It's a slippery slope. Once you look into what is being proposed as contrary science, it's hard to avoid Pielke-land.
This starts with the infamous "hot spot" controversy, which I've seen described as a "holy grail" by denialists but which I had little idea about. So I started out trying to make sense of the Klotzbach paper referenced by d'Aleo. Well, I got sort of befuddled in the middle, but RP Jr has piped up, with a very strong informal blog claim that the paper constitutes "Evidence that Global Temperature Trends have been Overstated" which is somewhat more readable, but still confusing.
This in turn refers us to RP Sr, "Why there is a Warm Bias in the Existing Analyses of the Global Average Surface Temperature" which looks like fairly straightforward boundary layer talks. But it starts right out with a reference to a peer reviewed Matsui and Pielke 2005: "Should light wind and windy nights have the same temperature trends at individual levels even if the boundary layer averaged heat content change is the same?" Again, a paper that seems to flicker in and out of comprehensibility for me.
But this last paper refers to Eastman et al 2001, claiming that
But this last paper refers to Eastman et al 2001, claiming that
As shown, for example, by Eastman et al. [2001] the addition of carbon dioxide to the atmosphere alters the downwelling long-wave radiative fluxes such that, in their idealized sensitivity experiments, the nighttime minimum temperature was increased, although the daytime maximum (and thus the temperature above the surface in the ‘‘residual layer’’ was essentially unchanged).
So we are basing an argument on this Eastman result. Yet, in the same paragraph 17 in M&P, we have
In the results reported in Eastman et al., a doubling of the carbon dioxide concentrations in the model resulted in a growing season, central Great Plains area-averaged increase of minimum temperature about 0.1 C.Doubling of CO2 leads to an average temperature increase of only 0.1 C ? OK, I'm trying to be patient, but now my BS detector is ringing pretty loud. Surely this is a typo? I mean, if the whole argument is based on a sensitivity 30 times lower than commonly believed, doesn't it become a circular argument? If everything we know is wrong, then we don't even know anything!
So what is this Eastman paper? Why it's Eastman, J. L., M. B. Coughenour, and R. A. Pielke (2001), "The effects of CO2 and landscape change using a coupled plant and meteorological model, Global Change Biol., 7, 797– 815." Not a single paper in this chain of reasoning outside of the Pielke domain.
Now what is this odd paper, which overthrows all of climate science, yet is published in a biology journal? It's a (wait for it) model study.
What's worse, it is a completely wrongheaded model study. It attempts to get a CO2 sensitivity from a limited area model run over a 14 degree lat x 14 degree lon domain. What's wrong with this picture? (Aside from the fact that its radiation model is the admittedly "rather simple" model of Mahrer & Pielke 1977?) Well the radiative equilibration time of the atmosphere is long compared to the residence of an air particle in a 14 degree lat/lon box. So how the hell are you supposed to set up the boundary conditions? Surely not from observations? That would pretty much, er prevent any warming from happening at all, right?
In order to simulate the 1989 growing season, the model uses a Newtonian relaxation method at the outer 3 grid points of the domain. Newtonian relaxation adds a tendency term to the prognosed quantity that drives it towards the observations from the NCEP reanalysis product (Kalnay et al. 1996).Not just the boundaries but the outer 3 grid points are dragged back toward observation at every time step on a 26 x 30 domain. That's 38% of the whole domain. Look. This is an experiment guaranteed to do about as close to absolutely nothing as possible. And guess what? It doesn't do anything!!!
Note that we have a chain, Klotzbach Pielke Pielke Christy & McNider 2009 to Matsui & Pielke 2005 to Eastman Coghenour & Pielke 2001 to Mahrer & Pielke 1977.
You can't fool me Mr. Feynmann. It's Pielke's all the way down.
The thing is, the Eastman piece, which seems like it might be crucial to the whole thing, is close enough to what I know that it's not a slog for me to get through it. And that piece is, err, what can I say, obviously wrong. A huge computational effort based on negligible computational skill. (No wonder these people don't trust models!)
(What of Mahrer & Pielke? Amazingly, Google scholar finds a copy. If I read it correctly, here is the actual radiation model used:
That's the whole thing. I doubt that is much good if radiation at the surface is actually the issue as in the present application. So the radiation model is likely inappropriate, but there's no great need to think deeply about it. It's moot.
There are reasons we don't use limited area models in climate studies of this kind. The Eastman result is already badly wrong.
There are reasons we don't use limited area models in climate studies of this kind. The Eastman result is already badly wrong.
So what are we to make of the whole enterprise? How much does Klotzbach Pielke et al depend on Eastman Pielke et al, and how much similarly confused reasoning is there in the whole stack?
Update: Roger Pielke Jr suggests in comments that
[I, mt] have badly confused "downwelling long-wave radiative fluxes" with the concept of "climate sensitivity." The first has to do with boundary layer meteorology and the latter with the equilibrium consequences of radiative forcing. These are completely separate effects of an increase in CO2 and yet you have mistakenly conflated them in order to try to discredit this paper
It is in fact very difficult for me to imagine myself conflating "downwelling long-wave radiative fluxes" with the concept of "climate sensitivity."
These are concepts I deal with every day, and they aren't even remotely alike. I would call the IR flux an energy density in watts per square meter, mapped to space (not just the boundary layer, of course). I would call the climate sensitivity, a global scalar, degrees C per (unit forcing), most often CO2 equivalent. I'd be interested to see where I might have confused them.
This seems like it itself must be a misunderstanding of something I was saying. I don't see how I could have conflated them. Like chop suey and donut holes, there is some very broad category relationship in my mental map, (foods associated with cheap restaurants, quantities associated with planetary physics) but the possibility of confusion is very small.
Note: RP Jr. also asserts that Klotzbach et al is not dependent on Eastman et al. by way of Matsui and Pielke, and that the relationship should be counted as minor supporting evidence.
61 comments:
See, now, on an anti-Climate Depot aggregator, this kind of post, released a day after the paper was announced, would be a first page feature. Except you'd have to have, like, a five line summary for the MSM, and maybe a cute graphic with a monkey.
But we're getting there.
Yeah, but I have a sinking feeling I might be wrong, and an even worse one that I am right and RP Sr will challenge me on it anyway. That may end up taking a long time.
Oh well. That's sure how it looks to me right now.
Somewhere in the dim recesses of my memory, there is a blogosphere rule about self-referential BS, esp from the emeriti. Eli, mayhap?
Nonetheless, don't the grownups in charge know this game already, and the API knows they know, so this is why they are gaming the vote perception with astroturf?? IOW: why do we waste smart people's time on the same ol' whack-a-mole?!
(word verif agrees, showing jacest)
Best,
D
This is the fundamental problem with arguing in public with deniers: They're playing a different game. They don't really care about being scientifically accurate, just putting on enough of a show for the passers by to fool a few more of them each time around into having doubts about AGW, or at least concluding that it's still a contentious issue so we should therefore avoid rushing into action that might not be needed anyway.
Lou, I don't think it's that simple. The McLean/deFreitas thing could well have been a co-production between someone with a simple result and someone with a non-scientific agenda. It certainly looks like the fingerprints of non-science, to put it mildly, are all over that one.
On the other hand, these RP Sr papers seem, though, full of striving and seriousness of purpose.
I'm far from sure this is intentional misrepresentation. It seems as likely that RP Sr thinks he has proven that the sensitivity is less than 4% of what the rest of us think. And, perhaps, that passing peer review in a biology journal is enough to establish his precedence on this result.
It still isn't clear to me what the more recent work means, or even what the word "bias" means in the recent work. And if things are as mangled as they so far appear to me, how he managed to round up four co-authors.
What this all tells us about peer review is another story.
I am starting to realize that when it comes to peer review, a lack of qualification is not a disqualification!
(a) they're leveraging arguments against the science that they don't believe in and don't care about as a drag on amelioration and mitigation, which they care about preventing, because they're apres nous le deluge about everything.
(b) they don't have to be consistent. they can say one day that peer review is broken or bogus, another that data gathering is a boondoggle, another that models aren't science, another that consensus is not part of science, and another day that only they are using sound science, and another day that they accept the science but want people to move at a wise, slow pace and not neglect other good causes, and then they can repeat the cycle. The real target has to be media, yet again. Media outlets that want credibility need to announce that if you lie to them, they'll stop giving you attention, and you can just go on Fox all the time.
Yeah, clearly the Eastman result looks a bit nonsensical... still, if all RPSr wants from it is the relative behaviour of max and min temps, it might still be legitimate.
About the idea of a "warm bias", that's Roger framing again. My impression is that the effect is likely quite real; but then, who expects temps measures 2m above ground to be very representative of anything? Isn't that also behind the effort to measure the changes in ocean heat content?
What matters is that where the surface stations are, is where we are. And the crops that we grow. And the ecosystems we are a part of. And the Arctic sea ice. It isn't "biased", it is precisely right; those other things, like lower troposphere temps from satellites, or ocean heat contents, while scientifically more useful, are not about us.
...and, on a linguistic note, it's not a "bias", it's a "scaling effect".
Michael,
You said, "Doubling of CO2 leads to an average temperature increase of only 0.1 C ? OK, I'm trying to be patient, but now my BS detector is ringing pretty loud. Surely this is a typo? I mean, if the whole argument is based on a sensitivity 30 times lower than commonly believed, doesn't it become a circular argument? If everything we know is wrong, then we don't even know anything!"
I'm curious. Where does the commonly believed number come from? I have read about this feedback, but I haven't found out where the commonly held belief was derived from.
Eastman is pretty weak timber to be constructing an argument that Parker's observational study is wrong.
Parker found no noticeable difference between temperature trends on windy days and non-windy days. RP Sr thereby concludes that Parker did the math wrong, based on Eastman. But Eastman is an experiment with meaningless counterfactual constraints.
And it seems to me that the differential between windy and calm days has a lot to do with the scenario they are constructing. In other words, the story they tell requires Parker to be wrong. But their argument is based on a very weird model experiment in a biology journal, while Parker's is based on observations.
Still, I have not one but two emails suggesting that I consider what Klotzbach et al mean by "bias". Indeed, it isn't clear.
It's almost as if they were saying that if you average the temperature at 2 meters height that is a biased estimate of the average temperature at 2 meters height (because it's windy some days and not others). Which is the opposite of a tautology.
Or, what?
What Marion said.
Best,
D
goodspkr:
Here's a page giving you an idea of some of the leading figures doing this work and how they do it. You can track them down from there.
The initial feedback-free estimate of about 1.2 C (if I recall right) for the Charney sensitivity (ocean/atmosphere/sea ice, all else fixed) is an almost elementary calculation of the direct forcing effect of CO2. The mechanism of positive water vapor feedback is very easy to understand but difficult to evaluate. One therefore expects the sensitivity to exceed 1.2; despite Lindzen's best efforts there are no strong candidates for attenuating feedbacks.
Full-featured dynamical coupled GCMs are one approach to evaluating
In my opinion, Annan and Hargreaves (GRL 2006) is an important milestone in thinking about this.
Please check their figure 1 for a sense of how the sensitivity is observationally constrained. Essentially this is the analyzed sensitivity with "error bars".
Note that the more streams of evidence are considered, the narrower the band of plausible sensitivity, which is the typical case. The evidence seems to converge around 3 C.
It is important to note that in the event of very rapid change, there may be couplings not represented in the Charney sensitivity. That is, all else may not be equal. For instance, rapid warming may increase methane levels via decay of formerly frozen tundra or release of clathrate deposits. The typically discussed sensitivity does not include such phenomena. So it might, in practice, be worse than represented.
Only in the light of the above should climate model sensitivity be considered. Climate models are tuned to represent contemporary climate. There are strong reasons to believe that they do so for reasons other than by chance. The best way I have to summarize those reasons without getting too technical about "degrees of freedom" is to refer you to a movie of model output.
You see, the models do more than just predict bulk properties like future temperature trends. They represent the dynamical evolution of the atmosphere well, and THEN are used, among other things, to extract bulk properties.
What Charney sensitivity do these models have? It turns out there is a range among the successful CGCMs, but they cluster around 3 C, perhaps leaning a little bit toward 2.5 .
Marion said:
"b) they don't have to be consistent. they can say one day that peer review is broken or bogus,"
I'm not sure you are interpreting what is being said correctly. Peer review will confirm that the paper is scientifically sound, but doesn't make the conclusions true.
While the people on the alarmist side seem to imply that it does at least for their side of the argument.
"another that data gathering is a boondoggle," I'll leave this one alone.
"another that models aren't science,"
Errr are you saying computer models are science? Or they are the equivalent of a study? Or are they tools that we can use to see what might happen given certain conditions. I would guess they are tools myself.
"another that consensus is not part of science,"
I don't think anyone is saying consensus isn't part of science. I think what they are saying is consensus isn't truth and it doesn't mean the science is "settled." I see many novices use this argument to try to end the conversation. But if consensus was truth and if it did settle the science, we would all be living in an earth centered universe.
"and another day that only they are using sound science, and another day that they accept the science but want people to move at a wise, slow pace and not neglect other good causes, and then they can repeat the cycle. The real target has to be media, yet again. Media outlets that want credibility need to announce that if you lie to them, they'll stop giving you attention, and you can just go on Fox all the time."
This is the problem that was discussed a few posting ago. How do you get your message across to the general public? Whinning about the other side isn't a good strategy. Blaming a sympathic press is foolish. Simply dismissing other studies is counterproductive.
What I see as your problem is that you assume what you know is truth (notice, I didn't say true and there is a difference). You therefore tend to present yourself and your theories from a somewhat arrogant position (your entire posting reeks of that, Marion).
I am a skeptic, but I do realize the science that you do have on your side. What I don't think you realize is the science you don't have on your side (what I see you do is simply dismiss it). Until you really try to understand this and address it, you will have a difficult time convincing the public.
"Peer review will confirm that the paper is scientifically sound, but doesn't make the conclusions true."
Apparently not.
I'll say it.
Peer review is bogus. Good stuff gets flushed. Bad stuff gets published. Better reviewers are overworked and worse ones available. Review is unpaid and itself unreviewed. There is very little motivation to do a good job, and increasingly it shows.
Consensus develops at formal and informal meetings, or on the web. The publication system as a metric of productivity is broken; that is why people are falling back not to a publication count but a citation count. But that's breaking down too.
See the Times Higher Ed article on the collapse of the journals. Also see the series on PhD comics, esp parts 2 and 4.
While there are dishonest people among the denial camp, most misinformed people are victims, not perpetrators. Please give naysayers among visitors the benefit of the doubt that they are not paid deniers.
I will shut them down once they get tedious, I promise. Even most of these are well-intentioned victims of the talking points machine, though. It's my job to make sure things don't get too boring. I don;t have any obligation to run them once they get repetitive,a nd I won't.
But I consider nasty a form of boring. Please don't get nasty, or I won't run your thing either.
Anyone showing up gives us a representation of how opinions get twisted. Let's try to untwist their minds gently, shall we? We may even succeed on rare occasion, which will give us something we can learn from.
But I promise that namecalling won't work.
Please do not assume that anyone who gets it wrong is acting in bad faith. I will remove any further ad hominem attacks on naysayers from the comment stream.
Michael is working towards this, but Michael is polite. The fact is that infinite amounts of crap get published. Crap has two colors, wrong and useless. Mostly the crap gets ignored (look at how many papers don't get cited at all**), so referees know that they usually don't do much damage by not being thorough.
** there are important publications that everyone knows about but don't get cited because they kill off a direction of inquiry. Eli has one of those
For Dano
"What you are doing here, and in your publications, and on Prometheus is to assert ownership of a series of issues, the latest of which is hurricane damage due to climate change. Your incessant self citation is a clear indication. I am certain you will reply that somewhere in a post somewhen you may have mentioned another’s work."
Eli, Professor of Prediction at the Kreskin Institute, predicts.
Roger Sr. will challenge you on his commentless blog
It's almost as if they were saying that if you average the temperature at 2 meters height that is a biased estimate of the average temperature at 2 meters height (because it's windy some days and not others). Which is the opposite of a tautology.
Michael, the idea as I understand it is that, on quiet nights, an inversion is created. Weather stations measure well inside this inversion, while, e.g., satellite measurements of the lower troposphere, or the vertical discretization of many GCMs, cut out a much thicker slice.
Now, the inversion is created by radiation cooling of the surface, which gets less effective with more greenhouse gases. Also on windy days, formation of a strong inversion is less likely due to mixing.
Martin, that's how I read it too. But what is a biased measure of what in case that is right?
"...if consensus was truth and if it did settle the science, we would all be living in an earth centered universe."
But the sun centered model is today's consensus. Are you arguing that its wrong?
"bias" is a key concept to the series of papers. It's pea-under-the-thimble stuff.
Occasionally the Pielke stuff gets rebutted, but more generally ignored these days. Eg citations of Pielke and Matsui.
James, "pea-and-thimble" seems to be brit for "shell game". Can you elaborate?
Are you saying that whatever meaning you attribute to "bias" it will always be something else?
"shell game"
Yes, although perhaps bait-and-switch would have been a better choice on my part, since the pea is always there, it just changes its meaning seamlessly. In context, "bias" would normally be taken to mean a measurement error but they clearly include real physical effects (such as CO2 warming!) in their use of the term, at least in one place. This seems to be a common Pielkian trait (to be fair, not unique to them) which is one reason why people have such difficulty interpreting their real positions on various topics.
I've been coming across this phenomenon fairly frequently. Another example is the local versus global CO2 forcing thing you pointed to in Eastman et al. RPSnr also has an odd new paper which appears to confuse synoptic short-term response to forcings (think weather modification) to the equilibrium sensitivity. Then the RPJnr thing where he apparently confuses the uncertainty of model predictions due to natural variability, with the uncertainty of measurements due to limited coverage and data processing.
MT, if I've understood your description of the paper correctly, the authors have applied the Christy 1.2 "deamplification" factor to the UAH and RSS satellite records.
As far as I know, this procedure has not made its appearance too often in the peer-reviewed literature. But Christy has been using it in his presentations and congressional testimony for the last few years.
The justification, of course, comes from expected tropospheric amplification in models.
CCSP (2005) noted the presence of short-term amplification in the tropospheric record, accompanied by the absence of long-term amplification in tropospheric trends.
But the Christy procedure reduces all positive anomalies, including all recent years (and raises all negative anonalies). And it also, of course, reduces the tropospheric trends at the outset.
I don't find this to be a justifiable procedure. Maybe if one wanted to "correct" for short-term amplification, it would be better to detrend (perhaps using a non-linear trend estimate like a lowess smooth) and then apply the 1.2 factor.
Then, at least, El Nino peaks would be lowered, and La Nina troughs would be raised in an intuitively sensible manner.
What would such a procedure say about tropospheric trends? Perhaps not too much. But it might show that some of the apparent divergence between observed and expected tropospheric trends is due to the La Nina end point in the series.
Meanwhile, to see the 1.2 solution in action, check out this post about the famous Alan Carlin/CEI/John Christy graph (as seen by millions on FoxNews).
http://deepclimate.org/2009/07/03/more-heavy-lifting-from-the-suppressed-alan-carlin/
Michael-
You have badly confused "downwelling long-wave radiative fluxes" with the concept of "climate sensitivity." The first has to do with boundary layer meteorology and the latter with the equilibrium consequences of radiative forcing. These are completely separate effects of an increase in CO2 and yet you have mistakenly conflated them in order to try to discredit this paper.
In daily temperature measurements a bias 0.1 deg C is important when assessing trends of the same order of magnitude.
Klotzbach et al. does not depend at all upon Eastman et al., rather Eastman et al. provides some confirmatory evidence. Throw it out if you like and the argument of Klotzbach et al. still holds.
Once you better understand your mistakes an apology would be appropriate.
Finally, look at the signal to noise ratio on this thread. Congrats, you are well on your way to becoming the anti-Morano.
To RP Jr.
If I ever understand the Klotzbach paper in such a way that it makes sense to me, I will certainly be at pains to apologize. I will, out of respect for your request, keep trying. So far it eludes me. Perhaps you could have intervened in one of the earlier postings while I was trying to figure out the context. We might be further along.
I am expressing doubt and reservations re Klotzbach, especially because two researchers I greatly respect also report confusion.
I will certainly revisit it keeping in mind that anything dependent on Eastman can be elided to little effect.
Sorry, I missed the earlier thread.
You have my email, and I have a thread up explaining the paper. If you have questions about it, please ask and I will answer.
Michael, you are much too nice and Roger depends on it. After what he did to you, Eli would think you would have learned. Eli would be wrong.
The point is that now, no matter what you decide he can quote your quiet response to his excuses. If this is war, you are a pacifist.
I'm not sure you're right, Eli. Propriety and politesse are not the same.
Michael, remember: aim for a fifth-grade reading level.
The PR guy has to be able to fool a fifth grader, to win in politics.
You have to be able to educate a fifth grader, to outdo the PR guy.
I suggest a focus group -- any grade-schoolers among your neighbors and friends you could call on?
As everyone has said and the PR guy has demonstrated -- they're playing a different game.
Look, we can now revisit the whole thing with a commitment from Jr to help us understand it. First of all, in the (I think most of us agree, unlikely) event that there is something to it, that is the right thing to do. Maybe there is just some communication glitch, after which the whole thing will come into focus.
Secondly, there's a sort of tacit abandonment of the Eastman claims.
Thirdly, there's the silly suggestion that I don't know a flux from a sensitivity. It will be interesting to see how that plays out; where that figures in my purported error.
In any case, I am not trying to debunk any claim. I am trying to understand what, if anything, all this hotspot business is about. The inactivists are making a great deal of it.
If there's nothing to it at all, somebody ought to have the gumption to say so. If there is something to it, it needs to be a lot clearer than it is.
Michael-
I won't make a habit of crosspost0ng blog comments but FYI Hank Roberts from this thread visited my blog with a question, which I appreciate, here is my answer.
"The short answer to your question is that we take a well-known and unsatisfactorily-explained feature of the temperature record -- a growing divergence between surface and satellite tropospheric temps -- and explore whether it is consistent with a previously alleged bias in the surface temperature record.
We offer up two hypotheses that follow directly from this line of thinking and then look at what the data show. What you are asking about is part of the answer to one of these hypotheses.
Santer et al. 2005 tried to explain the observed divergence and pointed to but ignored the option that we are exploring here. We think that it makes physical sense and is supported by the data."
Michael-
A few responses to your latest:
1. There is no "abandonment" of the Eastman et al. claims, as that paper is not cited in Klotzbach et al. nor was I involved in it. Eastman et al. is just not as directly relevant to Klotzbach et al. as you have purported.
2. You wrote "if the whole argument is based on a sensitivity 30 times lower than commonly believed, doesn't it become a circular argument?"
Presumably this comment is referring to "climate sensitivity". If so, it is a big error on your part as our paper does not discuss nor depends upon estimates of climate sensitivity, not even indirectly.
Since you are interested in communication let's see if we can communicate.
Roger, please explain the use or uses of the word "bias" in the Klotzbach paper. Presumably a bias means a systematic error in a measurement.
Exactly what measurement or measurements are proposed to be biased estimates of what physical quantity or quantities?
Thanks.
How does this work? I don't recall how to test this kind of expectation statistically. It's from RPJr's post on the subject. (Also asked there)
> the divergence that we observe is
> statistically significant in 3 of 4
> cases over land (that is, NCDC
> minus UAH, NCDC minus RSS, Hadley
> minus UAH) but not in any of the
> cases over the ocean, which is
> exactly what we’d expect
Since we're moving from Eastman back to Klotzbach, and since RP Jr is defending Klotzbach on his blog, let's take it there for now.
You have to give RP Jr credit for fastidiously posting every message that comes his way, including those from Mr CO2 Starvation Guy as well as myself.
But really peculiar conversation ensues. Meanings shift. You run into defenses posing as explanations; meaning shifts, context switches.
You noticed? Which is a great reason for not giving up control of the discussion. Also considering point 4 you now see the reasons not to be modest.
Well, he's likely pretty traffic-starved as a result of the recent boykoff.
Michael - excellent analysis and followup on Pielke's blog. Might I suggest you put up a new post here just focusing on the question of the *land* discrepancy they talk about (there is no statistically significant discrepancy in *global* surface/troposphere temperature trends, something Pielke Jr. conflates right at the start of his discussion) and exactly what might be issues with measurements of surface temperatures over land...
My questions on this:
* how do the satellites define "land" measurements, and could there be an discrepancy issue there (how well-defined is the boundary between land and ocean for what the satellites are doing)?
* dry and wet adiabatic lapse rates are significantly different, and moisture content is changing which changes lapse rates - can this be a significant factor in the discrepancy?
* Is some sort of boundary-layer effect like the Pielke's talk about actually realistic? E.g. Underground one gets salt-water intrusions into aquifers near ocean boundaries; is the issue a similar one in the atmosphere with ocean-air intrusions over land at high altitudes?
I think spending more time clarifying what they meant by "bias" is probably a waste - there are several statements in the paper that are clearly wrong given the understanding Pielke Jr. has provided of the content, but getting him to admit that is probably next to impossible, and in any case pointless since even a lay reader can pretty clearly see the difference you've delineated between measurement bias and this physical system effect.
Yup. Well, the answer I got when I asked about the statistics.
To me, hearing that three out of four on land (and none by sea) turned out statistically significant -- seemed an odd result.
He seems to say that fit their prediction, read the paper -- someone holler when it gets posted for the non-subscribers to read, please.
And, yeah, the assertion about confusing downwelling radiation with climate sensitivity seemed very strange.
Arthur: "getting [RP Jr.] to admit [error] is probably next to impossible"
Optimist!
Jr. quotes Sr.:
". . . a conservative estimate of the warm bias resulting from measuring the temperature near the ground is around 0.21 C per decade (with the nightime T(min) contributing a large part of this bias). Since land covers about 29% of the Earth’s surface (see), the warm bias due to this influence explains about 30% of the IPCC estimate of global warming. In other words, consideration of the bias in temperature would reduce the IPCC trend to about 0.14 degrees C per decade, still a warming, but not as large as indicated by the IPCC."
This assertion seems remarkable in that it purports to cancel most of the land surface trend. How is that not unphysical on the face of it?
"Klotzbach et al is not dependent on Eastman et al. by way of Matsui and Pielke, and that the relationship should be counted as minor supporting evidence."
Yeah, but is it dependent on Matsui and Pielke? (which is dependent on Eastman et al,)
Michael:
> But what is a biased measure of what
> in case that is right?
Yeah... good question. See my earlier comment.
...and BTW the rest of us understand perfectly well that your reference to "sensitivity" was shorthand for the surprisingly small radiative forcing reponse in the Eastman model. One has to be a bit dense, or a politician, to fail getting that.
One has to be a bit dense, or a politician, to fail getting that.
IMO no failure. There is a talking point to deliver and FUD to purvey, after all. That is my conclusion of the actions of paw and likely Jr as well. Hand-waving. And you won't pin either down.
DS
Hank,
An unformatted PDF of K et al is here:
http://www.climatesci.org/publications/pdf/R-345.pdf
I think Arthur is asking the right questions and any one of them could be profitably explored.
It's still worth trying to figure out what K et al mean by "bias". In that regard, I note that they cite surfacestations.org (Anthony Watts), but not the NOAA memo analyzing the Watts data set. The NOAA showed that the "good station" subset had an indistinguishable average warming trend from the larger U.S. network.
http://www.ncdc.noaa.gov/oa/about/response-v2.pdf
An interesting footnote: Stephen McIntyre at Climate Audit accused the NOAA of plagiarism when they initially cited surfacestations.org instead of the Watts paper published by the Heartland Institute.
An exploration of this and other plagiarism mysteries here:
http://deepclimate.org/2009/08/14/dropping-the-p-bomb/
I'm just going to leave this here, apropos of nothing in particular
Ouch!
If I may add a further note trying to understanding what the poorly-communicating Pielke's actually are talking about, Pielke Sr. has been relatively consistent in emphasizing the importance of heat content, rather than temperature, as a measure of warming.
Now I don't think we have particularly good measures of heat content (and Sr.'s attempts to limit the question to solely ocean heat have been a bit silly). In particular the (change in) heat content of the atmosphere itself is really quite small as temperatures rise, because the heat capacity of the atmosphere just isn't that large. The fact that the lower atmosphere can warm and cool by tens of degrees in a day tells you that already.
But the primary effect of an increase in radiative forcing is, surely, a steady increase in heat stored in Earth's system somewhere. If the imbalance between incoming and outgoing radiation is 2 W/m^2, then 2 W/m^2 must be getting stashed away. Can we measure it?
Part of that retained heat is going into melting ice - that should be pretty quantifiable. Part of it is going into ocean heating, which seems to be hard to measure (especially if a lot of the excess heat is heading into the depths through the thermohaline circulation or other overturning). And part of it must be going into warming land areas, gradually seeping deeper into Earth's surface. I think I've heard of borehole temperature measures - can they give us a sense of total heat content changes on land?
Anyway, if Pielke Sr. was actually really interested in the heat content question instead of continuing to kick up dirt, he'd surely have put some more serious effort into identifying the various possible heat stores. Atmospheric heat content is essentially irrelevant as far as I can see.
Arthur,
It is well known that essentially all of the heat energy (say 90%) accumulates in the ocean.
I agree that the heat capacity of the atmosphere is not a matter of interest. If the Pielke group's position is that this is all about confused estimates of the trend in atmospheric heat, they are talking about a bias in a quantity that nobody else is interested in.
There's little doubt that the imbalance is going into the ocean. Changes in atmospheric heat are small and, I'd guess, mostly at the seasonal time scale. But estimates of the actual imbalance quantity, which would help considerably in constraining models of the ocean, are not available. The DSCOVR instrument would be very helpful.
For what it's worth, the number I have heard for present day circumstances is 0.3 W/m^2 but I can't come up with a reference.
Hi James - nice blog post yesterday on this by the way!
I guess I'd heard assertions like that on ocean heat, but I've not seen clear evidence on the question. Not that I'm as familiar with the literature as many people are, so maybe it's clearly shown in some standard text that I just haven't seen yet.
For example, oceans are 70% of Earth's surface area, so one might expect just 70% of net radiative imbalance to go directly into oceans. And given the differences in land and ocean temperature variation and cloud/GHG concentrations over land vs ocean, I wouldn't be surprised by significant variation from that 70% number, possibly in favor of a higher fraction going to land than ocean. On the other hand, quicker land surface temperature response might keep the land more closely in radiative balance than oceans, so the heat flow might be expected to work out the other way too.
Does the "90%" come from evaluation of where heat goes in models, on general theoretical grounds, or is there some actual observational evidence on it? Could it be significantly different from 90% for short timescales, i.e. during the present early stages of CO2 increase?
And I think the ice question is an interesting one too. Latent heat of fusion of water is 333 J/g while heating 1 g of water 1 K takes 1 J. So each g of ice converted to water is absorbing as much heat as 500 times the same quantity of water warmed by 0.6 K.
I.e. if there was enough ice melted to raise sea levels by 5 cm, that's the same heat absorption as 25 meters of water warmed by 0.6 K.
Arthur,
Does the "90%" come from evaluation of where heat goes in models, on general theoretical grounds, or is there some actual observational evidence on it?
All three really.
Levitus et al 2001 is the canonical paper on heat content, it has been updated without major changes to the overall picture.
Could it be significantly different from 90% for short timescales, i.e. during the present early stages of CO2 increase?
It will in general depend on the time scale - on a daily to weekly scale, the heat content of the atmosphere may play a larger role (in fact heat shifting between atmos and ocean may cause opposite sign changes). I don't think anyone has looked at this in detail. Over really long term changes in large ice sheet may play a role - the amount of ice melting between last glacial max and today will probably come to a decent amount of heat, maybe comparable to ocean and atmospheric changes over that interval. But we aren't going to melt another 120m (sea level equivalent) of ice in the near future!
This is a long, confusing and arcane line of reasoning for a nonscientist like me.
The gist I'm getting out of it--which I wish you would clarify in a summary at the top of such posts--is that the global warming denier's arguments:
1) are largely circular
2) are insular because they rely on their own writers' conclusions from previous research,
3) are without verification by other reputable scientists and
4) come mainly from one scientist, Pielke.
So therefore the arguments advanced by deniers (which rely mainly on Pielke's papers) are highly suspect and not verified independently by other scientists.
Is that about right?
Please make it clearer to laypersons. I could use some ammunition to help educate the ignorant.
As for the crazies...well, they're beyond hope. We can pray that there aren't too many of them, and that they don't vote much. And don't carry their guns around to make their points clearer to peaceable people. (shiver!)
ThinkLife, when a scientist says something with little doubt in a scientific context,he or she had better have absolutely compelling evidence. So I can't really go as far as to assert those points.
This is a big problem, really. There is only so clear we are allowed to be. We sacrifice clarity for truth. It's an unfortunate constraint, and even the tiniest bending of that rule causes us endless trouble.
SO here's the best I can do.
Your fourth point is certainly wrong, though that is not to defend the quality of other anti-consensus efforts. A few groups of people claim to have evidence against the IPCC WGI consensus. Their positions tend to cluster in mutually exclusive clumps, though they never criticize one another.
But the particular group here does tend to do a lot of self-reference.
As for your other points, I will go so far as to say that there is evidence to support them.
- My perception is that this work is not highly cited by other scientists, certainly not in proportion to the attention paid to it on certain blogs. This would also tend to support insularity.
- I would say efforts to understand the work are rewarded with a very different sort of understanding than is typical of scientific results worthy of attention.
So, for replication of their work by others, that would require them to make their case clearly enough that others could even consider, and this they do not do in print nor in electronic conversation.
The strange uses of technical terms make the argument hard to pin down, so as far as I can say there may well be circular arguments as well as other invalid logic at work.
The strongest statement I can make with confidence is this: the way RP Jr and RP Sr discuss the Klotzbach paper in public is misleading. This seems pretty well established.
I don't suggest you sacrifice truth for clarity. That's what the deniers do, because their motives are often suspect and their tactics mirror their motives. We can't use their lying ways or it gets us, as you say, in trouble.
I do suggest a cogent summary of findings or conclusions at the beginning of each post--one in lay terms. It may take a lot of mental work to write these, but it's well worth it.
You want readers to grasp as much as possible, right? That's the reward of such labor.
Some suggestions to help keep truth and clarity:
"This isn't universal, but looks generally true..."
"Without getting bogged down by too much detail, I conclude that..._______. Here's why:
- bullet point
- bullet point
- bullet point"
"My general assessment is that...________. See below for full details and disclaimers."
Subheads make writing easier to digest
Subheads can help break up complex reasoning and make it more palatable--as per the example subhead above. They also summarize groups of paragraphs, making a piece of writing easier to digest.
Pardon me if I'm getting too pedantic here, but I learned these techniques through a combination of writing for employee publications and technical writing--most of which has to be understood by laypersons.
As a scientist writing for many audiences, you're actually practicing technical writing. It may be worth reading up on the basics.
I hope this helps!
Keep blogging--you're doing a great service. It's wonderful to see intelligent science out there and cast out on the Internet, even if I can't follow all of it.
Thanks TL. I struggle with those issues a lot, though sometimes it doesn't show.
James - thanks, Levitus is perfect. Obviously I hadn't paid much attention to the ocean side of things up to this point (looking at Ch. 5 of AR4 WG1 now, Levitus a co-author, look at that...) - it does seem pretty convincing. But then having read Pielke Sr. on heat uptake something's not making sense. Hmm, not that that's anything new with those guys...
> while heating 1 g of water 1 K takes 1 J.
Um, Arthur, wasn't that one calorie = 4.2 J?
Martin - oops, you're right, I was getting too carried away with metric happiness. Well that shrinks the ratio quite a bit - and suggests further why the Levitus 90%+ is right. Oh well...
Post a Comment