Forcing or Feedback?

Guest Post by Willis Eschenbach

I read a Reviewer’s Comment on one of Richard Lindzen’s papers today, a paper about the tropics from 20°N to 20°S, and I came across this curiosity (emphasis mine):

Lastly, the authors go through convoluted arguments between forcing and feed backs. For the authors’ analyses to be valid, clouds should be responding to SST and not forcing SST changes. They do not bother to prove it or test the validity of this assumption. Again this is an assertion, without any testable justification.

Now to me, showing that clouds respond to the tropical surface temperatures, either on land or sea, is a no-brainer, I’m surprised that the Reviewer would question it. But before I discuss that, let me digress for a moment to say that I think there are two simple changes in the reviewing process that would help it immensely.

1. Double blind reviewing. Right now, reviewing is just single blind, with the reviewer knowing the identity of the author but not the other way around. This is a huge, huge violation of the scientific norms, and it leads and has led to problems. A recent proposal to use double-blind reviewing in the awarding of grants by the National Science Foundation makes for interesting reading … http://www.nsf.gov/nsb/publications/2011/meritreviewcriteria.pdf This would put the focus back onto the ideas rather than the author.

2. Publish the reviews and the reviewer’s names along with the paper when it is published. This sunshine, this transparency, will have several benefits.  First, if a reviewer still has strong reservations about the paper, or if their objections have led to improvements or changes in the paper, this will be made visible when the reviews are published (presumably online) along with the paper. Second, we get to see who agrees that the paper is valid science, and who might not. Third, in the fullness of time, we will have a record of who was right and who was wrong. But I digress … the anonymous Reviewer wanted Lindzen to show that clouds respond to SST.

Not having read Lindzen’s paper, I don’t know what he said about the question. But for me, Figure 1 should convince even the most alarmist reviewer that clouds are responding to the surface temperature, both on land and over the ocean.

Figure 1. Monthly albedo averages from ERBE. Images show the N/S range of the geographical tropics (~ 23.5° N/S). Thin black horizontal line shows the Equator.

The upper panel shows the albedo for August (northern hemisphere summer) and the lower panel for February (southern hemisphere summer). The brighter areas show the regions with higher albedo, because there are more clouds. There are some things of note in the figure.

For example, look at Brazil and the Amazon in the lower panel (the bright spot in the lower left of the panel, below the Equator.). In February, in the southern summer, the heat creates lots of clouds. But in the upper panel, the southern hemisphere winter, Brazil has a much lower level of clouds.

The same thing is true about southern Africa. In the southern summer (lower panel) the albedo is high in southern Africa, with lots of clouds. In the southern hemisphere winter (upper panel), on the other hand, southern Africa loses its clouds, and above the Equator where it is summer we see increased albedo in the northern part of Africa.

Next, tropical Asia in August, when it is warm, has lots of clouds. But when winter comes, the clouds move with the summer down to the southern hemisphere.

Similar changes are generally occurring over the oceans as well, but in a more muted fashion. Again, this difference in effect shows that surface temperature is the driving factor, since the land temperatures change more with the seasons than do the ocean temperatures.

Now, we have three choices here:

1. Tropical clouds are increasing and decreasing  in response to the changes in surface temperature resulting from the sun’s motion, which alternately warms the areas north of and south of the Equator, or

2. The clouds are just coincidentally in sync with the sun by random chance, or

3. The clouds are making the sun move north and south of the Equator.

Your choice, but to me, that should satisfy even the most recalcitrant reviewer. It’s clear that the clouds are responding to the variations in surface temperature, with more clouds forming as the surface warms and less clouds when the surface is cooler.

w.

PS: Since we’re discussing clouds as forcing and feedback, you’ll excuse me if I take this opportunity to re-post the following from my “Thermostat Hypothesis” paper. In it I am describing how the clouds and thunderstorms work as a governor to stabilize the temperature:

The problem with my thought experiment of describing a typical tropical day is that it is always changing. The temperature goes up and down, the clouds rise and fall, day changes to night, the seasons come and go. Where in all of that unending change is the governing mechanism? If everything is always changing, what keeps it the same month to month and year to year? If conditions are always different, what keeps it from going off the rails?

In order to see the governor at work, we need a different point of view. We need a point of view without time. We need a timeless view without seasons, a point of view with no days and nights. And curiously, in this thought experiment called “A Day In the Tropics”, there is such a timeless point of view, where not only is there no day and night, but where it’s always summer.

The point of view without day or night, the point of view from which we can see the climate governor at work, is the point of view of the sun. Imagine that you are looking at the earth from the sun. From the sun’s point of view, there is no day and night. All parts of the visible face of the earth are always in sunlight, the sun never sees the night-time. And it’s always summer under the sun.

If we accept the convenience that north is up, then as we face the earth from the sun, the visible surface of the earth is moving from left to right as the planet rotates. So the left hand edge of the visible face is always at sunrise, and the right hand edge is always at sunset. Noon is a vertical line down the middle. From this timeless point of view, morning is always and forever on the left, and afternoon is always on the right. In short, by shifting our point of view, we have traded time coordinates for space coordinates. This shift makes it easy to see how the governor works.

The tropics stretch from left to right across the circular visible face. We see that near the left end of the tropics, after sunrise, there are very few clouds. Clouds increase as you look further to the right. Around the noon line, there are already cumulus. And as we look from left to right across the right side of the visible face of the earth, towards the afternoon, more and more cumulus clouds and increasing numbers of thunderstorms cover a large amount of the tropics.

It is as though there is a graduated mirror shade over the tropics, with the fewest cloud mirrors on the left, slowly increasing to extensive cloud mirrors and thunderstorm coverage on the right.

After coming up with this hypothesis that as seen from the sun, the right hand side of the deep tropics would have more cloud than the left hand side), I though “Hey, that’s a testable proposition to support or demolish my hypothesis”. So in order to investigate whether this postulated increase in cloud on the right hand side of the earth actually existed, I took an average of 24 pictures of the Pacific Ocean taken at local noon on the 1st and 15th of each month over an entire year. I then calculated the average change in albedo and thus the average change in forcing at each time. Here is the result:

Figure 2. Average of one year of GOES-West weather satellite images taken at satellite local noon. The Intertropical Convergence Zone is the bright band in the yellow rectangle. Local time on earth is shown by black lines on the image. Time values are shown at the bottom of the attached graph. Red line on graph is solar forcing anomaly (in watts per square meter) in the area outlined in yellow. Black line is albedo value in the area outlined in yellow. 

The graph below the image of the earth shows the albedo and solar forcing in the yellow rectangle which contains the Inter-Tropical Convergence Zone. Note the sharp increase in the albedo between 10:00 and 11:30. You are looking at the mechanism that keeps the earth from overheating. It causes a change in insolation of -60 W/m2 between ten and noon.

Now, consider what happens if for some reason the surface of the tropics is a bit cool. The sun takes longer to heat up the surface. Evaporation doesn’t rise until later in the day. Clouds are slow to appear. The first thunderstorms form later, fewer thunderstorms form, and if it’s not warm enough those giant surface-cooling heat engines don’t form at all.

And from the point of view of the sun, the entire mirrored shade shifts to the right, letting more sunshine through for longer. The 60 W/m2 reduction in solar forcing doesn’t take place until later in the day, increasing the local insolation.

When the tropical surface gets a bit warmer than usual, the mirrored shade gets pulled to the left, and clouds form earlier. Hot afternoons drive thunderstorm formation, which cools and air-conditions the surface. In this fashion, a self-adjusting cooling shade of thunderstorms and clouds keeps the afternoon temperature within a narrow range.

Advertisements

  Subscribe  
newest oldest most voted
Notify of

Willis: In Figure 1, what is the time period for Figure 1? Is that a specific year or are those the February and August averages for a time period?

cui bono

Thanks Willis. I entirely agree on double-blind peer review – that should cause the ‘pal-reviewers’ some problems.
As a complement to your work on cloud coverage, the average height of clouds has also been in the frame:
http://thegwpf.org/science-news/5052-global-warming-may-be-held-back-by-falling-cloud-levels.html
Cue Joni Mitchell…

Having lived and worked in the tropics I can say that convective clouds form over both ocean and land. Ocean formed convective cloud forms slower than land formed but still gives afternoon thunder storms, though not every day in the Maldives. Night time clouds did continue to form, and give rain, but this declined over night as the surface temperature cooled.
Having flown through the ITCZ several times I can confirm that it contains some very energetic convective clouds climbing to well over 70,000ft (21,336m). It would seem that the old claim that Cb clouds stopped building at the tropopause is wrong for the ITCZ.

Kasuha

This is a nice example on how by emphasizing a part of a sentence you can completely change its meaning. Are you really sure it wasn’t the “and not forcing SST changes” what the author considered important?
Apart of that, I kinda disagree that clouds respond to surface temperatures. Assuming constant supply of condensation nuclei, clouds respond to evaporation and evaporation is in fact negative feedback on surface temperature change – you get way more evaporation and way less temperature differences on sea than on land, for example.

Blade

Two excellent points.
The first one about double blind review is so damn obvious I am ashamed it never crossed my mind.

“Right now, reviewing is just single blind, with the reviewer knowing the identity of the author but not the other way around.”

Admittedly it might get tricky because of possible references within a paper back to earlier works of the author(s), but this is no excuse not to try.

AJB

Tiddly Typo: Brazil has a much lower leven of clouds.
[Mind you, unleavened clouds conjures up some very nice imagery. Maybe you should leave it as it is!]
[Thanks, fixed, and I do like the idea of unleavened clouds … -w.]

Stephen Wilde

I agree with pretty much all that but suggest that it is only part of the story.
The level of solar activity also changes the air circulation from the top down by processes not yet fully understood and that modulates the cloud response to SSTs on much longer timescales than the annual seasonal variation.
Longer even than the ENSO cycle and the 60 year Pacific Multidecadal Oscillation. Most likely on timescales between MWP and LIA and the present.
Climate change is the netted out consequence of the bottom up SST effects and top down solar effects both of which combine to alter surface air circulation globally and thus global albedo.
I also agree with Kasuha in that the rate of evaporation is at least as important as raw SST since water vapour is lighter than air and provokes additional convection without the need for any increase in the surrounding temperature.
In fact I think the apparent ‘lid’ on tropical SSTs that Willis has previously drawn to our attention is set at the point where the rate of evaporation rises to a level where convective uplift from the amount of water vapour present starts to exceed uplift from surface temperature alone.

Grey Lensman

I live in the ITCZ and its great fun, especially in peak periods, landing or taking off, in a jumbo jet wondering if you will make it.
The jungle retains vast amounts of moisture and due to its multilayer structure, has possibly far larger surface area than ocean to facilitate evaporation. Thus the rapid formation of storms over land as opposed to those growing over oceans .
It never ceases to amaze me how,at times, it can seen to suddenly get hotter at night. Usually occurs before night time rain. Perhaps indicative of humidity rising.
Watching, massive thunderbolts strike to road and trees just feet away from you reminds you strongly of your mortality. I saw a banana copse, across the road literally explode and get strewn all over the road, when struck by lightening.

Dave

Willis>
Is double-blind reviewing possible? I would have no trouble picking your writing from Anthony Watts’ or Steve McIntyre’s, let alone that of some of the alarmists, whether it has your name on it or not..

cui bono

Dave says (June 7, 2012 at 4:50 am)
Is double-blind reviewing possible? I would have no trouble picking your writing from Anthony Watts’ or Steve McIntyre’s..
——————–
Ah, but then S. Mosher and others had no trouble picking out Peter Gleicks writing from the fake memo, but apparently “he didn’t write it – official”. 😉

Willis
What if clouds show BOTH feedback AND forcing? e.g., forcing via Svensmark’s cosmic ray hypothesis?
How do we distinguish and quantify each?
With solar driving ocean temperature, note David Stockwell’s solar accumulation theory predicting a Pi/2 (90 deg) lag from solar forcing to ocean temperature response. e.g., 2.75 years for the 11 year Schwab cycle. Projecting tot he annual cycle suggests a 3 month lag from solar max/min to ocean temperature max/min. Presumably there should be a corresponding lag in clouds responding to temperature lagging solar forcing.
Recommend checking how cloud albedo magnitude follows this lag from the solar forcing to ocean temperature.
For direct solar forcing note Svensmark uses Forbush events to show impact from the Sun.
Effects of cosmic ray decreases on cloud microphysics J. Svensmark et al. 2012

Having lived in the tropics for a long time, within 100Ks of the equator. I can say the notion that convective cloud formation and precipitation is strongly tied to diurnal warming is wrong. It rains at all times of the day pretty evenly. The idea that tropical rainfall occurs in the afternoon results from places where sea breezes and orographic effects produce clouds and rain in the late afternoon.
Now to me, showing that clouds respond to the tropical surface temperatures, either on land or sea, is a no-brainer
I’ll suggest the response of tropical surface temperatures to clouds is rather more significant to climate change. We know there are effects on clouds: aerosols, GCRs that effect SSTs.

P. Solar

Willis, which paper are you referring to here?
Lindzen and Choi 2009 can be found here:
http://www.drroyspencer.com/Lindzen-and-Choi-GRL-2009.pdf
Lindzen and Choi 2011 which was a follow up addressing some of the criticisms is available here:
www-eaps.mit.edu/faculty/lindzen/236-Lindzen-Choi-2011.pdf
If it was another please specify.

P. Solar

Reviewer’s criticism of L&C:
Lastly, the authors go through convoluted arguments between forcing and feed backs. For the authors’ analyses to be valid, clouds should be responding to SST and not forcing SST changes. They do not bother to prove it or test the validity of this assumption. Again this is an assertion, without any testable justification.
The both are true seems to be patently obvious. Though someone ought to prove it formally, I have seen much less obvious stuff taken as given all though out climate research. Funny how they always get picky when they don’t “like” a paper’s findings.

P. Solar

Last thing I heard Dessler was trying to ridicule Spencer for suggesting cloud could be driving SST. Seems you can’t win.

P. Solar

Stephen Wilde: “In fact I think the apparent ‘lid’ on tropical SSTs that Willis has previously drawn to our attention is set at the point where the rate of evaporation rises to a level where convective uplift from the amount of water vapour present starts to exceed uplift from surface temperature alone.”
Good point, I had not made that connection and was a little sceptical of that claim when Willis made it. In fact I think the difference is that surface temperature only provides at negative feedback whereas uplift due to less dense water vapour is a self sustaining process. This makes it more than just a linear negative feedback. It’s like negative feedback with a tipping point!
That effect in itself would probable be sufficient to ensure climate stability on the upper end to a wide range of additional forcing.
I think Willis covered this in his posts on ARGO data. It was particularly clear in SH, there seemed to be an upper limit of about 38C of SST.

Pamela Gray

A forcing does indeed become a feedback, which becomes a forcing, which becomes a feedback. That’s what energy does. Perpetually stored energy in a highly variable intrinsic system is nearly impossible (and I say this because someone will tell me that such and such is a perpetually stored source of energy). So I would say that at some point, one can point to SST’s forcing cloud anomalies. But that is an arbitrary starting point ONLY. Cherry picked if you will. I can also say that clouds force SST anomalies. Both constructions are right and should be part of the background knowledge of basic science.

Ninderthana

Kashua,
I think that what Willis is showing is that both factors are occurring:
Of course surface temperature responds to cloud. Given a constant input of solar heat (e.g. stop the Sun in the sky), a cloud comes over and the surface temperature drops. Again if the Sun’s input remains unchanged, and the cloud moves away, the surface temperature goes back up again. And, of course the amount of cloud depends on the amount of evaporation. The more evaporation (e,g. in the tropics), the greater the amount of cloud that develops. Finally, its a no brainer that evaporation increases with increasing sea-surface temperatures, supporting the opposite contention that the amount of cloud responds to (sea) surface temperature.
I think that what Willis is saying is that from the unchanging perspective of the Sun, there is a clear
mechanism that stabilizes world temperatures that involves clouds forming from increased evaporation which is responding to increasing sea and land surface temperatures.

Ian W

John Marshall says:
June 7, 2012 at 3:32 am
Having lived and worked in the tropics I can say that convective clouds form over both ocean and land. Ocean formed convective cloud forms slower than land formed but still gives afternoon thunder storms, though not every day in the Maldives. Night time clouds did continue to form, and give rain, but this declined over night as the surface temperature cooled.
Having flown through the ITCZ several times I can confirm that it contains some very energetic convective clouds climbing to well over 70,000ft (21,336m). It would seem that the old claim that Cb clouds stopped building at the tropopause is wrong for the ITCZ.

Sounds like Gan.
It would seem that the old claim that Cb clouds stopped building at the tropopause is wrong for the ITCZ.
Not true. The tropopause is the top of the convective activity wherever the storms push it to. Its like a rolling sea and with discontinuities where the Hadley and Ferrel cells meet.
This is what Stephen Wilde is talking about as the Hadley Cell circulation increases in intensity the cells become larger and the jet streams get pushed away from the equator when the Hadley Cell circulation becomes weaker they shrink and the jet streams (and Ferrel Cells) move toward the equator. This is a larger scale longer term homeostatic response than the diurnal response that Willis has explained.

Clouds movement around equator is not random or only influenced by the sun. Clouds are electrically charged, so is the equatorial electrojet; intense tropical storms are partially result of interaction between these charged entities.
http://www.nasa.gov/centers/goddard/news/topstory/2006/space_weather_link.html
more at
http://www.vukcevic.talktalk.net/LFC20.htm
I think over-simplification of the events is not of assistance to understanding of complexity of the climate change.

Bill Illis

One issue with the 20S to 20N latitude band is that that clouds, rainfall, and SSTs vary considerably by longitude and by the state of nearby regions. The Walker Circulation, for example means that clouds and rainfall move east or west in the Pacific depending on the state of the ENSO. The last series of La Ninas has even left some Pacific Islands with very little rainfall and clouds while Indonesia and Australia have been drenched. El Ninos result in the opposite situation.

scarletmacaw

Willis: I though “Hey, that’s a testable proposition to support or demolish my hypothesis”.
Clearly you’re not a Climate Scientist. In CliScientology, testable hypotheses are not allowed.
Snark aside, I do agree with you. Negative cloud feedback was the obvious flaw in the cAGW snake oil pitch from the beginning.

JFD

I would like to extrapolate Grey Lensman’s point about the impact of surface area in tropcial bush on evaporation to land areas. For the last 50 or 60 years, vast areas of improved grasses have been planted both in suburban and urban settings, In suburban settings the grass is for fodder for food animals while in urban settings the grass is for landscaping.
The grass leaves provides surface for condensation at night. When the sun comes up the temperature of the leaves rise to the the evaporation point. The liquid water then changes from potential energy to kinetic energry as it evaporates and rises into the atmosphere. When the last liquid water evaporates the temperature of the leaves start to rise heating the air around them.
The water vapor, containing the latent heat of evaporation, rises to a level where temperature and other conditions are favorable for condensation of water vapor back into liquid water (rain). As a part of condensation the kinetic energy is transformed back into potential energy thereby releasing the latent heat as specific heat causing the upper atmosphere to heat up.
These energy and phase changes cause time lag in the system, transfer heat from ground level to the upper atmosphere and supplies water vapor for cloud formation. This effect would be for land areas only.
JFD

tty

I’m skeptical about double-blind reviewing. It isn’t really practical, at least in fields where there are relatively few people active. For example I can almost alway guess the names of the “anonymous” reviewers of my papers.
I’ve chosen the opposite alternative when reviewing other people’s papers. Many journals will let you chose whether you want to be anonymous or not, and I always select not to be, since I think that You should be willing to stand by your opinions openly. However, I realize that this is also difficult for many scientists who have to review papers by people who have e. g. influence over research funding.
Publishing reviews online together with the papers on the other hand seems a good idea with no obvious drawbacks. At least it would ensure that referees would have to produce reasonably rational and coherent reviews, which is not alway the case now.

beng

****
Grey Lensman says:
June 7, 2012 at 4:46 am
The jungle retains vast amounts of moisture and due to its multilayer structure, has possibly far larger surface area than ocean to facilitate evaporation.
****
I’d agree — even the deciduous temperate forest here in the growing season is IIRC similar in evaporative potential as open water, as long as it isn’t abnormally dry.

Eric

Willis
Are you sure about the reviewing process for Climate Science? If true, that is one major reform that should have happened long ago.
I know in my field, physiology/neurology, that all of the journals I have submitted papers to or reviewed for are double blinded. This forces the reviewer to judge the paper on its merits rather than on any personal issues he or she has with the authors.

Jknapp

Double blind sounds good, but won’t work. The field is sufficiently small that the reviewer will know the authors and/or their fields of research and will be able check with them to verify. Even if that is a breech of rules the current state of pal review implies that a simple “Is this your paper?” “no, then whose is it?”, won’t stretch their ethics much.

Hector M.

Willis, you may want to fix an obvious typo in the third paragraph below your two-panel map:
“The same thing is true about southern Africa. In the southern summer (lower panel) the albedo is high in South America, with lots of clouds. In the southern hemisphere winter (upper panel), on the other hand, southern Africa loses its clouds, and above the Equator where it is summer we see increased albedo in the northern part of Africa.”
In the first sentence, where it says “the albedo is high in South America” it should be “in Southern Africa”.
[Thanks, Hector, fixed. -w.]

Hector M.

About double-blind reviewing: As an author I usually do not know the name of my reviewers, and as a reviewer I do not know the authors’ name. In other words, double-blind reviewing is already in place. The only way around would be for editors to cheat, i.e. telling reviewers the name of the author (for the sake of gatekeeping) or vice versa (never either situation happened to me, but then I do not publish or review papers on climate science for a climate science journal: not my field). The violation of double-blind reviewing would be, in the journals I know, a subversion of the normal procedure and a case of editorial malpractice.

Jim Clarke

I agree with Pamela Gray:
“So I would say that at some point, one can point to SST’s forcing cloud anomalies. But that is an arbitrary starting point ONLY. Cherry picked if you will. I can also say that clouds force SST anomalies. Both constructions are right and should be part of the background knowledge of basic science.”
In the lower latitudes, particularly during the summer, surface temperatures seem to be forcing clouds. At higher latitudes, it is actually cloudier when the surface temperature is colder (winter) than warmer (summer), which is opposite of the tropics. Albedo is lower in the winter at high latitudes because the sun angle is so much smaller and temperatures are colder, not because there are fewer clouds.

Right on the double blind approach is the best but most costly. Today it is less about quality and more about quantity and grant generation. Grants result from catering to the preconceived ideas and bias of the granter. The business of science has always been some measure of balance between the “old boy network” and analysis objectivity. Double blind or not I think publishing the reviews along with the paper will go a long way to shaping everyone up if not solving the problem.

Gail Combs

I agree with Dennis Nikols P. Geo. and others Reviewing should be double blind but more important, the reviews should be published with the paper.

What’s bizarre about the reviewer comment is that, if it were true, that non-feedback cloud variations were confusing the signal, they would almost certainly skew the results in favor of positive feedback. And to some extent there probably are some non-feedback cloud variations that cause temperatures to change (Lindzen had lag correlation plots that hint this was going on in the short wave). But these create a positive feedback bias. Meaning that, if the reviewer’s suggestion were correct, Lindzen would have gotten too large a sensitivity.
I used atmospheric temps because the atmosphere itself will then determine the appropriate lag, not to mention the fact that clouds are not magically aware of the sea surface temperature below them, they respond to the temperature of their local environment (ie the atmosphere). I got strong negative feedback-even stronger if you variance adjust the tropospheric temps to match the surface. Me and another blogger tried to get an analysis like what I did published (that didn’t use quite the same method and got closer to no or weak positive feedback-probably biased to positive) the reviewers rejected it for what struck me as really stupid reasons. Some Bayesian bullsh*t (I am an unreconstructed frequentist) and some other objections that indicated they didn’t understand the paper.

Keith Battye

So once again we have the sun and water driving the temperature. Seems legit.
I particularly like the composite in fig 2 as it really does portray our reality from a different perspective.

I’ve reviewed for conferences with very thorough double blind reviewing, and IMHO it doesn’t make any real difference, except it gives the appearance of not being biased towards established authors. I don’t think anything is lost by double blind reviewing, but I don’t think there is that much to be gained. As others have pointed out, anyone that stands out from the crowd will be immediately identifiable from the content of the paper whether their names are on it or not. But as nothing is lost, why not?
Publishing the reviews with the reviewers names will end up with either timid ineffective reviewing (as the reviewers will be rightly concerned that if they give an incorrect review it will reflect badly on them) or the review process taking even longer (because the reviewer will have to take even longer on the reviews making sure that they are accurate). “Third, in the fullness of time, we will have a record of who was right and who was wrong.” this is the very reason this proposal will not work, if someones academic reputation hangs onthe quality of the review, then they will want to put as much effort into the review as they do their own papers, which is simply unfeasable, they need to get their own work done as well.
It seems to me that the best way of guarding against pal review would be for the journal to publish a list of the reviewers they have used each year (possibly with the number of reviews performed). Some journals (e.g. Neural Networks) already do this as a way of thanking the academics that have supported the journal.
However at the end of the day, the reviewers are only human, so flawed papers will always slip through the net, no matter how much effort is expended.
Which reminds me, I have a couple of papers to review…

Willis Eschenbach

tty says:
June 7, 2012 at 7:44 am

I’m skeptical about double-blind reviewing. It isn’t really practical, at least in fields where there are relatively few people active. For example I can almost alway guess the names of the “anonymous” reviewers of my papers.
I’ve chosen the opposite alternative when reviewing other people’s papers. Many journals will let you chose whether you want to be anonymous or not, and I always select not to be, since I think that You should be willing to stand by your opinions openly. However, I realize that this is also difficult for many scientists who have to review papers by people who have e. g. influence over research funding.
Publishing reviews online together with the papers on the other hand seems a good idea with no obvious drawbacks. At least it would ensure that referees would have to produce reasonably rational and coherent reviews, which is not alway the case now.

Thanks, tty. The NSF has gotten the first results from their trial of double-blind reviews of grant proposals, but unfortunately I didn’t snag the link when I read it and can’t find it now … grrr, any assistance in locating it welcomed. In any case, one of their comments was that in a number of cases, reviewers who thought they know who wrote one of the proposals were entirely wrong.
So yes, they can guess, and in some cases they might be right … but if you have to sign your name to your review and it is published, it would at least make you cautious about favoring or opposing something based on your guess about the author …
w.

KR

Willis Eschenbach – The issue referred to by the reviewer (assuming this was the L&C submission to PNAS) was that Lindzen and Choi required clouds to cause SST changes (as seen in cloud changes being related to SST, with cloud changes leading in their analysis), and the reviewer was objecting to that unsupported assumption.
Tied to this was the discussion in the earlier Trenberth 2010 paper (http://www.atmosp.physics.utoronto.ca/~jclub/journalclub_files/trenberth2010.pdf), and the following Dessler 2011 paper (http://geotest.tamu.edu/userfiles/216/Dessler2011.pdf), where Dessler notes:
“The observations show that larger negative slopes
exist when the cloud time series leads the surface temperature,
with mostly positive slopes when the temperatures leads the
cloud time series. Based on this correlation, LC11 conclude
that clouds must be initiating the climate variations.”

This also arises in the L&C11 AMIP model analysis, where they assume the same causality of clouds->SST, the same lead/lag relationship – even though those models, as written, have causality going in the opposite direction. In the models, that’s simply not what’s happening, and in the real world, L&C and Spencer appear to be the only ones holding this quite odd opinion you have criticized.

I believe you have misread those reviewer remarks. The issue is with L&C, not the reviewer.

Willis Eschenbach

Gotta say, the comments on double blind reviewing seem split between ‘It will never work’, and ‘In my field it’s the standard way it’s done’ … go figure. I don’t think both of those can be right, and one is theory while the other is practice.
w.

Willis, People always put their names in published book reviews, even highly technical books, so it must work.

Willis Eschenbach

KR says:
June 7, 2012 at 10:38 am (emphasis mine)

Willis Eschenbach – The issue referred to by the reviewer (assuming this was the L&C submission to PNAS) was that Lindzen and Choi required clouds to cause SST changes (as seen in cloud changes being related to SST, with cloud changes leading in their analysis), and the reviewer was objecting to that unsupported assumption.

Thanks, KR, but that seems like a very tortured reading of a clear either-or statement by the reviewer. S/he said:

For the authors’ analyses to be valid, clouds should be responding to SST and not forcing SST changes.

So near as I can tell, the reviewer is saying the exact opposite of what you claim …
w.

Anthony, the differences are that book reviews are not a check on the correctness of the book, just a rough guide to its overall worth. Book reviews don’t contribute much, if anything, to a scientists reputation.
REPLY: I was also thinking of technical books in science and industry, where facts do indeed have to be checked. For novels/fiction, yes your point is taken – Anthony

KR

Willis Eschenbach – L&C11 have clouds forcing SST’s, the reviewer is pointing out that assumption, and stating that the paper and analysis is invalid as a result of the L&C assumption.
I have, incidentally, read L&C09, the L&C PNAS submission, L&C11, the reviews, and a number of the rebuttal papers. L&C sensitivity results are calculated with temperature changes lagged behind cloud changes.
From L&C11:
“Based on our simple model (viz Section 4b of Methodology), this ambiguity results mainly from nonfeedback internal radiative (cloud-induced) change that changes SST.” (emphasis added)
in other words, L&C11 consider clouds a forcing, not a feedback.

Anthony, book reviews of technical books are not checks of technical correctness either. That is performed by reviewers engaged by the publisher to check the correctness of the book before it is published (I have done this myself – the reward for which was a book ;o). The published book reviews at least in my field are essentially justified recommendations (or otherwise), but not a validation of the content.

KR

Willis Eschenbach“…a very tortured reading of a clear either-or statement by the reviewer…”
The reviewer stated that:
Requirement for validity: SST changes cause cloud changes.
Invalid analysis: Cloud changes cause SST changes.
L&C11: Based on our simple model (viz Section 4b of Methodology), this ambiguity results mainly from nonfeedback internal radiative (cloud-induced) change that changes SST.” (emphasis added)”
I would agree that in isolation the reviewer’s comment might seem a bit ambiguous. In the context of the reviewed paper itself, however, I would consider it quite clear – the reviewer was pointing out that the L&C11 assumption, of nonfeedback cloud-induced changes forcing SST changes, was invalid. And based upon your initial post and the comments in this thread, you would seem to agree.

sophocles

Willis said:
3. The clouds are making the sun move north and south of the Equator.
==========================================================
ROTFL —-so THAT’s how it works!
A paper reviewer said:
For the authors’ analyses to be valid, clouds should be responding to SST
and not forcing SST changes.
=============================================================
An amazing comment.
Said reviewer should be taken out to tour some commercial greenhouses.
They regulate internal temperature (and shade) by drawing canvas curtains
over the glass roof panels. It might interest the reviewer to note the outside
surface of the curtains are invariably white. These cloud analogues definitely
FORCE the internal temperature.
For that matter, all homes/buildings with curtains are able to repeat this to a
lesser degree to regulate both incoming and outgoing heat for maintenance
of an even (comfortable) internal temperature.
One has to wonder where these idiots are found ….

KR says: “in other words, L&C11 consider clouds a forcing, not a feedback.”
L&C11’s lag correlation graphs for shortwave forcing showed that the correlations of the shortwave flux changes with sea surface temperature changes were stronger with leads or lags, and changed sign depending on the lead or lag. They interpretated this to mean that the feedback signal, which should be at zero lag, was being corrupted by non feedback changes in shortwave flux which would cause temperature changes, ie “force” them.The reviewer said of this interpretation:
“For the authors’ analyses to be valid, clouds should be responding to SST and not forcing SST changes.”
It’s a bit difficult to follow, but as far as I can tell, what this means is that, in order for there intepretation to be correct and give them the correct result, clouds should be responding at LC11’s selected lag time. Read this way, the reviewer’s comment suggest an ignorance of the arrow of time. Positive lags of course correspond to feedbacks (although perhaps not the full feedbacks).
Now what you seem to be saying is rather different: in order for LC11’s lagged results to be correct, it must be true that the zero lag result contain some cloud changes that are not feedbacks. True. You also imply this is at odds with Willis. I don’t know about that. So I’ll ask the question:
Willis, is it you opinion that clouds changing on monthly timescales should exlusively be a function of the temperature of the sea surface roughly contemporary to it? If so, your understanding is not compatible with LC11’s lead lag analysis, which requires that some changes in clouds are not responses to the temperature. If so, the feedback should be found at zero lag (although I think there are physical reasons to expect some lag independent of the cause/effect issue).

atarsinc

Willis, where has your thermostat been for the last forty years? JP

Matthew R Marler

I like your graphs. Have you made the data and code publicly available?
Double-blind reviewing of grant proposals poses a problem: a desideratum in awarding a grant to a team is whether the team has successfully carried out similar research in the past; over time, the better teams should get the greater funding amounts. Possibly not insurmountable; related, most successful teams cite a lot of their own work, so the reviewers know the team, or at least the PI almost surely.
I think that reviewers’ comments and authors’ responses might be made available as supporting online material, available to people who really want to know. Compared to the final product, the reviewer comments and intermediate drafts are usually not that interesting.

eyesonu

@ Willis
“PS: Since we’re discussing clouds as forcing and feedback, you’ll excuse me if I take this opportunity to re-post the following from my “Thermostat Hypothesis” paper. In it I am describing how the clouds and thunderstorms work as a governor to stabilize the temperature: …”
=====
I had not read that before. Excellent! Very well explained. Even a politician should be able to understand. Unless they are intentionally “double blind” in a common sort of way that would have an entirely different meaning than that which was discussed in the main post.

Andrew and Willis, without reading too much into what the reviewer said or meant, it’s pretty well established that,
– Clouds do not cause the bulk of temperature change that is caused during ENSO events (for example), though that’s an example of a dominant mode of tropical variability, and one that severely limits the usefulness of the method of choice that Lindzen has adopted for inferring “equilibrium climate sensitivity.” A number of independent authors have found this method to be unsuitable.
– There are entire books written on the subject of convection as well as tropical dynamics, and the factors giving rise to convection are considerably more complex than just knowing the surface temperature. Depending on where you go there may or may not be a strong diurnal cycle connected with heating of the ground by insolation. Some land areas feature midnight to early morning maxima of precipitation. There are also observed nocturnal maxima insconvection in regions over the ocean, where at least three our four mechanisms may be important (including destabilization caused by nighttime longwave radiative cooling) in addition to local effects such as complex terrain and sea-breeze circulations. The seasonal cycle has structure as well, associated with the movement of the ITCZ or monsoon onset/demise, again depending on location. Convection is modulated by synoptic variability (e.g, the MJO), changes in wind direction, hurricane activity, etc.
– It is not sufficient to consider only the shortwave component of cloud influence, since the greenhouse effect of clouds roughly cancels this influence. This is one reason there is no evidence for clouds operating to keep the tropics from overheating, or for providing a robust ‘thermostat mechanism.’ There are suggestions of this in the literature, but it has fallen out of favor.

Alex

@atarsinc Perhaps Willis termostat have a max and a min value they keep the temp within. Look at the earths temp over it’s life time tell me what you see, does it look like a stable system or not? Continents have moved, the sun has changed the atmosphere has changed and so forth, and during all this time temp has been within a narrow band.