Guest Post by Willis Eschenbach
I thought I was done with sunspots … but as the well-known climate scientist Michael Corleone once remarked, “Just when I thought I was out … they pull me back in”. In this case Marcel Crok, the well-known Dutch climate writer, asked me if I’d seen the paper from Nir Shaviv called “Using the Oceans as a Calorimeter to Quantify the Solar Radiative Forcing”, available here. Dr. Shaviv’s paper claims that both the ocean heat content and the ocean sea surface temperature (SST) vary in step with the ~11 year solar cycle. Although it’s not clear what “we” means when he uses it, he says: 
“We find that the total radiative forcing associated with solar cycles variations is about 5 to 7 times larger than just those associated with the TSI variations, thus implying the necessary existence of an amplification mechanism, though without pointing to which one.” Since the ocean heat content data is both spotty and incomplete, I looked to see if the much more extensive SST data actually showed signs of the claimed solar-related variation.
To start with, here’s what Shaviv2008 says about the treatment of the data:
Before deriving the global heat flux from the observed ocean heat content, it is worth while to study in more detail the different data sets we used, and in particular, to better understand their limitations. Since we wish to compare them to each other, we begin by creating comparable data sets, with the same resolution and time range. Thus, we down sample higher resolution data into one year bins and truncate all data sets to the range of 1955 to 2003.
I assume the 1955 start of their data is because the ocean heat content data starts in 1955. Their study uses the HadISST dataset, the “Ice and Sea Surface Temperature” data, so I went to the marvelous KNMI site and got that data to compare to the sunspot data. Here are the untruncated versions of the SIDC sunspot and the HadISST sea surface temperature data.
Figure 1. Sunspot numbers (upper panel) and sea surface temperatures (lower panel).
So … is there a solar component to the SST data? Well, looking at Figure 1, for starters we can say that if there is a solar component to SST, it’s pretty small. How small? Well, for that we need the math. I often start with a cross-correlation. A cross-correlation looks not only at how well correlated two datasets might be. It also shows how well correlated the two datasets are with a lag between the two. Figure 2 shows the cross-correlation between the sunspots and the SST:
Figure 2. Cross-correlation, sunspots and sea surface temperatures. Note that they are not significant at any lag, and that’s without accounting for autocorrelation.
So … I’m not seeing anything significant in the cross-correlation over full overlap of the two datasets, which is the period 1870-2013. However, they haven’t used the full dataset, only the part from 1955 to 2003. That’s only 49 years … and right then I start getting nervous. Remember, we’re looking for an 11-year cycle. So results from that particular half-century of data only represent three complete solar cycles, and that’s skinny … but in any case, here’s cross-correlation on the truncated datasets 1955-2003:
Figure 3. Cross-correlation, truncated sunspots and sea surface temperatures 1955-2003. Note that while they are larger than for the full dataset, they are still not significant at any lag, and that’s without accounting for autocorrelation.
Well, I can see how if all you looked at was the shortened datasets you might believe that there is a correlation between SST and sunspots. Figure 3 at least shows a positive correlation with no lag, one which is almost statistically significant if you ignore autocorrelation.
But remember, in the cross-correlation of the complete dataset shown back in Figure 2, the no-lag correlation is … well … zero. The apparent correlation shown in the half-century dataset disappears entirely when we look at the full 140-year dataset.
This highlights a huge recurring problem with analyzing natural datasets and looking for regular cycles. Regular cycles which are apparently real appear, last for a half century or even a century, and then disappear for a century …
Now, in Shaviv2008, the author suggests a way around this conundrum, viz:
Another way of visualizing the results, is to fold the data over the 11-year solar cycle and average. This reduces the relative contribution of sources uncorrelated with the solar activity as they will tend to average out (whether they are real or noise).
In support of this claim, he shows the following figure:
Figure 4. This shows Figure 5 from the Shaviv2008 paper. Of interest to this post is the top panel, showing the ostensible variation in the averaged cycles.
Now, I’ve used this technique myself. However, if I were to do it, I wouldn’t do it the way he has. He has aligned the solar minimum at time t=0, and then averaged the data for the 11 years after that. If I were doing it, I think I’d align them at the peak, and then take the averages for say six years on either side of the peak.
But in any case, rather than do it my way, I figured I’d see if I could emulate his results. Unfortunately, I ran into some issues right away when I started to do the actual calculations. Here’s the first issue:
Figure 5. The data used in Shaviv2008 to show the putative sunspot-SST relationship.
I’m sure you can see the problem. Because the dataset is so short (n = 49 years), there are only four solar minima—1964, 1976, 1986, and 1996. And since the truncated data ends in 2003, that means that we only have three complete solar cycles during the period.
This leads directly to a second problem, which is the size of the uncertainty of the results of the “folded” data. With only three full cycles to analyze, the uncertainty gets quite large. Here are the three folded datasets, along with the mean and the 95% confidence interval on the mean.
Figure 6. Sea surface temperatures from three full solar cycles, “folded” over the 11-year solar cycle as described in Shaviv2008
Now, when I’m looking for a repetitive cycle, I look at the 95% confidence interval of the mean. If the 95%CI includes the zero line, it means the variation is not significant. The problem in Figure 6, of course, is the fact that there are only three cycles in the dataset. As a result, the 95%CI goes “from the floor to the ceiling”, as the saying goes, and the results are not significant in the slightest.
So why does the Shaviv2008 result shown in Figure 4 look so convincing? Well … it’s because he’s only showing one standard error as the uncertainty in his results, when what is relevant is the 95%CI. If he showed the 95%CI, it would be obvious that the results are not significant.
However, none of that matters. Why not? Well, because the claimed effect disappears when we use the full SST and sunspot datasets. Their common period goes from 1870 through 2013, so there are many more cycles to average. Figure 7 shows the same type of “folded” analysis, except this time for the full period 1870-2013:
Figure 7. Sea surface temperatures from all solar cycles from 1870-2013, “folded” over the 11-year solar cycle as described in Shaviv2008
Here, we see the same thing that was revealed by the cross-correlation. The apparent cycle that seemed to be present in the most recent half-century of the data, the apparent cycle that is shown in Shaviv2008, that cycle disappears entirely when we look at the full dataset. And despite having a much narrower 95%CI because we have more data, once again there is no statistically significant departure from zero. At no time do we see anything unexplainable or unusual at all
And so once again, I find that the claims of a connection between the sun and climate evaporate when they are examined closely.
Let me be clear about what I am saying and not saying here. I am NOT saying that the sun doesn’t affect the climate.
What I am saying is that I still haven’t found any convincing sign of the ~11-year sunspot cycle in any climate dataset, nor has anyone pointed out such a dataset. And without that, it’s very hard to believe that even smaller secular variations in solar strength can have a significant effect on the climate.
So, for what I hope will be the final time, let me put out the challenge once again. Where is the climate dataset that shows the ~11-year sunspot/magnetism/cosmic rays/solar wind cycle? Shaviv echoes many others when he claims that there is some unknown amplification mechanism that makes the effects “about 5 to 7 times larger than just those associated with the TSI variations” … however, I’m not seeing it. So where can we find this mystery ~11-year cycle?
Please use whatever kind of analysis you prefer to demonstrate the putative 11-year cycle—”folded” analysis as above, cross-correlation, wavelet analysis, whatever.
Regards,
w.
My Usual Request: If you disagree with someone, myself included, please QUOTE THE EXACT WORDS YOU DISAGREE WITH. This prevents many flavors of misunderstanding, and lets us all see just what it is that you think is incorrect.
Subject: This post is about the quest for the 11-year solar cycle. It is not about your pet theory about 19.8 year Jupiter/Saturn synoptic cycles. If you wish to write about them, this is not the place. Take it to Tallbloke’s Talkshop, they enjoy discussing those kinds of cycles. Here, I’m looking for the 11-year sunspot cycles in weather data, so let me ask you kindly to restrict your comments to subjects involving those cycles.
Data and Code: I’ve put the sunspot and HadISST annual data online, along with the R computer code, in a single zipped folder called “Shaviv Folder.zip“
Tried to embed this image but failed in my previous post;
http://www.emeraldecocity.com/Pictures/Why%20we%20have%20summer%20and%20winter.jpg
From michaelwiseguy on June 7, 2014 at 9:41 pm:
Over billions of years the Sun has brightened, the planetary orbits have shifted while the asteroid belts and even moons formed and grew, we might even have had a rogue planet or two pass through the system, or be captured, or collide.
Then there’s that stupid continental drift, where our current polar regions aren’t the ones we had before, it was ocean at both ends.
Projecting back the possible effects of our current Sol-Terra distance variations to billions of years ago, or just millions, might no be that wise, guy.
W: “And even if that were not true, you still only get 163 years of “data” using your 2% solution … and from that you are diagnosing a 170-year signal. Does the name “Nyquist” ring a bell?”
Nyquist says you need at least two samples per cycle. How many monthly samples do you think there are in 163 years? This has nothing to do with Nyquist.
Willis: “I had quoted your exact words about the 170-year cycle, immediately above my comment
in the same post. You had said, and I had quoted:”
You quoted my words , then ignored them and reinterpreted. Now you re-quote me and chop out even more of what I said in an attempt to refute my objection. At least try to be honest, rather that twisting and misquoting.
Nowhere did I say there was a 170 “cycle” , in fact I explicitly drew attention to the length of the data in relation to that peak and explicitly said there was no grounds to interpret it as being cyclically repetitive.
For the third time, here is what you are choosing to ignore in order to criticise me for you misinterpretation of what you think I said:
“It should also be noted that is a clear anti-correlation with period of about 140 years and a lag of half that. That is a period of time, there is not enough data to suggest this is periodic as in cyclically repetitive.”
Now please stop pointless arguments based on misquoting and misinterpreting and get back to looking at the data.
kadaka (KD Knoebel) says:
June 7, 2014 at 10:34 pm
“Projecting back the possible effects of our current Sol-Terra distance variations to billions of years ago, or just millions, might no be that wise, guy.”
With all that going on, how can anyone think there should be no significant climate change except what man makes?
After numerous of these types of discussions, I’ve come to the conclusions that:
(1) Statisticians should not do signal analysis. They make so many basic mistakes it’s just painful to anyone versed in the art.
(2) If you are using “R”, you are probably a statistician. See conclusion #1.
So my immediate filter is “if it’s being modeled in “R”, it’s probably wrong”
You chose to use cross-correlation of hadISST in attacking Shaviv 2008, so I’ll repeat the question you forgot to answer relating to the key point you are trying to make in the article:
Without worrying about the FT of CC for the moment. What information do you think can be derived from cross-correlation. You used it to support your impression that there is no solar signal, so you must expect that something could be there that was not.
I’m guessing you were looking for a peak that is above your 0.2 threshold (but I don’t want to puts words into your mouth, so please correct me if that’s wrong).
How do you interpret ISST vs SSN peaks getting above 0.25 ?
http://climategrog.wordpress.com/?attachment_id=959
“http://climategrog.wordpress.com/?attachment_id=956”
Greg, where’s the source code? You can mess up the data so badly in so many ways (like decimating with a boxcar average filter like in the original article). We need to review the source code. (I note that your periodiogram of SSN matches mine though, so it’s probably right, or at least as wrong as mine).
My basic checklist for any signal analysis.
(1) was decimation done? Why did you bother introducing more error by decimation? FFTs on modern computers are very fast and there’s not enough reliable monthly history of any meteorological data set to make a modern computer even hiccup. If you have data sets with mismatched periods the proper procedure is to interpolate, not decimate. As a rule of thumb I always interpolate by 4x or 8x just to avoid any computation-induced issues. If you need to decimate please use a proper filter, not a boxcar average. If you want to decimate for display purposes make it the very last operation.
(2) was proper windowing done? You’ll spread sinx/x noise all over your FFT if you don’t.
(3) was the data zero-padded for better inspection of low frequency components? Note that you need about 8 cycles to get any decent quality low frequency information – the error is 1/period, and 1/8 means 12.5% error. For the ~300 years of sunspot numbers any supposed cycle any period discovered that’s > 40 years is highly suspect. It’s probably just noise. It’s not correlatable to anything else.
(4) was the data filtered before FFT to avoid aliasing?
(5) was any filtering done with a reasonable filter? e.g. a hamming filter or other linear-phase filter. averaging is wrong wrong wrong, and a clear sign you don’t know what you’re doing.
The original article that was reviewed failed every item in this checklist.
Nir Shaviv says:
June 7, 2014 at 2:00 pm
Thanks, Nir, good to hear from you.
Thanks. Have you been able to replicate their results? I’ve looked at some of those, and have either not been able to replicate their results, or have seen their results disappear (as happens with your SST claims) when longer datasets are used.
Nir, I agree that the result in Figure 6 shows about the right variation of the means of the folded cycles to fit your theory. As a result, you should be most suspicious of it, that kind of result is the most dangerous.
In any case, I also showed the 95%CI on those means. Given that the 95%CI barely crosses the zero point, I say that the means are not significantly different from zero. Do you agree, and if not, why?
And since the means are not significantly different from zero, I don’t see how you can claim that a cycle exists, unless it is a cycle of zeros.
You say that the appropriate statistical test is the “probability that the null hypothesis would give the blue line apparent in fig. 6, namely that a large fraction of the points would be as far as they are from the null signal.”
I can agree with that as far as it goes … but what is the “null hypothesis”? Red noise? If so, what kind? I don’t think you can do it with just a chi-squared test, that assumes white noise.
And you speak of an adjustment for autocorrelation using “the effective number of degrees of freedom”. However, all that adjustment can do is make the results less statistically significant … and they are already not statistically significant even without that adjustment.
Mmmm … were I in your shoes, I don’t think I’d advertise that attitude. Sounds a whole lot like “My mind is made up, don’t bother me with facts” …
I agree that my result shows only, and can only show, that the HadISST dataset does not support the claim of a solar signal. And since you have claimed that the HadISST dataset does support the idea of a solar signal, I’m sure that you would LIKE to “care less what the answer is” from my analysis … but I doubt if that is actually the case, you’re more of a scientist than that.
Your problem is my data is spotty, and your own analysis used ocean heat content data?? That’s special pleading.
First, the SST data from 1880 is entirely comparable, and arguably better, than the OHC data from 1955 that you used without demur. Hundreds and thousands of ships have been measuring SST for every oceanographic expedition ship measuring OHC. In addition, for OHC you need to measure in three dimensions, meaning you need more measurements and your results have greater uncertainty.
The other reason I say that your claim is special pleading is that if the full dataset supported your claim, you’d use it without compunction or question. How do I know that? Because it’s what I or anyone else would do with a full dataset that supports our claims. Live by the sword, die by the sword.
Not true at all. Both folded analysis and Fourier transforms clearly show the various ~11-year peaks of the solar cycle or their averages, and the solar cycle is definitely not regularly periodic. Claiming that a similar signal won’t show up in the same analysis of the climate data won’t wash. If that were true, we wouldn’t see it in the solar data either.
In addition, I don’t “assume that the solar cycle is strictly periodic”. I’ve never said anything like that. Please provide a quotation AS I REQUESTED to support that erroneous view of what I think.
Finally, I folded the data at the solar minima and plotted their values and their average from the solar minimum forwards for 11 years because you folded the data and looked forwards 11 years.
Look, Nir, you claimed that a signal was there. I’m not finding it. Let’s worry about the reasons and the details later.
One thing at a time, my friend. I try to keep my analyses focussed on one topic. As to your tidal records, you have not archived your data and your source (Douglas) is paywalled, so I am unable to replicate your results. Post up the 24 tidal datasets you used and I’m happy to take a look.
However, I’ve looked for the solar cycle in a lot of tidal records, and found nothing. See e.g.
The Elusive ~ 60-year Sea Level Cycle
The Sea Level Cycles Get More Elusive
Given that, I’m not sanguine about your sea level results, but I’ll take a look at them if you post the data.
Since your claimed cyclical variations in SST disappear as soon as we use a longer dataset, and since the folded means are not significantly different from zero whether we use the shortened or the full dataset, and since the cross-correlation doesn’t show a significant correlation at any lag whether we use the full or shortened dataset … I fear your quantitative analysis is wildly premature. First you need to show that a solar signal in the SST actually exists.
All the best, and again, my thanks for being one of the few scientists who is willing to come here and to defend your work against the “slings and arrows of outrageous fortune”.
w.
PS—This is far from my first venture into these sun-spotted waters. In addition to the above citations to sea level studies, see also:
The Slow Fourier Transform
Sunspots and Sea Level
Usoskin Et Al. Discover a New Class of Sunspots
Cycles Without The Mania
Sunny Spots on the Parana River
Cosmic Rays, Sunspots, and Beryllium
Basic signal analysis – based hypothesis: If the oceans are a low pass filter, it’s entirely possible that the sun affects ocean temperature, but you won’t see it in an 11 year cycle because that signal is filtered away. Alas, we don’t have enough data to see cycles much longer than 35-40 years. Time to stop looking for patterns in the noise…
Its funny willis.
These guys have this theory. And some like greg actually know a couple bits of math.
And you make a simple challenge.
1. show me the data set
2. show me the method
that would prove their case.
Of course, they point you at papers ( not data) and they blather on about methods ( usually with no code)
and they have all manner of wild goose chases to send YOU on to prove THEIR idea.
And when you fail to prove their idea, they blather on some more about how you should have done it
and where the effect might lie.
there be unicorns
you just cant find any.
Theodore White says:
June 7, 2014 at 3:19 pm
Theodore, in the head post I said:
I was totally serious. Let me repeat that, both for you and for others. This post is not the place for your pet theory, whether it’s about the ENSO or whatever other random subject you might light on, particularly when expounded in such excruciatingly boring detail as yours was. Please take it elsewhere.
w.
Peter Sable says:
June 7, 2014 at 11:01 pm
Basic signal analys – based hypothesis: if a system responds strongly to changes in input on a day-to-day basis, as the SST does, it is unlikely that it contains a strong low-pass filter …
w.
Woodfortrees provides an audio representation of their data. When I load the audio into Audacity, and plot the spectrum using Audacity’s built in spectrum analyzer, there is always a nice spike at 11 years. Coincidence?
“if a system responds strongly to changes in input on a day-to-day basis, as the SST does,”
Okay, I’ll just change it to a multi-bandpass filter 🙂 Sure the surface temperature changes daily, but the depths do not, and mixing between the two is going to be slower than daily The depths likely feedback enough to stop an 11 year cycle from being seen on the surface.
Frankly any hypothesis is possible that’s not obviously wrong – we don’t have enough data…
“http://climategrog.wordpress.com/?attachment_id=956”
Greg, now that I look a bit closer, it looks like you didn’t window – suspiciously looks like sinx/x noise in there…
Steven Mosher says:
June 7, 2014 at 11:11 pm
Thanks, Mosh. I’m glad some folks out there understand what I’m doing here. It’s a simple challenge, really. Show me the dataset and how to demonstrate the ~11-year signal in the data.
However, I think at this point I’ve heard every conceivable excuse for not being able to find the 11-year signal …
Regards,
w.
There are several people who claim to have seen a correlation between the duration of the “11 year” cycle the amplitude and the effect of warming. Shorter the length “11 year” cycles, have stronger amplitudes with greater effect on warming.
Some references:
E Friis-Christensen, K Lassen – Science, 1991
H Svensmark – 1998
SK Solanki, NA Krivova, M Schussler… – ASTRONOMY AND …, 2002
T Landscheidt – The Solar Cycle and Terrestrial Climate, Solar …, 2000
Peter Sable says:
June 7, 2014 at 11:24 pm
While the smiley is appreciated, I fear that such speculation and on-the-fly theory modification is all too typical of people in this discussion. Look, it could be so, it might be gamma rays … but I’m interested in evidence, not imagination.
Not true at all. The oceans, curiously, move in opposition to the atmosphere. By that, I mean that the atmosphere overturns during the day, and it stratifies at night. The ocean is the opposite. The upper layer of the ocean is stratified during the day, and it overturns at night. The surface cools by evaporation/conduction/radiation. Of course, cool water is denser than warm, so the surface water sinks and is replaced by warmer water from below. This nightly overturning insures the continual mixing of the upper layer, which is why that layer iscalled the “mixed layer”.
It’s absolutely possible that all of the molecules in my coffee cup happen to be moving upwards at the same time, too, so all my coffee might spontaneously jump out of my cup … but that doesn’t make it worth discussing. I’m interested in evidence, not some hypothetical possibility of the oceans acting as a multi-bandpass future. When you can show evidence that it’s actually happening, I’ll be happy to discuss it. Until then, not so much …
w.
You’re such a trooper Willis. I hope you decide to revisit this topic again in a future post. “Using the Oceans as a Calorimeter to Quantify the Solar Radiative Forcing”, or seeing the oceans as a capacitor like device or a simple power storage battery whatever. With over 332,519,000 cubic miles of water on the planet that’s about 352,670,000,000,000,000,000 gallons, it’s not an easy task to find that 11 year solar cycle signal in all that wet water.
Perhaps this would be a good time to revisit this previous thread of yours;
Ocean Temperature And Heat Content
Posted on February 25, 2013 by Willis Eschenbach
“That tells me that it takes about a thousand zeta-joules to raise the upper ocean temperature by 1°C.”
Darn, forgot the link for my previous post.
http://wattsupwiththat.com/2013/02/25/ocean-temperature-and-heat-content/
“However, I think at this point I’ve heard every conceivable excuse for not being able to find the 11-year signal …”
I think it highlights what one gets when one starts to look at data with no real “idea” of how the system works.
Its why I ask the question “why does anyone think the 11 year cycle show show up?”
If you start with a notion that “its the sun stupid” then you cant help but continue to insist that other people are looking in the wrong way or looking at the wrong thing
If you start with a notion that ‘damn, its really complicated, theres the sun and clouds and oceans
and GHGS, and …. and its all inter connected” then you wouldnt be shocked to find out that tiny
variations in the sun had no discernable effect” you conclude.. hey this thing is more complicated than a circuit or a steam engine.
Greg Goodman says:
June 7, 2014 at 10:05 pm
————————————
Proof does not exist? I have been running these experiments since 2011 –
http://i47.tinypic.com/694203.jpg
I am not lying to you. I am not a climastrologist. I work in engineering.
Just conduct the experiment as shown. Remember –
“Tell me I’ll forget. Show me I’ll understand. Let me do it I will KNOW.”
Type is cheap. Do the experiments and you will know. I have little interest in just publishing my results. I am more interested in others replicating experiments.
I consider the Internet to be a permanent record. If I present an empirical experiment, I have built it, I have run it. No if, no but , no maybe. I am not a climastrologist. I do not lie.
“What I don’t understand is why after 30y of intense investment in research no one in climatology seems to have tested this most basic physical question, physically”
Do the experiment and you will know why 😉
Nothing more to learn from this thread. I am leaving it to allow Willis and Mosh to tongue kiss. That’s not why I come WUWT.
Konrad says:
June 7, 2014 at 6:24 am
richard verney says:
June 7, 2014 at 12:49 am
//////////////////////////////////////////
Interesting. I agree that it appears that the atmosphere serves to cool the planet.
My point about wind swept spray and spume is this:
In Willis’ article ‘Radiating the Oceans’ Willis essentially cited the gross energy transfer budget, then said that equation balances, and then said if we remove DWLWIR, from such budget, the oceans would freeze, then said we know that the oceans are not frozen, QED the gross energy flow budget must be correct. That proves nothing, since the net energy flow budget also balances and it does this without DWLWIR. The fact that either or indeed both equations balances proves nothing.
So my point is, does all the claimed DWLWIR actually enter the ocean, because even if only 99% of it entered the ocean, over time (and we are taliking from the dawn of time that Earth first acquired oceans), the oceans would freeze if the gross flow energy budget governed and if only 99% of the claimed DWLWIR actually entered the oceans.
You correctly observed that some of the LWIR absorbed in the wind swept spray and spume would be re-radiated downwards towards the ocean. I agree, but of course, it is only some of the DWLWIR (re-radiation is omni-directional with say only (somewhat less than) about 50% being re-radiated in a generally downward direction). As we talk, in broad terms, about 1/3rd of the oceans are experiencing circa BF2, 1/3rd BF 4 to 5, and 1/3rd BF 6 to 8. In open ocean, it is rare to see less than BF2, unless in the doldrums. On the other hand, there will be large areas where severe storms are ravaging. Some of the wind swept spray and spume will find its way back into the ocean (along with any energy DWLWIR that it may have absorbed), but much of it won’t since it will evaporate or rise fuelling the cloudy conditions above (and with it latent heat changes will happen as phase changes take place).
So I am posturing that in the real world conditions that we see on planet Earth, not all the DWLWIR would get entrained in the oceans. Even if it is only few percent of DWLWIR that is essentially shielded from the oceans by the divorced layer of wind swept spray and spume which ravages above a not insignificant size of ocean, this has significant impacts on those who rely upon a gross flow energy budget for explaining why the oceans do not freeze.
Currents and wind play a large role in explaining ocean temperature profiles. Of course, wind also helps drives evaporation. In one of Willis’ arcticles on ARGO, he suggested that evaporation explained why the ARGO temperature data was capped at 30degC. Whilst evaporation does play a role, I suggested to him that it is not the sole reason since there are large ocean areas with higher than 30deg C temperatures commonly recorded (eg Gulf of Mexico, Red Sea, South China Sea, parts of the Indian Ocean, the ocean off the West coast of Arrica etc), I suggested that it is currents and wind which re-distribute temperatures from the equitorial and tropical seas polewards (and ocean overturning that helps distribute SWIR absorbed in the first 10 metres downward to depth) that is the main reason why the bulk ocean surface temperature is capped at 30degC.
I have seen your experiments, the results of which are extremely interesting, and could well be of signiificance. I have often thought that these experiments should be scaled up and replicated in laboratory conditions. There are plenty of large scale ship model tanks that could be used so that one has a large volume of water and a large surface area. It must be possible to rig something up so that the effect of the relevant LWIR bandwidth on water can be tested.
I have always been bemused why climate scientists do not attempt to experiment to test some of the theories upon which they rely. I don’t count computer modelling testing. That and the poor quality of data upon which they rely without questioning, and which data they frequently over stretch, is a poor testament of the quality and rigour of this particular head of science.
Peter Sable says:
“http://climategrog.wordpress.com/?attachment_id=956″
Greg, now that I look a bit closer, it looks like you didn’t window – suspiciously looks like sinx/x noise in there…
====
Yes, I know what you mean, that’s why I commented on it being odd. Both datasets have a long term rise, this is what causes the roughly triangular long term from in the cross-correlation, although this does reverse at the ends.
IIRC I used an extended cosine window on that plot. This generally provides better resolution but does need essentially flat data. Of course if you use something like a Kaiser-Bessel window in that which is quite heavy damping effect you are distorting the data to suppress a long term correlation that is real. Swings and roundabouts.
Here’s a quick lash-up comparing ISST vs SSN to ICOADS vs SSN , using KB windowing
http://tinypic.com/view.php?pic=nmzxo9&s=8#.U5QiwaKBwrQ
(To address Willis’ 2% criticism, both series are cropped at 1880, icoads is continuous with 20% cutoff in that range.)
This time plotted in frequency and amplitude scaled by 1/f as a concession red-noise arguments.
The low frequency peak has been bent down to 114y in period. Even with the 1/f scaling it is a significant part of the spectrum. Looking at the CC function the negative correlations are separated by about 140y, without being distorted.
http://climategrog.wordpress.com/?attachment_id=958
It is interesting that the ISST “reanalysis” data seems to almost completely remove the 11.4 year peak in ICOADS but correlates much more strongly with SSN at 10.2 and 5.25 years.
I don’t know the mojo that is used in deriving ISST so I can’t say which is better, maybe ISST is resolving some detail better than unprocessed ICOADS. I am a little suspicious of the how featureless its is around the zero lag tough:
http://climategrog.wordpress.com/?attachment_id=959