From the University of Hawaii ‑ SOEST:
Climate researchers discover new rhythm for El Niño
El Niño wreaks havoc across the globe, shifting weather patterns that spawn droughts in some regions and floods in others. The impacts of this tropical Pacific climate phenomenon are well known and documented.
A mystery, however, has remained despite decades of research: Why does El Niño always peak around Christmas and end quickly by February to April?
Now there is an answer: An unusual wind pattern that straddles the equatorial Pacific during strong El Niño events and swings back and forth with a period of 15 months explains El Niño’s close ties to the annual cycle.
This finding is reported in the May 26, 2013, online issue of Nature Geoscience by scientists from the University of Hawai’i at Manoa Meteorology Department and International Pacific Research Center.
“This atmospheric pattern peaks in February and triggers some of the well-known El Niño impacts, such as droughts in the Philippines and across Micronesia and heavy rainfall over French Polynesia,” says lead author Malte Stuecker.
When anomalous trade winds shift south they can terminate an El Niño by generating eastward propagating equatorial Kelvin waves that eventually resume upwelling of cold water in the eastern equatorial Pacific. This wind shift is part of the larger, unusual atmospheric pattern accompanying El Niño events, in which a high-pressure system hovers over the Philippines and the major rain band of the South Pacific rapidly shifts equatorward.
With the help of numerical atmospheric models, the scientists discovered that this unusual pattern originates from an interaction between El Niño and the seasonal evolution of temperatures in the western tropical Pacific warm pool.
“Not all El Niño events are accompanied by this unusual wind pattern” notes Malte Stuecker, “but once El Niño conditions reach a certain threshold amplitude during the right time of the year, it is like a jack-in-the-box whose lid pops open.”
A study of the evolution of the anomalous wind pattern in the model reveals a rhythm of about 15 months accompanying strong El Niño events, which is considerably faster than the three- to five-year timetable for El Niño events, but slower than the annual cycle.
“This type of variability is known in physics as a combination tone,” says Fei-Fei Jin, professor of Meteorology and co-author of the study. Combination tones have been known for more than three centuries. They where discovered by violin builder Tartini, who realized that our ear can create a third tone, even though only two tones are played on a violin.
“The unusual wind pattern straddling the equator during an El Niño is such a combination tone between El Niño events and the seasonal march of the sun across the equator” says co-author Axel Timmermann, climate scientist at the International Pacific Research Center and professor at the Department of Oceanography, University of Hawai’i. He adds, “It turns out that many climate models have difficulties creating the correct combination tone, which is likely to impact their ability to simulate and predict El Niño events and their global impacts.”
The scientists are convinced that a better representation of the 15-month tropical Pacific wind pattern in climate models will improve El Niño forecasts. Moreover, they say the latest climate model projections suggest that El Niño events will be accompanied more often by this combination tone wind pattern, which will also change the characteristics of future El Niño rainfall patterns.
Citation: Stuecker, M. F., A. Timmermann, F.-F. Jin, S. McGregor, and H.-L. Ren (2013), A combination mode of the annual cycle and the El Niño/Southern Oscillation, Nature Geoscience, May 26 online publication at http://dx.doi.org/10.1038/ngeo1826.
h/t to Dr. Leif Svalgaard
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Bob,
I think Greg Goodman’s point is that if you talking beat frequencies then you have to consider temperature anomalies ( i.e. ocean sea-surface temperatures oscillating about some long-term mean) and compare them to trade wind speed anomalies (i.e also about some long term mean wind speed).
Greg Goodman said: Yes, My gut feeling is that it’s ocean currents driving winds rather than the opposite. Tail shakes dog etc.
Not if your considering up-welling of cooler deep ocean water – in this case the changes in sea-surface temperature are driven by surface winds.
Thanks Bob,
http://climategrog.wordpress.com/?attachment_id=279
I don’t know if this is relevant to the wind patterns they were referring to but it looks similar to their figure 2 as far as I can see from the blurred preview.
Those peaks look like split pairs (I’ll work out what period is splitting them later but I’ll bet it’s the same thing).
taking the average frequency as the unsplit value I get periods of 19 months and 4.5 years.
Ian will like that.
Fits with my suggestion that the real frequency is half the value they are using but refers to 2*10 months (probably was rounded from 9.5).
“Also ran” peaks at 7.6 and 33 years. The two bumps around 0.4 correspond to circa 30 months.
Bang on.
Bob do you have any detailed knowledge of the wind patterns? Does what I roughed out about warm spot either marking strong winds and cross-over marking trough of wind intensity match what really happens?
If so it looks like we’re close to a periodic account of El Nino .
BTW the split peaks are about 3 and 5 years so this also explains the vague period attributed to ENSO cycle of being between 3 and 5 years.
Like I say if climatology had been build on engineering data processing and not econometric statistical methods we’d have worked this out 30 years ago.
Ian Wilson says: “I am assuming that Bob can show a correlation between the Trade Wind speed anomaly and the signed latitude of the location of the West-Pacific warm pool?”
We’re not concerned about anomalies; we’re dealing with absolutes. In other words, we normally think of ENSO as anomalies, but ENSO works from absolutes. The seasonal cycles in trade wind strength and direction in the Northwest and Southwest tropical Pacific oppose one another as I suggested in my earlier comment:
http://i40.tinypic.com/4hd3wx.jpg
Greg Goodman says: “Bob do you have any detailed knowledge of the wind patterns?”
Detailed? Nope. Sorry.
OK , I have other thing to do but I could not resist.
“30m” is about 31.7 months and is being split by 39 years. The other two are indeed being split by the same frequency: around 22.9 years.
Ian may like to play with that.
When you see that kind of structure and all the numbers tie you can be fairly sure it’s not “stocastic noise” 😉
BTW the counter part to the 39 years : Modulation beats: 2.6405 * 39.2375
Compare to the cycles I found by aligning eruptions yesterday:
http://climategrog.wordpress.com/?attachment_id=278
~11 / 4 = 2.75 years
Seek and you shall find.
Greg Goodman says:
May 27, 2013 at 4:30 am
I disagree. What the Wiki page says is all true, and readily demonstrated in my high school physics class, but the audible issue is that people can’t distinguish a negative signal from a positive. The beating signal alternates between positive and negative, and that’s where this idea of “full wave rectified” sound is coming from.
Look at the Wiki page again, especially the lower image at http://en.wikipedia.org/wiki/File:Beat.png and note the red low frequency sinusoid.
What you’re saying is analogous to saying that the positive and negative PDO phases have the same effects but differ from the neutral PDO.
Greg Goodman says: “Yes, My gut feeling is that it’s ocean currents driving winds rather than the opposite.”
Short reply: I disagree.
Long reply: The greatest convection occurs where the temperature is highest–in the tropics. The inrush of air from higher latitudes provides make-up air to the convection. And the winds in both hemispheres flowing toward the equator are deflected toward the west by the Coriolis effect. Those are the trade winds. The trade winds blowing from the Northeast to the Southwest in the Northern Hemisphere and blowing from the Southeast to the Northwest in the Southern Hemisphere cause the equatorial currents to flow.
However, the trade winds, ocean currents, the temperature gradients from east to west, and sea surface height from east to west in the tropical Pacific are all coupled. The trade winds and the temperature gradient provide positive feedback to one another–positive feedback that’s restricted by Mother Nature. And those interrelationships are impacted by other variables. If a large weather-related temperature anomaly works its way into the tropics along the eastern boundary currents from the west coast of the Americas, it influences the balancing act between trade wind strength, temperature gradients and sea surface height. And if there’s a large and persistent change in the wind patterns in the western tropical Pacific–caused by a tropical cyclone, for example–that change can be enough to destroy the balancing act. Then gravity takes over and causes the warm water in the west to slosh to the east, carried by a strengthened equatorial counter current, and there’s an El Nino. The trade winds in the west reverse, becoming westerlies, and now provide positive feedback to the El Nino.
Just a comment about my “split peaks” , it’s basically the same thing again . When a very long periodic variations interferes with a short one it produces a doublet, like the ones I instantly recognised in this spectrum.
This kind of effect is very well known in spectroscopy and acoustics. The unperturbed frequency is the average of the two peaks, The split gives the other one.
Just in case anyone is wondering what I’m burbling on about 🙂
The fact that the 4.5 doublet yields that same splitter and the 19 month doublet means it all hangs together.
I need to run though all this more thoroughly later but it looks coherent.
Bob says: “However, the trade winds, ocean currents, the temperature gradients from east to west, and sea surface height from east to west in the tropical Pacific are all coupled. ”
Yes , probably best not to over simplify.
I’m a bit allergic to all this talk of “sloshing” back and forth. Water does not “pile up” anywhere in it’s liquid state. If a wind blows on a lake all the dead leaves and flotsam pile up on once side and surface water is displaced. But the water does not pile up on the far shore, just the leaves.
If mean sea level is higher it must is more to do with the depression on one side and the high on the other that is driving the wind in the first place. Otherwise it would just flow north and south from the west Pacific and circulate.
I know this is the current orthodoxy and you are well read on the subject but I’ve seen enough dodgy climatology to be very mistrusting by now and this doesn’t ring true to me.
Seeing the apparent synchronising between major eruptions and tropical SST that I found yesterday, I’m more convinced that ever that there is an inertial driver at work here.
finding the crude 2.75 estimate in the volcano stack tie in with the central frequency of one of the main doublets here, just about clinches it for me.
Orchestra a little weak in the volcano section but a nice little symphony nonetheless …
If they truly have discovered the pattern, they should be able to make a prediction about when the next El Nino will occur. Have they?
Steven Mosher: you know if they tied the 15 month cycle to the orbit of juniper and saturn and sun spots people would have no doubts about a paper they havent read.
That’s clever.
If this paper ever comes out from behind the paywall, I hope someone here will alert us. The abstract is too skimpy.
Bill Inis: http://s17.postimg.org/noobqikhr/Trades_ENSO_Apr_2013.png
Interesting.
If you are discussing causality I’d want to look for lead/lag. My first impression was that Nino3.4 was lagging wind index. However, I realised this is yet another running bloody mean problem.
http://climategrog.wordpress.com/2013/05/19/triple-running-mean-filters/
Nino3.4 data has clearly been “smoothed”, and since there is a general downward trend in the data all the major peaks get bent to the right and give a false impression that wind is leading SST.
If I try to ignore the processing error, there seems to be no lag, but that needs an correlation lag plot to check.
If wind was driving SST by shifting away the surface layer it would require time. I take no lag to discount that but it needs checking with correlation lag on non runny mean bent data.
“When anomalous trade winds shift south they can terminate an El Niño by generating eastward propagating equatorial Kelvin waves that eventually resume upwelling of cold water in the eastern equatorial Pacific.”
Now this is the kind of mathematically precise language that cam make a geologist cringe. Or a salesman proud. Eventually? Exactly what sort of events? By any account these waves are shallow and they reflect with opposite sign.
Just one of a number of hedges. Combination tones? Once apron a time a serious violinist I can report that combination tones arise from inadvertent harmonics when the string is incompletely pressed, when two strings are engaged by the bow, and are the nature of harmonics in general.
Having just returned from a fortnight in the trade winds, I can also report from this definitely not statistically significant interval that the trades seemed far more drawn by the ITC than pushed by the synoptic circulation. Beastly in the afternoon, gone at night.
Greg Goodman says:
May 27, 2013 at 9:41 am
A well known (and often very detrimental) effect on the Great Lakes, and I assume on other long lakes is an effect called “seiche.”
http://www.isgs.illinois.edu/sections/engin-coast/lakemich-coastal-seiches.shtml describes one on Lake Michigan:
Seiches don’t really apply to the Ocean basins, but things like trade winds etc can drag a lot of water over thousands of miles and pile it up before the next continent.
Greg Goodman – regarding the split peaks – very good observation!
http://climategrog.wordpress.com/?attachment_id=279
I think that what you are trying to point out is that there are frequency peaks at 1.59 years and 4.43 years that are amplitude modulated by a higher frequency phenomenon present in the data.
What I find is that:
a) a 22.7 year period amplitude modulating a 4.43 year period in the (sea surface temperature?) spectra above would produce split peaks at 5.504 years and 3.707 years.The observed peaks are at 5.502 and 3.706 years.
b) a 22.3 year period amplitude modulating a 1.59 year period in the (sea surface temperature?) spectra above would produce slit peaks at 1.712 years and 1.484 years. The observed peaks are at 5.502 and 1.485 years.
CONCLUSIONS
1) The 22.3 – 22.7 year modulating signal is clearly the Hale cycle [assuming that they do not have a date series that just happens to be ~ 22 years long].
2) The 4.43 year period is just half the 8.85 year precession cycle for the lunar line-of-apsides with respect to the fixed stars.
3) And more controversially, I believe that the 1.59 year period (~ 19 months) is the 1.599 year cycle when Venus and the Earth align on the same side of the Sun.
I am hunkered down in my fox hole [flack jacket-on] ready for the mud to fly.
Ric , red what you quoted: “The wave or waves will travel across the lake until the seiche reaches shore, where it can be reflected and travel to the opposite shore.”
No where does it say the water “piles up” . It a wave that propagates.
Greg Goodman said earlier:
The author’s 1 year and 5 years will produce frequencies(1+1/5) and (1-1/5) per year, ie 1.2 and 0.8 per year. That gives 15 and 10 months. Like many doing this kind of calculation they seems forget (or not realise) the second is just as important.
AND
They are making the error of confusing the audible beat frequency with actual frequency. This is explained in Wikipedia page I linked above. The REAL frequency is half what they calculate. The ear can only hear acoustic intensity so the “beats” we hear are twice as fast as the actual oscillation.
So I double my previous results to give 20 and 30 months ( twice their 15m)
Now 30m is 2.5 years
http://climategrog.wordpress.com/?attachment_id=278
Looks like this deserves a closer look.
MY COMMENTS:
Couldn’t the 30 month period just be the Quasi-Bienneal Oscillation (QBO) and the 20 month period period 19 month period that you observe in the spectra of sea surface temperatures (?) posted earlier?
Could you post the details of how the spectra was generated please?
Lets say we use 1 year and 4.43 years instead of 1 and 5 years. This will produce frequencies(1+1/4.43) and (1-1/4.43) per year i.e. 1.226 and 0.774 per year.
That gives 14.71 months and 9.29 months.
If, as you suggest, they are using the beat frequencies rather than the actual frequencies involved, then we would have to double these periods.
Hence, we would get 29.4 months {the QBO is ~ 28 months) and 18.6 months [not far off the 19.1 months from your spectra].
Greg Goodman says:
May 27, 2013 at 8:29 pm
If your backyard floods in a seiche or a NYC subway tunnel floods in a storm surge, most people’s primary concern is not how the water lifted and moved in, but that it did. From the point of view of whoever is flooded, water piled up.
Umm, yours was the first mention of “piles up”. H.R. and Bob Tisdale said slosh, which is closer to the “standing wave” Wikipedia refers to seiches. IIRC, my high school biology teacher probably said slosh in reference to seiches on Lake Erie.
Typos Ian: “The observed peaks are at 5.502 and 1.485 years.”
” that are amplitude modulated by a higher frequency phenomenon present in the data.” lower frequency
1,2 and 3 seems fairly obvious propositions though I’d like to redo the spectrum by a different method and verify the numbers. I’ll also have to have a look at the trade winds index that Bill linked to , I think that is probably what they used in the paper.
I’d always thought ENSO cycle was a bit of “non cycle” but it seems there’s a lot more order in there than is at first apparent. (As is true of most things in climate).
It was a 32 year dataset I worked on. In any case rectangular modulation caused by the dataset length cropping a signal would not produce a nice clean doublet. That requires sinusoidal modulation.
Spotting these doublets and associated A.M. triplets is a valuable way to detect the influence of long orbit planets that have a period well out side the length of the data set.
http://climategrog.wordpress.com/?attachment_id=216
This pattern that I spotted in Arctic ice area data indicates a 12.83 year modulation. A figure I’m sure Ian will recognise instantly as a Jupiter – Neptune period.
Junipers, junipers! I told you they wanted junipers.
.
Greg,
Sorry, I often switch between periods and frequencies and so I get my “highers” mixed up with my “lowers”. Thanks for the correction.
It seems we live in interesting times.