NOAA's August global SST record is the result of one data set

Oookaay.
Soo…
How did they determine that the satellite data was biased, and not the other data source(s)?
Or were the simply cherry picking the data sets to confirm their own “observer” bias?

I guess my comments refer to their figure 1.

When are the NOAA climate scientists going to be “adjusted for bias”?

Please tell me you contacted them so they can “fix” their mistake?

The average global temperatures over the last ten years were showing a cooling trend, but I have discovered that this is because night-time temperatures were showing a downward bias. I have now removed them and I can now report out-of-control warming of the planet. It’s worse than we thought.

Stupid question here, but wouldn’t increased sea temps be indicative of the oceans releasing heat into the atmosphere and therefore cooling?

ooops, the term “increased sea temps ” should read “increased sea surface temps “

The temporal geography is a major clue that is being overlooked in the climate discussion:
“The bias was strongest in the middle and high latitude Southern Hemisphere”
How many blatant clues of this magnitude will it take before alarmists start considering natural factors in climate as factors in climate? (Maybe they are reading the literature selectively?…)

George E. Smith: You wrote, “Scuse me for asking, but what kind of BS graph paper are they using; that looks like some kind of phony logartithmic x axis scaling…”
Graph paper? It’s simply a standard time series graph that’s the output of EXCEL. I prepared it, not they.
Then you qualified your comment with, “I guess my comments refer to their figure 1.”
The source is listed in the post:
NOAA’s ERSST.v3b SST data was downloaded from the KNMI Climate Explorer:
http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere
Have at it.

Jeremy,
“In short, this argument is meaningless from all sides. As a believer, you cannot use it to demonstrate anything of value with regards to global temperature.”
No it is not meaningless. When a trusted government research organisation at one time merges data sets to get the desired trend, and later removes one of these datasets because “it showed less of a trend”, then It is not what it tells us about climate change so much as what it tells us about NOAA that this story is about.

Just a funny comparison for me… I got my M.S. in criminal justice, and one of the big problems that has occurred is how police departments across the country record crime numbers. What cheifs and politicians like to do is start recording crimes more stringently when they are going to start a new program i.e. community policing and then once they get the funds for said program begin “down grading” crimes. For example, if I stole a $1100 TV, they would report that as a financial loss of $700, so that it would reflect misdemeanor theft rather than felony theft. So in 5 years, they can show their amazing graphs showing how their program fixed the crime problem. If you really want to get depressed, check out what happened in Philadelphia in 1953.

Bob:
You have such a good grasp of things, and you are FAST on the “uptake” for various concepts, so I’m going to propose something here which I think needs to be considered.
I have been monitoring the NASA http://neo.sci.gsfc.nasa.gov/Search.html
website and find that the IR energy balance has been “negative” for some time.
I’ve discussed this with Dr. Spencer and he has noted that this is correct with the observed atmospheric cooling. He said, “It’s taken some number of hours, and running of a variety of modeling programs..but despite how counter intuitive, the negative IR balance is commensurate with the tropospheric cooling”. (I’m repeating this on memory right now.)
In any case, I have also asked Dr. Spencer if there is something to cause the oceans to become transitorily better absorbers. He has not had a chance to think this out (he really is a great THINKER and doesn’t respond “off the cuff” much of the time, because what he responds has THOUGHT and EFFORT behind it!).
But my (superficial) analysis is this: High sea surface temps = result of greater ABSORPTION around the globe. Accounts for negative energy balance currently observed.
Now here’s the kicker: This does not mean higher atmospheric temps.! Because there is a LOT of ocean out there below the “sea surface temp” region.
Thus a LOT of “thermal well”. Our current energy input to the sea surface could, and probably WILL be “swallowed up” by the oceans.
SO what we may have here, in the long run, is an even GREATER model/example showing the DOMINATING effects of WATER on the atmosphere, and the trivial nature of CO2. Water both in the vapor form and in the LIQUID form.
Mark Hugoson

The addition of satellite data caused problems for many of our users.

Has anyone requested details on that claim? FOIA on the communications?

Kelvin waves are a part of El Nino. They start out warm, throw that warmth into the air, and end with a cool tail as they propagate across the ocean. Kelvin waves are a good way of understanding that El Nino’s throw heat out of the ocean while La Nina’s hold it in under cool upwelled surface water. Kinda like a heat inversion where colder air above keeps warmer air trapped below.

Had the satellite data showed warming then they would have left it in the results for sure.

Anthony…
THE RESIDAL BIAS LED TO A MODEST DECREASE IN THE GLOBAL WARMING TREND AND MODIFIED GLOBAL ANNUAL TEMPERATURE RANKINGS.
Isn’t it residual?

Higher SST area means cooling oceans. Now it is for sure: There won’t be any El Nino this year:
http://weather.unisys.com/surface/sst_anom.html
If cloud cover keep increasing, along with GCR:
http://cosmicrays.oulu.fi/webform/query.cgi?startdate=1964/04/17&starttime=00:00&enddate=2009/09/17&endtime=00:26&resolution=Automatic choice&picture=on
SST colours will be blue again and then we’ll begin to miss those “old days” when NOAA’s maps were shining with red, oranges and yellows…

Mark Hugoson,
I’m not sure I understand your argument, but to clarify, are you saying there is ‘missing’ IR which is going into the oceans due to some unknown transitory effect?
Otherwise, energy is never ‘swallowed up’ by the oceans. Energy (heat) flows from the oceans into the atmosphere. Additional energy in the oceans will eventually be released into the atmosphere, all else being the same.
ENSO is known to have substantial effects on tropical ocean biology. It’s possible that biological effects are large enough to effect the (average) depth at which sunlight is absorbed in the ocean, and hence effect SSTs and the rate heat (from sunlight) is released to the atmosphere.
My understanding is that the reason ocean absorption of IR is minimal is due to the physical properties of water. I don’t see how a transitory effect could change this.

Let’s see… how can we get back to our global warming charts??? Yeah, let’s take these satellite data out… still not enough… let’s now remove those years… yes, we are getting there… ok, we will now square 1997 and repeat that we now have a full El Nino event even if the map is mainly yellow and blue with only tiny bit of red… WE HAVE IT FOLKS!!! Now let’s publish this result for August.

“”” Bob Tisdale (11:07:55) :
George E. Smith: You wrote, “Scuse me for asking, but what kind of BS graph paper are they using; that looks like some kind of phony logartithmic x axis scaling…”
Graph paper? It’s simply a standard time series graph that’s the output of EXCEL. I prepared it, not they.
Then you qualified your comment with, “I guess my comments refer to their figure 1.”
The source is listed in the post:
NOAA’s ERSST.v3b SST data was downloaded from the KNMI Climate Explorer:
http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere
Have at it. “””
Well thanks Bob; actually I subsequently guessed that the actual graphs migh be your plotting of their data. But what happened to the minor grid lines; am i just seeing them faded out, and not printed; or what.
And no disrespect to you was intended; reason for question on raw data, was to learn how many data points were actually available for each “cycle” of that weird fig one paired cycles graph. To what do they ascribe the fact that one cycle has a big swing, and the next just a minimal p-p range ?
The later graphs such as 4/5/6 I have no probelm with other than the minor time increment lines, but #s 1 and 3 both have that strange look about them; that is hard to equate to more or less random data.
Thanks for the link, I’ll go look to educate myself.
George

“My opinion/suspicion is that NOAA (along with other government agencies) are understaffed and overworked and are doing the best they can.”
From 2008, “NOAA’s total budget is $4 billion, with $1 billion of that going to all weather-related agencies”.
http://www.keysnet.com/210/story/17985.html
“R&D in the National Oceanic and Atmospheric Administration (NOAA) would climb 9.9 percent to $585 million in the House instead of a slight requested cut (see Table). The largest increases would go to oceans and climate research in NOAA’s Oceanic and Atmospheric Research (OAR). OAR R&D would climb 23 percent to $346 million; the Senate would provide even more for these programs.”
From here, ref their 2009 budget, http://www.aaas.org/spp/rd/doc08h.pdf
I don’t think funding is the issue.

Bob,
When I asked in the other thread if satellite SST data was removed, I was referring to whether it was removed from the ERSST data set last November.
In other words, did the removal affect only this year, last year or the entire record?

I do not understand why so many of you are compaining about the graph!
It is very clear and easy to visualize that a line fits half way between
the wide ones. Look at the year markings and see that the x axis
is linear, nothing important is missing.
If some of you are having that much trouble then print the graph out and
draw the lines in, after all most of your are the scientists, I’m not, and
I can make sense of it.

George E. Smith: You asked, “But what happened to the minor grid lines; am i just seeing them faded out, and not printed; or what.”
The minor grid lines appear fine to me. The graphs are reduced in size here at WUWT, which is one of the reasons I provide the direct tinypic links for the full-sized versions after each graph.
You wrote, “The later graphs such as 4/5/6 I have no probelm with other than the minor time increment lines, but #s 1 and 3 both have that strange look about them; that is hard to equate to more or less random data.”
The title blocks and the text in the post for Figures 1 and 3 state that they are SST data (not SST anomalies). The title blocks and text in the post for Figures 4, 5 and 6 state that they are SST anomaly data.
You asked I assume with respect to Figures 1 and 3, “To what do they ascribe the fact that one cycle has a big swing, and the next just a minimal p-p range ?”
George, “they” don’t explain the data. But here’s a similarly sized NINO3.4 SST anomalies graph for the same time span. Are you seeing the effects of ENSO?
http://i25.tinypic.com/5mc0wk.png

DR: You asked, “When I asked in the other thread if satellite SST data was removed, I was referring to whether it was removed from the ERSST data set last November.
“In other words, did the removal affect only this year, last year or the entire record?”
I misunderstood because in the other thread you had asked if they deleted the OI data. I would assume that the removal of the satellite data from ERSST.v3b should only have impacted the satellite period from 1981 to present. However, there were also other changes made between ERSST.v3 and ERSST.v3b well before that period that have not been explained.
http://i43.tinypic.com/6yfx8h.jpg
From my post here:
http://bobtisdale.blogspot.com/2009/05/recent-differences-between-giss-and.html

I’ve just noticed that the ERSST.v32 and the ol.v2 datasets have a strong seasonal signal. According to them, average SST increases by 0.5C from NH winter to NH summer. Which is strange because the SH has a lot more of the world’s oceans than does the NH, and consequently a seasonal signal should be in the opposite direction.
Which makes me think, there are large hemispheric biases in these datasets.

bugs (17:56:13) “I see a steady, long term temperature rise in all the graphs.”
What causes the “steady, long term temperature rise” you see?

bugs, you see a small short term rise (30 years) in temps. The question is, do such rises of such duration come and go historically? Is there anything unusual or alarming? Or is this like looking at the period June-October and saying that you see a definite steady decline? Without realizing that the climate is seasonal?
I don’t know. But I don’t think there is much point getting too worried about small 30 year trends UNLESS there is evidence something unusual is going on in a timescale of centuries or millenia.

Hey, I think I am now catching on to this stuff.
Now that I have my grid lines back, I can see that the “global SST” graphs (1 & 3) have a big dip almost right on the year lines, and a small dip half way in between; so can I conjecture that the big dip is the SH winter, and the little dip, is the NH winter; winter in this case being defined as the time of lowest SST (locally).
And yes I would expect the anomalies to wander aimlessly around as they seem to.
Dang it is hard keeping up with you guys. Of course I’m still not sure what it all means, since everybody seems to have his favorite data set.
Hey I like Gaia’s data set; she gets it all right everywhere all the time. Gaia is a really great exeperimental climatologist.
IMHO.
George

Yes, Bugs, no one debates that temps in the last ten years have been hotter than the previous 20. However, temps in the last 10-13 years haven’t changed much, and seem to be on a declining trend, which frankly, worries me a good deal more than any warming.
In many areas, average temps increased by up to 2 deg in the 1920-1940 period. So did the polar bears die? Was it a catastrophe? Doesn’t seem to have been. Why then, if allowing it would happen, would another 2 deg matter now? What’s different from 1920? Is there something ‘special’ about current temperatures?
OT: Am I missing something, or are there only four posts on Real Climate in the last month? Is anyone manning the barricades?

Yes, Bugs, no one debates that temps in the last ten years have been hotter than the previous 20.
Is this whole surface stations project a waste of time?
REPLY: Is determining the quality of any set of measurements a waste of time?

Funny how we are expected to ignore 30 year warming trends but jump for joy at the current two year trend of Artcic ice extent.
So let’s also ignore the 30 year trend and look at the 100 year or even 200 year trend.
See where I am going?
And if we are to accept that temperatures should be expected to rise with the current weak El NIno why should we not also expect temperatures to cool after the strong El Nino of 1998? I can’t get my head round that one.
In my opinion it’s time the cherry pickers and nit pickers packed it in and looked at the long term benefits of global warming. The opening up of the North East passage and possibly soon the North West passage to commercial shipping will be immensely valuable to global trade. And look at all that fertile frozen land in Canada and Russia ready for cultivation.

What interests me is that there clearly has been a significant rise since 1980, and that it is not simply turning right back down although we know the Sun is in a blue funk and land temperatures are down. These graphs are an easy message for warmists to interpret as “it’s obvious” etc. But I want to look at another factor: slow thermal inertia buildup effects from the ocean. This also shows up IMO in Ferdinand Engelbeen’s observation that although recent Greenland summers have been cooling, the Greenland mean temp has not really changed. So that when the Sun starts to “cool” the cooling surface is still overbalanced by the ocean depths returning the signals of warming. Like Akasofu described in his pic of a 60-year (?) cycle overlaying a steady return from the Little Ice Age. We know that the ice cores show CO2 changes lagging temperature changes by ~800 years which would seem to correspond to the thermal inertia effects of the global thermohaline current.

i was cooking tea tonight . I put some fat into my frying pan put the pan on the element truned the power on .After about 3min the fat started to melt. I turned off the power after 6min and the fat still was melting . It was about 30min after i turn the power off that the fat stoped melting and started to harden again.It was back to it normal state in about 70min .
So may be after a time when the sun was giving of more heat. That it has gone less active and we are seeing as it were the fat still melting . And say 7years later the sea temps get back to normal?. ( 6min of xtra heat. then power off takes 70min to get back to normal)

Corrections to previous post (I hit the submit early)
“rcord” should be “record”
“strenghten” should be “strengthen”
The last line, i.e. “Following the 1940s PDO shift” should read
Following the 1940s PDO shift the temperatures dropped fairly quickly. By 1951, most of the cooling had already taken place. We need to face up to the fact that a (possible) CO2 signal may reveal itself in the coming years.

I hope you don’t mind Anthony. I posted this in an earlier thread but I see you have moved on.
From the NOAA press release, just in time for Copenhagen.
Expect more research corruption and spin as the Copenhagen free loaders convention draws near.
What the NOAA data distorters didn’t want you to see.

The June-August worldwide ocean surface temperature was also the warmest on record.

Only if you remove this.
http://i599.photobucket.com/albums/tt74/MartinGAtkins/UAH-Jun2Aug.jpg

The worldwide ocean surface temperature of 62.4 degrees F was the warmest on record for any August,

Ditto.
http://i599.photobucket.com/albums/tt74/MartinGAtkins/UAH-Aug.jpg

John Finn: “We’re a long way into this minimum and there’s been no discernible change. Peak activity was in the early 1990s. SC23 was nothing to shout about yet the warmest years have all been recorded during that cycle.”
=============================================
As another post alluded to in it’s discussion of melting fat, the climate doesn’t have an instaneous response to solar activity levels. Another key factor is the DURATION of activity peaks and valleys, not just their magnitude. If you look at cosmic ray levels (as modulated by solar activity and as proposed by Svensmark as a key driver of climate), it is interesting to note the peaks and valleys and their duration:
http://tinyurl.com/q2mpkg
What’s interesting is when you track the cumulative difference between the point in time cosmic ray level and the average. If Svensmark is right, you would expect temperatures to be higher as the inverse of the cumulative difference increases (more rays, more clouds, cooler earth). Well looking at the results (plotted from the start of a posttive trend in late 1966):
http://www.geocities.com/mcmgk/Cumulative_CR_Inverse.jpg
this is what you get:
– You’d expect temperatures would have peaked in the early 1990’s which just might have happened had it not been for Pinatubo.
– Then you get another peak through the early to mid 2000’s before you’d expect temperatures to start dropping off over the last couple of years (which they’ve done)
Of course there are other factors in play (ocean thermal lag, PDO/AMO, El Nino’s and La Nina’s) but the bottom line suggests that focusing on a solar peak in the early 1990’s is the wrong perspective.
Given that CR levels are still at low levels not seen since the Oulo record began, I would expect more cooling once the current El Nino fades. After all we hit a UAH anomaly of zero in June before the El Nino kicked in!

“”” Paul Vaughan (23:07:23) :
Re: George E. Smith (18:48:18)
The big dip is NH winter.
You might find figure 1 here interesting:
Each month gets a different average when you change base-periods …so it’s not just one shift of the whole curve, but rather 12 shifts. “””
Thanks for the clarification Paul, and the nifty links; but now I am totally confused.
I thought that instead of recording real temperatures (like in Kelvins or even deg C) that Climatologists like to plot anomalies, being the difference between the actual measured (anomalous) temperature, and the “correct” temperature that it was supposed to be based on the average over some 30 year time period (base period).
If I understand what you seem to be saying now, is that the current temperature is essentially compared to what it was on the corres[ponding day some arbitrary time from the past, and that the “expected” true val;ue changes for each data point, daily, weekly monthly annually or whatever.
Why on earth would anyone suppose that the sequence of actual temperatures from some past era was in any way “normal” ?
No wonder it is impossible to relate any of this temperature data, to thermal energy flows, or air/ocean mass flows that result from real time temperature differences across the planet.
The whole thing sounds to me like an exercise in self flaggelation.
George

Mark (09:01:44) :
John Finn: “We’re a long way into this minimum and there’s been no discernible change. Peak activity was in the early 1990s. SC23 was nothing to shout about yet the warmest years have all been recorded during that cycle.”
=============================================
As another post alluded to in it’s discussion of melting fat, the climate doesn’t have an instaneous response to solar activity levels. Another key factor is the DURATION of activity peaks and valleys, not just their magnitude. If you look at cosmic ray levels (as modulated by solar activity and as proposed by Svensmark as a key driver of climate), it is interesting to note the peaks and valleys and their duration:
I’m not suggesting it should have an instantaneous response – though in some of the solar/climate reconstructions the response is pretty near instantaneous, and in the case of the Dalton Minimum temperatures actually fell before the DM started. The so-called link is iffy.
The Little Ice Age is supposedly due to low solar activity. If true, then the LIA should be clearly defined – except it’s not. There is no firm agreement when it started or when it ended.
http://tinyurl.com/q2mpkg
What’s interesting is when you track the cumulative difference between the point in time cosmic ray level and the average. If Svensmark is right, you would expect temperatures to be higher as the inverse of the cumulative difference increases (more rays, more clouds, cooler earth). Well looking at the results (plotted from the start of a posttive trend in late 1966):
Svensmark appears to be the last throw of the dice. But the trend in GCR is just not there. Ok there has been a slight rise in recent years but the GCR plot looks absolutely nothing like the surface (or satellite) temperature plot.
http://www.geocities.com/mcmgk/Cumulative_CR_Inverse.jpg
this is what you get:
– You’d expect temperatures would have peaked in the early 1990’s which just might have happened had it not been for Pinatubo.
– Then you get another peak through the early to mid 2000’s before you’d expect temperatures to start dropping off over the last couple of years (which they’ve done)
What are the cumulative values as a percentage of the average values. I suspect some selective scaling has gone on here. I’m not convinced your link means very much.
Of course there are other factors in play (ocean thermal lag, PDO/AMO, El Nino’s and La Nina’s) but the bottom line suggests that focusing on a solar peak in the early 1990’s is the wrong perspective.
Given that CR levels are still at low levels not seen since the Oulo record began, I would expect more cooling once the current El Nino fades. After all we hit a UAH anomaly of zero in June before the El Nino kicked in!
An anomaly of zero simply means the average temperature for June in the 1979-1998 period. Also, bear in mind that there appears to be an odd dip in May/June anomalies in the UAH record. John Christy has posted on this issue on WUWT. In any case, any temperature dip was due to the lagged response to the near La Nina conditions a few months earlier.

John Finn (11:03:30) : “the LIA should be clearly defined – except it’s not. There is no firm agreement when it started or when it ended.”
Based on what? The Team? How do you know that the start and end dates of the LIA are “disagreed upon”? What conflicting evidence doesn’t point to anything specific? Sorry, but having read CA I have seen a lot of bunk demolished about the paleoclimate of the last thousand or so years…

Lucy Skywalker (00:37:21) “lagging […] by ~800 years “
I recommend being careful with that unless you’ve run some really thorough diagnostics on a battery of statistical analyses. (Cautionary note: People in these threads toss around misguided ideas about lags on a regular basis. Bear in mind that most of the lags being discussed have a distribution – and that, furthermore, that distribution is not random.)

Andrew (12:55:43) :
John Finn (11:03:30) : “the LIA should be clearly defined – except it’s not. There is no firm agreement when it started or when it ended.”
Based on what? The Team? How do you know that the start and end dates of the LIA are “disagreed upon”? What conflicting evidence doesn’t point to anything specific? Sorry, but having read CA I have seen a lot of bunk demolished about the paleoclimate of the last thousand or so years…
Ok – when was it? Define the period for the LIA.

V Rig (18:02:24) :
John Finn: I am referring to the below graph John. I did not do a least square fit but after 40 years the temp. anomaly is currently .2C from the average. A straight line fit to the data gives about .15-.2C departure from the average over the time period.
It’s actually ~30 years – not 40. The current August anomaly is +0.23 above the 1979-1998 average. However, while most of the post-2000 anomalies are ~0.2 above the average, most of the pre-2000 anomalies are below the average. Since the start of the record, there has been an increase in temperatures of between 0.3 and 0.4 deg. UAH state that the decadal trend is ~0.13 deg. You can see this if you scroll down to the bottom of this link
http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt
At 0.13 deg per decade the total warming over the 30 year period is 0.39 degrees. GISS, Hadley and RSS show ~0.5 deg warming over the same period.

Philip_B: You wrote, “I realize I am getting off the seasonal signal in SSTs. Just trying to understand how it all works.”
A good reference page: The NOAA Physical Oceanography Division
Journal Publications webpage has links to more than 90 papers, many of them covering SST variability and the impacts of atmospheric circulation on SST.
http://www.aoml.noaa.gov/phod/docs.php

Bob Tisdale (01:50:54) ” http://www.aoml.noaa.gov/phod/docs.php “
Thanks for that link Bob – found this:
Wang, C.; Lee, S.-K.; & Mechoso, C.R. (2009). Inter-hemispheric influence of the Atlantic warm pool on the Southeastern Pacific. Journal of Climate, in press.
http://www.aoml.noaa.gov/phod/docs/Revision_Wang_etal_2009.pdf
“[…] the effect of the anomalous AWP is to strengthen the regional Hadley-type circulation from the AWP region to the SEP […] may play a role in the initiation of a cool phase of ENSO.”
“Almost of all state-of-the-art coupled ocean-atmosphere models exhibit serious errors in the form of a severe warm bias in simulated SSTs over the SEP […] Many state-of-the-art climate models suffer from serious climate drift in the annual cycle of atmosphere-ocean processes in the AWP region.”

Re: Lucy Skywalker (01:32:31)
Hi Lucy, My comment is about lags in general. As for the bit about “non-random”: Too many people are eager to cut-corners with untenable assumptions of randomness. (The discipline of statistics needs to keep itself highly-valued – we should all understand their need to market the dogma, which floats on a raft of assumptions and only rests upon a solid foundation in the abstract (i.e. imagination). The stuff can be useful for some purposes, but it isn’t necessarily the best way to make decisions about geophysical phenomena that are not well-understood.) To keep this exchange practical, I suggest using qualifiers like “roughly” & “generally” to deflect potential criticisms.

Re: Chris Knight (14:33:28)
Chris, I dug around to find some Sea Ice time series. When I compare them with LOD I find that generally LOD leads by about a month. The factors affecting LOD (e.g. AAM) appear to go beyond just sea ice. This exchange has been helpful in sharpening holistic perception and developing new questions.
–
Polar Sea Ice Data:
http://polynya.gsfc.nasa.gov/seaice_datasets.html

Re: Chris Knight (05:28:40)
Based on your comments, I suspect you may find the following papers interesting:
Sidorenkov, N.S. (2005). Physics of the Earth’s rotation instabilities. Astronomical and Astrophysical Transactions 24(5), 425-439.
http://images.astronet.ru/pubd/2008/09/28/0001230882/425-439.pdf
Zhou, Y.H.; Yan, X.H.; Ding, X.L.; Liao, X.H.; Zheng, D.W.; Liu, W.T.; Pan, J.Y.; Fang, M.Q.; & He, M.X. (2004). Excitation of non-atmospheric polar motion by the migration of the Pacific Warm Pool. Journal of Geodesy 78, 109-113.
http://202.127.29.4/yhzhou/ZhouYH_2004JG_PM_Warmpool.pdf
http://www.shao.ac.cn/yhzhou/ZhouYH_2004JG_PM_Warmpool.pdf (alternate)
[I think you’ll recognize the pattern in the graph at the top of the 3rd page.]
Yan, X.H.; Zhou, Y.H.; Pan, J.Y.; Zheng, D.W.; Fang, M.Q.; Liao, X.H.; He, M.X.; Liu, W.T.; & Ding, X.L. (2002). Pacific Warm Pool Excitation, Earth Rotation and El Nino Southern Oscillations. Geophysical Research Letters 29, 15685-15690.
http://adsabs.harvard.edu/abs/2002GeoRL..29u..27Y
Also, see the works of Yu.V. Barkin.

sorry, not AAM, but global sea ice extent

Warmmmmmmmmest oceannnnnnns in 120 yearrrrrrrrrrs!
Read all about it!

Paul, when these threads get pushed down the list, they begin to be like private chats!
Sidorenkov’s book:
http://eu.wiley.com/WileyCDA/WileyTitle/productCd-3527408754.html
The price is a bit steep for me, but I will order it from my public library.
Another book to read for a historic view from 30 years ago might be:
The Earth’s Variable Rotation: Geophysical Causes and Consequences By Kurt Lambeck
As for Barkin, I’ve read “UNITY OF RHYTHMS OF EARTH ROTATION, GRAVITY AND GEODESY VARIATIONS: THEIR NATURE AND INVERSION PHENOMENON”, he has papers on many aspects of planetology – have you any specific recommendations?
In the few works I have read, neither of the above authors seem to count sea ice, as opposed to the Antarctic and Greenland ice sheet variations, as being a major player in the shift of rigid mass from pole to pole and thus the inertia of the rotating earth. See fig 1 here:
http://syrte.obspm.fr/journees2004/PDF/Sidorenkov.pdf
The difference between the observed and theoretical curves again shows the pattern we discussed above.
There is however a move away from core-mantle-lithosphere differentials (which are not directly measurable) to surface hydrosphere-cryosphere-atmosphere interactions (which are becoming more so) affecting the rate of the earth’s rotation.
Walter Munk has long commented on the role of the moon in tidal friction and braking of the earth’s rotation, and has also reversed the timeline back to when the moon was orbiting as close to the earth as physically possible, near the Roche limit, shortly after the moon’s formation. The tides would have been formidable, and the heating effect of the tidal friction is an excellent replacement for the lack of insolation from a “dim young sun”. See:
http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?1968QJRAS…9..352M&data_type=PDF_HIGH&whole_paper=YES&type=PRINTER&filetype=.pdf
However there is something wrong with the calculation – about 2billion years of history is missing. Could that be accounted for by losing a liquid ocean to either a vapour hothouse or Ice snowball?
Enough of gravity for now.
One more effect of sea-ice caps is the temperature of the overlying atmosphere, and thus the composition of gases, and the overall density of the polar atmosphere. Under the coldest conditions over the Antarctic (and probably over the Arctic in winter), there is almost no water vapour nor carbon dioxide – no greenhouse effect, and there is also the ozone hole, so there is effectively no restriction on OLR, and no heating of the atmosphere by infrared.
The ozone hole also represents a local increase in TSI, in the form of short-wave UV reaching the surface. This may just be reflected by snow and ice, but must be absorbed by the southern ocean.