

Um, well, there you go. There was some other stuff too, but it was less exciting.


* The last tie I slagged of Hansen


  1. #1 M

    On heat mixing into the deep ocean:

    AR4 did state that “Comparing observed thermal expansion with those AR4 20th-century simulations that have natural forcings indicates that ocean heat uptake in the models may be 25% larger than observed, although both could be consistent within their uncertainties. This difference is possibly due to a combination of overestimated ocean heat uptake in the models, observational uncertainties and limited data coverage in the deep ocean (see Sections, 9.5.2, and”

    However, more recent research seems to suggest that this conclusion is dependent on which ocean heat estimate is used: Sokolov et al. 2010 ( finds that if Levitus 2005 is used, models are found to overestimate ocean heat uptake, but if Domingues 2008 is used, models underestimate. I don’t know where the current best guess on ocean heat uptake lies in comparison to those two papers…


  2. #2 Steve Bloom

    What I found most interesting was the concluding overt appeal for the necessary sat to be funded outside government. I wonder if that’s aimed at anyone in particular. Presumably SpaceX would be in a position to do the launch cheap. The sat itself?

    Almost as interesting was the glimpse or two of some inside-baseball stuff on modeling.

    Re talking to other groups, he seems to have done that, noting in acknowledgements “We thank Gokhan Danabasoglu, Tom Delworth, and Jonathan Gregory for providing data and papers defining responses of the NCAR, GFDL and Hadley climate models(.)”

    [Yes, as I noted. But the problem is the sequencing in the paper - at the point where he is giving his "3 reasons" he hasn't mentioned that. It may be bad editing. It might be a brain-dump problem. But it looks like he made up his mind first and talked to the other groups after -W]

    Re the partial reliance on his physical intuition, well OK, but I think you should give him a little credit on that. I realize it’s a bit jarring from the POV of how papers are traditionally done to see Hansen focusing in the last few years on what amount to synthesis papers with an overtly political component (which neccesarily call for intuition), but arguably the times call for it. If the quality of the co-auths he’s able to attract are any indication, lots of people seem to be endorsing the approach.

    Re JC, you seem to imply that she could have done some sort of expert analysis of the paper had she bothered. IMHO no, as she just plain lacks the chops.

    [I could have been clearer. I agree with you: it is classic Curry: put up the paper with no analysis (which she either can't do, or can't be bothered to do) in the hope that someone else might do the work for her. But, well, reading her comment thread tells you all you need to know about the chances of that happening -W]

    M, it looks to me as if Hansen may have been using the most up-to-date OHC material he could get his hands on. Again in the acknowledgements, he thanked “Gregory Johnson, John Lyman and Sarah Purkey for providing ocean heat data and preprints of their papers (and) Sarah Purkey for computing the sea level rise due to Southern Ocean warming excluding overlap with Argo data(.)” I can’t see what else that could have been referring to.

  3. #3 Thomas

    You do know that there is a cat listed as co writer on a scientific paper? I think the incident was described in “A random walk in science”. A physicist had written a paper using ‘we’ throughout, and a colleague told him the journal he intended to send it to wouldn’t accept that since he was the single author. Since this was before the days of word processors he thought it too cumbersome to rewrite the manuscript using ‘I’ so he added his cat as an author. Sent out preprints with his signature and a pawprint too :-)

  4. #4 Scribe

    Bill, be sure to occasionally edit another wikipedia called SourceWatch, which is free of the industry-funded nonsense that goes on at wikipedia. Look at the page on Anthony Watts, for example. Tells it like it is. It’s getting some quite high Google scores lately too.

  5. #5 Eli Rabett

    For Eli, the key to understanding Hansen is to accept that he has an understanding of the climate system that he cannot verbalize very well, and if it tells him something that extends beyond the models he will take his intuition.

    So far his gut has been a lot better than, for example, Richard Lindzen and RPSr. who do the same thing.

    [Is that damning with faint praise? More seriously, a paper based around intuition, errm, well, you see the problems -W]

  6. #6 Gareth

    With luck, a following wind, and the provision of facilities at the quake-damaged University of Canterbury in ChCh (they’re working on it), I’ll be interviewing Hansen on Friday morning for The Climate Show and Hot Topic. If anyone’s got succinct questions they want to ask about the thrust of this paper, or the previous one on ice sheets/ocean temps/Eemian etc, post them in the comments here, and I’ll see if I can work them into the chat. Nothing too technical, though – got to think about the audience (and the interviewer). ;-)

    [Fun for you! What I'd ask (and which, happily, isn't too technical) is something like "how do you attempt to balance scientific rigour and intuition? From reading this paper, it is clear (well it says so explicitly) that intuition has a place. But how can other people be expected to be convinced of results that depend on intuition? -W]

  7. #7 Steve Bloom

    OK, I went back and re-read the paper, and I think you’re leaning on Hansen’s use of the word “intuition” a bit too much, William. Would you have batted an eye if he’d said the values selected were the “most plausible” within the stated constraints (and there were some, note, so it was by no means just a matter of “intuition”)? In any case he didn’t claim the results as definitive, rather just that they were grounds for prioritizing getting the needed obs (which he first pointed out in 1992, so we can give him a little credit for impatience here).

    [I'm not sure quite how much it matters exactly which word he uses. What matters is that he doesn't want to use the values his model provides, for reasons that are weak. And I think you have the consequences wrong: he uses this to deduce that aerosol cooling is larger than we think, which then has the discussed consequences -W]