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 …
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Judith, my thanks as always for your willingness to defend your work. This should not be an occasion for thanks, it should be the norm, but unfortunately, in climate science your honesty and courage are far too uncommon.
Onwards to your points.
Judith Curry says:
August 18, 2010 at 8:18 am
Unfortunately, the issue I raised was that the HadISST data is no good before about 1980 for most of the Southern Ocean. As a result, the plot of the SST anomalies 1950-1999 thus doesn’t clarify the issue at all.
How the study was funded is meaningless to me. The issue is whether it is valid, not who paid for it.
Several points are at issue here. First, given that there is no valid data before about 1980 for most of the Southern Ocean (particularly around Antarctica), there is no “general pattern of warming”. In fact, if we look at the post 1980 data, there is a general pattern of cooling which increases from north to south. This pattern is supported by the UAH MSU lower troposphere temperature data. Here are the trends by latitude band for the period since 1980:
Figure W1. Temperatures of the Southern Ocean surface and the atmosphere above the Southern Ocean.
As you can see, rumors of “accelerated warming” are greatly exaggerated. The area where the Antarctic sea ice is located is cooling, cooling, cooling …
Well, as near as I can tell, the Southern Ocean around Antarctica has been cooling for the last thirty years, and the sea ice extent has increased … so I’m not clear why we need some further “plausible physical mechanism” to explain that. You talk in the paper about the “the seeming paradox of the observed increasing total Antarctic sea ice area for the past three decades (0.89 × 105 km2 per decade; Fig. S3)”. Since the ocean around Antarctica has been cooling for three decades, and the atmosphere above the area around Antarctica has been cooling for three decades … why is more Antarctic sea ice a “seeming paradox”?
Well, titling your paper “Accelerated warming of the Southern Ocean and its impacts …” when the Southern Ocean is not warming, much less showing “accelerating warming”, strikes me as an extravagant claim, but YMMV.
“Imperfect tools” is a very generous definition of the situation immediately around Antarctica (the location of the sea ice), where there is no data at all prior to 1980. You are missing two thirds of the data for the period you are studying. For me, missing data is not “imperfect”. It is missing, and thus, useless to test the hypothesis.
Whoa, whoa, whoa. Where does your paper show that regarding Antarctic sea ice the models simulations “generally agreed with observations”? You show two models that kinda sorta resemble the Southern Ocean data since 1950 … but as we know, the data prior to 1980 is non-existent so that resemblance is spurious. In any case, two models is hardly “general agreement”. Where is the study that shows that the models do better than random at simulating the Southern Ocean sea ice? You may have mentioned it in your paper, but I couldn’t find it.
Thank you, Judith, your points regarding the tone of the discussion are very valid.
w.
Well let me cast my self adrift on the thin ice (Southern Ocean style) where only fools (like me) don’t fear to tread.
For those non mathematicians who have no idea what Orthogonal Functions are; here’s my stick on a sandy beach explanation.
Everybody understands the function y = sin(x) which goes from 0 to +/-1 back and forth as (x) goes from zero to 2.pi.
Draw yourself a sin “wave” and you can immediately see that the total area for a complete cycle is exactly zero.
OK; now we could aslo have a function y = sin(2x), or y =sin (3x) and so on ad infinitum.
This set of functions; y = sin (nx) (for n = zero to infinity) is a set of orthogonal functions.
What the heck does that mean ?
If I take any one of these functions (y = sin(7x)) and multiply it by itself to get y = sin^2 (7x), and now plot a graph of that, I will discover that the total area included in one complete cycle of x = zero to 2pi is not zero but has a definite non zero value.
If I take any two different ones of these functions and multiply them; y = sin(7x).sin(13x) and integrate that, I find that the area is still zero so long as the two functions are different.
So that is the general property of “Orthogoanal Functions” The integral of the product of any two different functions of the set over some svery specific interval (0 to 2pi for the sin function), is zero, but the integral of the square of any one function of the set is finite; over that same interval of course.
There are a whole slew of mathematical functions that have this property; and any such set can be used to synthesize ANY continuous function over some interval.
The most common form of this of course is the Fourier analysis of cyclic functions into a series of “harmonics” ;but Bessel Functions, and Chebychev Polynomials (special case of sinusoids) LaGuerre Polynomials and on and on; have all been use to synthesize other functions. I’m not going to go into exactly how you figure out the correct sequence to use; but the common method relies on that orthogonal property, because I can multiply every term in my series by one of the functions and do the integrations; and every darn one of them is going to disappear, except for the one I used as my multiplier; and that will get me the correct coefficient for that term.
People who calculate fields in electron Optics and such like make a lot of use of Orthogoanl Functions. Ortho-Normal simply means that the integral comes out to something nice like 1.0 for example.
But empirical has me bamboozled so you are on your own out there now so be careful !
Laughed out loud (literally – & involuntarily) when I read this:
“[…] just an illusion […]”
http://news.discovery.com/earth/antarctic-sea-ice-growth.html
Perhaps the writer is being overzealous in misrepresenting what Drs. Curry & Liu are saying.
Judith Curry says:
August 18, 2010 at 12:20 pm
Judith, again I say that the data during the first three decades of your study is neither “deficient” nor “sparse”. It is non-existent for the area south of 55°S, the area of most interest, the area of the Antarctic sea ice, the crucial area under discussion. And unfortunately, it has been infilled with bogus meaningless values. So it is neither “sparse” nor “deficient”. It is missing and has been replaced with incorrect imaginary numbers. As a result, any analysis of the period 1950-1999 will be wrong. Incorrect. False. Invalid. You can’t do a fifty year analysis when you are missing the first thirty years of data, no matter what method you use, EOF or otherwise.
That is the nettle that neither you nor your co-author seem to have grasped.
I respect Dr. Curry.
I don’t agree with her but I respect her.
She’s probably taken more crap from the CAGW cognoscenti than from the “World is warming but we don’t have the answers” crew.
I have much less respect respect for people like Richard Telford, telling Willis to download “the outputs of models”!
Isn’t that what got us into this crap in the first place?
Dave.
Tamara,
The sea surface temperature undergoes an annual cycle (warmer during the summer). During the winter, the surface temperature of the water between the ice floes stays very close to the freezing point of sea water (about -1.7C). During the summer, the ocean warms to slightly above freezing, particularly after the ice has melted (much of the energy goes into melting the ice). In a warmer climate, the winter temperatures remain the same, but the melt season is longer (and the sea surface temperatures get somewhat warmer during the summer).
Snow that falls on sea ice accumulates on the surface of the sea ice. If the snow gets very heavy, it can weight down the flow and the snow can get wet and form snow/ice. But mostly the snow sits on top of the sea ice until the snow melts in the summer. The more snow on top of the sea ice, the slower the ice will melt. Nearly all of the Antarctic sea ice melts each summer, except for that which gets trapped in various embayments.
The rainfall doesn’t do all that much to the sea ice (other than speed up the summer melt), it is mainly the absence of snow accumulation that speeds up the ice melting.
Dr. Curry, I guess Willis’ analysis boils down to two questions:
1. Given that data before 1980 is questionable at best, wouldn’t this paper and its conclusions have been strengthened by using 1980-2000 as the study limits (or 1980-2010) rather than starting it in 1950? Why was 1950 chosen as the start date instead of 1980?
2. In a similar vein, why was the study extended as far North as 40 deg S? How do the SSTs 20 degrees from the edge of the ice have an effect on the ice extent?
Thank you for the time spent answering our questions.
Dr Curry,
why didn’t you investigate the impact on the deep water upwelling by an increase of the wind stress?
You state that the anular mode circulation around Antarctica is stronger now than in the past. But you failed to analyse which impact that atmospheric circulation change could have on sea temperature and salinity, And this analysis is a pre-requirement before to look for some P-E change.
p.solar says:
August 18, 2010 at 2:12 am
“thompson Denis et al 2009 GEOPHYSICAL RESEARCH LETTERS, VOL. 36, L20704, doi:10.1029/2009GL040104, 2009
This (paywalled) paper also pretends to find warming in Antarctica. The main line is that Gomez dome at the base of the antarctic peninsula shows similar temp rise to a station at the tip of the peninsula where climate is clearly determined by surrounding ocean (so they report it is continental temp rise) . ”
I analysed the Gomez ice core data, kindly provided to me by Dr Dennis, soon after this study was published. The Gomez data shows insignificant correlation with any of the nearby cloudmasked AVHRR satellite 1982-2006 ground temperature data used by Steig et al in their 2009 Nature paper on Antarctic temperatures.
The Gomez study quotes a high correlation with temperatures at Faraday station (nearer the tip of the Antarctic peninsula). However, whilst there is a strong correlation with Faraday during the 30 years 1947-1977 (R-squared 0.361) there is a negligible correlation during the subsequent 29 years (R-squared 0.06).
This evidence implies that the dO18-temperature relationship is not stable over time and/or that the data reflects temperatures that occurred far away from Gomez and/or that the latest few decades of ice-core data is not reliable. I invited Dr Dennis to comment on these findings, but he never did so.
Willis, I don’t disagree that there are issues with the Southern Ocean data, although there is arguably sufficient data in the midlatitudes of the Southern Ocean, which is where the warming is observed. Also, climate models have been simulating a decrease in Arctic sea ice and no decrease in Antarctic sea ice for the latter half of the 20th century, in general agreement with observations.
Is this paper the “last word” on this subject? Of course not. I hope it stimulates more data set building and hypothesis testing, using data and models. This is a topic that deserves more attention.
Willis your 2.37
Can we make the other obvious points that as well as there being very little data from which to extrapolate, the amount of water being sampled when there IS data is an absurdly small proportion of the Ocean AND water of different temperatures do not mix very well.
This afternoon I took three readings in a fifty yard stretch of the English Channel 400 yards from my door and they varied by 3.5C.
To believe we can parse the very sparse information available to a tiny fraction of a degree and come up with anything meaningful is not credible.
Sorry Dr Curry as I admire the way you speak your mind and try to build bridges but the study has little practical merit.
tonyb
In looking at the first two so-polar pictures, I see that the horizon seems to be in the middle of New Zealand; which would put it at about 40 deg S. I know that Wanganui is about 36-37 deg S.
And the color chart says that the “trend” is about 0-0.1 deg C per decade.
So that tells me that these plots are not SSTs; but some sort of “anomaly”
And there’s the rub.
It’s impossible to tell from anomalies, whether the snow should melt when it lands in the ocean or not.
So I guess we have to look elsewhere for SSTs to find out where the freeze line is or was.
I’m also curious; now that everyone wants to know what EOFs are what has been written as to the validity of EOF analysis; if the sampling rules for sampled data systems are violated.
It seems to me that the entire HADCrut or GISSTemp data is corrupted by aliassing noise due to gross undersampling both in time; but especially in the spatial sampling; so the Arctic, and Antarctic are not renowned for their surplus of sampling stations.
So I’m curious as to the validity of whatever EOFs purport to deduce in the Antarctic ?
Dr. Curry:
It would be easy to suggest, from you paper, that increased preciptitation would slow down the Southern Sea Ice Melt, as the time spent in Antarctic Day/Night remains fixed. As the jets move more poleward, they are also taking very cold air back northward, and as the mositure mixes with that frigid air, it will deposit yet more snow, yes? You mentioned a circulation pattern called the ACC. Also, recent data suggests that weak solar activity is also involved in moving jets. So, that’s 2 causes of jet shift.
If the Antarctic Sea Ice defies the model simulation and continues to expand and melt less each year, is the ACC then subject to interruption?
This is what I mean by exploring the alternatives. With models only showing a single facet of climate change, the other possibilities go unresearched.
Judith Curry says:
August 18, 2010 at 3:05 pm
The only place we see warming in the Southern Ocean from 1980 to the present is the northernmost latitude band (40°-45°S). The HadISST data shows that the rest of the Southern Ocean is cooling. See my post above.
Since Antarctic sea ice is increasing and has recently hit a record high value, how is the models’ “no decrease in Antarctic sea ice” in “general agreement” with observations showing record highs? General agreement would be both showing increasing sea ice but with different rates. One showing record highs and one showing no change at all is not “general agreement” to me.
In addition, I have still not found a study showing the agreement between models and observations. I find other claims akin to yours, like this one from the NSIDC:
I loved the last part. Antarctic sea ice is unimportant to the climate system because it occurs during the winter … but Arctic sea ice is somehow different despite the fact that it is nearer the pole and thus affected more by the winter. But I digress. My point was that I can’t find a study that supports your claim that models in general agree with the changes in Antarctic sea ice.
Total layman here.
“With regards to my concerns about the quality of the SST data, I have only been digging into that the last several months.”
But we published and drew conclusions anyway.
I thought we were talking science here?
And if you can’t support your paper here, get out of the kitchen dears, we have CO2 listed as a hazardous here in the USA for God’s sake, and cap and tax legislation using studies like these as the “facts”. Folks here have been more than generous.
Anthony,
I respect completely, on your blog, that you have decided to exercise (what appears to me) some special case of commenting style/restraint for the post of Judith Curry & Jiping Liu as opposed to many other posters who took a full spectrum of style/unrestraint commenting without complaint and with good grace.
I am confident you have your well thought-out reasons. Though, honestly, I do not understand why you do so.
I am sorry if it may be inappropriate to ask whether you will use this special approach to all future WUWT posts going forward?
Again, I appreciate the open and fair venue you have provided here.
John
REPLY: It simply got out of hand on the other thread http://wattsupwiththat.com/2010/08/16/georgia-tech-on-resolving-the-paradox-of-the-antarctic-sea-ice/
…which I’ve closed comments on, and redirected commenters here, now that we have the full paper and additional supplemental data. Sometimes it simply pays to remind people to be courteous in the heat of the moment. – Anthony
EM Smith wrote:
“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.”
As it happens, the Southern Ocean is more, or less my backyard. I’m not much of a sailor, so I do not often sail on those waters though I have from time to time. Mostly, the water has not been horizontal enough for the snow to settle on it. Right now there’s a patch just north of Casey with wave heights between 10 and 15 metres, wind speed between 48 and 64 knots.
I have no intention of investigating whether the snow is settling on that “hot”, salty water. I shall maintain my incredulity concerning the reported phenomenon over a cup of black, unsweetened Italian coffee. I make no doubt that I might believe it if I substituted opium, or ganja for the coffee. But that’s not likely to happen. Gits prefer sipping sauvignon blanc next to a warm fire while contemplating the prospect of more snow in the coming week 😉
Judith Curry wrote:
“[…] midlatitudes of the Southern Ocean […]”
Dear Dr. Curry,
This could be a source of important misunderstanding, so please quantify “mid” (as used in the preceding context) as precisely as possible. Thanks if you can clarify.
Sincerely,
Paul Vaughan, M.Sc.
–
A note to “nonalarmists” who [humorously] seem alarmed by the paper:
Drawing attention to the Southern Ocean is the way to go. I wish Drs. Curry & Liu much success in raising the profile of this important neglected region. I wish Dr. Liu far more stable personal income.
–
Willis, You make several good points, but I would advise against going so far as suggesting that Southern Ocean data is “nonexistent” before the satellite era. It is, however, fair to laugh heartily at the pre-satellite HADISST “data” for 60-90S. Those folks would be well-advised to take out some of their garbage.
Myself, I find it somewhat hard to follow that we can reliably extrapolate temperature over any sizable distance in a constantly changing (and maybe evolving) climate system.
It’s one thing to have say a closed largely static system (like a tank of water or a solid mass) and being able to extrapolate the temperature of the whole from sparse samples. Its something else completely different in a highly dynamic system – you need to get to a point of sufficient coverage to ‘capture’ the underlying temperature behavior, otherwise the error bars go off the scale and the exercise becomes pointless.
The other thing I find odd with the extrapolations I’ve seen is that they assume consistent ‘validity’ across all of the globe. There are well known and described climate ‘sub systems’ (gulf streams, etc) that ‘skew’ the temperatures in certain areas due to their effects – extrapolating across these effects must be highly dangerous to accuracy.
Dr. Curry,
With respect to a previous question I asked: I understand your arguments relative to winter antarctic ice, however the summer ice minima has risen aproximately 1.5 million sq km since 1980 from CyrosphereToday. While the area has not gone up as much, it too is showing a noticeable rise which means the ice is not melting back as your theory assumes?
Willis Eschenbach says: “Mosh, from what the study says, what I show is the source data from which the first EOF shown in Fig. 1 (a) is constructed … what am I missing here?”
Sorry. I’ve been out of touch today. First thing this morning I got started on a complementary post to this, but never got it finished.
ICOADS is the source dataset for ship and buoy readings used in the Hadley Centre and NCDC SST data. The NCDC and Hadley Centre then infill the remainder of the data. I provided a link above to maps I created that show the limited observations for Januarys spaced a decade apart from 1950 to 2000. There’s nothing special about January, other than it occurs during the Southern Hemisphere summer, so one might expect some traffic occassionally at high latitudes. There is, as you can see, very little source data for the latitudes included in Liu and Curry. Here’s the link to the maps:
http://i37.tinypic.com/t8x4ox.jpg
Regards
“”” Pompous Git says:
August 18, 2010 at 4:02 pm
EM Smith wrote:
“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.” “””
Well that is totally amazing.
I could sware that I wrote those exact words just a few minutes ago; and lo and behold E.M. Smith goes and posts the exact same thing; that is truly amazing.
I think I should go out and buy a lottery ticket.
As a long time lurker who rarely posts, many thanks are in order for this thread. Especially to Anthony Watts for maintaining this site, which allows so may to explore these issues in an open & thoughtful manner.
And thanks to Judith Curry. For having the insight to engage in the manner she has, & for being wide open to discussion. Even though at times highly critical.
As one that has excoriated Dr. Curry once right here at WUWT, on a thread she had engaged on, I’ve come to appreciate she has qualities many of her contemporaries don’t possess. Among those, most importantly, is courage.
Thanks also go out to Willis Eschenbach, Jiping Liu, Steven Mosher, George E. Smith, Bob Tisdale & a handful of others that have brought this issue forward, or added depth into this discussion.
Within the valuable context so thoughtfully provided by so many above, these issues are becoming much more rich & have much more value now, than most of the belligerent close minded pap that we’ve seen so much of in the past.
To save space, I’ll end by saying I find many interesting points, & agree with much of the above. But in the general sense, the thoughtful comments made by Cassandra King truly are exceptional. She’s summed my overall thinking to such a near perfect degree, I wonder now if somehow I have a mind clone?
Bill Tuttle, richard telford, George E. Smith, Max:
As I just noted to Willis above…
*************
ICOADS is the source dataset for ship and buoy readings used in the Hadley Centre and NCDC SST data. The NCDC and Hadley Centre then infill the remainder of the data. I provided a link above to maps I created that show the limited observations for Januarys spaced a decade apart from 1950 to 2000. There’s nothing special about January, other than it occurs during the Southern Hemisphere summer, so one might expect some traffic occassionally at high latitudes. There is, as you can see, very little source data for the latitudes included in Liu and Curry. Here’s the link to the maps:
http://i37.tinypic.com/t8x4ox.jpg
*********
So it’s not a matter of averaging 50% of the grids. You’d be lucky if you could find data in 10% of the grids on a good month.
http://i48.tinypic.com/2i7spwy.png
And I’ll add another note, since 1982, the Hadley Centre has also used satellite data which has much better coverage from 90S-40S. But the NCDC still relies in ship and buoy data. They included satellite data in their ERSST.v3 data then deleted it a few months later due to, assumedly, political pressure, because the satellite data lowered the Global SST readings by about 0.03 deg C and changed the order of record years. More on this in my post detailing the SST datasets:
http://bobtisdale.blogspot.com/2010/07/overview-of-sea-surface-temperature.html
Regards
Willis, read the papers regarding how the missing ocean data is filled in. The method seems reasonably robust for the period since 1950. But the point of our paper does not rest on the detailed accuracy of the observed surface temperatures.
We pose a hypothesis about how the upper ocean, sea ice, and hydrological cycle interaction in the Southern Ocean. There are many ways that hypotheses can arise: inferences from observations, identification of patterns, imagination, etc. You then test the hypothesis with available information. In this case it is observations and model simulations. The evidence that is available supports our hypothesis. If say the SST has greater uncertainty than we think it does, well then there is less support for our hypothesis.
When you say garbage in garbage out, you are not understanding the scientific process. We posed a hypothesis, we tested it using data and model simulations, which support the hypothesis. If the data are bad and the model is wrong, that doesn’t falsify the hypothesis, it reduces the support for the hypothesis. So it doesn’t make any sense to say that our hypothesis is incorrect because there are holes in the sea surface temperature data set.