Data Comparisons

Alarmists support Genocide!

105 Comments

Brandon has closed his survey and beginning to apply “Lewandowki” methods to the data. It appears ‘the data show’ that those who think “global warming is real and also a threat” (aka “alarmists”) tend to support genocide. Or at least they do if we use the sort of data interpretation Lewandowki uses in his papers. Also, it turns out that the same group of people tend to believe they are never wrong; once again: this is using the sort of data interpretation Lewandowki uses More here . Note that the interpretive method is to decree that “belief” X predicts a tendency to believe Y if a positive correlation exists between X and Y and to state this as “those who believe X tend to believe Y” even if nearly everyone who believes X thinks Y is untrue.

105 thoughts on “Alarmists support Genocide!”

  1. I think that description of of the method is correct. I can’t swear to it though. That sentence keeps making my head swim.

  3. Ah, yup. That’s pretty much right. The methodology doesn’t require a positive correlation though. You can do the same thing with a negative correlation just by flipping what Y you say X will predict.

  5. Brandon–
    Sure. But I think they way I word it it’s a positive correlation. The negative correlation between X and Y ‘lets’ one say “those who believe X tend to believe notY”.

    The main difficulty with Lewandowski’s claims is that the math doesn’t map into what concepts the words convey. Obviously, if only 1 person in the entire study believed “Y”, a correlation will exist. And possibly that might, under some math be “significant”. But clearly, X doesn’t predict much of anything. Possibly those who believe Y are more likely to believe X– but even that can’t be said because you didn’t get enough who believed Y to get a valid sample of the beliefs held by those who believe Y!

    (Though I would suggest that to some extent “significance” is incorrectly decreed whenever the number who believe Y is 1 or even very few. )

  6. It’s worse than that. You can have a “statistically significant” correlation with zero people who believe Y. That’s what got me interested in this issue. I was filtering out data from Lewandowsky’s first survey based on various criteria to see what would happen. Out of curiosity, I decided to filter out all people who believed in a conspiracy. I found there was still a statistically significant correlation between skepticism and believing in the conspiracy. It turns out the methodology lets you conclude a group of people believes Y even if your data doesn’t show anybody believes Y.

    Statistical significance via this methodology borders on meaningless. Calculating a correlation is basically doing a linear regression on the data. If you have no data beyond a certain point (say nobody picked 4 or 5), the linear regression will extrapolate results to fill it in the parts that are missing. That extrapolation has no basis.

    The fundamental problem with this methodology is it assumes two data sets have a linear relationship. If they don’t, any results you get can be borked. Think about all the problems you can imagine with doing linear regressions. They pretty much all could apply here. Linear regressions can be valuable tools, but you have to establish they’re appropriate for the data you’re using.

    Lewandowsky et al don’t do this. They don’t even try. They just wave their hands and pretend their descriptions prove calculating a linear regression on their data set makes sense.

  7. Brandon, Lucia,

    In your opinions, is it at all likely that this sort of mistake (constructing a study this way to draw Lew-ish conclusions) could be made honestly, or is it reasonable to think that anyone doing this must know they’re foisting off garbage on those reading the study?

  8. In re-reading my question I see that it might seem silly, but I mean it seriously. I’m told for example that some ‘psychics’ don’t honestly believe they are frauds, for example, they honestly convince themselves they are psychic and are truly mystified when their ‘powers’ fail a controlled test. What I’m trying to figure out is if this is like that; could somebody do a test like this and honestly believe they’ve setup a real test, or does the setup of the test demonstrate that they must know they’re being dishonest?

  9. Oh, it can definitely happen inadvertently. Abusing correlations like this isn’t limited to Lewandowsky et al. A lot of people don’t understand what the math means. They have a vague idea of what the process means, they use it and they reach conclusions they think are right.

    What bothers me is there are people who do understand what the math means but still abuse this methodology. Michael Wood is an example. I discussed my correspondence with him here. Read this remark:

    Whether it continues as a linear relationship to the top of the scale or, counterintuitively, the shape of the relationship changes at some point, is something that would be best answered with a more conspiracy-minded sample, and unfortunately those are quite hard to come by.

    He recognizes the nature of the math. He accepts its possible the linear relationship he found for his data wouldn’t hold if extrapolated out. Despite this, he acts like it doesn’t matter.

    That’s obscene.

  10. Mark Bofill

    In your opinions, is it at all likely that this sort of mistake (constructing a study this way to draw Lew-ish conclusions) could be made honestly,

    By somebody, somewhere? Sure. I’ve known lots of stupid people. Some have advanced degrees.

  11. Brandon

    is something that would be best answered with a more conspiracy-minded sample, and unfortunately those are quite hard to come by.

    Heck, he even seems to understand that finding a sample with sufficient conspiracies would be difficult. Of course it is: Because in reality, skeptics are not conspiracy theorists. After all: there are lots of skeptics. If they were conspiracy theorists, it would be easy go round up a conspiracy-minded sample because conspiracy theorists would be easy to come by!

  12. Michael Wood didn’t claim skeptics are conspiracy theorists. His paper wasn’t about views on global warming. His paper claimed to prove conspiracy theorists believe in multiple, often contradictory, conspiracies. I became interested in it because Lewandowsky cited its results, and they were so striking. (Unbeknownst to me at the time, Steve McIntyre had the same reaction and also wound up contacting Wood.)

    Wood’s methodology is arguably worse than Lewandowsky’s. Wood asked a bunch of people if they believed in various conspiracies. Of course, most of them said no for many or all conspiracies. That produced a positive correlation between them. Wood cited that as proving conspiracy theorists believe multiple, contradictory conspiracies. It’s basically the same as Lewandowsky’s methodology, just dumbed down. It’s even easier to spot why his doesn’t make sense.

    For example, imagine if I went around asking people: 1) Are you Christian; 2) Are you Buddhist; 3) Do you believe in the Flying Spaghetti Monster? Naturally, people would say no to two or all three. Applying Wood’s methodology, that’d mean I just proved believers in the Flying Spaghetti Monster tend to be Buddhist Christians. It’s even more fun to do with sport teams. We could conclude people are die hard fans of 12 rival teams.

    My mind breaks a little every time I think about the fact people get payed money to write papers like that.

  13. Brandon–
    Ahh… Ok. I remember. So, that let you do belief in the conspiracy on a 1-5(?7) scale… right?

    So, say you were agnostic on lots of stuff. For example you didn’t follow the lady di story for reasons ranging from “not born at the time”, “didn’t care”, “were working on your phd and didn’t read paper at time” or so on, and you decided you were uninformed about her death, on a scale of 1-5 with 1 “sure conspiracy theory not true” and “5” sure it’s true, you answered “3” corresponding in your mind to “Have no idea”, you might answer:

    For :
    1) Queen acting alone sent out hit man: 3.
    2) UK government acting w/o Queen authorization sent out hit man: 3.

    This would then be deemed “simultaneously believing in mutually exclusive conspiracies” when what it really meant was “doesn’t know and knows he has now knowledge”.

    Absent data or information, it is perfectly reasonable to fail to exclude hypotheses even if hypotheses that are kept are mutually exclusive. This happens all the time. Someone could say, “Lucia has one or two cats. I’m not sure which.” Of course I either have one or I have two. That’s two hypotheses and they are mutually exclusive. You just don’t know which is true.

  15. lucia, yup.

    The most troubling part of all this is a person can usually predict what the results will be before anyone takes the survey. I predicted most of the correlations I found with my survey. Heck, I picked the questions I picked in order to get many of them. I knew what results I wanted, and I chose questions designed to get them.

    It’s bad a methodology with practically no validity is approved of. It’s mind-boggling a methodology that allows one to pick his or her conclusions on a whim is approved of.

  16. Brandon, your survey makes the general point of how easy it is to produce some rather vague questions that lead to whatever conclusion you might seek – and unfortunately it is done all the time. I think where the discussion got lost on Lewandosky was that it too quickly got into methodology arguments when it was rather obvious that the premise was wrong and silly. I think luke warmers and skeptics waste way too much space discussing details in these cases and in the process appearing to except the basic premise.

    While I am criticizing my fellow posters at these blogs and on the same subject, I was rather surprised how easy Richard Tol was left off the hook at these blogs on his badly flawed analyses of the 97% survey. Again, while he was critical of the survey, he got into the details of the methodology and got it wrong and appeared to miss the fact that the basic premise was crap. Some on the other side like to attempt to turn these wrong headed analyses of the details into evidence that the basic premise is correct. M. Mann comes to mind.

  17. “Lucia has one or two cats. I’m not sure which.”

    That’s nothing, Schrodinger has one or no cats, and cannot know how many he has.

  18. Kenneth Fristch, I was surprised by that too. I think you were the only person other than me to criticize Tol on his initial mistakes. Not only did most people say nothing, people (like Carrick) spoke up to support him in his mistakes.

    If I didn’t know better, I’d think Tol screwed things up on purpose to create an easy target to knock down. Cook et al were able to respond to Tol and act like that responded to their critics as a whole. One of my favorite quotes deals with this:

    The most perfidious way of harming a cause consists of defending it deliberately with faulty arguments.
    -Friedrich Nietzsche

    I’m just glad you were participating in the discussion with him on this site. I think you gave him a lot more slack than he deserved, but you actually looked at his argument. I think you were about the only one.

  19. I know what you mean lucia, but:

    Someone could say, “Lucia has one or two cats. I’m not sure which.” Of course I either have one or I have two. That’s two hypotheses and they are mutually exclusive. You just don’t know which is true.

    Are you sure you don’t have three cats?

  21. I thought that Schrodinger had two cats, one alive and one dead. His problem was trying to sort out which was which as they both appeared to be there at the same time.

  22. Kenneth Fritsch:

    While I am criticizing my fellow posters at these blogs and on the same subject, I was rather surprised how easy Richard Tol was left off the hook at these blogs on his badly flawed analyses of the 97% survey.

    None of us do things perfectly, but “badly flawed” is not a characterization I would agree with. Weakly supported is probably more accurate.

    There were flawed assumptions in your criticisms that people pointed out to you. Given your inability to concede even simple well-documented errors, I eventually just gave up.

    But continue with your superiority dance if you must. It makes you look like a prick, but go for it.

    Brandon:

    Not only did most people say nothing, people (like Carrick) spoke up to support him in his mistakes.

    That’s simply not an accurate statement. When I tried to have a rational discourse with you, I was met with childish name calling, attempts to divert the conversation and even intellectually dishonest arguments.

    I’ll point people
    to this link and let them decide.

  23. Carrick (Comment #122654)
    January 16th, 2014 at 10:21 am

    “There were flawed assumptions in your criticisms that people pointed out to you. Given your inability to concede even simple well-documented errors, I eventually just gave up.”

    I totally disagree that you provided well-documented errors. In fact I gave up on your providing documentation. It involved the validity of using a chi square test on a distribution that was not normal.

    My larger point here is that when you have posters with good statistical and mathematical skills sometimes the tendency is to get into the nitty gritty of the methodology used in a flawed paper and ignore or give short shrift to the validity of the basic premise of the authors. I see this quite frequently with criticisms of papers dealing with temperature reconstructions that deal with methodology and tend to ignore the more basic problem of selecting temperature proxies after the fact.

  24. Kenneth Fritsch:

    I totally disagree that you provided well-documented errors. In fact I gave up on your providing documentation.

    I said it is well documented, not that I wrote primer on it. I gave you plenty to start from… Just a statement that “hey you need to look at this!” should have been enough. We are adults here after all and this isn’t debate class. You should know how to perform research to address issues that are raised, without having to have everything spoon fed to you.

    My larger point here is that when you have posters with good statistical and mathematical skills sometimes the tendency is to get into the nitty gritty of the methodology used in a flawed paper and ignore or give short shrift to the validity of the basic premise of the authors.

    And that is the fundamental problem here. You seem to view the fundamental role of the analyst as “nit picky”.

    Computing a statistic is frankly a rather pedestrian thing to do, especially given how many packages there are out there to hold our hands for us. It’s in the interpretation that you as an analyst are providing added value.

    That’s where I had problems with what Shub did. It may seem like a minor thing to say that the p-value associated with Cohen’s kappa is not diagnostic in rejecting a hypothesis, but at a fundamental level, that is everything. Knowing what the p-value really means and what the limits of it is, especially when your statistical model isn’t a good descriptor of the data you are handling.

    In any case, we have spent a lot of time looking at the issues with using correlation to select proxies here, so I’m not sure why you brought that up. It’s natural that we would be interested in other points than ones that directly affect the validity of a paper, since most of us are constantly looking for ways to improve our art.

  25. Carrick, why would you link to an unrelated discussion rather than the discussion being referred to? Before we divert ourselves, let’s try to look at what was actually brought up. Your first response to my criticism of what Richard Tol said about Cook et al was here. It said nothing more than:

    Brandon, seriously, you’re obviously wrong on this one.

    I’m grappling with how you don’t see it.

    Your next response consisted of you saying nothing relevant while misrepresenting my criticism of Tol:

    Anyway, if you’re going to claim the data as presented to the rankers was heteroscedastic, that’s yet another flaw in the study…. but color me unconvinced on that. It looks to me like Cook et al got that one right.

    The reality is I had never disputed the raters were presented the abstracts in a random order. I pointed out you were responding to a strawman, and you began overreacting:

    Strawman arguments? Wtf.

    That’s not exactly a charitable way of reading this disagreement: You are accusing me now of an intellectually dishonest argument because I don’t understand what you are trying to say, partly because you seem to be moving the goalpost all over the playing field.

    I pointed out I didn’t believe saying someone was arguing against straw man arguments implies intellectual dishonesty on their part as it can happen by mistake. Your next comment had you storm off with this commentary:

    Brandon, I’m done on this one.

    You have this pattern of becoming very hostile when anybody challenges anything you say, and you started out needlessly bellicose to begin with.

    You absolutely make no sense to me, but I find I don’t care.

    Laterz.

    The reality is you were completely wrong on the technical issues. From your writings, it seems apparent you didn’t understand the argument that was taking place. My points were perfectly correct, a fact Richard Tol tacitly acknowledged by quietly dropping the argument I criticized – it vanished from his writings somewhere around his third draft.

    Beyond that, I said nothing remotely hostile to you. I was not mean or rude in any way. This is despite the fact you entered the discussion by simply waving your hands at me and insisting I was wrong with the very bellicose behavior you accused me of. You then projected your own behavior onto my neutrally worded comments to paint me in a negative light, refused to address my defense of myself and stormed off.

    Perhaps I behaved poorly in some other discussion months later on a different site. However, what I said about the discussion I referred to was completely accurate. Moreover, in that discussion, you behaved in the practically the same way you accuse me of having behaved elsewhere.

    Incidentally, you shouldn’t accuse me of “childish name calling” when I’ve never called you any name. For all your complaints about me in the discussion you linked to, you never accused me of having done so. You’d be hard pressed to find an example of me having resorted to “childish name calling” anywhere.

    If you’re going to blame the failure of “rationale discourse” on someone, you shouldn’t say things about their behavior which are completely untrue.

  26. Maybe we don’t need to argue about who more correctly evaluated Richard Tol’s stuff and stick to more current stuff like Brandon’s survey?

  27. The specifics of my discussion with Richard Tol involved the use of the Lung-Box test for autocorrelation of a time series. He claimed that it was not a valid test because the distribution was not normal. I respectfully disagreed. The Lung-Box test uses a Chi Square test and further Chi Square tests do not require a normal distribution. The larger problem with his analysis of the 97% paper was that he was using circular arguments.

    http://en.wikipedia.org/wiki/Ljung%E2%80%93Box_test

    I have no problem with posters exercising their technical skills in analyzing a paper. I have a problem when they fail to acknowledge (or perhaps realize) that the basic premise of the paper is wrong.

  28. lucia, agreed. I shouldn’t have even written that last comment. I was just annoyed. I don’t like being falsely accused of using “childish name calling” and “intellectually dishonest arguments.”

    For something more topical, yesterday Greg Laden responded to me on Twitter about me linking belief in global warming with supporting genocide:

    .@Corpus_no_Logos @wattsupwiththat @STWorg @tan123 It is an asinine and offensive thing to say. You should be ashamed of yourself.

    Only today he said:

    .@Corpus_no_Logos @wattsupwiththat @STWorg @tan123 Well, to date I’ve seen no such study. Just your tweets.

    I responded by pointing out links to the discussions were in my tweets, then:

    @gregladen It’s cheeky to demand someone apologize for what they say without even reading it. @wattsupwiththat @STWorg @tan123

    I’m not sure he’s ever going to catch onto the fact I was being facetious when I associated the two. I think there’s going to be a hard time getting people on his “side” to even look at the results, much less think about or discuss them.

    By the way, can you embed tweets here? I would have just dropped links, but I’m not sure if that works here.

  30. I guess tweets won’t automatically embed here.

    I was going to edit that last comment to reflect this, but the edit feature isn’t working for me. The box which pops up doesn’t have a text box or buttons. All it has is an X in the upper right corner and a weirdly aligned “close” by it.

    Is anyone else having trouble with it?

  32. lucia:

    Maybe we don’t need to argue about who more correctly evaluated Richard Tol’s stuff and stick to more current stuff like Brandon’s survey?

    Since it was inappropriate to drag me into the middle of this to start with, that’s fine with me.

  33. In case anyone was interested in the above exchange, I wrote up an explanation of the disagreement about Richard Tol’s argument. I don’t want to pollute this thread further so I posted it here. If anyone thinks I’m wrong, they’re welcome to post there explaining why.

  36. Eli Rabett is using the wrong animal for his alter ego. He’s more like a pigeon: it flies in, sh1ts on your head and flies away again.

  37. Does anyone know what Eli Rabett means by that? What is soon to be published?

  39. Is saying that ‘skeptics’ are sustained by those who create, publish and promote a belief in a deliberate ignorance? Does he believes you to be one or both types of skeptic? Is he taking a jab at the “soft” sciences? Only Eli knows, and that is just the way he wants it.

    Eli seems to delight in what he considers ignorance in others and takes great pains to avoid being plainly understood. Hopefully, he behaves better in the classroom and laboratory.

  40. [ Lucia: Edit / Delete do not seem to function properly ]

    Is he saying that ‘skeptics’ are sustained by those who create, publish and promote a belief in a deliberate ignorance? Does he believe you to be one or both types of skeptic? Is he taking a jab at the “soft” sciences? Only Eli knows and that is just the way he wants it.

    Eli seems to delight in what he considers ignorance in others and takes great pains to avoid being plainly understood. Hopefully, he behaves better in the classroom and laboratory.

  41. One can always look him up on rate my professor (I don’t think the reviews are very good but I haven’t checked in a while).

    But Joshy is nothing if not great at being offputting and uninteresting.

  43. Hey Lucia it seems like a while since we’ve had a proper sciencey posting. you know with graphs that show how bad models are and such like. Can we have one? i’m getting withdraw symptoms.

  45. Re: HR (Jan 18 19:53),

    If you’re starved for sciency things about the badness of models and our lack of understanding how the climate actually works, Science of Doom has a multi-part (up to twelve now) series on glacial/interglacial transitions.

    Here’s a quote from a paper reviewed in Part Ten:

    Milankovitch (1941) postulated that the driver for this cooling is the orbitally induced reduction in Northern Hemisphere summertime insolation and the subsequent increase of perennial snow cover. The increased perennial snow cover and its positive albedo feedback are, of course, only precursors to ice sheet growth. The GCMs failure to recreate glacial inception, which indicates a failure of either the GCMs or of Milankovitch’s hypothesis.

    Of course, if the hypothesis would be the culprit, one would have to wonder if climate is sufficiently understood to assemble a GCM in the first place. Either way, it appears that reproducing the observed glacial–interglacial changes in ice volume and temperature represents a good test bed for evaluating the fidelity of some key model feedbacks relevant to climate projections.

    [my emphasis]

  46. “lucia (Comment #122649)
    January 16th, 2014 at 7:40 am

    Brandon,
    Currently, I have two cats.”

    I think you’ll find that currently, two cats have you.

  47. Well Chris Colose is doing part 1, of a two part series, in which he shows how people can create really stupid models by treating oscillating influxes interacting with a complex dynamic system as a flat, average, equilibrium state.

    http://climatephys.org/2012/07/31/the-water-vapor-feedback-and-runaway-greenhouse/

    In part two he is going to show that everything written in Part 1 is rendered stupid by the fact that the Earth is a rotating sphere and that ‘average’ fluxes are a non-thermodynamic description of energy. It is worth are read, as it is not often you get to see such a masterclass of stupidiy, but Chris Colose manages to do it with out letting the reader know that everything presented is wrong, including the word ‘the’.

  48. Doc, I think he may still be a student, and thus probably hasn’t reached peak-arrogance yet. I hope none of his academic committee go on-line to ask him questions anonymously.

  49. I would love to see the ‘climate scientists’ model something like the heart or lung, or even oxygen transport in mammals. Average all the fluxes, define the O2 levels independent of blood flow rate and add finite and infinite sinks.

  51. Earlier I was reflecting on the fate of Holofernes and the escape of Judith; patterns can be very deceptive.

  53. Doc, O/T, but I recall a thread where I think you mentioned cardiogram patterns/traces, and the difficulty of using them to accurately make diagnoses (by electronic or human means). Do you have some leading references I could read? I have a slight quasi-professional interest.

  55. Michael, there is a huge literature on ECG interpretation in non-invasive cardiology. It is way out of my area I just talk to the people on the third floor. This is a website of Frank G. Yanowitz, of Utah
    http://ecg.utah.edu/
    He give a nice overview of the general pattern recognition that underlays diagnosis. Have a read of the ‘simple’ cheatsheet:-
    http://ecg.utah.edu/pdf/
    Some of the cardiologists, with 30 years or so under their belt, have a quick glance and know.
    This is a recent editorial in Circulation: Arrhythmia and Electrophysiology. (2013; 6: 2-4)
    http://circep.ahajournals.org/content/6/1/2.short
    ‘Computerized Interpretation of ECGs Supplement Not a Substitute’
    I quote
    “A large international study compared the performance of 9 ECG-C programs with that of cardiologists in interpreting ECGs in clinically validated cases of various cardiac disorders. The percentage of ECGs correctly classified by the ECG-C was lower than that for the cardiologists. The median sensitivity of the computer programs was also significantly lower than that of the cardiologists in diagnosing left ventricular hypertrophy, right ventricular hypertrophy, anterior myocardial infarction, and inferior myocardial infarction.”

    It is a bit humbling, when you consider how much we know about hearts, that people do better than software.

  56. Colose’s article seemed pretty reasonable to me (though I’ve not read it in depth). I’m guessing from his comment DocMartyn never read the post, or if he did, did not understand it.

    I think the fundamental premise is “what would be needed to force our Earth into a runaway greenhouse state”, not “what are the likely consequences of human generated warming if allowed to continue unchecked”.

    It actually provides a rebuke to alarmists:

    ome claims have surfaced (e.g., by NASA’s Jim Hansen) that a runaway greenhouse is possible if we burned all the CO2. Unfortunately, there is no evidence in the planetary science literature to support the claim, and it can be dismissed based on the fairly trivial fact that the amount of sunlight that Earth absorbs does not even come close to the limiting OLR values typically found in the literature (usually > 300 W/m2).

    Hansen should have known better, IMO. Colose obviously does.

  57. “…if we burned all the CO2.”
    It’s pedantic, I suppose, but the equating of CO2 with “carbon” in all the political talk — e.g. it’s not a CO2 tax, it’s a carbon tax; nobody talks of an economy’s “CO2 intensity”, it’s carbon intensity — was bound to end up with someone using CO2 when they mean carbon. Precision in language is an asset.

  58. Hi Carrick,

    I agree, the Colose post is “ho-hum”, and not terribly alarming. It just lays out the standard climate science POV on water vapor feed-back. He appears to avoid the controversy of the ‘missing hotspot’, sticks to the modeled (theoretical) atmospheric profiles, and does not mention the uncertainty of how atmospheric moisture content influences cloud albedo and overall energy balance. It reads to me like a Wednesday afternoon seminar put on by a PhD candidate (designed to show his adviser he has been doing his homework 😉 ).

  59. Carrick, if you live on flat world, where the influx of energy is constant, rather than strobing, then Colose’s near-equilibrium model will suffice, but if there is a diurnal cycle, it is complete bollocks.
    So do you live on a rotating planet where the influx of solar energy follows a truncated sine-wave or not?

  60. Doc–
    Why do you think the simplification of treating heat flux as constant over the full surface of the earth as bollocks?

    Under many circumstances, engineers will treat a heat addition as “constant” even though it’s oscillating. For example: If one wants to model a turkey cooking in an oven, one will often treat heat addition by the coils as constant even though the heating element actually goes on and off. One does similar things for simmering sauces on a stove top or home heating air conditioning and ventilation systems. For many calculations it would be pointless to complicate the model by including turning the heating and ventilation system on and off and on and off. You get practically the same answer doing the problem the more complicated or simple way– and it is generally better to use a simplification when it works and is adequate to discussing the problem at hand.

    When we simplify by treating heating sources as constant vs. oscillating, we have ways figure out if the approximation is adequate for the purpose intended. The simplification is not simply “bollocks” out of hand. It seems to me that Coloses approximation is fine for the purpose intended. But if you can explain why you think it is not fine for the purpose intended maybe I can understand your diagnosis that the method is “bollocks”.

  61. If the Earth were like the moon, i.e. low surface heat capacity and low rotation rate, then using averages wouldn’t get the right answer for the global average temperature by quite a lot. But the Earth rotates fairly rapidly and the surface heat capacity is high. The diurnal variation of sea surface temperature is quite small. It’s larger over land because the heat capacity is lower, but not enough to make much difference in the overall energy balance or the global average temperature.

  62. “Doc–
    Why do you think the simplification of treating heat flux as constant over the full surface of the earth as bollocks?”

    Allow me to explain how one needs to examine the kinetics of the system and not just examine the gross thermodynamics.

    Lucia, I have just been in touch with the guys at Magrathea and they have accepted a commission.
    They are going to place two identical planet Earth/moon systems in orbit around Sol(II).
    The first one is going to be given a day of exactly 365.25 standard days in length (Earth long).
    The second is going to be given a day of exactly 0.1 standard days in length (Earth short).

    What will be the average global temperatures on the two planets and what will be the changes in the three states of water on Earth long and Earth short, compared with Earth just right?

    a prior to an equilibrium calculation one must ask if the system is at, or near, homogeneity; is it well mixed.
    The further from local equilibrium the system is, then the less applicable the equilibrium approximation. The equilibrium approximation bedevils the climate science field.
    These people have never spent a summer day at the Mediterranean or the Mexican Gulf and observed what water and temperature actually do.

  63. DeWitt–
    Precisely. Those are the factors one considers.

    Doc

    What will be the average global temperatures on the two planets and what will be the changes in the three states of water on Earth long and Earth short, compared with Earth just right?

    The average temperature for the real earth will be very close to the rapidly rotating planet. This is because given properties of the earth (mentioned by Dewitt above), the earth’s surface temperature approaches the limit of “rotating infinitely fast”. In contrast, Mercury is “same side facing sun all the time.”

    The further from local equilibrium the system is, then the less applicable the equilibrium approximation. The equilibrium approximation bedevils the climate science field.

    Precisely. And for the temperatures involved, the earth happens to fall in the limit where approximating heat flux as constant if fine.

    These people have never spent a summer day at the Mediterranean or the Mexican Gulf and observed what water and temperature actually do.

    Sure. And this has nothing to do with rotation, it has to do with the inclination of the earth. And before you get to concerned about the fact that day time and night time temperature differ, remember that when doing these things, the correct temperatures are in absolute not celcius or farenheit. So consider the difference in day/night temperature relative to the average in an absolute temperature system like Kelvine or Rakine. Of course a difference of between 40F and 80F seems large to us. When looked in % terms, the difference between 458.67 F + 40 F or 458.67 F + 80 F is not so large.

  64. DocMartyn,

    Look at DeWitt’s comment, posted 4 minutes before yours. To which I would add: a part of the diurnal variation over land is due to local nighttime temperature inversions near the surface…. the vast bulk of the atmosphere has little diurnal temperature range, so the variation in the rate of heat loss to space over land is less than the daily surface temperature swing might suggest.
    .
    I have spent lots of time on the open ocean; the daily temperature swing is not very large compared to land.

  65. OK, let us take the little bit
    the earth’s surface temperature approaches the limit of “rotating infinitely fast”

    If we had no axial tilt, and thus no seasons other than that induced by our elliptical orbit, I would accept this. However, we have ice sheets that cool surface waters and inject it into the bottom of the oceans. The movements of hot, highly saline, surface waters moving from the equator to the poles being transformed into cold, dense salines, that fall to the abyss and make their way back to the equator are the main heat transport process. The state transitions of water, from the poles to the equator, are what makes Earth, Earth. Treating the whole system as flat equilibrium, without acknowledging you are making pretty rubbish, is an intellectual trap. It cons you into believing you know what you are dealing with.
    It is far, far better to deal with one area, say the equatorial Pacific, and model the processes of energy transduction there.

  66. “SteveF

    a part of the diurnal variation over land is due to local nighttime temperature inversions near the surface…. the vast bulk of the atmosphere has little diurnal temperature range, so the variation in the rate of heat loss to space over land is less than the daily surface temperature swing might suggest.
    .
    I have spent lots of time on the open ocean; the daily temperature swing is not very large compared to land.”

    I was more interested in how much water vapor the atmosphere can support during a full 24 hour day. In summer you tend to have super saturated just pass 1:30 in the afternoon and this collapses into rain storms 3-4 hours later and again 11-13 hours later. The conversions of hv into latent heat, an increase in air pressure, lifting a mass against the gravity well, then its conversion into sensible heat, the drop in pressure and its free fall are all rather complex.

  67. Doc,
    I realize you don’t accept this. But I’m also aware that you have no experience in heat transfer, thermodynamics or kinetics problems.

    The movements of hot, highly saline, surface waters moving from the equator to the poles being transformed into cold, dense salines, that fall to the abyss and make their way back to the equator are the main heat transport process.

    These movements on the earth’s surface are one of the factors that makes the approximation you consider “bollocks” a good approximation. It’s called “mixing”.

    Treating the whole system as flat equilibrium, without acknowledging you are making pretty rubbish, is an intellectual trap.

    Not for the problem Colose is discussing. Those factors matter if you are trying to get weather predictions of to discover how temperature varies on the surface. But those temperature differences are small relative to the mean temperature (in absolute) and for the problem Colose is discussing they don’t matter.

    I realize you might be interested in problems other than the one Colose is discussing. But for the issue he is discussing the simplification is fine.

  68. The conversions of hv into latent heat, an increase in air pressure, lifting a mass against the gravity well, then its conversion into sensible heat, the drop in pressure and its free fall are all rather complex.

    Since the troposphere is in hydrostatic equilibrium over large scales and close to the adiabatic lapse rate, very little work is needed to lift or lower a parcel of moist air.

    What increase and drop in pressure? Surely you weren’t taken in by Makarieva. Warming air only increases its pressure at constant volume. But there is no such constraint in the atmosphere. That’s why chemists use constant pressure thermodynamics even though the equations aren’t as elegant as those at constant volume.

  69. “These movements on the earth’s surface are one of the factors that makes the approximation you consider “bollocks” a good approximation. It’s called “mixing”.”

    Er no. These actual movements of hot and cold water are the opposite of mixing.

    Take a model ocean tube in the equator and thermally isolate it from its surroundings, so there are no flows sideways.
    Set the tube of sea water at 4 degrees and switch on the sun, stand back an wait.
    The water in the tube will eventually arrive at a near-equilibrium state, it will become iso-thermal.
    If you use a constant heat flux, the bottom will eventually come to thermal equilibrium with the surface. If you oscillate you flux seasonally, you will have the bottom temperature near the surface temperature found on the winter nights (like lakes and the Mediterranean).

    In real ocean columns the dissimilar currents of hot and cold water produce a stratified temperature gradient. The temperature gradient of the oceans can be used to perform work:-
    http://en.wikipedia.org/wiki/Ocean_thermal_energy_conversion

    The asymmetry of ocean circulation; of cold brines into the deep ocean and the movement of hot brines along the surface to the poles, is the opposite of mixing. There is a real energy gradient being continually renewed, which is generated by the differential temperatures and densities of the surface waters at the poles and equator.

    It is true that one can have standing gradients, including temperature gradients, that represent near equilibrium; both the atmosphere and the oceans have temperature gradients, and yet the atmosphere is close to being well mixed and is essentially at equilibrium, but I maintain that the oceans are not. The ‘suck it and see’ test for deciding of a gradient is far from equilibrium is to see if one can get work out of it. Trying to use the temperature difference between the atmosphere at seal level to 1 km up is futile, however, one can perform work, and even generate electricity, by using the temperature difference between the surface and 1 km down.
    Tracking fluxes is the key to understanding what is going on in non-equilibrium, dynamic, steady states.

    (I do steady state kinetics and thermodynamics for a living).

  70. SteveF:

    I agree, the Colose post is “ho-hum”, and not terribly alarming. It just lays out the standard climate science POV on water vapor feed-back.

    I saw it as being more planetary physics than terrestrial physics, associated with the question of when a planet might encounter a runway greenhouse gas effect. Probably publishable though, assuming there is enough new material, in an astrophysics journal.

  72. Doc

    These movements on the earth’s surface are one of the factors that makes the approximation you consider “bollocks” a good approximation. It’s called “mixing”.”

    Er no. These actual movements of hot and cold water are the opposite of mixing.

    Er … The actual movements of hot and cold water are mixing. The notion that they are unmixing — or leading to larger gradients that would occur without these motions– is ridiculous. Among other things, the claim these motions would unmix would involve a violation of the 2nd law of thermo.

    Take a model ocean tube in the equator and thermally isolate it from its surroundings, so there are no flows sideways.

    You think there is no “sideways mixing” when water flows in the ocean? Rest assured, some “sideways mixing” occurs. Moreover, even with no “sideways” mixing, water moving from a warmer location (e.g. the equator) to a cooler location (north pole) and being dumped in that location is “mixing” on a planetary scale.

    of cold brines into the deep ocean and the movement of hot brines along the surface to the poles, is the opposite of mixing.

    This is not the “opposite of mixing”. The cold brine later dumps into warmer waters where it combines with them. The hot brine later upwells in a cold location and combines with that water. This is “mixing”.

    (I do steady state kinetics and thermodynamics for a living).

    You clearly don’t do fluid transport. You work in medicine right? Or pharmaceuticals?

  73. “Er … The actual movements of hot and cold water are mixing. The notion that they are unmixing — or leading to larger gradients that would occur without these motions– is ridiculous. Among other things, the claim these motions would unmix would involve a violation of the 2nd law of thermo”

    Lucia. I know you are a very smart lady and so I offer you a thought experiment.
    Let us take a triangular tank, thermally isolated, containing salt water. in cross-section it has this aspect, but upside down

    http://etc.usf.edu/clipart/36500/36521/tri11_36521_lg.gif

    At point A we have a heat sink maintained at minus 6 degrees and at point C we have a heater/heat sink that is at plus 20 degrees.
    Prior to the insertion of the heat sinks the water is at a uniform temperature, call it 10 degrees.
    Add the two disparate heat sinks and wait. Within a short time you will start to see the formation of currents and the stratification of temperature by depth. Cold brine will sink to the bottom and warm brine will travel on the surface from the heat source to the heat sink. The water movements are being powered by the difference in temperature between the heat source and the heat sink.
    After a while, depending on the depth of the tank, the bulk water that the bottom of the tank will consist of cold brine and there will be a layer of mixed water, of lower salinity at the top. this will happen without an air layer, but an air layer would allow an air current to flow.

    Such temperature stratification is a manifestation of the properties of saline; cold saline sinks and warm saline floats. You would not get a stratification using pure water as the ice forming would sit on the surface.

  74. By the way, the ocean gyres are driven primarily by winds, not the thermohaline circulation. They would exist even if there were no downwelling at high latitudes. The thermohaline effect is not sufficient to drive circulation because, like in the atmosphere, very little work can be extracted from vertical flows caused by density differences when the potential temperature gradient is small. The major circulations in the atmosphere are driven by the pressure gradient force caused by the temperature difference between the equator and the poles. The Coriolis Effect then bends the flow and you get jet streams, trade winds, etc.

    The thermocline in the ocean exists because cold water upwells pretty much everywhere to replace the downwelling water at high latitudes (hydrostatic equilibrium again) and heat diffuses downward from the surface by eddy diffusion.

    [insert snide comment here]

  75. You would not get a stratification using pure water as the ice forming would sit on the surface.

    So the density of pure water is not a function of temperature and pressure? Oh, puhleeze. You should really stop digging now.

  76. Hey guys.

    Sorry to interrupt your exciting discussion of atmospheric… whatever. I just wanted to let you know I’ve proposed an explanation for why global warming proponents might be reasonable in their support of genocide. The idea came to me when I decided to try to offer an olive branch.

    Or when I was watching The Arrival last night. I’m not sure which.

  77. “DeWitt Payne

    So the density of pure water is not a function of temperature and pressure? Oh, puhleeze. You should really stop digging now”

    Of coarse it is, how ever in a simple tank model using pure water one would not have a heat gradient with cold at the bottom and a warmer surface.

  78. Doc

    The water movements are being powered by the difference in temperature between the heat source and the heat sink.

    I’m not sure what your point is. You seemed to claim there was no mixing and now proceed to provide a thought experiment in which mixing occurs. Moreover, your own words describe the mixing. But you’ve organized the problem in a way that makes it rather opaque.

    Yes. Water will begin to move and hot and cold water will intermingle at different locations and by different amounts. This is “mixing”. Heat transport from the hot point to the cold point would be enhanced relative to what happens if the fluid could not move. This is due to “convection” which involves “mixing”

    In contrast, if the material in the tank were solid with identical heat capacity, density and such like, the material would not move and things would not “mix”.

    Note: Because you made a thought experiment where temperatures are fixed at A and C, heat fluxes are changed when we change the contents from liqud to solid in our ‘thought experiment’.

    If in contrast we made this more like earth and fixed the magnitude of the “heat source” at A at Q= – 1 W/sec (making it a sink and the “source” at C at Q= +1 W/sec (making it a source) we would set up a similar flow, and but the temperature difference between points A and C would be different in the thought experiment where the tank contains a solid vs a liquid with equal thermal conductivities. The temperature difference will be lower when the tank contains a liquid due to the mixing (i.e. convection) that arises when the fluid flows but which does not arise when the fluid does not flow.

    Note: convection involves “mixing” or “intermingling” of packets fluids. And it is happening in your thought experiment even if though you see temperature gradients.

  79. The oceans are similar to the tank. The vast bulk of water at the bottom is cold, at about 4 degrees. The surface is where heat enters the oceans and also where it exists. For the vast majority of the oceans the surface is warmer than the depths. This heat gradient is standing. The reason that the depths do warm up to the ‘average’ surface temperature above is the continuous influx of cold water to the bottom of the ocean, from the poles.
    The system is a dynamic steady state and not an equilibrium. Treating the system as an equilibrium is wrong. The heat gradient is maintained by the difference in the temperature between the surface waters at the equator and poles.
    Similar heat engines function when you have a temporal heat potential, such as the case when you have a rotating planet. The surface responds to the actual energetics, during a daily cycle, and does not behave as if it is held at ‘average’ flux. This is especially important in the water cycle where the phase transitions between solid, liquid and gas occur at temperature levels that the surface temperature travels during the course of a day, at least in some places on the planet.
    Near equilibrium approximations are OK for rough and ready rules of thumb, to give you a reasonable guesstimate of your system. However, doing this sort of thing to work out water amplification effects of a change in CO2 levels is bollocks, it is especially bad if you use a known, defined, unit of time as the abscissa of a plot; using days as a unit of time for a flat, non-rotating planet is taking the piss.

  80. Of coarse it is, how ever in a simple tank model using pure water one would not have a heat gradient with cold at the bottom and a warmer surface.

    You would have density gradients in the tank because pure water reaches a maximum density at about 4 C. At that point flow to the bottom of the tank would stop. There would be a horizontal temperature gradient at the surface, but it’s not at all clear that it would be enough to cause flow.

    Complications ensue because the ice that forms can’t float, it will be attached to the cold sink and will be maintained at a temperature below freezing. That will raise the level in the tank. In fact, because the thermal conductivity of ice is higher than water, most of the tank, assuming it’s insulated and not too large, would eventually freeze because the surface of the ice will be below the freezing point of pure water.

  81. Doc

    Of coarse it is, how ever in a simple tank model using pure water one would not have a heat gradient with cold at the bottom and a warmer surface.

    This claim is incorrect. Heck, for a given point (e.g. A, B, c etc) the temperature at the top would differ from that at the bottom unless you included the specific heat flux to prevent that occurring. In a real experiment, this would require a thermostat (which is already required at your constant temperature points in your model.)

    The oceans are similar to the tank. The vast bulk of water at the bottom is cold, at about 4 degrees. The surface is where heat enters the oceans and also where it exists. For the vast majority of the oceans the surface is warmer than the depths. This heat gradient is standing. The reason that the depths do warm up to the ‘average’ surface temperature above is the continuous influx of cold water to the bottom of the ocean, from the poles.

    (a) The tank would be more similar to the oceans if defined the heat flux at points rather than defining temperatures. This is because the incoming radiation from the sun is not affected by the tempearture of the ocean.
    (b) Mixing (convection) happens in both the tank and the ocean.
    (c)

    he continuous influx of cold water to the bottom of the ocean, from the poles.

    If your point is this occurs: Of course. It’s “the mixing” is said occurs and which makes Coloses simplyfing assumption of anisothermal not too bad. This cold water them mingles (i.e. mixes) with local water. It is also why the Colose’s assumption of uniform temperature is not a bad approximation. That was my point.

    The system is a dynamic steady state and not an equilibrium

    Yes. So what?

    Treating the system as an equilibrium is wrong.

    No. It’s not “wrong”. It’s a simplifying assumption that gives close to correct answers in certain circumstances. In this case, the mixing from the ocean and the atmosphere makes it closer to correct.

    You are bringing up a lot of features that are are not ‘thermodynamic equilibrium” but which are irrelevant to diagnosing whether Colose’s analysis is useful. I realize that if you are a chemist working on stuff in test tubes these issues may loom large for you. But people working on larger systems deal with “steady state”, “pseudo-steady state”, “dynamic equilibrium” all the time. And quite often treating them as pure steady state problems for 1 D systems is fine.

    Anyway: Obviously no one is claiming thermodynamic equilibrium. Heat transfer is assumed to occur in Colose’s problem. So stop dithering about “thermodynamic equilibrium”.

  82. Re: DocMartyn (Jan 20 16:30),

    The oceans are similar to the tank.

    Not really. You would need a fan blowing from C to A to drive circulation at the surface. Better would be a fan blowing along one edge of the tank so you get a gyre.

  83. DeWitt,
    I assume he means it’s “like the ocean” in the sense that features he wants to discuss are “like” but other things are different.

    Of course you are correct that much of the circulation on the oceans is wind driven.

  84. Lucia & DeWitt,

    What would happen to the ocean if cold water didn’t sink at the poles? I guess the vertical temperature gradient would somewhat resemble that on land…warmer at great depths? Then what would the wind- driven currents do – pile water up at the poles until the potential hill was great enough to counter the wind pressure?

  85. Re: bill_c (Jan 20 18:49),

    Your scenario implies that the density of water is not a function of temperature. Are you going to include salinity and pressure as well? Note that there is bottom water formation in the Mediterranean Sea because the prevailing winds blow surface water through the Strait of Gibraltar, causing the average level to be above the Atlantic. The water evaporates and the increased salinity and density cause it to sink and flow back out into the Atlantic through the Strait below the current flowing in.

    Anyway, if the density of water is a constant, then there is no reason for the temperature gradient to look anything like the atmosphere, where the density is a strong function of temperature and pressure, leading to the dry adiabatic lapse rate equal to g/Cp because gases cool when expanded adiabatically. But if water were truly incompressible, then the temperature is not a function of pressure and there is no unique adiabatic rate. A packet of water will remain where it is when raised or lowered regardless of the surrounding temperature.

    This doesn’t mean that there wouldn’t be circulation. There would still be oceanic gyres causing differences in sea level. So there would still be downwelling where the sea level was higher and upwelling where it was lower. What effect that would have on temperature profiles is beyond me.

  86. DeWitt, did you ever notice that the ice sheets during ice ages cover the northern and not the southern hemisphere? Perhaps there is some reason for that associated with snow albedo feedbacks? Just sayin

  87. FWIW, Chris’ does miss something about which Hansen is well aware. If global warming weakens or vanishes the tropopause, water vapor can pass into the stratosphere and higher, where it can be dissociated by VUV radiation to hydrogen and oxygen. Hydrogen is light enough to escape to space. This is thought to have happened at Venus leading to the runaway. The Earth is a bit too far away from the sun for this to be a serious issue until the Sun gets hotter/bigger, but it is the runaway mechanism.

  89. DeWitt,

    The scenario I was going for was the truly incompressible water. I didn’t mean to imply it would be like the atmosphere. More like a continent, except there would still be diffusion. Why would there be downwelling and upwelling? Assuming the winds blew constantly in the same direction, I would think the sea level would adjust to the dynamic pressure.

  90. If the TOA is where outgoing and incoming radiation flux are the same, how can there be a true radiation imbalance when there is no change in the total energy in and out. If there is no true imbalance then there can be no missing heat.

  91. Re: Eli Rabett (Jan 20 20:21),

    Not only that, but the ice sheets are mainly in North America, not Asia. The most logical explanation I’ve heard for that is the geography of northern Canada. As a contiguous land mass, Asia would tend to get hotter in the summer than the mix of land and water in northern Canada. The permanent snow line would then be much further north. Then if Baffin Bay remains open due to a shift in ocean currents, you could have lake effect type snow on a large scale.

  92. Re: angech (Jan 20 21:38),

    If the TOA is where outgoing and incoming radiation flux are the same, how can there be a true radiation imbalance

    Simple. They’re not the same. Increasing CO2 reduces outgoing radiation while not affecting incoming radiation, to a first approximation. One can think of it as raising the effective emission altitude for the emission wavelengths in spectral regions where the transmission of LWIR photons from the surface is effectively zero. In the troposphere, temperature decreases with altitude. Emission decreases with decreasing temperature.

  94. bill_c (Comment #122722)
    January 20th, 2014 at 6:49 pm

    What would happen to the ocean if cold water didn’t sink at the poles? I guess the vertical temperature gradient would somewhat resemble that on land…warmer at great depths?

    I can’t imagine that would happen, since the heating would still take place near the surface and eddy diffusion would be mostly down gradient, hence you might guess some kind of stable stratification would be the steady state.

  95. Re: DeWitt Payne (Comment #122712)

    By the way, the ocean gyres are driven primarily by winds, not the thermohaline circulation…The thermohaline effect is not sufficient to drive circulation because, like in the atmosphere, very little work can be extracted from vertical flows caused by density differences when the potential temperature gradient is small…

    The thermocline in the ocean exists because cold water upwells pretty much everywhere to replace the downwelling water at high latitudes…and heat diffuses downward from the surface by eddy diffusion.

    Of course, “eddy diffusion” needs to be “driven” by something as well. 😉

  96. “DeWitt Payne
    Your scenario implies that the density of water is not a function of temperature. Are you going to include salinity and pressure as well? Note that there is bottom water formation in the Mediterranean Sea because the prevailing winds blow surface water through the Strait of Gibraltar, causing the average level to be above the Atlantic. The water evaporates and the increased salinity and density cause it to sink and flow back out into the Atlantic through the Strait below the current flowing in.”

    Actually that is not what happens. During the summer months there is a build up of highly saline water on the surface, kept afloat by being warm. During the winter the surface brine cools and becomes more dense that the underlying water. The surface brines sink, forming whirlpools (mentioned in the Odyssey).
    At the bottom of the Med are layers of dense brines, layered by year density. These brines exit the Med into the Atlantic through the straits of gibraltar via a siphon.
    There are two major currents through the strait, evaporated water is replaced by a surface current from the Atlantic into the Mad, drive by the sea level difference. The counter current from the Med into the Atlantic is driven by the salinity difference between the density of the two waters.
    Here is a cartoon graphic which shows the general movement of water in the Atlantic
    http://myweb.cwpost.liu.edu/vdivener/notes/NADW.gif
    Warm waters, of high salinity, travel the surface to the pole, cooling along the way. During the winter night they cool and sink, then travel back to the equator. Annually you get another deep layer. The ocean is stratified, like a layer cake, and a new layer is slid into the bottom annually.

  97. DocMartyn (Comment #122734)

    Here is a cartoon graphic which shows the general movement of water in the Atlantic
    http://myweb.cwpost.liu.edu/vd…..s/NADW.gif
    Warm waters, of high salinity, travel the surface to the pole, cooling along the way. During the winter night they cool and sink, then travel back to the equator. Annually you get another deep layer. The ocean is stratified, like a layer cake, and a new layer is slid into the bottom annually.

    DocMartyn, I think you have to imagine the deep water flow as a combination of vertical and horizontal motion. If you envision the seasonal deep water formation as simply sliding a new layer “into the bottom” each year, then you would be ignoring the horizontal movement completely and assuming that the water immediately fills the bottom layer!

  98. Re: DocMartyn (Jan 21 07:02),

    The ocean is stratified, like a layer cake, and a new layer is slid into the bottom annually.

    Cite please.

    I see no evidence of annual stratification in the literature. The ocean is stratified, just not annually. The main stratification is by temperature and salinity resulting in the thermocline and the pycnocline. The ocean is not like a Pousse Cafe with annual layers. That drink is made from the bottom up starting with the highest density liqueur. You couldn’t do it by starting with the lowest density and pouring the higher density ingredients through the lower ones unless they were immiscible. But that’s not true for sea water.

    The downwelling flow generates turbulence and mixing as do the horizontal flows in the deep ocean. If your model were correct, the 14C age of deep ocean water would be fairly recent. It’s not. In the Atlantic where there is a lot of downwelling at high northern latitudes, the age of the water below 1500m is about 600 years at high latitudes and gets older as you move south. In the Pacific, most of the basin water is older than 2,000 years. The deep ocean is also less saline than the surface.

  99. DeWitt,

    Over most of the ocean you are correct, but in the Arctic Ocean, the annual freezing of the surface produces a new high-salinity layer each year; the Arctic ocean really does have a layer cake structure! Everywhere else, no.

  100. DeWitt Payne (Comment #122731)

    Your reference says:

    In some coastal areas of the ocean (and large lakes such as the North American Great Lakes), the combination of persistent winds, Earth’s rotation (the Coriolis effect), and restrictions on lateral movements of water caused by shorelines and shallow bottoms induces upward and downward water movements. As explained above, the Coriolis effect plus the frictional coupling of wind and water (Ekman transport) cause net movement of surface water at about 90 degrees to the right of the wind direction in the Northern Hemisphere and to the left of the wind direction in the Southern Hemisphere. Coastal upwelling occurs where Ekman transport moves surface waters away from the coast; surface waters are replaced by water that wells up from below.

    I’m still not convinced that wind shear alone is enough to drive an overturning. Wind shear plus planetary rotation, OK. What happens when the wind shear is parallel to the rotation, like in ENSO-neutral or la Nina conditions? I think water piles up on the high side. You get geostrophic flow, OK. But does ENSO-neutral drive a parallel overturning, or is the warm pool mostly a static phenomenon held in place by the trade winds?

    Put another way, the surface of the water would slope to the point where the potential gradient was equal to the pressure from the wind time the friction coefficient.

    The surface gyres would exist without the thermohaline component of the circulation. But would the overturning?

  102. @DeWitt Payne (Comment #122730)
    Since atmosphere is a gas, and gas easily changes volume when temperature changes, how much does the atmosphere change volume in the presence of temperature changes? To what extent is this accounted for. SLR is allegedly changing dangerously to a thermal expansion, so certainly we should be hearing more about atmospheric thermal expansion in the age of dangerous climate change.

  103. Re: hunter (Jan 23 03:48),

    Nobody’s dwelling will be affected if the 500mbar pressure level moves up a few meters. OTOH, atmospheric expansion is a positive feedback, if I remember correctly. Meteorologists are well aware of the change in volume of the atmosphere with temperature.

    Only about half, and maybe less, of sea level rise is from thermal expansion. The rest is from loss of mass of the Greenland and Antarctic ice sheets plus other land based glaciers. There’s even a contribution from using water from slow filling aquifers.

Comments are closed.