Guest Post by Willis Eschenbach
There’s a much-cited paper (129 citations) from 1994 called “On the Observed Near Cancellation between Longwave and Shortwave Cloud Forcing in Tropical Regions” by J. T. Kiehl (hereinafter Kiehl1994), available here. The paper makes the following claim (emphasis mine):
ABSTRACT
Observations based on Earth Radiation Budget Experiment (ERBE) satellite data indicate that there is a near cancellation between tropical longwave and shortwave cloud forcing in regions of deep convective activity. Cloud forcing depends on both cloud macrophysical properties (e.g., cloud amount, cloud height, etc.) and on microphysical properties (e.g., cloud particle size, particle shape, etc.). Hence, the near cancellation in the tropics could be due to either the macrophysical or the microphysical properties of these clouds, or a combination of these effects.
Now, to me that’s a pretty curious and surprising claim. The paper says that both in the Indonesian Region, as well as over the entire tropical Pacific, deep convective tropical clouds have no net effect on radiation, ostensibly because the longwave (positive) and reflected shortwave radiation (negative) cancel each other out. Let me call this the “cancellation hypothesis”. Kiehl supports the cancellation hypothesis inter alia with his Figure 1:
Figure 1. The first figure in Kiehl1994. The vertical axis shows shortwave cloud forcing, which is the amount of sunlight reflected by clouds in watts per square metre (W m-2). The horizontal axis shows the longwave cloud forcing, which is the change in top-of-atmosphere longwave forcing from the clouds. By convention, the reflected solar radiation is shown as negative, presumably because it is cooling the earth. Data is from April 1985.
And that looks pretty convincing … but, despite the 129 citations of the paper, I’m a suspicious fellow who believes firmly in the famous fallibility of experts. So I thought I’d use the CERES data and see if the cancellation hypothesis held up. Figure 2 shows that result, for the same Indonesian Region used by Kiehl.
Figure 2. Replication of the Kiehl study, using the ten-year April averages for the specified region to remove annual variations. Each square symbol represents a 1°x1° gridcell (N = 1500).
Instead of one single month’s data, I’ve used the averages of the ten Aprils in the CERES dataset.
Now, as you can see from Figure 2, the Kiehl hypothesis of cancellation has a big problem. The CERES results do not bear out Kiehl’s claims in the slightest. Instead, they support my hypothesis that increased tropical clouds cool the surface. As you can see, on average the loss from the reflected sunlight is about 20% or so greater than the gain from increased IR. This means that there is no cancellation. Instead, the clouds have a net cooling effect.
The paper goes on to say:
One feature of the cloud radiative forcing obtained from the Earth Radiation Budget Experiment ( ERBE) is the near cancellation between the longwave cloud forcing and the shortwave cloud forcing in tropical deep convective regions. This result was clearly shown by Kiehl and Ramanathan ( 1990) for the Indonesian convective region (here reproduced in Fig. 1 ). Further analysis of tropical deep convective regions of net cloud radiative forcing indicates that this is a ubiquitous feature that occurs over either ocean or land regions.
This is a more expansive claim than the one in the Abstract. Here the paper says that the cancellation happens over both land and ocean. So I thought I’d divide the Indonesian Region shown above into land and ocean regions. In addition, rather than average the months as in Figure 2, Figures 3 & 4 show all available April data for the entire time period. First, Figure 3 shows the land. I have colored the points by surface temperature of the gridcell.
Figure 3. As in Figure 2, but showing only the land. N=1,320.
Well, this shows that whatever might be happening in the Indonesian Region in the way of a linear relationship between reflected shortwave and longwave, it is definitely NOT happening over the land. The land shows little in the way of any relationship between shortwave and longwave.
Figure 4 shows the ocean data for the same Indonesian region.
Figure 4. As in Figure 2, but showing only the ocean. Note that the scale is slightly larger, to include all of the individual data. N=17,736
Well, this clarifies matters somewhat. First, about 40% of the land gridcells, but only about 10% of the ocean gridcells, have longwave cloud forcing greater than the shortwave forcing. Next, the land is pretty tightly clustered, with no apparent pattern. The ocean is different. Over the ocean the longwave is proportional to the shortwave … but it is a long ways from cancelling out. Instead, there’s a net cooling of -13.7 watts per square meter over the region. And the amount of cooling increases as the forcings increase. By the time the cloud reflections are up to 100 W m-2, the longwave is only up to 75 W m-2. That is a cooling from the clouds of about 25 W/m2, and not a cancellation under any meaning of the word.
In addition, there are a number of the warmest gridcells (red) which are on the left of the group (lots of reflection, little longwave).
Kiehl goes on to show his Figure 2, which compares the entire tropical Pacific region, from 10N 140E to 10S 90W. He shows a different kind of graph for this region, viz:
Figure 5. Kiehl Figure 2, showing the longwave and shortwave cloud forcing separately as functions of temperature.
Based on this graph, he makes the even more expansive claim that the cancellation of long-and shortwave radiation occurs across the Pacific, viz:
This figure illustrates that the cancellation between these two forcings occurs not only in the western tropical Pacific region of Fig. 1 but also across the entire tropical Pacific Ocean region. Even in regions of colder SSTs (298 K) the cancellation is apparent.
I found his style of graph in Figure 2 to be notably uninformative regarding the purported Pacific-wide cancellation. So I repeated his Figure 1 style of graph using the Pacific data, to see whether the forcings actually cancelled all across the Pacific. Figures 6 and 7 shows that result, for land and ocean, and reveals that there are large problems with this second claim as well.
Figure 6. As in Figure 2, but covering a larger area, the entire tropical Pacific Ocean as specified in the title. This figure shows land only. N=348.
Again, there is no clear pattern over the land, merely a cluster of data.
Figure 7. As in Figure 6, but covering ocean only. Note the slightly larger scale than in Figure 4. N=63,132.
This actually is pretty interesting. First off, again the ocean and land are different, with the land results clustered as in the Indonesian Region. Regarding the ocean, in only 5% of the gridcells does the longwave ever exceed the shortwave (area above the central diagonal line). In order for Kiehl’s cancellation hypothesis to be true, the average of a number of gridcells over an area would have to fall on the central diagonal line … which is clearly impossible with only 5% of them above the line.
Nor do things get better when we look at the entire dataset, and not just the April data. Figure 8 shows all of the data for the Pacific-wide area shown in Figure 7 (ocean only).
Figure 8. All months of data for the Pacific-wide area as in Figure 7
Next, rather than cancellation, there are a whole lot of gridcells where the longwave is smaller, and often much smaller, than the reflected shortwave. When there is solar reflection of a hundred watts per square metre, the longwave is only around sixty watts per square metre, for a full 40 W/m2 of cooling. And even on average the cooling is nearly 20W/m2 … not what we call cancellation on my planet.
Finally, his claim that “Even in regions of colder SSTs (298 K) the cancellation is apparent” is not upheld by the data. The temperature of 298 K [25°C] is shown in blue in Figure 7, and it is the farthest from cancellation of any of the data.
CONCLUSIONS
The main result is that the CERES data clearly and emphatically falsifies the cancellation hypothesis. In general, the longwave and shortwave are far from cancelling each other in the tropical deep convective areas.
A secondary result is that this clearly shows how the politicization of the field has affected the scientific process. Kiehl’s claims were very tempting to the theorists and modelers, because cancellation meant that they didn’t have to concern themselves with the deep convective processes, aka thunderstorms—the could simply repeat Kiehl’s claim that the shortwave and longwave cancelled each other out. And as a result, when more detailed data became available, the original claims of Kiehl1994 were never questioned.
Now, all we need is some automated method to notify the 129 people who cited the cancellation hypothesis in other scientific papers that the rumored cancellation has been cancelled for the duration …
All the best,
w.
Wonderful work & a great way to start off the new year.
My compliments, Willis.
This is a data-based confirmation validating your hypothesis moving it to a fully-fledged theory. The cooling effect is so strong that it can be seen on infrared camera trained on a storm path. CERES provides the data and your calculations the methodology.
You should at least write a a Letter to the Journal.
Sciences prizes are given to all the wrong people these days.
Talented amateurs such as yourself should consider writing a paper for peer review and publishing in a journal such as Nature. There have been many amateur scientists who advanced knowledge, no reason I can see for you to not to attempt to join their legion.
Typos abound, sorry about that, no way to edit once posted!
[Fixed. -w.]
Excellent analysis.
Willis, that is a great piece of work that needs to be widely distributed and published in a peer reviewed journal. It needs to be in a place where it can’t be ignored or casually dismissed as the work of a loathsome “denier”.
John Robertson says:
December 30, 2013 at 8:57 pm
Thanks, John. Been there, done that, I’m one of the few people with absolutely no scientific credentials to have a peer reviewed piece of writing published in Nature in the last while. And while you may be right, I am generally averse to publishing in the journals. In part, this is because of the difficulty of getting things past the pal-review process. In larger part, it is because of the sometimes glacial slowness of the whole thing.
Mostly, though, the reason is that I feel that I can have more effect on the ongoing scientific conversation by publishing here. My strong sense is that not only do I reach more people on WUWT. I also reach the people that count, the movers and the shakers on both sides of the climate aisle, all of whom read WUWT.
Be clear that I’m not saying that publishing in the journals is bad or wrong. I just think that I can have a much greater sway publishing here in a timely manner, than publishing four months from now in some journal that not many folks read.
All the best,
w.
Greetings from Oz. Excellent article, Willis, which seems to fairly demolish the fudged claim by Kiehl. I also agree with your thoughts re publishing in peer reviewed journals vs WUWT, despite the high likelihood that the Other Side will use it against you anyway. However, as you say, the outcome you want is to make the article available in a timely (and unimpeded?) manner to All those interested in this field, not just a select few. Great work, please Soldier On!
Confirmation bias trumps intuition. Furthermore climatologists rarely have good intuition…
1. Is there a reason why only April is considered?
2. You conclude that the longwave and shortwave do not cancel. Can you expand on your conclusion? What effect does the 20 W/m2 cooling have on the CAGW claims, or on the energy balance calculations?
Another nail in the coffin of “Global Warming” as imagined by the Warmistas.
Those on “The Team” should be completely embarrassed by now.
Good work, Willis! A most enjoyable read!
Lance Wallace says:
December 30, 2013 at 10:23 pm
1. Is there a reason why only April is considered?
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Perhaps to rightfully compare the data.
Lance Wallace says:
December 30, 2013 at 10:23 pm
Yes, because that was all that was considered in the original Kiehl paper. When showing that there is an error in a paper, I always strive to match the original analysis, so that people can compare apples to apples.
Perhaps. I think that the cooling caused by the onset of cumulus clouds is one of the ways that the global temperatures are regulated. This regulation occurs because the clouds have a very strong cooling effect, and as the temperatures increase, we get more clouds.
Mmmm … well, insofar as the CAGW claims include the idea that tropical clouds are not part of the regulatory system because of the claimed “cancellation”, it would make them wrong.
The main effect of the Kiehl 1994 paper was to provide a scientific reason for people to ignore clouds, and to ignore folks like myself who think that tropical clouds are a key component of the climate system. People could just say “Well, Kiehl showed that the tropical clouds have no effect on the radiation balance”, and dismiss the fact that the tropical clouds are important.
Regards,
w.
I partularly like the way you linked the (now falsified) Kiehl94 Cancelation Hypothesis to making the Global Climate Models simple enough to be constructed.
If Kiehl94 is falsified, then the GCMs are falsified. Replacement GCM must be made so complex the compute power would need to increase explosively, if chaos doesn’t make them impossible in theory. The computer waste heat alone might be a significant source of Global Warming 😉
Willis
If recall correctly, in your last post you pointed out, giving a variety of reasons, that monthly averages were suspect. Yet the basis of this article is monthly averages.
You might, while you are having a look at the CERES data, have a look at the discrepancy between expectations about the nexus between lightning frequency and increasing temperatures, and what is happening. It seems to me that something is amiss.
Nice work,clouds matter. Does this mean those papers citing this work now get a note attached ?
Danger assumption no longer valid?
Seriously how does that work?
Nice one, Willis! Very well done.
climateace says:
December 30, 2013 at 11:11 pm
Ace, good question. I’d like to have hourly data … and people in the place of eternal perdition would like a cold beer … but what I have are monthly data so that’s what I use. Above, I show both the average of all Aprils in the dataset, and also the individual data for each gridcell for April.
No idea what you’re referring to. I also don’t know where there is a good global database of lightning strikes, but then there’s lots of things I don’t know …
w.
john robertson says:
December 30, 2013 at 11:21 pm
Unfortunately, how that works is generally that it doesn’t work well at all. There is no central repository, no way to get a note attached. It’s one of the big holes in the way we do science.
I’ve often thought that Google or someone like that should set up a database of all the scientific papers and what other papers they link to, and then set up some kind of system to have changes ripple to the affected papers …
w.
Good work Willis.
My first question of seeing their fig 1 is why only one month of one year to draw conclusions that will be applied to models trying to hindcast and forecast CENTURIES of climate. I smell the foul odour of selection bias is lurking nearby.
“Instead of one single month’s data, I’ve used the averages of the ten Aprils in the CERES dataset.”
A fine way to get a broader look than one year spot reading. Now what effect does selecting a different month make?
Is this cherry picking ? Much of this seems to relate to their previous paper from which fig 1 was drawn.
From your fig 4 I would say the cooler end of the range (blue) look fairly linear though certainly not 1:1 ratio. The warmer end looks decidely non-linear , with LW becoming less and less as the magitudes rise.
This would support your claim that the overall feedback effect is non linear.
The hot end of the pacfic data seems to be about the same though it is impossible to tell what the blue dots show in fig 7 since they are pretty much masked by the warm data. Maybe you need to break them out to a separate graph.
Their 1990 paper from which most of the thinking derives.
www-ramanathan.ucsd.edu/files/pr53.pdf
Their 1990 paper is worth reading. It takes a much more thorough look. However, I do not find anything resembling their fig 1 which is claimed to come from that paper.
“Any change in climate which affects the relative distributions of these could types will change the cloud forcing , especially in the tropics.”
I know Eistein said that science lacked common sense BUT really one can feel the difference on a cloudy day.
Sorry , that’s not the right 1990 paper though it includes Ramanathan.
http://onlinelibrary.wiley.com/doi/10.1029/JD095iD08p11679/abstract;jsessionid=B0E0E9942C33A4AE438E70D67588B781.f01t02
“the Indonesian region where deep convection is present, we consider the statistical correlation between the longwave cloud forcing and the shortwave cloud forcing from the ERBE data and CCM. Results indicate a near cancellation between the SWCF and LWCF for these regions, whereas the model predicts a net cooling. Another major area of discrepancy is over the North Atlantic and Pacific oceans where ERBE shows that clouds significantly reduce the solar heating of the oceans. While the model simulates this cooling, the magnitude is underpredicted by more than a factor of 2.”
Since the claimed cancellation (which is actually that near to cancelling) is based on one month of one year. I doubt the objectivity of applying that result for all months. The prev. paper I linked shows huge variation across the year.
Greg says:
December 31, 2013 at 12:17 am
Good question, Greg. As you might imagine, in the course of writing this, I’ve wondered that myself.
Dang, bro’, let me draw breath. I just finished this paper. I thought about putting in a larger analysis of annual data, but then I realized that I should follow the KISS principle and discuss exactly the areas and times that Kiehl discussed. I’ll get to the rest unless someone else gets there before me … or unless my monkey mind finds something else shiny to focus on.
Turns out, April is fairly typical, and adding the other months doesn’t change the results a lot. Here’s a first cut on annual data …
I’ve added this graphic to the head post as well.
Thanks for the reference to that paper, good stuff.
I’ve redone Figure 7 with a finer dot size, so that you can see that the blue is actually clustered on the left side of the graph, and not much of it is masked by the warm data.
Regards,
w.
OK , Willis , I’ve just gone over what you did again and I get the picture of what Kiehl et al are misrepresenting.
The temp variation in the indonesian region is very small but does roughly match the same segment of Pacfic data. It’s somewhat away from 1:1 slope they try to impose on the data, especially at higher power end.
Your figures 7 & 8 show that the disparity is huge at the lower end of the temp range around 25 deg C.
Far from cancellation, the ratio is more like 10:1 , intermediate temps show about 3:1
What would be worth determining is whether these different bands are really just temperature dependant or whether they reflect goegraphical differences.
By dumping everything together as in Figure 5.( Kiehl Figure 2) they are obscuring rather than clarifying the relationship. The spread is not due to noisey data, it is unresolved variables. Temperature and likely geographic distribution.
“I’ve added this graphic to the head post as well.”
Heh, I was wondering how the hell I missed that the first time I read the article 😉 Thanks.
Dear Willis, An excellent analysis of the possible workings of a thermostatic atmosphere. This is a way forward to improve science: repetition of the experiment is the best “peer-review”. Much obliged.
Happy New Year to you all.
I think you have a slip up on the graph labels. Several read cooling in “Wm-1”
Willis,
Your link to the paper, Kiehl 1994, in the intro only links to Fig. 2.
Ah – no it doesn’t. Sorry. WordPress for Android seems to bring up the last link viewed instead of the download, but I get ghe pdf when I view the post through Chrome.
A pictures really are worth a 1,000 words. Thanks Willis.
This seems like great work, but it’s hard to judge without doing a lot of research.
Specifically, it appears that the first Kiehl figure uses several-times-per-day ERBE data, whereas Mr. Eschenbach’s depicts once-per-month CERES data. Do those sources both really purport to measure the intensity of all the radiation flowing both ways while the satellite is positioned at the reported location? Is the downward radiation measured when the upward is? Or are the numbers the result of some long train of mathematical inferences we should question?
(I actually doubt that the Kiehl data are much less inconsistent with that paper’s stated conclusion than Mr. Eschenbach’s data are. But it would help if someone familiar with these data sets could explain what they really represent.)
‘the could simply repeat Kiehl’s claim that the shortwave and longwave cancelled each other out. And as a result, when more detailed data became available, the original claims of Kiehl1994 were never questioned.’
its always a good way to not get an answer you don’t wont, to not ask the question in the first place .
Poor science of course but typical for politics or religion , so you can see why its approach that finds a happy home in climate ‘science’
Willis says
and as the temperatures increase, we get more clouds.
Henry says
careful there, with that statement, especially in the tropics.
you have to qualify it
currently we are cooling
http://www.woodfortrees.org/plot/hadcrut4gl/from:1987/to:2014/plot/hadcrut4gl/from:2002/to:2014/trend/plot/hadcrut3gl/from:1987/to:2014/plot/hadcrut3gl/from:2002/to:2014/trend/plot/rss/from:1987/to:2014/plot/rss/from:2002/to:2014/trend/plot/hadsst2gl/from:1987/to:2014/plot/hadsst2gl/from:2002/to:2014/trend/plot/hadcrut4gl/from:1987/to:2002/trend/plot/hadcrut3gl/from:1987/to:2002/trend/plot/hadsst2gl/from:1987/to:2002/trend/plot/rss/from:1987/to:2002/trend
and we know this cooling is coming from the top (you can see this from the drop in maximum temps.)
As the temperature differential between the poles and equator grows larger due to the cooling from the top, very likely something will also change on earth. Predictably, there would be a small (?) shift of cloud formation and precipitation, more towards the equator, on average. At the equator insolation is 684 W/m2 whereas on average it is 342 W/m2. So, if there are more clouds in and around the equator, this will amplify the cooling effect due to less direct natural insolation of earth
You follow?
So as the temps. decrease you get more clouds in around the equator whereas at the higher latitudes it will become both cooler and drier, at most places (but not all)
Joe Born, the 1994 Kiehle paper Willis links at the top, refers to using ISCCP cloud data that is monthly averages too, I don’t know where you get “several-times-per-day ERBE data” from. They must be using monthly averages by the time they are working in terms of clear-sky minus all-sky data.
CERES is from sun synchronised polar orbital craft, that have about 14 orbits per day IIRC.
What Willis presents here is a pretty cursory view and this sort of question does require more thorough treatment but I don’t see an obvious problem here. Very valid points to raise though.
With all the data and data-gathering expertise you have now, it should be possible to answer the cloud feedback question.
How does LW cloud forcing and SW cloud forcing change with temperature. And these cloud forcings have Anouilh variability that they are the major determinant of the ALL forcing variability.
To add the extra dimension of understanding, this needs to be done versus “Time” as well. Then, in the Pacific, the ENSO will completely dominate how the cloud forcings change, with a lag of 2 to 3 months versus Nino 3.4 and there will be a spatial difference in which Indonesia will be opposite to the International Dateline region.
With all the Ceres data, you will be able to answer these important questions now, unlike Kiehl who has a theory to defend despite the data.
Willis,
Congratulations – looks a pretty impressive result to me which also negates the ‘positive cloud feedback’ stance of the IPCC. The higher SST the stronger the radiative cooling.
Your result also confirms the global result for cloud forcing from CERES. If you take the monthly global average cloud radiative forcing then the net effect is cooling -21 watts/m2 [1]. This means that a change in global cloud cover of ~2% is a larger forcing than CO2.
Furthermore it is simply assumed that changes in cloud cover are a response to warming – a feedback. However suppose that clouds themselves can also be a forcing driven by natural cycles such as AMO/PDO. I strongly suspect this is the real reason for the rapid warming phase observed from 1970 – 2000 followed by the current pause in warming 2000-2030. As a result climate sensitivity is about half that predicted by models.
[1] Richard P. Allan, Combining satellite data and models to estimate cloud radiative effects at the surface and in the atmosphere, RMetS Meteorol. Appl. 18: 324–333, 2011
Willis,
I can see an imbalance there favouring –ve forcing, however with things convective I think of it in terms of what goes up must come down and I’m sure we have all observed that the nature of deep convection (unless associated with high level divergence) – the ITCZ is not – then the two generally balance out. Now clearly, the ITCZ is a fairly tight zone of convergence and one would expect a majority of uplift – but what are the consequence to the N/S extent of the ITCZ when it’s at a max. Is the convergence zone narrowed? – which would result in more SW absorption to the N/S than when convection was weaker.
You see what I mean? There are consequences of deep convection other than just looking at the direct cloud forcing. We have to look at areal extent of the influence of that convection.
Another consideration is in WV transport polewards (N and S) of the ITCZ. This moist outflow is obviously evident as anvil/Ci cloud initially but after evaporation does the heightened upper Tropospheric humidity serve to increase a LW warming effect to the surface, that would have been less in the absence of those deeply active convective episodes (eg monsoon and MJO responses)?
Greg:
Thanks for the response. I inferred the “several-times-per-day ERBE data” from the fact that Fig. 1 has a large (>>30) data points for a single month. From your response, I assume you take those points as being different locations, not different times for the entire region.
Another inference I draw from your response is that, rather than total radiation, they classified location-time combinations as cloud vs clear and used only data from the former: obviously an inexact and somewhat subjective segregation, and one that different platforms are likely to perform differently.
This is not a criticism of Mr. Eschenbach’s work, which I think is great. I guess what I’m saying is that a great many posts don’t really tell us what we’re looking at, even if we survive the barrage of inscrutable acronyms. Particularly for a popular audience, it would be helpful to give the layman a little help.
I hasten to add that Mr. Eschenbach is (in comparison with, say, Steve McIntyre) pretty good at that. But even he could do a better job at telling us what the data actually are. Otherwise, as I just demonstrated, it’s too easy to fill in the blanks incorrectly.
Beautiful Willis! Thank you.
Joe: “Another inference I draw from your response is that, rather than total radiation, they classified location-time combinations as cloud vs clear and used only data from the former:”
I suggest you read the papers rather than let me provide a résumé. “Forcing” is all-sky minus clear-sky, so they use all.
Willis,
Re your over-land analysis, is it possible that cumulus cloud formation over forested land, such as the Amazon basin, is equal or even exceeds that over the ocean? Sunlight on dark green leaves and needles likely elevates the temperature of the leaves and needles more than it does ocean water. So long as the trees can suck up water from the dirt as fast as it evaporates, might cumulus cloud formation be greater? But then, trees may well have a mechanism to limit evaporation in full sunlight and keep it below that of a water body.
You may wish to sort your data into forested vs non-forested land areas to see if there is anything there.
One of the best things about this study, apart from disproving the Climate Model inputs, is that it matches what we all experience, ie it is cooler, sometimes much cooler during the day time when it is cloudy.
Which is something that the original study contradicted.
Really good work Willis, I hope this is the start of something big in changing the current Scientific “Concensus” on how the climate actually works.
A C Osborn says
ie it is cooler, sometimes much cooler during the day time when it is cloudy.
Henry says
Your statement is true, but…
at night time clouds keeps earth warmer, specially in winter.
The GH effect, you know….
It becomes a big puzzle, does it not?
“I hope this is the start of something big in changing the current Scientific “Concensus” on how the climate actually works.”
Yes, I heard there was even some light snow in hell this week too 😉
“Your statement is true, but…at night time clouds keeps earth warmer”
That’s separate to what is shown by the above data but may be important depending upon what conclusion are being drawn. Reducing outgoing can happen 24h / day , whereas cutting down incoming only affects one side of the globe at any time.
Willis says, “As you can see, on average the loss from the reflected sunlight is about 20% or so greater than the gain from increased IR. This means that there is no cancellation. Instead, the clouds have a net cooling effect”
Also please consider that even if they did cancel the affect would still be cooling. The reason is the residence time of the energies involved. The SW from cloud reflection is lost, meaning it no longer goes into the ocean, where it can penetrate up to 800 feet. The SW energy which does shine on the tropics stays within the earth’s land, ocean, earth system for days, months and years. This means todays sunlight in the tropics, is CUMLITIVE to yesterdays, last weeks, last months, and last years.
The LW energy mainly accelerates the evaporation of the surface, and had a much shorter residence time.
David’s Law… “There are only two ways to affect the energy content of a system in a radiative balance. Either change the input, or change the residence time of some aspect of the energy within the system.”
I encourage all to think in terms of residence time of the energies entering our earth, land and atmospheric system.
Looks good, Willis. But in your position I think I’d want to know what of one (or more) of the more reliable big hitters on the mainstream side actually thought of your work. I’d add an abstract and a reference list and ask, say, Dr Ramanathan (a) whether he would be willing to critically read it and suggest changes and improvements and (b) which journal in his opinion might be interested in publishing it.
There’s a lot of people out there who would sit up and take more notice if you had someone like Ramanathan taking your work seriously.
As it is – well, I for one would like to read what Dr R had to say about your findings. He’s no fool and seems honest.
The Guardian reports a new paper claiming a rise of 4 degrees by 2100 due to warming reducing cloud formation and therefore less reflection of solar radiation.
More model based alarmism.
Note, comma added… Also please consider that even if they did cancel, the affect would still be cooling.
Clearly the residence time of energy entering the tropical ocean varies on turbulence, clarity, currents, specific solar WL reaching the surface, which varies with solar cycles, etc. One clear demonstration of this affect is the annual seasons. The TSI is about 7 percent stronger during the SH summer, but the earth’s atmosphere cools down. However this does not mean the earth cools down. Yes, the NH albedo increases, removing energy from the atmosphere. However the SH has vastly more ocean then land, and that TSI enters the oceans, below the surface also removing energy from the atmosphere for a time.
Does the earth (Land, ocean, atmosphere) gain or lose energy during the SH summer, (Plus seven percent increase in TSI) as compared to the NH summer? So far as I know, no one has answered that question, which would appear to inform a fairly basic understanding of our climate. We may know less then we think.
Lovely work Willis.
The net radiative cancellation identified by Kiehl 1994 played a significant role in validating Hadley Centre’s climate model HadGEM1 – which was used in the IPCC Fourth Assessment Report.
In “The Physical Properties of the Atmosphere in the New Hadley Centre Global Environmental Model (HadGEM1). Part I: Model Description and Global Climatology
G. M. MARTIN, M. A. RINGER, V. D. POPE, A. JONES, C. DEARDEN, AND T. J. HINTON”, P1293 (Journal of Climate http://journals.ametsoc.org/doi/pdf/10.1175/JCLI3636.1 ) we have:
“The most marked improvement in the net CRF simulation in HadGAM1 is the representation of the near cancellation of shortwave and longwave CRF over the Tropics, particularly the tropical oceans (Kiehl 1994).”
Note: HadGAM1 is the atmosphere-only version of HadGEM1.
This means if Kiehl 1994 is wrong about CRF cancellation – then a significant error was baked into HadGEM1 and AR4.
I think the Fig. 2 of this paper confirms what Willis has done here.
https://www.eumetsat.int/website/wcm/idc/idcplg?IdcService=GET_FILE&dDocName=PDF_CONF_P41_S5_FUTYAN_V&RevisionSelectionMethod=LatestReleased&Rendition=Web
Willis,
Very interesting. Two comments:
1. Do papers which site the original 1994 paper also site more recent papers on the same subject? Andy Dessler’s papers on net cloud feed back and Troy Masters followup paper questioning the Dessler results seem related to your efforts here.
2. It might be informative to generate a 3-D graph with ocean temperature as the third axis and map a best-fit 2-D surface instead of the individual data points.
Forgot the one critical aspect which invalidates the analysis…
Coldish +1
Not be deterred by reality Grauiad gets 2014 off to a flying start with “at least 4C by 2100″.
http://www.theguardian.com/environment/2013/dec/31/planet-will-warm-4c-2100-climate
What is meant by “near cancellation”?
David A: “The LW energy mainly accelerates the evaporation of the surface, and had a much shorter residence time.”
That is a good point. So it may well get sent straight back up in the form of convection.
There is an unspoken assumption in all this cancellation/ not concellation stuff, that a watt is a watt, irrespective. Whereas a watt in the first 100um is not that same a watt penetrating 30m or more.
Very good point.
HenryP says: December 31, 2013 at 5:46 am “Your statement is true, but…
at night time clouds keeps earth warmer, specially in winter.
The GH effect, you know….”
I agree, which is why I specifically included the “daytime” in what I said, I remember a Summer in the UK in the 70s when we had literally weeks and weeks of cloud cover both day and night, but it didn’t rain.
That was the summer that never was, the temperatures hardly ever got above 70F and it was extremely depressing, it would actually have been better if it had rained occasionally.
George Steiner says: What is meant by “near cancellation”?
It means please ignore the biggest incertainty in climate modelling and go somewhere else so we can pretend it’s all based on fundamental physical laws and not back of envelop guesswork.
It means we have no idea how all this adds up so we’ll _assume_ it does not matter.
It means take one small area of ocean and look at one month of one year then draw conclusions for the whole of climate that we can extrapolate 100 years beyond the data.
Excellent question and analysis Willis!
It is a classic example of why independent replication of research is so critical as the replication confirms or falsifies research.
In the above case, that’s a solid thumping falsification of Kiehl1994.
The issue brought up is that you are almost required to publish.
a) To officially falsify Kiehl1994.
b) To cement your research as a fundamental cornerstone in climatology
c) To force climate research to recognize both their folly and their need to rethink research into forcings.
Publishing in Nature or whatever is not required. Only that you publish, perhaps even a paper here on WUWT, Climate Audit, wherever.
I’m sure that a number of worthwhile peers would sign on as peer reviewers and solidify the paper as ‘peer reviewed science’.
Willis,
Interesting stuff for sure! I think, to better ready this for any kind of peer review/critique, you need to address a few essential points. That being, “why is there such a difference between your results and those in 1985?” As well as introducing new data, as you have, you should also be able to say what may have been wrong with the old data/methodology. Leaving it open to the reader to jump to conclusions will result in a lot of conclusions and questions.
Discussions for example,
Data collection method between Ceres vs. ERBE…
Is there any upgrade to the sensor technology used to collect the data? What are the differences in collection methodology?
The Kiehl Paper looks at cloud top height/ temperature/ high ice albedo ..for example and you don’t discuss or address these implications.
Was the sample year 1985 in any way special?
Perhaps there was better correlation in 1985 – 25 years later, maybe it has changed significantly?
Here you would need to look at the yearly variation in the 10 year data set that you present.
Does any one of them look like 1985? Is there a gradual shift with time? Or was 1985 a high aerosol year? More or less cloud formation due to some other effect? Kiehl looked at variations in cloud height/formation as well as Enso and SST variations that you have not even addressed yet.
Anyway a good start I think, but you need to improve on the old work by saying what they missed , what you found instead and what is still missing in your work. I realise also WUWT is just a forum for shooting out (down?) new ideas and not a Science journal.
Good work, checking the uncheckable.
If one went through AR5, I wonder how many referenced works have been withdrawn or repudiated?
Have you considered colourcoding the temperature data not by temperature, but by date of capture? Where there is a background effect that has a time element, as in the PDO, you will see a shift in data within the dataset reflecting the other parameter.
Climate: Cloud Mixing Means Extra Global Warming
Doubling of atmospheric carbon dioxide points to the higher end of warming estimates.
A decline in ocean cloud cover projected in climate models points to more than 5.6°F (3°C) of global warming coming in this century, on the high end of past global warming estimates, warn climate scientists in a new study. (See also: “Global Warming Effects Map.”)
“This degree of warming would make large swaths of the tropics uninhabitable by humans and cause most forests at low and middle latitudes to change to something else,” says Steven Sherwood of Australia’s University of New South Wales, who led the study.
http://news.nationalgeographic.com/news/2013/12/131231-climate-sensitivity-doubling-carbon-warmer/
April only is enough. The slice of data is essentially the same period of the year for all points and for the time interval it produces over 60,000 points. That is enough for an accurate eyeball analysis of the claim that the shortwave and longwave cancel. Good work Willis.
My goodness, it’s like you are using some crazy method that puts the conclusions of one study to the test to see if the results are reproducible. You should call this the scientific method or something along those lines. Quick, everyone make the climate science community aware of this new technique so that they too can discard their method of ‘accept everything you read if it fits the agenda’. But seriously, good work Willis. CERES data is certainly turning out to be a bane to the CAGW church’s agenda.
Joe Born says:
December 31, 2013 at 4:45 am
Yes, the Kiehl paper identifies them as being the 2.5°x2.5° gridcell averages.
The way it’s done is slightly different than that, but close. They classify the instantaneous data as clear or not clear. The cloudy data is calculated as the difference between the “clear” and “all sky” conditions. And you are right that all such divisions have gray areas.
Joe, I strive to make my work accessible to everyone from the interested layman to the scientific specialist. How much and exactly what information to include are tricky questions. Too much, and the reader bails out. Not enough, and they want a better explanation of e.g. what the data are, or what the logic is, or …
In addition, what seems crystal clear from this side of my eyeballs is far too often somewhat vague or not specific enough when seen through the eyes of others.
In general, if I’m writing about a certain scientific paper, I don’t try to cover everything in the paper. I try to extract the most relevant quotes and data. Then I put in a link to the paper, and leave the interested reader to go as far as they wish.
All the best, and thanks as always for your comments.
w.
Paul Carter says:
December 31, 2013 at 6:33 am
That is an absolutely classic find, many thanks. I’m sure if we looked at the 129 citations of Kiehl’s work, we’d find some other interesting stuff …
w.
Kirk c says:
December 31, 2013 at 9:53 am
Say what? I have no idea what might be gained from that. A scientist is under no obligation to discover and identify the errors of other researchers. It is enough to falsify their results by presenting solid, accurate data and analysis.
In this case, being a suspicious fellow, I suspect that the issue was cherry-picked data. I mean, he used exactly one month’s worth of data to make his far-reaching claims, you do the math …
But how would I prove that, Kirk, and what difference would it possibly make? I neither know nor care what went wrong with their work. As long as nobody can falsify my work, the exact location where they went off of the rails is of little interest other than for the history of science.
w.
Willis,
Perhaps you can clear this up for me, not being an expert……How does your conclusion relate to (or go beyond and/or confirm) the conclusion published by Lindzen and Choi, 2009, “On the Determination of Climate Feedbacks from ERBE data [and CERES data]” (Geophy. Res. Lett. v.36, p. L16705-..)? From Lindzen-Choi abstract: “It appears, for the entire tropics, the observed outgoing radiation fluxes increase with the increase in sea surface tempertures (SSTs). The observed behavior of radiation fluxes implies negative feedback processes associated with the relatively low climate sensitivity…Results also show, the feedback in ERBE is mostly from shortwave radiation while the feedback in the models (GCMs) is mostly from longwave radiation….constituting a very fundamental problem in climate prediction.” They illustrate that the total feedback, LW+SW was less than minus one, which I interpret as more heat is lost than gained, or opposite of climate models.
Did you correct for the known bias in the CERES data? Also, did you filter the data to exclude grid cells dominated by low, non-convective clouds? (Both issues are discussed in the paper below.)
http://journals.ametsoc.org/doi/abs/10.1175/1520-0442%282004%29017%3C3192%3ACRFIPA%3E2.0.CO%3B2
The bias adds about 10 W/m^2 to the cloud forcing, making it appear as though clouds cool more than they do (which would change your figure in yellow above). The failure to filter out low, non-convective clouds would obviously bias your data (because they cool more than they provide LW forcing), and is a first-order QA thing you should do if you are seriously considering submitting this for publication.
Anyway, the Iris Hypothesis is a couple of decades old so this is not really new stuff. Lindzen fought, and lost mostly, this battle a decade ago.
A C Osborn says
http://wattsupwiththat.com/2013/12/30/cancelling-the-tropical-cancellation/#comment-1519009
Henry says
True enough, I found the wave in New & Old England runs opposite of the global sine wave
meaning they get the clouds and the “WARMER” weather during a cooling period and “COOLER” weather during a warming period.
We are currently cooling, so the weather would be depressing.
Just take a holiday to South Africa, where the sun always shines (most of the time)
Wishing you all at WUWT a blessed 2014
“I neither know nor care what went wrong with their work”
Maybe just the 85 La-Nina?
Willis:
Fascinating.
So looking at the data honestly reveals:
1) cooler SST, not much heat goes to space. Fewer clouds more solar heating.
2) warmer SST, a lot more heat goes to space. More clouds, but less solar heating.
Hypothesis
3)Very Hot SST (hypothetically 35C), much more heat goes to space. Much more clouds, difficult for solar heating. Perfect cancellation or better with much warmer SST?
Willis well done a nice bit of work. Your response to Kirk is also spot on but unless you publish in peer-reviewed journals it’s – unfortunately – not going to be taken seriously.
You say peer-review isn’t worth the effort but a letter to the editor (although sometimes there is a cutoff for response) might be. Good stuff all the same.
ed.
You miss the point here. The purpose of the post was to examine the hypothesis which was accepted in the 1994 paper. He has falsified it. And yes Lindzen has showed that the data suggested their was some self-regulation. I don’t think, however, that they had such a weight of data and yes there may be limitations to the approach (as you say cloud types, weather conditions etc.) – but just look at the weight of data. In MHO he’s falsified the arguments put forward in the Kiehl paper. Don’t be so mean spirited.
*sigh* I wish I knew how to read scatterplots. 🙁
Ah Science! It’s beautiful when it’s in action.
Willis Eschenbach says:
December 31, 2013 at 11:27 am
…. Say what? I have no idea what might be gained from that. A scientist is under no obligation to discover and identify the errors of other researchers….
Willis,
Thank you for taking the time to respond. Of course, your data stands on its own and you don’t “have” any obligation to prove them wrong or find fault with their methods…. but it helps a lot.
It demonstrates that you, at least, understand the argument (science) and that you can point at the flaws, because you understand it just a bit better than they did. This is new data you presented nearly 30 years in the future. They did not have this available at the time. “Here is the new proof and this is why”.
I don’t think it’s a question of “why do I need to” .. but rather “will it help my case?”
You are defending your position on a topic. Like a lawyer in a court room , not only be sure to make your own case, but refute the other side at the same time.
Science and data stand on its own. Yes, but you have to tell them all why we should change and the update the science.
In the end you’re trying to sway a jury (of your peers) and simply saying “I’m right. Here are my crime photos – draw your own conclusion”. You must also say “they were mistaken because they lacked “……. ” and here is new evidence and our best guess at the truth”!
It represents at least half your battle. You don’t “have to”… but if you want to win you “have to”.
Science is about proving your data is more correct and all others are misplaced or lacking and why. It’s also about trying to prove your own data is wrong and being unable to.
I wish you a very Happy New Year.
Keep up the good work.
Thank you Willis. Another piece of the puzzle in a very, very unsettled (and unsettling too) thing called “climate science”.
Whole lotta cancellation happening and overdue and in the pipeline. A repository of falsified or challenged papers would be a good start on bringing them to M&S (Movers and Shakers) attention. “Oh, that paper is in the Cancelled Conclusions database”. The Kiss of Death!
The paper Ed cited above indicates that Kiehl was arguing in his 1994 payer that the near cancelation seen in his data was a coincidence, rather than something fundamentally reliable, no?
ed. says:
December 31, 2013 at 12:02 pm
My hypothesis (involving cumulus and cumulonimbus) and this analysis have nothing to do with Lindzen’s Iris hypothesis, so it’s not clear what you mean.
w.
Kirk c says:
December 31, 2013 at 5:48 pm
Thanks, Kirk. I agree with your claim that it in general it “helps a lot”, but you have not replied to the issue I raised. I suspect that the problem is cherry-picked data. How would I prove that? And what good would it do even if I could prove it? How would it help a lot to show that?
w.
Willis, your lucid and relevant work here is a good illustration of my comment on the earlier ‘Peer review/Troll’s last hiding place’ thread, that critical review is perhaps decamping from the established journals to (parts of) the blogosphere. Some people are voting with their feet.
You fall firmly in the ‘could not replicate’ stream of review that should be part of the scientific mainstream. What is missing is a means of referring to your work plus the assembled comment stream as a citation.
It is is a real problem with this medium, and what is needed possibly is a neutral digest which can be lodged in a standardised location. Once anyone sees your work being cited it may give them courage to cite it in turn. At that point, Bingo, you’re in the Canon as you should be rather than an unacknowledged gadfly…
Stuart B
Willis Eschenbach: “Joe, I strive to make my work accessible to everyone from the interested layman to the scientific specialist. How much and exactly what information to include are tricky questions. Too much, and the reader bails out. Not enough, and they want a better explanation of e.g. what the data are, or what the logic is, or …”
Amen.
I may not know much, but I think I do know quite a bit about technical exposition, and in many if not most cases the overwhelming majority of the time I spent on it was dedicated to deciding what to include (usually very little) and what to exclude (usually most of it). The “right” answer, of course, depends on who the audience is. In my case, I usually had a much better idea of who the audience was than you can here, and I still got it wrong lamentably often. So I can’t tell you where to draw the line.
I’m merely giving you one data point: one member of your audience–who, incidentally, is usually too lazy to read the whole referenced paper and gather the relevant data–can often benefit from more data description. This doesn’t necessarily mean that including such a description would be a net benefit. But maybe that data point will help you decide.
Willis, please consider responding to my cogent questions and comments up thread…
David A says:
December 31, 2013 at 5:59 am
and two comments below that .
Thanks in advance
================================================
Thank you Greg, yes, not all watts are equal.
Willis
Great insight. For publicity to get to the 126, try Drudge Report.
Recommend submitting to Retraction Watch under a new category of “Highly Cited but Falsified”.
As always, Willis, your work is a succinct and merciless standard with which to gauge status quo theories. I especially appreciate the scatter plot, which I had never seen used to such persuasive effect until your color-coded plots appeared on the scene. No linear, poly or other interpretive fit needed (the choice of which tends to bend data to one’s own ends). I like the way, over your many articles, how science gets revealed and consolidated – a work in progress approach.
Interestingly, the “near cancellation” theory gets greater support with increasing SST. Perhaps, at the sharp maximum limit of 31C (?) that you earlier revealed, the Kiele94 is “nearer”.
I’m with Bill Illis here. I think you have enough to put the final touches on a solid body of work on the earth’s response to heating.
Bill Illis says:
December 31, 2013 at 4:09 am
“With all the data and data-gathering expertise you have now, it should be possible to answer the cloud feedback question.”
Hi Willis,
Would you be willing to publish the code? And also the data as you have sliced and diced it – although that is probably taken care of in the code. Some people are claiming they don’t have a good enough understanding of what you did to affirm it.
Thanks.
jim2 says:
January 1, 2014 at 8:26 pm
The code is here … but it is a snarled tangle that is not only not user-friendly, it is actively user-aggressive, has the table manners of a wolverine, and needs to be beaten severely about the head and shoulders.
It has been used to produce maybe five posts at this point. Yeah, I know, I should do proper revision control … but I’m only one guy. I need minions. Or at least grad students. I do all of the computer programming, and all of the research, and all of the wandering around the property gazing at the sky and thinking about the theoretical underpinnings of the work. Plus I work at a day job remodeling houses, and I like to spend time with the gorgeous ex-fiancee …
In any case, there it is. As to the data, it’s here in R format, 168 megabytes of the variables as arrays …
Regarding how I analyzed it, well, it’s just a scatterplot. Take the CERES dataset called “cre_sw” (shortwave cloud radiative effect), it’s in the data listed above, and do a scatterplot with the CERES dataset called “cre_lw” (longwave cloud radiative effect) … you’ll need to mask out the areas you’re not interested in.
Regards,
w.
David, you’ve asked above that I comment on your ideas, viz:
David A says:
December 31, 2013 at 5:59 am
I fear that I didn’t answer because you seem to have a bee in your bonnet about something you call “energy residence time”, and I didn’t want to get into another endless discussion.
Short version. Energy residence time doesn’t change the temperature at all. Here’s an example. Suppose we have a brick in a very, very, very well insulated container. We measure the brick’s temperature with the built-in thermistor. 30°C.
Then we come back in an hour, and measure the temperature again. Still 30°C
Then we come back in another hour. All this time the brick is in radiative balance. Still 30°C.
Now, a few questions about the energy that has been residing in the brick for the two hours of the experiment:
• What is the residence time of the energy in the brick in this example?
• How is it calculated?
• How can you tell one bit of energy from another, to decide how long it has been residing in the brick?
• Is the residence time different after an hour elapses?
• Did the residence time change over the period of the experiment?
• What would be a sign that the residence time had changed?
• Suppose we open the door of the container, and the brick starts to cool. Has the energy residence time gone up, down, or stayed the same?
Serious questions …
w.
Willis you’re showing how it can be done, and I appreciate your efforts. I’m further informed, thank you. There is more to the story of atmospheric circulation than meets the eye, isn’t there? Now, how about that magnificent cooling engine, eh?
@ Willis Eschenbach says:
January 1, 2014 at 11:58 pm
I’ve seen worse code and I’m glad you have a life!
Thanks.