Le Stern Nouveau est arrive!

Though of course I havent read the whole thing or anywhere close. I wonder if anyone ever will. Maybe it will be fun reading for Christmas! Or maybe not…

Looking at Part I. First science nugget: “a doubling of pre-industrial levels of greenhouse gases is very likely to commit the Earth to a rise of between 2 – 5°C in global mean temperatures.” Hmmm – where does that come from? 1.5-4.5 is the conventional range, has Stern rounded it up? Or is that IPCC 2007? And “Several new studies suggest up to a 20% chance that warming could be greater than 5°C.” – hmmm, sounds like James stuff. Perhaps I should admit my biases and say that I find James’s analysis quite convincing… maybe its just his forceful personality.

Before I get back to that, I notice “If the Greenland or West Antarctic Ice Sheets began to melt irreversibly, the rate of sea level rise could more than double, committing the world to an eventual sea level rise of 5 – 12 m over several centuries.”. Errrm… centuries? Current SRL is 2-3 mm/yr, ie 20-30 cm/century. Double that to 40-60 and you’re a fair few centuries into the future before you hit 5m, let alone 12. SRL is the “great white hope” of impacts, since its unequivocally bad (at least I’ve never seen anyone assert it to be a good). 5m is SRL in a millenium might well cause problems, true, but I’m not really happy looking that far ahead – tech could do anything by then.

Back to the sensitivity… scroll down to page 9, box 1.2. Which includes “Some studies, e.g. Annan & Hargreaves (2006), have used statistical techniques to estimate climate sensitivity through combining several observational datasets (such as the 20th century warming, cooling following volcanic eruptions, warming after last glacial maximum).” Errm, but doesn’t point out that A+H gives a low prob to anything over 5 oC. I’m not sure where all the other lines on the graph come from, or whether they should be considered reliable or not. Clearly, only a few of them give any kind of probability to > 5 oC.

And if you get down to table 1.1 (page 12) the range of T is enormous. The problem here is the same one that has been around for ages: there is a large range of plausible temperatures and you can slate your language how you like to get emphasise different estimates. The trick is to find some sensible way of weighting the different estimates to say something useful. If Stern has managed to do that, I haven’t got that far yet – maybe its in part II. My initial impression, especially from box 1, is that Stern is putting the weight of his words onto the high-end scenarios, and I’m not sure this is justifiable from a neutral point of view.


  1. #1 llewelly

    a doubling of pre-industrial levels of greenhouse gases is very likely to commit the Earth to a rise of between 2 – 5 C in global mean temperatures.” Hmmm – where does that come from? 1.5-4.5 is the conventional range,

    1.5-4.5 C is the normal range for C02 sensitivity of the ocean-atmosphere system – not including ice sheet or biosphere effects. Do they think ice sheet and biosphere effects could add another 0.5 C ? I haven’t read the report at all. (Anyone have a link?) I admit 2-5 C does look like a rounding…

    SRL is the “great white hope” of impacts, since its unequivocally bad (at least I’ve never seen anyone assert it to be a good).

    Not SRL per se, but the claim that the melting of Greenland will free up fertile new farmland has showed up in denialist comments on RC several times.

  2. #2 llewelly

    Oh, I missed your link, sorry. … And the link to part 1 is broken – it has an excess asterisk at the end.

  3. #3 llewelly

    That is, your link to part 1 has an extra double quote (not an asterisk) at the end.

  4. #4 Eli Rabett

    I think you are missing the point. When does the disintigration of the Greenland ice sheet become inevitable. Once it goes, you get ~ 5 m.

  5. #5 llewelly

    Eli, the urls in your links got lost.

    As far as 5m SRL goes, it is one thing for it require millenia, and quite another for it to arrive in centuries.

  6. #6 Tim Worstall

    Their starting pooint is the A2 family of scenarios from the SRES.

    All of their subsequent calculations are based on that, the worst case one.

    [For some reason this comment got flagged for moderation, not sure why. Hmmm, it would have been nice if they had said A2 out front -W]

  7. #7 Carl Christensen

    It’s funny to see the usual skeptics screeching about the economist Stern, yet when their own economist Lomborg comes up with far crazier stuff, they rally around & circle the wagons etc. The sad reality is that climate science has quite a bit of people from other fields (economists, lawyers, politicians etc) making grandiose pronouncements, and indeed the scientists within the field come from a huge area, all trying to make themselves relevant (from old geezers like Gray & Lindzen through to a younger generation all eager to jump on the bandwagon, or take contrarian stances to get heard).

  8. #8 Eli Rabett

    There are a two issues mixed together in sea level rise.

    The first is, at what point does ice sheet collapse become irreversible.

    The second is once the first point is reached, how long does the collapse take.

    The TAR and (AR4??) are very conservative on both issues. There is increasing evidence that they are much too conservative. Here are the links

    Government science agencies are always conservative, in the good sense of the word.

    [Hmm, I'm not convinced by the "inc evidence" bit -W]

  9. #9 Eli Rabett

    Have you read the EOS articles on glacial flow in Greenland from satellite measurements 18(3) 8, 2006? I am much more cautious than you on this matter.

  10. #10 Adam

    Another point on SLR (SRL?) is how much rise do you need before damaging events become catastrophic (probably locally, obviously).

    E.g. (extreme case maybe) if a cyclone were to hit Bangladesh again, how much worse would it be if the sea level is 5cm, 10cm, 1m etc higher when it hits? I realise that there are other factors involved, but even across a range of scenarios it could be taken into account.

    More generally, how much do relatively minor and and non-catastrophic GW-induced changes affect (magnify or diminish) other problems or disasters that may occur despite it?

  11. #11 James Annan


    That’s a valid point, but OTOH one might reasonably ask whether a more effective strategy is to increase the resilience of the Bangladeshis to the risk of flooding (which already happens regularly with devastating results, and which will continue to happen irrespective of AGW) or to basically command the sea to stop rising, Canute-style, through GHG emission controls, even though a continuing rise is guaranteed for long after the surface temperature is eventually stabilised (if indeed it ever is).

    Of course the warming affects a wide range of environmental threats globally, whereas increasing the resilience of the Bangladeshis only deals with that one threat. Plus, it is not an either/or situation, and the best answer may include a bit of both. But still, it bears thinking about.

  12. #12 Eli Rabett

    Why James, you know the answer is to do both…

    For the simple reason that adaptation is more useful in the short run and mitigation more useful in the long run.

  13. #13 Adam

    I would certainly agree that both is the answer. Ignoring the general and focussing just on the specific means you have to do more for each specific problem tackled. All those extra specific fixes may make it seem better to have carried out a more general mitigation.

    On the flip side, how about other issues that may exacerbate non-catastrophic GW effects? For instance if a river starts to dry up, you get conflict amongst the countries through which it flows – conflicts that may already exist because of rising demand (this example provides a circular problem). It’s easier said than done to just the countries involved should avoid the conflict (we’ve been saying that for a long time already).