UPDATE: 8/18 10:30AM I spoke with Dr. Judith Curry by telephone today, and she graciously offered the link to the full paper here, and has added this graphic to help clarify the discussion. I have reformatted it to fit this presentation format (side by side rather than top-bottom) While this is a controversial issue, I ask you please treat Dr. Curry with respect in discussions since she is bending over backwards to be accommodating. – Anthony
===========================================================
[Update] My thanks to Dr. Curry for showing the graphic above, as well as for her comment below and her general honesty and willingness to engage on these and other issues. She should be a role model for AGW supporters. I agree totally with Anthony’s call for respect and politeness in our dealings with her (as well as with all other honest scientists who are brave enough to debate their ideas in the blogosphere). I also commend the other author of the study, Jiping Liu, for his comments below.
However, as my Figure 2 below clearly shows, any analysis of the HadISST data going back to 1950 is meaningless for the higher Southern latitudes. The HadISST data before about 1980 is nonexistent or badly corrupted for all latitude bands from 40°S to 70°S. As a result, although the HAdISST graphic above looks authoritative, it is just a pretty picture. There are five decades in the study (1950-1999). The first three of the decades contain badly corrupted or nonexistent data. You can’t make claims about overall trends and present authoritative looking graphics when the first three-fifths of your data is missing or useless. – willis
===========================================================
Guest Post by Willis Eschenbach
Anthony has posted here on a new paper co-authored by Judith Curry of Georgia Tech, entitled “Accelerated warming of the Southern Ocean and its impacts on the hydrological cycle and sea ice”. The Georgia Tech press release is here. Having obtained the paper courtesy of my undersea conduit (h/t to WS once again), I can now comment on the study. My first comment is, “show us the data”. Instead of data, here’s what they start with:
Kinda looks like temperature data, doesn’t it? But it is not. It is the first Empirical Orthogonal Function of the temperature data … the original caption from the paper says:
Figure 1. Spatial patterns of the first EOF mode of the area-weighted annual mean SST south of 40 °S. Observations: (A) HadISST and (B) ERSST for the period 1950–1999. Simulations of CCSM3 (Left) and GFDL-CM2.1 (Right): (C, D) 50-year PIcntrl experiment (natural forcing only),
Given the title of “Accelerated warming”, one would be forgiven for assuming that (A) represents an actual measurement of a warming Southern Ocean. I mean, most of (A) is in colors of pink, orange, or red. What’s not to like?
When I look at something like this, I first look at the data itself. Not the first EOF. The data. The paper says they are using the Hadley Centre Sea Ice and Sea Surface Temperature (HadISST) data. Here’s what that data looks like, by 5° latitude band:
Figure 2. HadISST temperature record for the Southern Ocean, by 5° latitude band. Data Source.
My first conclusion after looking at that data is that it is mostly useless prior to about 1978. Before that, the data simply doesn’t exist in much of the Southern Ocean, it has just been shown as a single representative value.
So if I had been a referee on the paper my first question would be, why do the authors think that any analysis based on that HadISST data from 1950 to 1999 has any meaning at all?
Next, where is the advertised “Accelerated warming of the Southern Ocean”? If we look at the period from 1978 onwards (the only time period with reasonable data over the entire Southern Ocean), there is a slight cooling trend nearest Antarctica, and no trend in the rest of the Southern Ocean. In other words, no warming, accelerated or otherwise.
Finally, I haven’t even touched on the other part of the equation, the precipitation. If you think temperature data is lacking over the Southern Ocean, precipitation data is much worse. The various satellite products (TRMM, SSM/i, GPCC) give widely varying numbers for precipitation in that region, with no significant correlation between any pair (maximum pairwise r^2 is 0.06).
My conclusion? There is nowhere near enough Southern Ocean data on either side of the temperature/precipitation equation to draw any conclusions. In particular, we can say nothing about the period pre-1978, and various precipitation datasets are very contradictory after 1978. Garbage in, you know what comes out …
I browsed comments and saw that no one had yet commented on the use of an “orthogonal” model. I have not read the paper, but I want to highlight how the use of an “orthogonal” model, as I understand the term, should be close to a fatal flaw.
The paper attempts to take a few data sets – the “empirical” aspect – and develop a model: how much to weight each of the sources of data when using them to predict some outcome. So, ultimately, the method has the same intention as a linear regression: figure out how to best weight two predictors to predict values of a third.
By using an “orthogonal” model, the calculations will not compensate for covariance (with a lower-case “c” so maybe I should say “covariation,” or “co-mingling” ) in the predictors. In contrast, an “oblique” model has the assumption that the various predictors are co-dependent.
For weather and climate phenomena, the various empirical sources of data have to be assumed to be inter-related, no? Temp relates to rainfall, barometric pressure relates to rainfall, and so no? No?
I know this from factor analysis in behavioral and social sciences – we might try to predict some behavior based on related predictors such as using both annual income and years of education. We know these two predictors are dependent upon each other to some degree, so we would simply develop any “factor analysis” with a model that allows the predictors to be oblique, to have some common variance.
Here is the problem – and I have read Mann 1998, and I have not yet analyzed it enough to determine how orthogonal or oblique they allowed their principle components analysis to be – but this could be an issue in Mann 1998 as well- —
With orthogonality, you force the weighting to be done with the aspects of these inter-related variables that are unique, thus magnifying the influence of a modest portion of the variance of any predictor / not taking advantage of the full variance available to predict.
I believe that, mathematically, this will end up necessarily with a result that excessively amplifies at least one predictor / contributing “empirical” variable.
These models are developed to predict the future. Just as it is a small issue to be 3 degrees off-course on a short trip but it is a huge problem on a long trip, the prediction will incerasingly be “off” as we use an erroneous model to predict the future: predict temp, rainfall, ice, whatever.
Do I have this use of the term “orthogonal” correct?
Can anyone else put this issue into clear language? I have tried to be clear, but it is not easy.
Please be patient. We are working on a plot of SST anomalies (on the same scale as the EOF plot) that will clarify the issue that Willis raises.
The funding for the project is acknowledged in the paper, it is funded partly by two grants, one NASA and one NSF. The funding pays Liu’s salary and also supports graduate students. Georgia Tech pays 100% of my salary, i don’t need grants to support my own salary.
With regards to my concerns about the quality of the SST data, I have only been digging into that the last several months. The concerns are greatest for the period prior to 1950, and there is relatively little data in some parts of the Southern Ocean prior to the satellite era. The general pattern of warming in the mid latitudes and slight cooling in the high latitudes is the key issue with regards to SST. We explore how the atmospheric hydrological cycle, ocean temperatures and sea ice extent interact. We then do a “what if” experiment with scenarios of 21st century warming to consider how this interaction would change in a perturbed environment.
Our paper compares model simulations with available observations (we consider two different data sets) in an effort to unravel the physical mechanisms that determine Antarctic sea ice extent in response to climate variability and change.
We have identified a plausible physical mechanism that seems to make sense. Science is about trying understand how things work.
We have made no extravagant claims in either the paper or the press release. I would not label this as a “warmist” paper. It talks about the increase of Antarctic sea ice, which is hardly a talking point for alarmists.
Yes, climate models are imperfect and there are deficiencies in SST data sets particularly in the first two decades of the period that we examine. So we have imperfect tools to test our hypothesis. Others will examine this problem from different angles. Eventually we will have better data sets and better models to work with. That is how science works.
This paper raises an issue that climate researchers should pay more attention to. Since the climate model simulations of antarctic sea ice generally agreed with observations, climate researchers would say “consistent with” without really understanding the mechanism. And we definitely need more and better data in the Southern Ocean.
With regards to the behavior of WUWT in all this. Anthony sent me an email yesterday informing me of the post, which I appreciated. Willis raised an issue that needs to be addressed in terms of the SST data, and he did this in a professional manner. With regards to comments on the previous thread, they were generally a disgrace to WUWT. The comments are being discussed by other blogs e.g. http://shewonk.wordpress.com/2010/08/17/curry-curries-no-favour/
They do not reflect well on WUWT, and my recommendation to Anthony and the moderators is that they should reconsider what kind of comments they want here.
If this exchange becomes one big rant, I won’t bother. Please keep your rants to the other thread, and lets try to have a serious discussion on this thread. But please be patient, as this is a very busy week for us.
As an occasional guest poster here, I’m hoping things tone down here a bit. I think a lot of criticism is coming from people who haven’t read the paper and is based on the use of the phrase ‘future global warming.’
I have no doubt Judith will be along to join her colleagues in speaking with us in the comments section here. But until she does, remember a few things.
Some global warming has occurred. It may indeed continue. If it does, it will affect oceans as well as land. It’s a very legitimate area of study.
Willis was first out of the gate with strong criticism. Rather than piling on, shall we wait for a response to Willis? I’m certainly curious.
As one who was in the Navy in the 70s and involved in experiments involving temperature measurements the old-fashioned way, I’m acutely aware of the possibility of measurement error. But so is Judith.
Let’s wait a bit, shall we?
Arno Arrak says:
August 18, 2010 at 7:12 am
When they start outputting Rohrschach blots and call them maps I quit. Somehow they think that because their computer can draw these blots the reader should be impressed…
Precisely, that’s the deal. That’s is what Prof.Khabbibulo Abdusamatof calls “Hollywood Science”
due to the theoretical freedom that the top physicists would seem to have—is actually quite rigid and dogmatic. There are certain things you do and certain things you do not do. Superstring theory is prestigious. Looking at basic algebra is not. Looking into the distant future is progressive. Looking at old dusty papers is not. Tying esoteric theory to time travel and science fiction and Star Trek and the Dalai Lama is au courant and cool
http://milesmathis.com/pre.html
richard telford: August 18, 2010 at 7:30 am
Since no demonstration I make against the creed would ever be accepted by the true believers here, I instead offer a simple way that Eschenbach could try to falsify the claims made in the paper instead of relying on incredulity.
In other words, you can’t address my questions without looking foolish.
He appears to prefer his incredulity, perhaps his readers do to, than risk finding out that his test of reasonableness is little more useful that extispicy. The other reason for not attempting the analysis myself is that Eschenbach finds so many thinks incredulous that there is not time in the day to test them all.
Take a deep breath and try again. Your typos preclude understanding your point.
I will however point out that the average of several measurements is much more precise that that of the original measurements.
I will however point out that your statement is not invariably correct, particularly when those measurements are spread so widely apart in both time and distance, and when instrument and calibration error combined are 250% greater than your claimed accuracy.
ZZZ says:
August 18, 2010 at 6:27 am
I began reading your post with genuine interest, but about 5 or 6 sentences in, my eyes began to bleed. Walls of text on web forums are extremely bad form. Please, in future, make use of paragraphs to better organize your virtual thoughts.
~~~~~~~
On topic, thanks to Willis for this analysis. It’s disappointing to see such a paper from Ms. Curry. I have respected her willingness to courteously engage skeptics in the past. But, this paper reeks of severe AGW overstretch placing her firmly in the camp of other odious figures in the AGW debate (Mann, et. al.).
“”” For the last half of the 20th Century, as the atmosphere warmed, the hydrological cycle accelerated and there was more precipitation in the Southern Ocean surrounding Antarctica. This increased precipitation, mostly in the form of snow, stabilized the upper ocean and insulated it from the ocean heat below. This insulating effect reduced the amount of melting occurring below the sea ice. In addition, snow has a tendency to reflect atmospheric heat away from the sea ice, which reduced melting from above. “””
Well the above was C&P directly from the Georgia Tech page.
So I read it; ten times actually; and it forms a picture in my brain. In this picture I have the Southern Ocean which is a favorite place for round the world sailors; and it consists of salty water and other stuff. So there are clouds above this salty water; and if it is cold at times it snows over the Southern Ocean and the snow falls in the salty water; and being fresh water the snow floats on the salty water. Funny thing is I have never heard of any round the world sailor who ever mentioned finding snow on the water while down there in the Southern Ocean; but I’ll Take Dr Curry’s word for it if she says there is snow on the water.
Now this snow is sitting like a blanket on top of the cold top salty water; and below that there is this warmer deeper salty water.
Suddenly out of nowhere we have some sea ice and the hot salty water down below it can’t get to the bottom of the sea ice to melt it because of the snow that is floating on top of the salty water.
Maybe I have it all wrong; perhaps the snow is on top of the sea ice rather than the salty water; and the hot ocean salty water is below the sea ice, but can’t get heat to the bottom of the sea ice to melt it, because of the snow sitting on top of the sea ice.
I’m going to go and get a cup of Capuchino; and then I’m going to come back and read that paragraph from Georgia Tech for the eleventh time; because I know there’s a trick in there somewhere !
Tom Fuller, global warming will affect oceans and land? That is an interesting statement. If made to say that the flora and fauna IN the oceans and on land will be affected by global warming, I would not disagree. We certainly know that oceanic temperature oscillations have direct and predicable affects on sea life.
However, it you mean to say that anthropogenic global warming will affect oceans and land in terms of their temperature, I cannot go there with you. The ONLY way oceans can anthropogenically warm at the depths typically measured for ENSO purposes is through longwave radiation, since greenhouse gasses do not absorb and emit shortwave. However, the equation for longwave heating of ocean bodies down to about a meter (while LW is generally considered a surface only penetrating energy source, ocean mixing could theoretically send some small fraction of it down below the water’s surface) results in warming that will not be enough to change the temperature outside the normal 1/2 degree bands. Thus AGW cannot affect oceans.
http://www.ce.utexas.edu/prof/hodges/site2006/documents/thermodynamics.pdf
Dr Curry and Dr Liu – Many thanks to both of you for coming to the forum, rolling up your sleves, and sitting down with the WUWT Crowd. It is perhaps a very disturbing initial experience but one that offers so much in return. There are very few places where you are welcomed with open minds, arms, and hearts and challanged and torn apart for what you say at the same time -quite like your own immediate families. Be assured that we hold great respect for each of you. Rest assured we will challange everything you say. What’s the benefit to you? Let’s call it an educational experience that very few have the guts to even think of attempting. I assure you that you will be the better for it. In giving so much of yourself you will gain so much in return.
PS: When dealing with the WUWT Crowd you will no doubt meet a few you’ll wish you hadn’t. They too teach.
Please, everyone, read Judith Curry’s and Tom Fuller’s comments above and think a little before weighing in.
Let’s try and bear in mind what Judith Curry has been trying to do recently and demonstrate that we get the point here. Rational, constructive debate is the way forward.
I have lurked here for a long time but never post (I hear best when not talking), but this has prompted me to break silence.
Dominic
Judith Curry says:
August 18, 2010 at 8:18 am
Dr. Curry, you have always been open minded and fair in your work and published papers. While some here have been derogatory towards you in light of the Eschenbach critique, many have also defended you. Anthony has a very open policy in regards to comments, so both will appear.
I hope you take into consideration that not all commenters are Anthony or Willis, and that some are going to be more ascerbic than others in their comments, but your commenting is always a welcome addition to the posts.
Thank you for clarifying your paper, and for commenting. I look forward to your posts as they seldom (if ever) seem to be defensive, and mostly are explanatory.
And another thing – manners. Dr Liu is kind enough to drop in and provide more info about this paper and gets dealt with rather snappily. I think someone should have the courtesy to say: ‘thanks for taking the time to contribute to this thread Dr Liu. Much appreciated. Now, about the potential drawbacks of an orthogonal modeling approach with such sparse SST data… etc’
There, not so hard.
Dominic
I can’t resist commenting on this, even though I have seen some reference to it in Anthony’s original thread:
From the GT Press Release:
“This increased precipitation, mostly in the form of snow, stabilized the upper ocean and insulated it from the ocean heat below. This insulating effect reduced the amount of melting occurring below the sea ice. In addition, snow has a tendency to reflect atmospheric heat away from the sea ice, which reduced melting from above.”
This just does not make sense to me. Does snow behave differently in Anarctica than it does in the US? In my experience, when snow falls on water, even water at 0 degrees C, the water absorbes into the snow crystal, making it ice and water, the whiteness of the snow and thus its reflective properties are nulified. Snow’s insulative properties are at least partly due to the air incorporated into the structure it forms when packed, which again gets nulified when it fills with water.
I note the worthy defence of Judith Curry by some posters who claim that because she is amenable to communicating with sceptics she should be somehow immune from criticism and her work should be immune from comment. This response is I feel mistaken and does not take into account the nature of this site.
The obvious flaws in this paper are so obvious that they really do merit comment, I mean an accuracy of +- 0.06 oC on the back of such sparse data does seem on the face of it a ridiculous claim does it not?
If the layman and interested amateur can pick up the flaws so easily and quickly then it has to said that in laymans terms the paper ‘aint all that’ is it? Here I think we see the inherent beauty of this site in operation, we have interested ordinary people able to access and view and dissect scientific papers and learn from others as they do, this is no dusty boring ivory tower where the new climate science high priests can discuss how many angels can dance on the head of a pin or churn out half arsed or ill thought out work without fear of criticism. This site is real and immediate and wholly accessible to ordinary people very keen to understand what is happening in climate science today.
The simple fact is that climate science has been able to influence society via a largely tame media while the actual climate science ‘product’ has largely been immune from harsh criticism and any sort of quality control and this has produced a field of endeavour wholly unable to accept or learn from or respond to critical analysis from within the ranks of science at large let alone from the ordinary people so greatly affected by the work of the climate science community.
Lets face some facts shall we? Climate science has grown too fast,become too rich and been able to access and control the MSM channel to the population at large and in the headlong rush to present a (un)scientific consensus it has produced some very poorly researched and produced work reliant in large measure on computer models and statistical manipulation of raw data.
Climate science is being forced to confront its own deeply ingrained flaws and failings and it is being exposed not by a compliant media or others in the climate science community, it is being forced to confront these inerrant failings by ordinary people and amateurs and outsiders who have no stake in the cosy and insular world of climate science.
I can understand the reluctance to address ones own failings and flaws and shortcomings, few enjoy the experience of having their work criticised by peers let alone ordinary people, yet this cold shower is exactly what climate science needs whether they realise it or not!
The more I learn the more I realise how little I truly know which instills in me the desire to learn more, I have no wish to destroy others but I will not stand by and say nothing if I see fault and error in others.
“”” Bill Tuttle says:
August 18, 2010 at 8:54 am
richard telford: August 18, 2010 at 7:30 am
Since no demonstration I make against the creed would ever be accepted by the true believers here, I instead offer a simple way that Eschenbach could try to falsify the claims made in the paper instead of relying on incredulity.
………………….
…………………..
(Richard)
I will however point out that the average of several measurements is much more precise that that of the original measurements.
(Bill)
I will however point out that your statement is not invariably correct, particularly when those measurements are spread so widely apart in both time and distance, and when instrument and calibration error combined are 250% greater than your claimed accuracy.
Well an average is just that; an average. Unless the quantity being measured is stationary; meaning that it’s real value is presumably not changing; then one would expect the value of that quantity to change from time to time, and hence from measurement to measurement. So each of those measurements if they had infinite precision would be expected to have different values. and those real quantities could be represented with high precision by the measured values, if the measurement method is accurate enough. There’s no reason to expect the average of several readings of a variable quantity to be more correct than each of those readings was at the time it was taken.
We can say; due to an argument by Galileo Galilei, that at some time between the first measurement, and the last measurement, there must be an occurrence of any possible value that lies between the extreme high and low values that were observed; and hence there must be a point at which the variable quantity had a value that exactly equals the average of the measured values.
But unfortunately; we have no idea when that occurred; so the most we can say is that at some point the value will equal the average; but there’s no reason to believe that is a better number to use, than the actual observed values were at the time they were measured.
If the quantity being measured is not a variable but some fixed value such as Boltzmann’s Constant for example; then perhaps the average of a number of measures of it could be more precise; well unless our measurement system had some systematic error that we were not aware of. But I would hardly say the average was much more precise, since if the errors were randomly distributed; then at best we would expect the error of the average to diminish by the square root of the number of measurements; assuming of course that nothing changes during that number of observations.
Statistics can give you a different answer from what you measured; it isn’t automatically a better answer.
George E. Smith says: August 18, 2010 at 9:28 am
George, I think you are on to something here. Every time I have put snow on water, the snow gets wet. It’s the darndest thing! On the occasions when the water was of a temperature such that the snow did not melt, the snow acted like an iceberg with a small fraction above the surface of the water. But that which was above the water was still wet. And to think for the past fifty years I thought the insulating capacity of snow was the air between the flakes. Boy, did I have my head screwed on wrong!
Regardless of the paper’s insights and ultimate research value, it is legitimate to complain about the title – “Accelerated warming of the Southern Ocean and its impacts on the hydrological cycle and sea ice” – in light of Dr. Curry’s comments:
The two don’t match up in at least a couple of ways:
1. Differential warming, maybe; “Accelerated” warming is unproven until the data are sufficiently tested for reliability and uncertainty.
2. It should be “modeled impacts”, not just “impacts” which implies actual measurements over time rather than computer-generated “what-ifs”.
We should be patient to hear a defense of the paper, but frankly, it looks like it came out of the oven undercooked with the nutritional values improperly labeled.
These people are absolutely desparate to show warming in Antarctica. Explain to me how warming water leads to the highest ice extent on record? Like Steve said, a complete 3 ring circus, and PT Barnum was right all along…
Jiping Liu: Thanks for stopping by.
You wrote, “The corresponding principle components (time series) show a substantial upward trend (statistically significant at the 99% confidence level). Thus, the observed SST pattern in the Southern Ocean during the second half of the 20th century is dominated by a broad-scale warming that accounts for one third of the total variance (28% for HadISST and 29% for ERSST).”
That may be true, but the SST anomalies of the mid-to-high latitudes of the Southern Hemisphere (90S-40S), the Southern Ocean in you paper, have been declining since the mid 1990s.
http://i36.tinypic.com/am83mb.jpg
Also, since the vast majority of the Southern Ocean data is manufactured by NCDC and Hadley Centre, aren’t you really performing PC analyses on data that was created using similar statistical methods? A quick look at the spatial coverage of ICOADS data reveals that almost all of the data prior to the satellite era was manufactured. And since the ERSST.v2 dataset does not use satellite data, the NCDC is still infilling most of the data.
http://i37.tinypic.com/t8x4ox.jpg
Bob Tisdale, thanks as ever for providing cogent data analysis. I was rather hoping you would have something to say about this.
Dominic
I think we should mind our manners lest this thread degenerate into the insults, name-calling, vitriol, ad homs, etc. of RC, CP and OM. I, for, one, appreciate the willingness of Dr. Curry and Dr. Liu to make their work available and contribute to this blog. There are certainly points in the paper that need to be discussed/explained but this should be done in a civil, collegial,and objective manner.
This has to be a deliberate [snip] by Judith Curry to see if the MSM/BBC etc. will adopt this news and run with it with NO Critical thought. She must have known it would get a critical review here !
Judith Curry: You wrote, “With regards to my concerns about the quality of the SST data, I have only been digging into that the last several months.”
In my comment above to Jiping Liu, I linked maps of ICOADS observations from 1950 to 2000 for the Southern Oceans, south of 40S, for Januarys in 1950, 1960, 1970, 1980, 1990, and 2000. The source data is basically nonexistent:
http://i37.tinypic.com/t8x4ox.jpg
If the “what if” experiment is a model, I would think that it should be reviewed by a University level statistician prior to publication. Just sayin.
Judith Curry and Jiping Liu,
Thanks for joining the discussion. I am sure that you have developed a sufficiently thick skin to deal with the less pleasant comments.
The press release is misleading. There is no “paradox” – your own data shown above and in fig 3a of the paper shows cooling near the poles, and the land data shows no warming except on the peninsula. So there is no need for contorted arguments about how warming can produce more ice!
The title of your paper is also misleading. There is no ‘acceleration’. Again, your own data, fig 2a and fig 3c in the paper, show that the warming is DEcelerating – in fact virtually flat in the last decade. The only acceleration is in the computer models that merely reflect the beliefs and assumptions of the climate scientists who designed them.
Then there is the central issue that Willis raises. Why do your graphs show warming when his, directly taken from the data, show no warming? Why is it that climate scientists are so reluctant to present the raw primary data, preferring instead to put it through some mangle like EOF, SVD or PCs? If there are holes in the data, fine, show this. It is almost as if you are deliberately trying to arouse suspicion. When will climate scientists learn that is their own behaviour that is primarily responsible for increasing levels of skepticism?