There have been lots of comments on this one already, so it's in pretty good shape, but I wanted to get it up for official review before its publication on Monday.

Coincidentally, John emailed Monckton about a week ago asking several questions, and he never responded.  So this time he can't chide us for not asking for his sources.  All we can do is proceed as best as we can based on his blog post, which is only fair, because that's what the readers are forced to do as well.

Monckton Misrepresents Scientists' Own Work (Part 1)

Dana, I'm not sure I agree with this

"Douglass et al. (2007): deals with tropical troposhere temperatures and does not even discuss climate sensitivity.  In fact, the adiabatic lapse rate, which causes tropical troposphere warming, is a negative feedback.  Thus if it's missing, it actually undermines the low climate sensitivity argument.  Monckton also mischaracterizes the 'hot spot' as a fingerprint of human warming - it is not."

I know I've talked about this before, but we can keep learning, right? ;)

The 'hot spot' is a representation of changes in the lapse rate, which is a negative feedback. But by being there it also increases the quantity of water vapour (in order to be a 'hot spot' there must be condensation, so there must be plenty of vapour there). Soden & Held, 2006 found that the water vapour + lapse rate feedback was completely independent of the size of any hotspot.

Well, they didn't mention the hotspot iirc, but they showed the relationship between LR + WV feedback. Afaik the existence or nonexistence of the hotspot shouldn't affect climate sensitivity either way.

"Wentz et al. (2007): dealt with precipitation, not temperature changes or climate sensitivity.  The paper finds that while water vapor has increased as expected, precipitation has increased more than predicted by climate models.  This in no way supports low climate sensitivity, certainly not threefold lower than model estimates, as Monckton claims in his misrepresentation of this research."

Also I think this needs checking.

More condensation/evaporation could reduce climate sensitivity.

As I understand it: in terms of the hotspot it's not relevant, because there's a tight radiative/convective balance in the tropics and very high humidity. This is what Soden & Held found.

However, if there were more precipitation due to things like the storm tracks, then you might get the condensation in heavy storms (i.e. cooling of the surface/warming of the atmosphere, which a negative feedback) which wouldn't affect the long term, large scale water vapour content as much.

I haven't had time to read the paper, but I think this is also worth checking.

MarkR, the main problem with Douglass et al. (2007) is that the test they use is obviously absurd, I have no idea how it got through peer-review.  A model with perfect physics and infinite temproral and spatial resolution and an infinite ensmeble, to properly characterise tthe effect of internal variability, would be guaranteed to fail the Douglass test.  Who would use a test that is guaranteed to fail the best possible model?

It would also be worth pointing out that many of these papers have revieved peer reviewed rebuttals and give links to them (as in the Scafetta post).  If monckton hasn't mentioned them, then that is again a case of selective reading.

Dikran - it wasn't so much a comment on Douglas et al, as on the science talked about afterwards. Models seem to say the total feedback is independent of the 'size' of any hot spot. And it's reasonably easy to see why from basic physics, so it should be a robust result.

i.e. no hot spot isn't necessarily evidence for low sensitivity, because it would be cancelled out by a smaller increase in atmospheric water vapour.

Mark R, you may be over interpreting Soden and Held 2006.  In particular, they write:

"It is well known that climate models tend to maintain a fixed tropospheric relative humidity as they warm.  The modest changes in relative humidity that the models do generate are worthy of study, but they are too small to substantially modify the increase in column integrated vapor resulting from the increase in saturation vapor pressure. The data in Fig. 1 do not raise any concerns in this regard, over the tropical oceans at least. It is perhaps worth emphasizing  that  column integrated vapor is dominated by the lower troposphere, whereas infrared water vapor feedback is dominated by the upper tropical troposphere (see Held and Soden 2000). Our focus here is not on water vapor feedback nor on climate sensitivity but on the hydrological response given a lower-tropospheric temperature change."

Not only does this disavow any discussion of effects on climate sensitivity, but because of the different weighting of water vapour in the upper and lower troposphere on the green house effect, it is probable that the lapse rate feedback and the water vapour feedback will not co-vary if changes in relative humidity differ vertically in the troposphere.  For example, an increase in water vapour in the lower troposphere coupled with no increase in the middle and upper troposphere would result in a tropospheric hot spot cooler than that predicted by the models, but with virtually no change in the radiative balance due to increased water vapour.  That scenario would definitely suggest a lower climate sensitivity than that suggested by models.  Scenarios with reduced lapse rate feedbacks but higher than expected water vapour feedbacks can also be devised.

Unfotunately I do not know which balance the actual humidity data support.  However, Dana's claim that, "Thus if it's missing, it actually undermines the low climate sensitivity argument" is accurate provided that it is understood that any such change will also have effects on the water vapour feedback which tend to counteract the reduction in the lapse rate feedback.  Perhaps that sentence couls be revised to reflect that fact, but I cannot think of suitable wording at the moment.  Nor is it clear that the quailification is necessary in so brief a mention of the issue.

I'll just say Douglass doesn't say anything specific about climate sensitivity.  That should be sufficient.  Wentz says this:

"The difference between a subdued increase in rainfall and a C-C increase has enormous impact, with respect to the consequences of global warming. Can the total water in the atmosphere increase by 15% with CO2 doubling but precipitation only increase by 4% (1)?Will warming really bring a decrease in global winds? The observations reported here suggest otherwise, but clearly these questions are far from being settled."

Basically it's just not clear about the implications regarding climate sensitivity.

Dikran - good idea, I'll add the peer-reviewed response papers.  If I miss any let me know.

MarkR fair enough, it just surprises me that anyone can think they can retain any credibility after referencing a paper as bad as Douglass et al, I can't think off-hand of a worse paper published in a reputable climate journal.

Hmm, Tom, I think I misreferenced. The graph I was sure was in there... isn't. I'll have to go back over my climate physics notes to find the real one (I'm off to Edinburgh until Tuesday so it'll be a while!)

The hot-spot is not necessarily a case of 'no hotspot, weaker negative feedback'. As the atmosphere warms then specific humidity (what matters for radiation) goes up. More evaporative cooling (interpreted as the negative lapse rate feedback) means more warming aloft and more vapour up there too, so a warming feedback.

I was convinced I'd seen a paper that graphed that properly: LR versus WV feedback parameters for different models. And it was found that they roughly cancelled & the feedback parameter almost always came out as 1.2 W m-2 K-1. However, maybe I dreamed that. In which case I have to go find some other calculations before we can state with confidence whether the hot spot is linked to a more positive or negative feedback. (or has no effect)

I made those changes.  Should I wait to hear further from you on the hot spot as a feedback before posting, Mark?

Mark R, are you thinking of this graph:

If so, Chris Colose cites is as being from:

 Soden, B.J., I.M. Held, R. Colman, K.M. Shell, J.T. Kiehl, and C.A. Shields, 2008: Quantifying climate feedbacks using radiative kernels. Journal of Climate, 21, 3504-3520

http://journals.ametsoc.org/doi/pdf/10.1175/2007JCLI2110.1

I havn't read it for the relevant sections as yet.

I am uncertain about the climate sensitivity section.

The particular point that Monckton makes is in this exchange:

"Cook: “Monckton also repeats a myth … that most climate sensitivity estimates are based on models, and those few which are based on observations arrive at lower estimates. The only study which matches Monckton’s description is the immensely-flawed Lindzen and Choi (2009).”

Reply: I am not sure what qualifications Mr. Cook has to find Professor Lindzen’s work “immensely flawed”. However, among the numerous papers that find climate sensitivity low are ..."

(Original emphasis)

So, Monckton is not asserting directly that there is no concensus on climate sensitivity.  Rather he is asserting that, contrary to Cook, there is more than one study based on observations which conclude that there is low climate sensitivity.  Technically this is correct, but it is a blatant goal shift.  In his original debate he states that (16:46 on the video) that:

"Why do we think that we're going to suddenly get 3.3 Celsius for a doubling of CO2 concentration this century - that's the IPCC's central estimate - or 5.1 - which is your [Australia's] government's central estimate - when all the science done by measurement and observation rather than by models, suggests just one Celsius degree?"

(My emphasis)

Despite that, however, his specific claim is most naturally read as a dirict rebutal of dana's claim that there is only one paper meeting Monckton's description.  So interpretted, his point is correct and is established by his examples.  Complicating matters, however, his specific claim (that the papers cited are among numerous papers that find low climate sensitivity) is:

1)  Lowers the bar from supporting a climate sensitivity of 1 degree C per doubling of CO2 to finding low climate sensitivity;

2)  Is innacurate in that there are not many other papers supporting even low climate sensitivity (although no doubt a few more could be dredged up); and

3) Not all of the papers cited support a low climate sensitivity, either directly or indirectly.

Further, Monckton's citations is more nuanced than just the initial claim.  For example, he writes:

"Paltridge et al. (2009), who found that additional water vapor at altitude (caused by warming) tends to subside to lower altitudes, allowing radiation to escape to space much as before and greatly reducing the water vapor feedback implicit in a naïve application of the Clausius-Clapeyron relation"

Without reading the original paper I cannot comment on that.  Certainly the mechanism is plausible for reducing the WV feedback, although not for eliminating it in that excess Water Vapour due to greater evaporation rates will still be lifted to the upper troposphere before "subsiding", thereby increasing upper tropospheric water vapuor and hence forming a positive water vapour feedback.  What is more, it is foolish to pick out individual feedbacks and say that because an adjusted estimate is lower, therefore the total climate sensitivity is lower.  Such arguments assume that either there is no error, or any error will make the estimate more negative for other feedbacks.  Of course, it would be bizzare for all error estimates to be wrong in just one direction.  More importantly, many climate sensitivity determinations are empirical in nature and hence make no assumption about the force of individual feedbacks.  Altering the estimation of individual feedbacks therefore has no impact on the empirical estimates of climate sensitivity. 

More importantly for my present purpose, there is nothing in Monckton's use of Paltridge et al which is contradicted by our comment on it.  Therefore our treatment of the individual papers is insufficiently nuanced.

Further notes:

I had difficulty finding the relevant sections of various papers.  Indeed, I was unable to fin where in Wentz et al 2007 it states the implications for climate sensitivity are unclear.  It would be helpfull to add as notes the various quotes form these papers supporting the primary claims made about them.

Minor issues:

Tol 2009 link dead.  Also, it appears to be hosted at SkS.  We must be certain there is no violation of copyright in doing so.

"The temperature change also occurred..." The specific temperature change cited by Monckton ... (ie, neither we nor he are referring to the entire YD event)

We need to state that it is our emphasis after the Met Office quote.

Link to Church et al 2011 in caption for figure 4 is dead.

Fixed the Tol and Church links.  Tom, Monckton said:

"the real scientific debate is about how much warming extra CO2 in the air will cause. There is no “consensus” on that"

That's what I was responding to.  It's a later point, but on the same issue, so I just grouped it with the sensitivity discussion.  Anyway, we don't want to turn this into a point-by-point debate, we want it to be more of a general scientific overview.

Hence I think the Paltridge response is fine.  Paltridge's conclusions (and thus Monckton's extrapolations from those conclusions) were based on flawed data, which has since been updated by subsequent studies.  That invalidates Monckton's argument.

Wentz said the implications of their reserch on global warming are unclear (see the quote in this comment).  Hence that should mean their implications regarding climate sensitivity are also unclear.

In general, as we know Monckton is no climate expert, my feeling is that he shouldn't be trying to extrapolate conclusions from scientific papers that the scientists themselves don't make.  Thus to me, the fact that Wentz, Paltridge, Douglass, etc. didn't say anything about climate sensitivity in their papers is a sufficient to Monckton's claims.  Certainly the influence of the hot spot and larger than expected precipitation on climate sensitivity is a tricky question, but it's therefore also not a question that Monckton should be trying to answer on his own (I added a comment to the post saying as much). 

Thus here I think it's sufficient to simply point out that the papers draw no conclusions about climate sensitivity, and thus Monckton is misrepresenting them by claiming otherwise, and hence I think the post is okay to publish in its current state, without having to try and tease out the answers to these questions ourselves.

Oh, and here's my thumb.

I don't like the discussion on Wentz.  Let me clarify:

What you said:

[Wentz et al] dealt with precipitation, not temperature changes or climate sensitivity.  The paper finds that while water vapor has increased as expected, precipitation has increased more than predicted by climate models.  The implications regarding global warming (and thus climate sensitivity) are unclear, as stated in the paper.  It certainly does not imply that climate sensitivity is threefold less than expected, as Monckton claims in his misrepresentation of this paper.

What Monckton said:

Wentz et al. (2007), who found that the rate of evaporation from the Earth’s surface with warming rose thrice as fast as the models predicted, implying climate-sensitivity is overstated threefold in the models;

Important things to note:

• Nowhere did Wentz et al ever mention climate sensitivity - the word senstivity only appears once in the whole paper+supplement, and doesn't have to do with ECS/TCS; there should be a better way to nuance what you mean to say than what you have now, it seems you're saying they mentioned sensitivty explicitly;

• Monckton is WRONG about evaporation being the variable different between observations and models, it was precipitation (indicating he didn't read the paper, even aside from his comments about sensitivity);

• The total water content of the atmosphere has increased at the same rate predicted by models;

• OT a bit, and not pertinent to the discussion I think, if precipitation is higher than the models say, does that really have no implications for the hot spot (Mark?)?

• If precipitation is higher, could that mean that there is a greater loss of heat to space than the models predict?  Where does the latent heat of that phase change go to?  Or, does the atmosphere actually contain more heat than the models predicted, implying a potentially higher climate sensitivity?

• Drive home the point that changes in the hydrologic cycle are not the only feedback, so his 3/3 comparison is falsely applied.

Alex make some good points.  Monckton is completely misrepresenitng this paper and showing that he does not understand the hydrologicla cycle.  The paper's findings are anything but encouraging as they indicate that the models are being too conservative in simulating the acceleration of the hydrological cycle as the planet warms and water vapour levels increase-- this does not bode well for future extreme precipitation events.   

Also from Wentz et al. (2007):

"The SSM/I data set extends from 1987 to 2006. During this time Earth’s surface temper- ature warmed by 0.19 ± 0.04 K decade–1, according to the Global Historical Climatology Network (10, 11). Satellite measurements of the lower troposphere show a similar warming of 0.20 ± 0.10 K decade–1 (12). The error bars are at the 95% confidence level. This warming is consistent with 20th-century climate-model runs (13) and provides a reasonable, albeit short, test bed for assessing the impact of global warming on the hydrologic cycle."

So during the time that the models were understimating the increase in precipitation, the modelled rate of increase in global temperature (mostly due to an increase in GHGs) was consistent with the observations both in the surface and satellite records. This does not support the models' having a climate sensitivity that is too high, nor does is support in any way the assertion that the current best estimates for equilibrium climate sensitivity to doubling CO2 is three times too high!

Is there a reason some papers don't have hyperlinks to them too?  Spencer and Braswell 2011, for instance.

I thought that was basically what I said about Wentz.  That was the intent anyway, but I'll try to clarify.

Some papers I linked to our posts about them, which in turn link to the papers themselves.  Hence the link is in the discussion rather than linking the paper reference itself, but in each case there's a link.

Smack those points down though, the evap. v. precip. one (Monckton didn't read - don't really have to say those words, but that's why this is important, yet more misrepresentation) and the point about the models getting the total water content change right.

New Wentz write-up:

Wentz et al. (2007): dealt with precipitation, not climate sensitivity.  The paper finds that while atmospheric water vapor has increased as expected, precipitation has increased more than predicted by climate models.  Note that   Monckton wrongly claims Wentz et al. concluded evaporation has risen 3 times faster than expected; water vapor was on par with expectations, and it was precipitation that has risen faster than expected.  The implications regarding global warming (and thus climate sensitivity) are unclear, as stated in the paper (in fact the paper does not even mention climate sensitivity).  Monckton claims that because precipitation has increased roughly 3 times faster than climate models expect, this means climate sensitivity is 3 times lower than expected.  This conclusion is wholly unjustified; even if we knew the implications of increased precipitation on climate sensitivity, the hydrological cycle is quite obviously not the only global warming feedback.

This post is scheduled to be published in the morning; I think it's ready, but I can hold off if others want to continue reviewing it.

Maybe:

The implications regarding global warming (and thus climate sensitivity) are unclear

to:

The potential effects regarding global warming (and implicitly climate sensitivity) are unclear

I think that this is a bit clearer in identifying that the paper didn't actually mention climate sensitivity; it's not as seemingly orthogonal (there's my once a blue moon usage of that word) to your following parenthetical comment.

I like it otherwise.  I'm going to do another check through, but it looked good before to me.

Okay thanks Alex.  Change made, let me know if you catch anything else.

Well since I have access, I made two minor changes - took out the WUWT short near the beginning (it was unnecessary as the acronym is not used again in the post), and changed an i.e. to e.g.

Suffice to say, looks good to me.

The link to WUWT has been removed but the (linked above) in sensitivity section has not.

Seeing I'm a co-author and all, I should probably admit that I like the post.

Unless it was fixed Brian, I still see the link.

Hehe thanks Tom, and Alex.  I'm going to go ahead and publish.

tropical troposhere temperatures should be "tropical troposphere temperatures", with a "p"

Thanks KR.

Tom, I think that was the paper I was looking for! It's over a year since I read it, I guess I need to go through it again and make sure I didn't make any mistakes. Thanks for finding it for me again :)