ENSO forecast based on tidal forcing with an Artificial Neural Network
Investigation submitted by Per Strandberg
Here on this page, you are going to find evidence that tidal forcing is one of the most important, if not the most important driver for ENSO variations. That tidal forcing could be the main explanation for ENSO variations was something I stumbled upon when I examined possible ENSO drivers.
After the previous results which I got when I was using an Artificial Neural Network ANN and where I did an analysis of the correlations between the global mean temperature and possible forcing drivers, which can be viewed here, I turned my attention to the ENSO index by looking into the Multivariate ENSO Index (MEI).
One thing I found when I analyzed the result from correlations to ENSO was that there is a strong correlation between variations in Earth’s rotations both to the global mean temperature and to the ENSO index. What we are talking about here are small variations in Earth’s rotations, which are in the order of milliseconds.
One other factor with correlation to the ENSO is of course SOI, but I also found correlations to SST, PDO, and the Kp and Ap indexes.
From this, I concluded that either it is ENSO which is driving changes in Earth’s rotation or it changes in Earth’s rotation, which is causing variations in ENSO or more likely it is some combination of both.
Proof that ENSO and variations of Earth’s rotation are proportionally correlated to each other has been known for some time. This can be seen here.
The mechanisms which tie ENSO and variations in Earth’s rotation together are caused by sea current changes, changes in trade winds or by displacements of water between the equator and slightly higher latitudes. This all makes sense.
The water currents in the northern hemisphere follow a clockwise pattern, and in the Southern hemisphere they follow a counterclockwise pattern because of the Coriolis effect. The trade wind and the currents near the equator are moving to the west. However the Current closest to the equator called the equatorial counter current move to the east. Still deeper at depth down to 200 meters at the equator an ever stronger current is moving to the east.
The behavior of this current of the Equatorial Pacific is shown on this page by Bob Tisdale.
The only mechanism by which ENSO can be driven by changes in Earth’s rotation is by variations in the tidal force.
My next step was to try to include tidal forcing in my ANN.
I then got three problems, which I had to overcome.
Firstly: I had to find data over the position and distance to the Moon and to the Sun. Eventually, I found software from which I could get this data, although it gave limited information and I was only able to print out time and position when the Sun and the Moon were closest and farthest from the Earth and with the Moon I could also calculate the time and position of the new moon, the full moon and the moon nodes. The Moon nodes are the location where the Moon cross over the ecliptic plane.
Secondly: I had to find the formula for the tidal force vector and implement this into my software.
Thirdly: I had to figure out what features in the tidal forcing which could affect ENSO. I had to experiment with different configurations based in my limited and rather crude data. To do this, I had to make complicated trigonometrically calculations in order to get the right value of the tidal force. Eventually, I got good correlations between ENSO and the tidal forcing. By this time, I had figured out which features in the tidal forcing that were causing this correlation. However it was not a direct correlation with ENSO, rather it was a correlation with the derivate signal of ENSO, i.e. it was affecting the rate of change of ENSO. The correlation to the change of rate in Earth’s rotation, on the other hand, is direct. This means that tidal forcing is causing the rate of Earth’s rotation to either speed up or speed down. The rate of rotation is then responsible for changes of the ENSO index. One reason was that it was difficult to identify, which features, which cause correlations. This was because each tidal forcing point I use the sum of monthly calculations. The size of the tidal forcing changes each and every day and how to summarize this data the right way into useful functions, which can be used to construct values that could create good correlations were difficult.
Of course, the tidal force is not the only factor which drives ENSO, but it is the most influential factor.
To test if that would be the case I ran my network with the right tidal forcing data. I also included feedback loops back in the network from the output ENSO values to some of the input nodes. After some testing and individual adjustments of the internal components in the artificial network, I got good results. Following on my earlier experiment of the ANN on the mean global temperature I trained the ANN from late 1978 up to the end of 2004. I used the time from 2005 up to the late 2011 for test the calculations, in order to find the minimal error function.
 |
This is the result I got. The exciting thing with this result is that it is possible to make forecasts for much longer times into the future. Today’s predictions use computer models and are only able to make credible predictions 4 to 5 months into the future. While in my case, using my ANN calculations based on tidal forcing can be made for forecasts for an almost unlimited time because the Moon and Sun’s positions into the future are known in advance. Although, I have to stress that with the predication so far it is not possible to get the last figure right. Currently, it is only possible to make an estimate with a relative high likelihood at any date if ENSO are going to be positive, negative or neutral. However, as can be seen here the predictions are not always correct. The main large El Niño events of 1982 and 1998 can clearly be seen, but the large magnitude of these events can not be predicted.
 |
I later made a ENSO forecast from late 2011 up to 2020. I cannot show the result here because of proprietary reasons. This picture shows the test period and some of the forecast which ends in early 2013.
Note, however that the calculations from this graph, the ENSO index uses ENSO feedback values which all are from estimated ones. Those are not the real ENSO values.
Now Look at: the previous graph with the whole time span from 1979 up to 2011! On this graph, look at the beginning at the 3 first years from the start of 1979. These 3 first years have all exceptional good correlation to the real ENSO values. The difference with the start values in this case is that I use real ENSO values for the feedback values in the network calculations which are going into the calculations with values before the graph begins. This is because in my ANN, I use for every calculation point values which goes 3 years back in time and I must use real input values for values before my first calculated value.
If I would make a forecast for the next years using current real ENSO values from 3 years back and up to the current date, my forecast would be greatly improved and would be much better than forecast made with current computer models. There were 2 important events that happened the years just after 1979. The first was the eruption of El Chichon in 1982 in Mexico. The second was the unusually strong el Niño event between 1982-1983.
My calculated values after 1982 of ENSO tend to come out of phase after and around 1982, and the ANN seems not to be able to handle strong El Niño’s very well. After some years, the estimated ENSO value deteriorate somewhat mainly because of errors in the feedback caused by the inertia in Earth’s rotation. In contrast to computer model forecasts, I don’t use data from the Tropical Atmosphere Ocean TAO network which is a NOAA measurement network of buoys in the tropical Pacific that deliver real time data which feeds these ENSO models with real time data.
 |
Here is a result from the same program, but as input it uses variations in the Earth’s rotation instead of tidal forcing. As you can see, the correlation to the Earth’s rotation makes the result much better. However in contrast to tidal forcing, future changes in Earth’s rotation is unknown.
 |
Here is another graph from the same program with feedback but this time the input signal is only from SOI Southern Oscillation Index. As expected SOI is closely related to ENSO.
 |
Here is a repeat with the same program, but this time with a combination of tidal forcing, changes in Earth’s rotation, SOI, Kp and the Ap indexes. As you can see, this result is similar to that of the previous with only SOI.
The next step I plan to take is to use the ANN with real time data and make more accurate ENSO predictions for the next 3 to 4 years in to the future. I also want to test with real ENSO input data for several time periods in order to evaluate statistically how the good predictions can be based on real time ENSO feedback input data for the beginning.
After that, I want to improve on my result by using more precise and accurate tidal calculations. I have found a program from which I can make precise calculations of the Moon and the Sun on a daily basis. Other factors I plan to look into are the mechanism of the Kelvin wave, Walker circulation and MJO which all should influence ENSO to some degree.
So far I have only looked at ENSO. I can easily switch to SOI, NOI and changes in Earth’s rotation and use that as an output for predicting ENSO with the ANN.
Conclusion from my result is that tidal forcing is as a major factor in ENSO forcing. I now have gotten new questions. Compared with other causes how important is the effect from tidal forcing? Is it possible to find an increasing effect from tidal forcing by improving the tidal data I use? Is it, for example, possible to identify tidal forcing as the cause for the strong El Niño of 1982 and 1998?
It may be possible to get better ENSO results by using predictions based on SOI, NOI, Earth’s rotations or by starting from tidal forcing only. I’ll test and see. Also, ENSO and SOI are parameters for which there exists long historical data records. By, using a longer time span for training and testing, the accuracy of predictions based on ANN should be improved.
I acknowledge that it is not easy to find correlations between tidal forcing without testing out the right feature and by using am ANN. However I do find it very strange that no scientist to my knowledge has been looking into a possible connection between tidal forcing and ENSO in any depth.
As can be seen from what the IPCC writes about ENSO predictions, they do not have a clue. The current data models that are in use can only predict with any accuracy 4 to 5 months into the future. When it comes to the ENSO drivers, these researchers think chaos theory and random noise are the mechanisms which explains the causes of ENSO changes.
However Cerveny, R. S. and J. A. Shaffer (2001) et al. in the report, The Moon and El Niño, Geophys. Res., writes about the Moon cycles and ENSO, where they find correlations between the solar cycles and ENSO.
To me at least, it seems that the solution to long range ENSO prediction has for a long time been right in front of the eyes of these researchers, but nobody has taken up the challenge to figure it out.
I see the same reason why the climate community at large has not studied tidal forcing as an explanation for ENSO variations and why none TSI solar forcing as an additional cause for climate forcing ignored. The primary reason is that they have had their education in meteorology, atmospheric physics, thermodynamics or in computer science. Most of them are specialists in a few narrow disciplines, and as such they prefer only to apply knowledge from the fields they know. They are not generalists and display strong resistance for applying knowledge from other area from which they lack knowledge. Then add to that group thinking, peer pressure and lack of funding for research in alternative causes of climate change and this explains the current one-sided situation. This is one of reason that predictions made with computer simulations are failing.
ANN are seldom used in climate science. There are some exceptions. One is research done by Dr William Hsieh from the University of British Columbia who uses this technique for ENSO predictions, but to my knowledge without using any tidal forcing. To learn more about how ANN works and how I have implemented this technique in climate investigation, Click here
@ur momisugly willis
You disparaged him.
He presented his thoughts, and you attacked him.
Personally.
It bothers me.
Genuine science is reproducible and makes testable predictions. There are a number of statements that disqualify this work as science, such as
…I later made a ENSO forecast from late 2011 up to 2020. I cannot show the result here because of proprietary reasons…
Per Strandberg what you have done is very encouraging, I would be interested in whether you found a stronger signal in the phase relationship tidal effects or the declinational tidal effects (from the slowly changing 18.6 year Min/max culmination angle of the declinational component) or the 27.32 day long period from each declinational cycle. It is the declinational tides in the atmosphere, and oceans that has the most effect on changes in meridional flow patterns, and I would think that it would be a prime candidate for a mechanism for the effects you are seeing.
http://research.aerology.com/aerology-analog-weather-forecasting-method/
detailed results from my line of research into global circulation patterns,and related weather effects, have shown a better repeatability when using either a 54 day long two cycle period or a four cycle 109.3 day period. Best results when looking at a 6558 day long pattern that considers the inner planet tidal harmonics as well.
Have a look around my research pages, if you have the time for some mental stimulation on other possibilities than you have looked at so far.
Per Strandberg (@LittleIceAge) says:
January 20, 2013 at 4:04 pm
Gosh, Per, with that reassurance I feel so much better. The fact that you believe in the neural network that you built is so convincing, what can I do but fall in line?
On a more serious note, the problems with ANN are well known, so I fear your reassurances ring hollow. The main issues are that they are difficult to train, often landing on local rather than global minima. They are also very problematic to interpret, often cannot outperform traditional methods, can get locked into minor or meaningless cycles in the data, and are subject to catastrophic failure.
Can they work on a given problem? Perhaps, but they were developed a while ago, people (including myself) have worked on them for years, and they haven’t made much inroads into the world of modeling the real world. That should tell you something.
So when you come in, say you are using neural networks, and refuse to say anything about them, I fear that people will indeed “question their usefulness”. After all, you haven’t given us anything to indicate that you’ve been the guy that finally made ANN work on a complex real-world problem …
w.
Arno Arrak says:
January 20, 2013 at 2:29 pm
I disagree that tidal influence has anything to do with ENSO. ENSO is made possible by the fact that in the Pacific both equatorial currents are blocked and as a result the warm water they carry piles up to form the Indo-Pacific Warm Pool, the warmest water on earth. ………. They always occur in pairs and talk of El Nino-like or La Nina-like periods is just pure nonsense. The natural period of the oscillation is determined by the dimensions of the ocean basin and is about five years. In practice this can vary because there are other things happening in the ocean that can influence timing.
Fascinating stuff, Arno.
But it tends to make me think Per may be onto something with his tidal forcing of ENSO …. you are saying that the oscillation is related to the dimensions of the basin. If you are correct perhaps it (tidal forcing) is one of the “other things happening” which influences timing. ie if waves are moving within a basin, speeding and slowing of the movement of that basin must affect the wave motion.
Sidorenkov in his publication “Physics of the Earth’s rotation instabilities” focuses mainly on “temporal variations in the atmospheric angular momentum and their contribution to the instabilities of the Earth’s rotation” as he discuses irregularity in the rotational motion of the earth. I am surprised that there is not more discussion in there on ocean currents, it seems it would be more significant.
Much of the discussion is on variations in LOD (Length of Day – measured in milliseconds!) but he also points out; “It is not only the Earth’s angular velocity that varies. Our planet has small nutations relative to the axis of rotation. Therefore, the points at which the axis intersects the surface of the Earth (instant poles of the Earth) are in motion. They travel on the Earth’s surface around the middle pole in the direction of the Earth’s rotation, i.e., from west to east.”
If you are correct in your theories Arno, then it seems it must follow that the variations in the earth’s rotation may have some influence in the periodicity and perhaps path of the ‘waves’ you describe.
Physics of the Earth’s rotation instabilities N. S. Sidorenkov (Hydrometeorological Centre of Russia, Moscow, Russia) Online Publication Date: 01 October 2005
Astronomical & Astrophysical Transactions, 24:5, 425 – 439
URL: http://dx.doi.org/10.1080/10556790600593506 (full PDF is available here)
markx says:January 20, 2013 at 7:40 pm
“…If you are correct in your theories Arno, then it seems it must follow that the variations in the earth’s rotation may have some influence in the periodicity and perhaps path of the ‘waves’ you describe…”
e my statement above: It should also be said that it may be nigh on impossible to separate cause and effect here… given the atmosphere, the oceans and the earth itself are all subject to these tidal effects, and differential movement of these different masses is interconnected.
u.k.(us) says:
January 20, 2013 at 4:54 pm
No sir, no way he presented his thoughts. He presented the most vague, misty, obscure, and handwaving of claims. I asked him about his thoughts, raising eight different serious scientific issues.
He refused to tell me his thoughts, flat out refusing to discuss the eight individual issues that I raised.
So no, UK, I didn’t disparage his thoughts, how could I? He wouldn’t tell me his thoughts.
As to whether I attacked him “personally”, since he has given us absolutely nothing scientifically to discuss, I could hardly falsify his scientific claims—he had made none. All I did was, I pointed that out. Yes, that’s personal, I suppose, to point out that he is just blowing smoke … I’m afraid I don’t see how it can be otherwise.
It’s personal, but it’s not a personal attack, UK. It is a statement of fact. He is useless to me, because he is unwilling to do what every honest scientist does—reveal his data and methods. But presumably he wants the scientific imprimateur of WUWT, so he has published this nonsense here, and then declines to defend it.
I’m afraid, UK, that when anyone goes to comment about Per’s kind of action, trying to gain scientific traction by publishing on WUWT while refusing to answer questions or provide any transparency, well, whoever comments on that kind of BS is going to be talking about Per’s personal actions. No way around it.
And no, I’m not happy about his actions. Posting here and refusing to answer questions? That’s not the WUWT way, and Per knew that coming in.
Finally, UK, the relationship between the distribution of the ocean water and the length of day (LOD) has been known and discussed for years. In addition, the fact that the ENSO alteration affects the LOD has been known and discussed for years.
Now, Per pops up and says, Hey, the ENSO isn’t affecting the LOD. The LOD is affecting the ENSO. Well, maybe so. To show that, first Per has to show that the current explanation (conservation of momentum, the “skaters’ arms” phenomenon) is inadequate to explain the change in the LOD.
Has he done that, UK?
Then Per needs to show some kind, any kind, of evidence that causality runs from the LOD to the ENSO, rather than the other way around. A Grainger Causality analysis, time delays, something.
Has he done that, UK?
In fact, he has done nothing of the sort. He has made a very doubtful claim, offered no evidence to support it, then refused to defend it … and you are busting me for pointing that out? I’m sorry to discuss his personal actions, but he’s given us nothing else to discuss.
w.
I must agree with Willis, there is nothing here if you don’t disclose your calculations or your forecast [same problem Piers Corbin has],apart from the many obvious errors committed.
Back of the envelope calculations for:
A. What is the angle from vertical for a deceleration of 2 ms/day in 14 days?
B. What will be the change in elevation of the surface across the Pacific Ocean do to that deceleration?
From Deviation of Day Length from SI day (Wikipedia) it seems logical that the greatest rate of change in deceleration occurs during half a lunar orbit (14 days) and the chart says the maximum change is 2 ms / day in a short period.
Radius of Earth at Equator: 6,378,000 m
Seconds per solar day: = 86400 Sec/day
Seconds per Sidereal Day = 86164 Sec/sid.day
Circumference at Equator = 40,074,156 m
Velocity of Surface at Equator = 465 m/sec
Assumed maximum deceleration = 2 msec/day in 14 days
Average increase in day over 14 days = 1 msec/day
total delay in 14 days = 14 msec
Distance traveled in 14 ms at equator: = 6.51 m
What constant acceleration nets a distance of 6.51 m in 14 days?
Number of days = 14 days
seconds in 14 days = 1,209,600 seconds
distance traveled = 1/2 * a * t^2 under constant acceleration; a = 2 * dist / t^2
Distance = 6.51 m
a, lateral acceleration = 8.90E-12 m/sec^2
The acceleration must be zero at the start and the end of 14 days (cyclic).
So a better estimate is constant acceleration for 7 days covering half the distance, then reversed acceleration for another 7 days to cover the other half.
half distance: = 3.26 m
seconds in 7 days = 604800 seconds
a, lateral Acceleration = 1.78E-11 m/sec^2
g, acceleration of gravity = 9.77E00 m/sec^2
a/g = 1.82E-12
Angle from vertical: 3.76E-07 sec of arc
offset of center of acceleration from center of earth = 1.16E-05 m
by similar triangles, change in sea surface elevation across pacific ocean (19000 km) during deceleration.
distance across Pacific = 19,000,000 m
B) Change in sea level across Pacific = Dist * a/g = 3.46E-05 m
So a 2 ms/day increase in LOD over 14 days, will create an surface equatorial lateral acceleration that will make the surface of the Pacific Ocean just 0.35 millimeters higher in the east (Panama) than the west. (Jakarta)
May I suggest 0.35 millimeters is much smaller than a molehill?
Unless someone can point out a blunder here, I think changes in the length of day have insignificant manifestations in anything climatological.
It is wonderful to observe and enjoy an active discussion of proper scientific method.
I am humbled in the presence of great minds.
The story of the artificial intelligence getting the images of tanks wrong can be found in the documentary “The Machine That Changed The World”, in part 4 I think. By my recollection though, it was a difference between sunny days and overcast days of the differences in identifying correctly or not the tanks. (I watched the original US airing of the series and still have them recorded on video tape).
http://topdocumentaryfilms.com/the-machine-that-changed-the-world/
Willis…”and you are busting me for pointing that out”
No, that was not what UK was trying to point out.
I have to say I had the same view as UK. I was disappointed with your first post on this thread. UK did not say you should not point out the fallacies of his analysis. It was the way you did it that bothered UK and me. Go back and read your post. Now go read richards post (defending you). Richard did it without disparaging the author. You chose a different path. Not a big deal really, we are all grown ups here mostly, but I was taken aback by it. I have come to respect your writing here. You have a way with the word. I would have expected better than that from you.
Now if the author had disparaged you in the past or treated you with disrespect, I could see where it might have been tempting to return the favor. I assume that was not the case here.
There are lots of folks here with different levels of expertise and non expertise. I think for the most part we are in the same camp, we are seeking the truth. The author sounded like he was someone putting his own time into finding answers down a path he chose, thought he might have found a winning argument, and wanted to share it with the rest of us. Just because some takes a chance to stick there neck out, no reason to disparage him when you decide to chop it off. There are ways to tell someone they might be wrong without making them feel disrespected, you can just ask my wife.
If you feel so strongly that the article was inadequate, maybe your beef should be with Anthony for posting it.
With all respect to all, Steve
‘rgbatduke’ and ‘Willis Eschenbach’ say what I was planning to say but more elegantly.
Per Strandberg: For all practical purposes I’m not going to disclose in detail precisely how I proceeded or what type of optimization I made or what specific feature which it is that is responsible for the connections between the tides and ENSO variations.
The only comment I’m going to make on that issue is that the optimizations which worked best make full sense based on physical principles.
After all, this discovery has a certain monetary value. Not necessarily for me so much, but sure for others.
Color me unimpressed. Until Willis Eschenbach’s technical criticisms of your work are addressed fully and openly, there is no reason for anyone else to accept your personal testimony that you believe the result.
Go Home says:
January 20, 2013 at 9:49 pm
Go Home, I raised eight separate scientific issues regarding his claims. Rather than deal with the issues I raised, he cut and ran. Now that’s bad enough, but rather than deal with the issues I raised, you and UK want to lecture me on my tone. You don’t care about the gaping ragged holes in his scientific claims, you don’t care that he has cut and run, you pay no more attention to the scientific issues I raised than Per did.
Instead, you want to lecture me on my tone, no scientific issues for you, I guess.
Look, Go Home, I have no use for businessmen trying to pass themselves off as scientists, and I’m happy to tell people so. I see that you want me to blow in Per’s ear and tickle his stomach, or something like that, so he stays happy. Sorry. Not gonna happen. I don’t care if he’s happy. He comes here masquerading as a scientist and refuses to ask questions, claiming business reasons, I’ll call him on it.
Per knows better than that, he reads WUWT, he knows about scientific transparency. So when he doesn’t want to be transparent, when he doesn’t want to reveal his methods, when he refuses to reveal his predictions, yes, I’ll bust his chops.
You keep trying to cast this as Per the poor scientist trying to get his voice heard and me cruelly jumping on him. That’s not what’s happening. Instead, we clearly have Per the businessman trying to masquerade as a scientist, in order to provide some kind of scientific imprimatur for what is no more than a quarter-baked claim.
To recap:
• I object to him passing himself off as a scientist while refusing to act in a scientific, transparent manner.
• I object to him not revealing his ENSO prediction(s).
• I object to him using WUWT for his business purposes.
• I object to his refusal to answer scientific questions regarding his theory.
Finally, yes, Go Home, I know that there is likely some nice, California-approved, sensitive New Age way to tell a man that he is not welcome to come in to WUWT, make pseudo-scientific claims, and then refuse to answer questions about his claims. I suspect that either you or UK could tell Per that, in a way that wouldn’t ruffle his feathers in the slightest.
What you guys fail to understand is that I want to ruffle his feathers. I don’t want him to be at ease while refusing to answer questions, I want him aware of the scorn with which I regard that kind of cheap trick. I don’t want him walking away thinking he got away with it. I want him to walk away hanging his head, and take his business reasons with him. I want to discourage him from this course of behavior.
And most assuredly, I don’t want anyone else to try this kind of nonsense, posting up some claims on WUWT and then running away when people ask questions. If making Per squirm about his actions serves to dissuade one person from doing the same, I’ll consider it a win.
You and other people keep assuming that I pick my tone at random, or that my tone reflects my instantaneous emotional condition. It is rare that either one of those is true about me. I pick and choose my words, and the tone of my words, with great care. It was my intention to be aggressive and unbending with Per. I really don’t like that kind of action, and despite my attempts to be a reformed cowboy, I retain enough of my upbringing to simply tell the truth about that.
w.
Very nicely done. You may find the lunar tidal calculations / formulas in the paper from this link:
http://chiefio.wordpress.com/2013/01/04/lunar-cycles-more-than-one/
This paper:
http://www.pnas.org/content/97/8/3814.full
calculates tidal forces on several different time scales, most longer than ENSO, but the cycles considered might help you tune things more closely (the lunar orbital changes are on several time scales and causes).
These folks:
http://www.appinsys.com/GLobalWarming/SixtyYearCycle.htm
find a sixty year cycle, for example. It pretty much explains the 60 ish year weather cycles and ocean (PDO etc.) changes.
So “watch that moon!” 😉
ENSO has a period of approximately 5 years.
Earth’s magnetic field (generated in the liquid core) has two dominant periods 21.3 and 5 years. 21.3 years can be directly linked to the Hale cycle, because exactly 21.3 years is the period of the sunspot magnetic cycle since 1880.
http://www.vukcevic.talktalk.net/EMFspectrum.htm
It is unlikely that the ENSO is able to ‘shift’ the liquid core circulation, more likely is the other way around, considering the mass ratios of the corresponding components.
In support of the ‘outward’ hypothesis it can be demonstrated that the tectonics of the sub-equatorial Pacific correlates with the ENSO time line.
http://www.vukcevic.talktalk.net/ENSO.htm
Some of the commentators are evoking ‘predictability’ as a proof of a hypothesis.
That may not be always possible, in such a case science is reliant on the backdated correlation, as it often is, e.g. by using ice core sampling data, magnetic polarity in the mid Atlantic ridge, etc.
‘Predictability’ is desirable but not essential to the instituting of a new understanding of a physical phenomenon.
Obviously, Willis needs no help in articulating the true spirit and specific values of the scientific method. Bravo, Willis!
And Braddles too, for instantly becoming skeptical when Per announced his revolutionary ENSO predictions for the next decade were top secret. One doesn’t need PhD in physics to smell, uh, confirmation bias. The common sense of an illiterate fishmonger will do just fine.
Btw, Anthony is perfectly on topic by making his venue available for bleeding edge new ideas. WUWT is a forum for free and open debate, not a realclimate mouthpiece for micro-polished propaganda. It is entirely predictable that some ideas presented here will not withstand thread scrutiny.
Just to be sure, none of us can have any idea whether Per is on to something based on the sales pitch he has presented anymore than we can be sure that a Ford is the best car for your money based upon the glossy brochure at the dealership. Being ever the optimist, I had hoped Per had found the Masonic key to ENSO while reading his post. Sadly, his form stinks of the same post-mod trust-me appeal-to-authority rot that WUWT and Climate Audit, et al, stand as bulwarks against.
Nevertheless, this is the kind of wide ranging, highly instructive, madly unpredictable thread that makes WUWT the most sage science blog on the bloody planet.
We learn best from error deconstructed.
The only mechanism by which ENSO can be driven by changes in Earth’s rotation is by variations in the tidal force.
———-
Prove it!
I haven’t a clue, but my first guess is that changes in the earth’s rotation rate are contolled by conservation of angular momentum, when there are mass redistributions due to earth quakes, ocean currents etc.
ENSO is associated with changes in atmospheric pressure and hence changes in sea level. Thus it seems plausible from this effect alone could affect rotation rates. The question is do the numbers add up?
The danger of doing this kind of correlation study is running foul of good old correlation is not causation.
Willis Eschenbach says:
“On a more serious note, the problems with ANN are well known, so I fear your reassurances ring hollow. The main issues are that they are difficult to train, often landing on local rather than global minima.”
ANN with backpropagation is a commonly used method. I have used backpropagation. It is a fast method but suffers from problems. It is inflexible, and results in most cases get stuck in local minima.
Therefore, I don’t use backpropagation.
Instead, I randomly pick weights. Then I change the weight by adding a step. The size of this step is then chosen randomly within a set of limiting borders. Which makes it possible for the program to jump over local minima.
Then if, the result improves I keep the new weight or if it doesn’t I keep the old weight.
By doing this, I eliminate the risk of getting stuck in local minima.
True, this method is slower, but I avoid getting lost in minima. It almost always converge and I avoid the risk of divergence in the training set.
This is reminiscent of Nils-Axel Morner’s claim that tiny variations in LOD (one part in 10^8) have an effect on climate. Adding to Raseys analysis, a few points:
First, is it possible for reversible tidal effects to influence ocean currents? Keep in mind that the measurable effects on LOD, those of fortnightly frequency, are about 99.99% reversible–that is, terrestrial angular momentum is retained, angular velocity is recovered. Which is no say no energy has been transferred to or from the hydrosphere. That tiny fraction of tidal friction which transfers the earth’s angular momentum to the moon–mainly due to semi-diurnal tides–is right at the edge of the IERS’s ability to measure. Clearly the world’s rivers have a far greater–but still negligible–capacity to affect ocean currents, than irreversible tidal effects.
The tidal force is the weakest of measurable terrestrial influences–a gentle sloshing across the ocean basins of a fraction of a mm is sufficient to raise the water along the shores. The tidal bulge is wave, not a current, and a weak wave at that. The idea that it governs ENSO is as far fetched a claim as ever there was. Why waste statistical analysis on conceptually flawed hypotheses?
A further minor point: whereas the ocean is stratified by density, tidal effects occur at all depths, but their cumulative force is no greater than the sum as observed at the surface. Still the slight motion is spread through all depths, hence the mm claim. –AGF
Science is (or should be) a full-contact sport. You figuratively bash the brains out of opponents/other players & expect no less in return. Afterwards, you shake hands & maybe even have a beer w/him at the local bar.
Advice — get thicker skins and improve your work…
agfosterjr says:
January 21, 2013 at 7:17 am
“Keep in mind that the measurable effects on LOD, those of fortnightly frequency, are about 99.99% reversible–that is, terrestrial angular momentum is retained, angular velocity is recovered. Which is no say no energy has been transferred to or from the hydrosphere.”
It does not follow. Angular momentum and energy are entirely different things. Many spin stabilized spacecraft, for instance, are controlled by dissipating energy to damp nutation, while angular momentum stays constant.
I’d recommend using a feedback neural networks (NN) – not to be confused with back propagation or error correction – also known as a recurrent NN – instead of a feedforward NN.
Feedforward NN is for static pattern classification.
For a time series, you need a feedback NN – or RNN.
Take a look at
Evolino
http://www.idsia.ch/~juergen/evolino.html
which is now part of PyBrain.
Here are some of the listed advantages:
1. Recognition of temporally extended patterns in noisy input sequences
2. Recognition of the temporal order of widely separated events in noisy input streams
3. Extraction of information conveyed by the temporal distance between events
4. Stable generation of precisely timed rhythms, smooth and non-smooth periodic trajectories
5. Robust storage of high-precision real numbers across extended time intervals of noisy input sequences
Also, since the stratosphere is thermally stable (warm on the top and cold on the bottom) I’d recommend using the phase of the QBO winds as an input. The QBO is actually an oscillation – unlike ENSO.
RE: my 9:18 pm back-of-the-envelope calc above.
It turns out I did blunder in my last step.
B) Change in sea level across Pacific = Dist * a/g = 3.46E-05 m (OK)
So a 2 ms/day increase in LOD over 14 days, will create an surface equatorial lateral acceleration that will make the surface of the Pacific Ocean just 0.035 millimeters, 35 microns, higher in the east (Panama) than the west. (Jakarta).
[I was too large by a factor of 10 — slipped a decimal point]
That is less than the thickness of 1.5-mill painter’s plastic – stretched halfway around the world.