Willisgate, Take 2

October 10th, 2013 by Roy W. Spencer, Ph. D.

So, I have apparently opened a can of worms with my challenge to Willis Eschenbach to read up on subjects a little more before he claims to have invented something new. Unfortunately, Willis took my post as a swipe against citizen scientists, which it wasn’t. As I said, I “applaud” people who are willing to get their hands dirty with the data. I also said I consider Willis a very sharp guy, and he is a gifted writer.

Anyone who read my post could see I was not faulting citizen science. (My showing a sharp-looking Homer Simpson as a citizen scientist was insulting, but showing professional scientist Jim Hansen being arrested wasn’t? C’mon.)

For example, Anthony Watts (who is a meteorologist) has forced NOAA to quit pretending it has pristine surface temperature data, and has helped reveal what a mess the surface thermometer adjustments have become (the adjustments are *always* in the direction of more warming? Really? [btw, that’s a criticism of NOAA’s adjustments, which always seem to produce more warming]).

Similarly, Stephen McIntyre (along with Ross McKitrick) have shown that if you analyze tree ring data and statistically *assume* there is a hockey stick shape to the data, you will get a hockey stick shape to the data. (Sorry if I’m oversimplifying what Steve has shown).

Now, both Anthony and Steve have formal technical backgrounds, which I don’t believe is necessary… a person can be self-taught, which is what Willis is. And that’s fine.

But in Willis’ case, as far as I can remember, he has not revealed anything that we did not already know. Now, that’s fine if this is just a part of Willis’ personal education. But over the last several years he has put forth ideas which he has claimed as his own, when in fact they have been previously published…even leading to multi-agency field experiments in the case of “his” Thermostat Hypothesis.

In science, you find out what has been done before, and then you either build upon it or show where it is wrong and do it differently. And, you don’t advance as your theory what has been done before.

Why should I care what Willis does? Because the way he presents his analysis leads readers to assume that what he is presenting is new, when in general it is not. *I* then have to deal with e-mails asking what I think about “his” latest theory. I have to explain to people that either “we already knew that”, or “that data plot doesn’t demonstrate what he claims, and here’s why”.

This is not a new problem, and I have blogged on it before. Now, since I have been talking in generalities, and Willis has asked for specifics, here we go again (I have covered some of this ground before)…

The Thermostat Hypothesis
In 1991, Ramanathan and Collins advanced in Nature their theory of surface temperature regulation by deep moist convection in the tropics. This became known as the “Thermostat Hypothesis”, which led to a field experiment (CEPEX, 1993). Yet, on WUWT, you will find Willis talking about the Thermostat Hypothesis as being ‘his’ theory. For scientists, this would be a major faux pas.

Of course, moist convection on average cools the surface of the Earth…even Kiehl & Trenberth’s much-maligned global energy budget diagram shows this average rate of cooling to be about 100 W/m2, compared to an average solar input of 240 W/m2. But how convection might change in a global warming scenario (i.e., its regulation ability) is a very complex topic, and one which has been studied for decades by hundreds of scientists. Willis is just scratching the surface of a very large body of existing knowledge.

For example, even climate models say convection will increase with warming, leading to even more surface cooling…yet, the models still amplify the warming. This is because surface evaporative heat flux is only one of a myriad influences on the surface energy budget, and therefore on surface temperatures.

Now, the Thermostat Hypothesis might just be a case of Willis not doing his homework first. And if Willis figured out this theory on his own, why should I care?

Because it leads to the mistaken impression among readers (read the comments on WUWT) that professional climate diagnosticians are too stupid to figure out what this citizen scientist has done, when in fact it *has* been done before. This puts me in the VERY unusual position of defending mainstream climate scientists…and, as my readers know, most of the mainstream have disowned me. But I will give credit where credit is due, and mainstream climate scientists have learned (and published) a lot over the years…some of them just have a bad habit of claiming silly things like “proof” and “95% certainty”.

Again, I am not against citizen scientists figuring out something new. But don’t give people the impression that this stuff hasn’t been done before, unless you are familiar with the literature and know that to be the case.

Instead, build upon what we already know. Anybody who can download a dataset and plot graphs in Excel can claim this or that about what the graph means. But it is rare that anyone discovers something new and significant; almost always the data can be explained based upon what we already know.

Some think I am being harsh in my criticism. But After years of having to answer questions about Willis’ latest ideas, I frankly don’t know what else to do. I have previously tried to keep it low key.

I don’t want to dissuade Willis from contributing to the science. But contributing to the science requires more due diligence than plotting graphs and leaving readers with the impression that the graphs show something new or unexplained (in this case, demonstrating the differing water vapor greenhouse effect between tropics and high latitudes, and the fact that the tropics export heat to the high latitudes).

Willis, you write really well, you are a smart guy, and you can make complex subjects more understandable. Don’t dilute those talents by leaving readers with the wrong impression.

239 Responses to “Willisgate, Take 2”

Toggle Trackbacks

  1. Dr. Roy, having reread all the comments, let me re-open the discussion to make my position clear.

    My hypothesis is that a variety of emergent phenomena, from thunderstorms to dust devils to the La Nina pump pushing warm water to the poles, regulate the temperature of the earth.

    Regarding thunderstorms, my main insight was that it is the timing of the daily emergence of the thunderstorms that is the main regulator of the temperature.

    I find Dr. Roy’s claim that this is long-known to be curious, in that he hasn’t provided a single link to back up his claim. Not one.

    Now, I like and respect Dr. Roy, he’s one of my scientific heroes. But in this case, he’s way off the rails. My theory is NOTHING like that of Ramanathan, who Dr. Roy claims preceded me. I find NOTHING in Ramanathan’s work about either emergent phenomena in general or more particularly the timing of their emergence. Instead, he talks of something he calls a “super greenhouse effect”, something I don’t discuss at all.

    Here’s Ramanathan’s hypothesis, from their paper “Thermodynamic regulation of ocean warming by cirrus clouds deduced from observations of the 1987 El Nio”

    “Observations made during the 1987 El Nio show that in the upper range of sea surface temperatures, the greenhouse effect increases with surface temperature at a rate which exceeds the rate at which radiation is being emitted from the surface. In response to this ‘super greenhouse effect’, highly reflective cirrus clouds are produced which act like a thermostat shielding the ocean from solar radiation. The regulatory effect of these cirrus clouds may limit sea surface temperatures to less than 305 K.”

    I said NOTHING about cirrus clouds, or about the rate of increase of the greenhouse effect with temperature. Not one word.

    So Dr. Roy, on what planet is that Ramanathan paper about cirrus clouds somehow prior art to my hypothesis about the timing of emergence of thunderstorms, in which cirrus clouds and a “super greenhouse effect” are not even mentioned???

    So, Dr. Roy, the ball’s in your court. Please either link to someone discussing the thermoregulatory effect of the timing of the daily emergence of the cumulus field and thunderstorms, or admit that you accused me unfairly.

    Dr. Roy, I ask because I can’t overestimate the damage that your unpleasant claim has done to my reputation. To this day, almost ten years after you posted this, people STILL say something like “Willis, you can’t be any good, Dr. Roy said you’re a plagiarist and he’s a REAL scientist, not a citizen scientist like you.”

    And yes, I understand that if you parse what you wrote very, very carefully with the most favorable state of mind, you didn’t directly accuse me of plagiarism but it sure reads that way to far too many people who read it.

    As I said above, and you never responded to:

    “Dr. Roy, I have to assume that you dont understand the implications of what you are writing. You said:

    ‘Ive previously commented on Willis thermostat hypothesis of climate system regulation, which Willis never mentioned was originally put forth by Ramanathan and Collins in a 1991 Nature article.’

    Note that you didnt say I didnt know about R&C1991. You said I never mentioned it. This assumes that a) I knew about it and further assumes that b) I was concealing it.

    I was not the only one who noticed this. Several other people commented that they found your accusation of plagiarism out of line. Thats a very reasonable reading of your words.”

    Getting accused of plagiarism, or alternately of not doing my homework, is getting old. I greatly dislike putting you on the spot, and I say this in all friendshipit’s long past time to pull out your claimed citations to prior art about the timing of of the daily emergence of thunderstorms, or publicly state that you were wrong.

    I asked you for such links upthread nearly a decade ago. You ignored it. Almost ten years gone and I’m STILL waiting for your links.


    • Roy W Spencer says:


      I re-read your original post, and my response here, and I don’t know that I would change anything.

      Let me quote from your original article:

      “I propose that two interrelated but separate mechanisms act directly to regulate the earths temperature tropical cumulus and cumulonimbus clouds. Cumulus clouds are the thermally-driven fluffy cotton ball clouds that abound near the surface on warm afternoons. Cumulonimbus clouds are thunderstorm clouds, which start life as simple cumulus clouds. Both types of clouds are part of the throttle control, reducing incoming energy. In addition, the cumulonimbus clouds are active refrigeration-cycle heat engines, which provide the necessary overshoot to act as a governor on the system.

      A pleasant thought experiment shows how this cloud governor works. Its called A Day In the Tropics.

      I live in the deep, moist tropics, at 9S, with a view of the South Pacific Ocean from my windows. Heres what a typical day looks like. In fact, its a typical summer day everywhere in the Tropics. The weather report goes like this:

      Clear and calm at dawn. Light morning winds, clouding up towards noon. In the afternoon, increasing clouds and wind with a chance of showers and thundershowers as the storms develop. Clearing around or after sunset, with an occasional thunderstorm after dark. Progressive clearing until dawn.

      Thats the most common daily cycle of tropical weather, common enough to be a clich around the world.”

      -end quote-

      When I originally read that, I was taken aback, because (1) the convective cloud “Thermostat Hypothesis” had already been advanced and debated, and (2) the diurnal cycle is NOT evidence for negative feedback. (Climate models with strong positive cloud feedbacks also have diurnal cycles).

      Of course clouds and convection help keep the Earth’s surface cooler than if such cloud systems did not exist. Everyone knows that. So, when you say they “regulate” Earth’s temperature, let’s be sure to make it clear you mean regulate (in time) over the long term in terms of feedbacks upon any imposed change in the Earth’s radiative energy budget. Otherwise, you aren’t proposing anything new.

      This then implies you are talking about clouds and precip systems as negative feedbacks. You use the diurnal cycle in tropical convection as an example, but it isn’t. First of all, the diurnal peak in oceanic convection occurs at night, not during the daytime. Only near islands (or over the continents) does it shift to the afternoon, which is a local sea breeze effect around islands. If you would like references about the nighttime maximum in maritime convection, I can provide those, too. It’s caused by solar absorption during the daytime by the upper troposphere, which raises the temperature there more than the ocean surface warms (due to its larger heat capacity), which then stabilizes the atmosphere slightly during the afternoon. At night, the troposphere radiative cools, the ocean surface remains warm, and convection increases. Your experience to the contrary must be an island effect.

      But whether the diurnal maximum occurs during the day or night, it doesn’t matter. It’s not evidence of negative feedback on *change* in the climate system. Your “Thermostat Hypothesis” about tropical clouds (no matter whether a “super-greenhouse” effect is or isn’t involved… I think Ramanathan created that as a gimmick to get buzz for his work) has been debated well before you brought it up. Climate models that produce large amounts of warming also have a diurnal cycle.

      If you really cannot find the “Thermostat Hypothesis” papers on your own, I will dig the references out for you. But instead, you introduced a concept as something “new” that really wasn’t new. And as a result, I had to field numerous questions about it.

      The bottom line is you were proposing a convective cloud “Thermostat Hypothesis” and using the tropical diurnal cycle in convective activity as a way to support a negative feedback argument. It doesn’t. In fact, no feature of moist convection (diurnal cycles, warm pool physics, etc.) provides evidence of negative feedback. I’m not saying there can’t be a convective cloud thermostat, just that (1) the evidence you presented did not support it, and (2) lots of work had already been done on cloud feedbacks, and smarter people than you or me have been unable to find clear support for such a stabilization effect.

      If you really want to revisit this issue, write something up and I will post it on my blog.

  2. Dang, my friend, miss the point much? Let me make it real simple.

    You accused me of plagiarism, saying I “didn’t mention” what you claimed was the prior art of Ramanathan.

    But there was no reason for me to mention it. Ramanathan said that there is a “supergreenhouse effect” that works by changing the amount of cirrus clouds.

    My hypothesis, whether right or wrong, did not mention either one of those.

    So just how in the hell am I plagiarizing Ramanathan?

    And please, don’t say you didn’t accuse me of plagiarism. DOZENS of people, including people in this thread, have agreed that they read your words that way. I sure did.

    And I get accused of plagiarism to this very day based on your false accusation.

    Now, you might not have meant it that way. Fine.

    But you sure as hell have NEVER admitted that. You’ve just let it stand, and now you want to pretend the issue is whether the peak rainfall is at night or during the day??

    I implore you, either support your damn ugly claim of plagiarism or make it clear that you didn’t mean that. My hypothesis had and has NOTHING TO DO with Ramanathah’s hypothesis.

    I ask you to clarify this for a simple reasonbecause I’ve spent almost a decade defending myself from people who claim you did accuse me of plagiarism.