Hansen again. He seems to have got bored with saying climate sensitivity is 6 oC, and now agrees with JA that its 3 oC. Hurrah. Less excitingly, his estimate for all-ice-sheet melt on 2-3 x CO2 is now "it would take some time", which has the virtue of vagueness.
But far more excitingly, he is now pushing the idea that Earth could be heading towards a Venus type runaway greenhouse effect, apparently with added forcing as small as 10-20 W/m2. (over present day or over pre-industrial? What CO2 level would that be? Not sure, but In my opinion, if we burn all the coal, there is a good chance that we will initiate the runaway greenhouse effect. If we also burn the tar sands and tar shale (a.k.a. oil shale), I think it is a dead certainty. Perhaps we are expected to guess what Hansens estimate of total coal reserves is). Sadly he seems to have omitted the reference to the paper where the fine details have been published, but I'm sure that just a matter of time. I'm sure I can leave the likes of JA and RMG to rip the fine details to shreds.
As an encore, he has yet another go at defining tipping points. There are where Climate forcing (greenhouse gas amount) reaches a point such that no additional forcing is required for large climate change and impacts. Try to play that around D-O events (removing the unreasable restriction that forcing has to be from GHG amount).
There are some tantalising headlines about intergenerational equity, but no substance.
- Log in to post comments
Perhaps we are expected to guess what Hansens estimate of total coal reserves is). Sadly he seems to have omitted the reference to the paper where the fine details have been published, but I'm sure that just a matter of time.
Likely this one...
[Reading off fig 4(a), that would be about 600 ppmv. Thats not +10 W/m2, so its pretty hard to understand what he is on about. What do you think he means? -W]
Have I missed something... haven't Hansens point always been: its 3 oC, his estimate for all-ice-sheet melt on 2-3 x CO2 `?
[http://scienceblogs.com/stoat/2008/03/hes_at_it_again.php etc etc -W]
3 C is the Charney sensitivity, which excludes slow feedbacks, such as changes in vegetation
http://www.realclimate.org/index.php/archives/2008/04/target-co2
"Paleoclimate data show that climate sensitivity is ~3°C for doubled CO2, including only fast
feedback processes. Equilibrium sensitivity, including slower surface albedo feedbacks, is
~6°C for doubled CO2 for the range of climate states between glacial conditions and ice-free
Antarctica."
Hmm I'm confused... what is the new thing here?
[Where is your quote from? Thats what I thought he *used* to be saying, and doesn't appear to say in his latest -W]
The slides you link to say 3 C for fast-feedback sensitivity and 6 C if you include slow ice albedo feedback. Which is the same as what he's been saying, as far as I can tell. You make it sound like he has changed his mind on the sensitivity. He may have changed his mind on how fast the slow feedbacks are. But as far as I can tell, he's sticking with the same magnitude for climate sensitivity.
[I can find "Between the depths of the last ice age and deglaciation of Antarctica climate sensitivity to a specified greenhouse gas change is doubled to about 6 C for doubled CO2 because of the surface albedo feedback." Thats rather different to what he was saying before. In that form, its explicitly irrelevant to the current debate. Before, he was trying to claim the 6 oC was of some relevance for the future. Or were you reading some other text? -W]
Regarding the "runaway greenhouse" effect, 10 to 20 W/m^2 above the current forcing corresponds to 2^(10/3.7) to 2^(20/3.7) times current CO2 levels, by my calculation, assuming 3.7 W/m^2 for 2xCO2. If current CO2 is 387 ppm, then that corresponds to an additional 2500-16400 ppm necessary to trigger a runaway greenhouse (according to Hansen). IIRC, 2500 ppm is about what you get if you burn all known fossil fuel reserves (~ 5000 GtC, including sands and shales). IIRC, it's something like 1700 GtC with "standard" reserves (coal/oil/gas), which is, I don't know, maybe another 700-800 ppm worth?
Thus, I don't see how "burning all the coal" can approach 10 W/m^2 of forcing. Maybe Hansen meant 10 W/m^2 above pre-industrial? Then that would be 2^(10/3.7) * 280 = 1800 ppm more CO2. (I'm ignoring non-CO2 forcings here, and carbon cycle feedbacks.) Still more than what coal could provide, unless my GtC->ppm conversion is off.
I hadn't read through the whole presentation when I first commented. Looking through it, Hansen seems to be saying that burning all of the coal could, and burning the unconventional will trigger a clathrate release, given the proximity of the two statements in the presentation:
What is different about the human-made forcing is the rapidity at which we are increasing it, on the time scale of a century or a few centuries. It does not provide enough time for negative feedbacks, such as changes in the weathering rate, to be a major factor.
There is also a danger that humans could cause the release of methane hydrates, perhaps more rapidly than in some of the cases in the geologic record.
In my opinion, if we burn all the coal, there is a good chance that we will initiate the runaway greenhouse effect. If we also burn the tar sands and tar shale (a.k.a. oil shale), I think it is a dead certainty.
This is decidedly muddled, to be sure.
"Between the depths of the last ice age and deglaciation of Antarctica climate sensitivity to a specified greenhouse gas change is doubled to about 6 C for doubled CO2 because of the surface albedo feedback."
In the new one... the above was in the old... so the devil is in the details of not specifying time ranges?
[I would say so. If H is no longer pushing 6 oC as relevant to the future, thats a strong difference, IMO -W]
Footnote 3, page 8:
Anyone understand what Hansen means by this? I commented several times on realclimate a while back pointing out that you can't use glacial-interglacial temperature/CO2 changes to estimate climate sensitivity today without further assumptions about how climate sensitivity depends on climate:
[self quote]
Is Hansen arguing that the climate sensitivity to 2XCO2 - yes, an oxymoron, but it pays to be clear - is the same today as in the LGM? If so, why?
[Err no, this is him (very sensibly) arguing that it *isn't* the same as now, which is more of the abaondoning of the 6 oC stuff -W]
He can't be arguing that *all* contributions to climate sensitivity are different today than in the LGM, otherwise his calculation of 0.75C/W/m2 is invalid, since he obtains it by dividing the temperature difference between the LGM and today by the forcing difference from all sources.
Obviously ice albedo is less sensitive today than in the LGM (which is the crux of your 6C remark), but I can't see where Hansen argues, for example, that the forcing multiplier due to water-vapour feedback is the same today as in the LGM. If it isn't, then the 3C figure suffers the same objection as the 6C figure.
[I can think of at least one reason why the water-vapour multiplier would be different today: the tropics are already saturated, whereas they were probably not at the LGM]
Testing the comment (the Hansen quote William wanted a source for is easily found, just paste a chunk of it into Google or Scholar, but I can't get it to post)
Hansen 1988 was 4.8 C
Mugwamp, to me Hansen is arguing that the system has significant hysteresis, that the path it follows to a particular state, under a forcing scenario is not the path it follows back from that state if the forcings are reversed. Logically this is not true for small excursions else there would be no climate, but logically also, it can (and appears to be true) for larger ones.
Eli explains it all:
What Hansen means is fairly simple.
First of all a runaway greenhouse effect means the oceans vanish. That is a pretty standard definition from what we understand happened on Venus. Once that happens you lose a lot of hydrogen to space and can never reform the water, the oxygen goes to CO2
Second, the 3C sensitivity comes from ice cores and such extending back ~ 800,000 years. He considers this as a result of what he calls the fast feedbacks, mostly greenhouse gases and aerosols. What is interesting here is that he puts much narrower limits on the Charney sensitivity 3 +/- 0.5 C. There is an interesting discussion about aerosol forcing uncertainties.
Third, he gets the slow sensitivity from looking at the difference between the ice free early Cenozoic (35-60 million years ago) and the icy and icing later times. The additional forcing is ascribed to albedo differences when there is no ice. That is the 6 C.
Fourth, he says that Antarctic ice is but a memory if CO2 > 450 ppm.
Fifth, and this is the key point, he says that modeling show climate sensitivity is a function of forcing. J. Geophys. Res.110, D18104 (2005) In particular if forcing goes below -10 W/m2, the sensitivity goes to the roof as most of the water vapor freezes out, which is what yields snowball earth scenarios. The Earth eventually digs out of this by decreasing weathering raising ghg concentrations and forcing. (see page 23)
OTOH (playing economist), if forcings increase (for example, burning all the coal) then at about + 10 W/m2 the climate sensitivity ALSO gets much bigger, in that case because the oceans evaporate. Hansen uses the words boil away, but WTH.
Eli sees a couple of problems. First of all, at the high end the model (GISS E) blows up, so it may not be much use at the hot extreme. Second, there is a real question as to whether the albedo change is also being counted in the oscillatory ice-age determination of the Charney sensitivity, so, in effect there may be some double counting going on.
Sixth RTFR
Eli, the problem with Hansen's tight 3+/-0.5 C is it is based on an unjustified assumption, viz:
dT / dF is independent of climate state.
Here T is equilibrium temperature, and F is forcing.
For Hansen's argument to work, he requires the equilibrium change in temperature as a result of an incremental change in forcing to be the same today as it was at the LGM (and for all times in-between). In other words, if I throw in an extra 1W/m2 of forcing today, I get the same temperature rise as an additional 1W/m2 produced at the LGM (ignoring "slow feedbacks" eg ice sheet changes).
The truth of this assumption is not self-evident to me. One can imagine all kinds of ways in which the climate state might affect dT / dF, for example differences in the amount of water-vapour already resident in the atmosphere.
mugwamp, clearly what Hansen is saying is that dT/dF IS NOT independent of the climate state, but that within broad limits (-10 W/m2 < F < +10 W/m2) it is only weakly affected. See p23 of the presentation. It is only outside of those limits that dT/dF goes through the room.
The exact slope of dT/dF is going to vary a bit depending on other things, but broadly put not very much. Remember 1MY ago is not so long on geological scales that you need to get the result you are arguing to.
Eli, Hansen is saying dT/dF is dependent on aspects of the climate state that influence what he calls the "slow feedbacks" (eg ice albedo). But for his 3C calculation to be valid he needs dT/dF to be largely independent of aspects of the climate state (eg water vapour) that influence the "fast feedbacks".
You say: "The exact slope of dT/dF is going to vary a bit depending on other things, but broadly put not very much."
Can you justify your claim?
Intuitively, it seems to me that some of those "other things" might have a substantial influence on dT / dF. For example, today all outgoing infrared radiation at the surface in the tropics is absorbed by the water vapour in the air, so at least to first order, adding more water vapour to the tropics is not going to have much of an impact on the heating there. However, at the LGM, the tropics were likely drier than today, hence increasing the water vapour may have had a much larger impact.
Go look at the talk, p 23
Eli, page 23 shows modeling results. Hansen's justification for having "nailed" sensitivity at 3 +/- 0.5 is that it is based on paleo studies, not modeling. And his argument that the Earth's climate becomes more sensitive as it gets colder is about the slow feedbacks (ice albedo), not fast feedbacks.
So we're back to square one: where is Hansen's (or anyone's) justification that dT/dF is independent of climate state (equilibrium T, fast-feedback F)?
Could anyone please explain where I'm going wrong:
a) Manmade sulphate aerosols don't make it to the stratosphere. They stay in the troposphere and are thus short-lived (1-2 weeks).
b) As a consequence, their effect on temperature is rather local, over the emission points and downwind from them. Since we know where sulphates are being emitted (most notably in China and India) we should be able to detect their cooling effect over these areas, relative to other parts of the world. This doesn't seem to be happening.
c) Hansen's proposition that a large part of the GHG-warming is being masked by anthropogenic aerosols contradicts all the above.
Thank you and happy new year.
[You may want someone more expert than me on this but... your 1-2 week timespan may well be too short, and your estimate of their dispersion in that timespan too low as well. Historically, their main sources are europe and the US, and I'm not totally convinced that they aren't still (do you have any sources?). And what is the source for your assertion that there is no correlation between aerosol and temperature? As for (c), its not Hansens proposition, its everyones -W]
You forgot to mention that "everyone" also gets to freely choose different aerosol values to get their models to fit the data (*). Who wouldn't believe in a universal fudge-factor? An unfalsifiable theory is a publishing bonanza!
(*) Jeffrey T. Kiehl, 2007. Twentieth century climate model response and climate sensitivity. GEOPHYSICAL RESEARCH LETTERS, VOL. 34
[Like it says, there is a range of uncertainty for this forcing. That doesn't mean people can freely choose their own values, nor that people have -W]
The modeling studies agree with the paleo 3 +/- 0.5 result over a range of forcings +10 to -10 W/m2. Note also that on the low end (snowball earth) p there is paleo evidence for the 6 oC estimate, it is just that the Earth has not been there in a long time. The paleo evidence is a validation of the modeling which Hansen uses to extrapolate the modeling studies to above +10 W/m2 and below -10 W/m2
Thank you William. I don't remember what my source for a) is. But I do remember that it didn't sound controversial at the time. I may be wrong though. FWIW, where I live smog doesn't go up much. At an elevation of 2500-3000m the air is crystal clear and you can see the ugly, brown cloud well below you. It vanishes as soon as it rains.
As for b) and c) I think that, paradoxically, one of the best sources could be Hansen's own GISTEMP (it doesn't matter much whether Asia has overtaken USA+Europe in sulphates or not). If, indeed, we are talking about a significant masking effect of aerosols I would expect to see some coolness over the industrialized areas on this map (on the assumption that a is not totally wrong). It doesn't show up. The coolest areas are generally far away from the emitting points.
[What you are seeing there is the coolest areas being mostly over the oceans. To disentangle this properly you'd need to do much more work -W]
Oh yes they have. Kiehl shows there is an inverse relation between model climate sensitivity and total anthropogenic forcing (including aerosols). In Kiehl's own words:
In other words, if your model is too sensitive, up the aerosol content. If it is too insensitive, reduce it.
Even Hansen finally admits that you cannot infer climate sensitivity from 20th century modeling. Without a justification for why dT/dF is independent of climate state (for fast-feedback processes), his paleo numbers are equally bunk. Mann's hockeystick has been completely discredited. And we've seen no warming for a decade now.
Those who would exploit this issue to impose their own vision of how we should live are rapidly running out of ammunition. Better get those carbon taxes in place before the general public wake up!