Reader Eric Worrall writes:
I was playing with Wood For Trees, looking at the relationship between Pacific Decadal Oscillation vs global temperature (Hadcrut 4), when the following graph appeared.
The interesting thing is PDO in this graph appears to have predictive skill for changes in global temperature – the changes in PDO appear to match changes in global temperature, once the graphs are normalised, but temperature lags PDO by around 5 years.
Source: http://goo.gl/hzOxW
Is it all just coincidence? Bad endpoint choice? Or does it in fact have some predictive value?
Readers are invited to weigh in.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
rgbatduke says:
June 5, 2013 at 7:40 pm
That will not happen because it would reduce entropy and violate the second law of thermodynamics.
Ah, would you care to prove, justify, or otherwise support this assertion? I’m not defending the assertion that heat sequestration is occurring on a grand scale, but it is surely occurring all of the time on some scale, even if only the minor scale of sunlight warming the ocean by day and the heat being released at night. I cannot think of any good reason to think that this or any process akin to this would either reduce entropy or violate the second law.
——————————–
Well, I’m not sure what Manfred was thinking, but I thought his original idea was something a little different. If the energy that would have caused 1C atmospheric warming had it not been sequestered in the deep ocean GOES into the deep ocean, you’re never going to get 1C of warming from that energy back out of the water again. I guess you can get some of it back out with the right assumptions. Generally speaking though, if the ocean is colder and denser than the atmosphere, wouldn’t most of the energy that goes into it both 1) cause less heating to the dense water than it would cause applied to the atmosphere and 2) tend to stay there unless the ocean becomes warmer than the atmosphere?
I could be making a mistake, and I’d be most grateful for a clear correction if I am.
Thanks
Eric Worrall
Here is my correlation graph between PDO and detrended 21-years running mean of the global mean surface temperature.
http://www.woodfortrees.org/plot/jisao-pdo/mean:252/normalise/plot/hadcrut4gl/mean:252/from:1910/detrend:0.8/normalise
It shows the PDO is a leading indicator of the global mean surface temperature.
It agrees with your result.
<blockquote]Why is it so very difficult to say “we don’t know”?
rgb
A question I’ve brought up frequently. Often in the form “It is ok to say ‘I don’t know'”. And I don’t rule out that we can’t know. What we do know is the Earth has been a frozen snowball and an ice-free water planet at various times, so those are the known natural extremes. We know CO2 content has been very much higher and very much lower, and that it is uncommonly low now, and that the minor recent increases are having a positive influence on most things. We know none of the doomsday predictions are validated by observation. We need to stop allowing this bogieman to destroy our economy and quality of life and focus on creating jobs, not carbon tax schemes.
Steven Mosher says: “PDO is temperature.”
No, it is not.
Eric Worrall: You’ve presented an argument that’s been around for a few decades.
Unfortunately, the PDO does not represent the sea surface temperature of the North Pacific. The sea surface temperatures of the North Pacific north of 20N, where the PDO is statistically derived, is inversely related to the PDO.
There is no mechanism for the PDO to vary global temperatures.
See the posts, working backwards in time, here:
http://bobtisdale.wordpress.com/2013/05/14/multidecadal-variations-and-sea-surface-temperature-reconstructions/
And here:
http://bobtisdale.wordpress.com/2011/06/30/yet-even-more-discussions-about-the-pacific-decadal-oscillation-pdo/
And here:
http://bobtisdale.wordpress.com/2010/09/14/an-inverse-relationship-between-the-pdo-and-north-pacific-sst-anomaly-residuals/
And here:
http://bobtisdale.wordpress.com/2010/09/03/an-introduction-to-enso-amo-and-pdo-part-3/
About time Bob Tisdale showed up. My two cents, the PDO is tweaked by volcanoes and Svensmark’s GCR effect. The effects take hundreds of years to cycle out and temps stay within a couple of degrees either way. Things look strange when ALL of the ocean cycles break sync, ala Swanson and Tsonis, but still temps never get too far out of hand. Just a hunch.
As I “knew” he would, Bob Tisdale weighs in.
Time to listen.
Ah, how I love playing with the graphs at WoodForTrees. Even more fun with the Normalize function.
Here you can see the obvious correlation between the Atlantic Multi-decadal Oscillation and the Arctic sea ice extent:
http://woodfortrees.org/plot/nsidc-seaice-n/from:1979/mean:13/normalise/plot/esrl-amo/from:1979/mean:13/normalise/scale:-1/plot/nsidc-seaice-n/from:1979/mean:13/normalise/trend/plot/esrl-amo/from:1979/mean:13/normalise/scale:-1/trend
Make the melt prediction for this year with that!
Except starting around 2004, AMO diverged from the Arctic sea ice extent, their tends went different ways. It’s just like they warned us, the climate is getting freaky!
(Note: due to the 13-mo running average, it’s necessary to overshoot a half year when specifying the intersection.)
http://woodfortrees.org/plot/nsidc-seaice-n/from:1979/to:2004.5/mean:13/normalise/plot/esrl-amo/from:1979/to:2004.5/mean:13/normalise/scale:-1/plot/nsidc-seaice-n/from:1979/to:2004.5/mean:13/normalise/trend/plot/esrl-amo/from:1979/to:2004.5/mean:13/normalise/scale:-1/trend/plot/nsidc-seaice-n/from:2003.5/mean:13/normalise/plot/esrl-amo/from:2003.5/mean:13/normalise/scale:-1/plot/nsidc-seaice-n/from:2003.5/mean:13/normalise/trend/plot/esrl-amo/from:2003.5/mean:13/normalise/scale:-1/trend
However, Arctic sea ice should lag the AMO. If the AMO makes more ice favorable, it still takes awhile for the thin first-year ice to survive and become the longer-lived multi-year ice. With some work, a four year lag gives great results, up to the 2004 freak-out.
http://woodfortrees.org/plot/nsidc-seaice-n/from:1979/to:2008.5/mean:13/normalise/plot/esrl-amo/from:1979/to:2004.5/mean:13/normalise/scale:-1/plot/nsidc-seaice-n/from:1979/to:2008.5/mean:13/normalise/trend/plot/esrl-amo/from:1979/to:2004.5/mean:13/normalise/scale:-1/trend/plot/nsidc-seaice-n/from:2007.5/mean:13/normalise/plot/esrl-amo/from:2003.5/mean:13/normalise/scale:-1/plot/nsidc-seaice-n/from:2007.5/mean:13/normalise/trend/plot/esrl-amo/from:2003.5/mean:13/normalise/scale:-1/trend
Voila! The trends match near perfectly until 2004, when the Arctic Ocean goes haywire.
Now that I have conclusively demonstrated the global climate underwent a hidden chaotic transformation around 2004, it is clear that further research is required to analyze and project the degree of catastrophic impact that will result. We can only hope the federal government will pursue this matter on behalf of the general public, as required by law.
Release the funding!
The 61 year Jup-Sat Scafetta cycle is missing in the discussion
as climate driver and the PDO as resulting effect…..
rgbatduke,
“why they WERE either sudden, or gradual as the case may be, or gentle and moderate”
I personally believe this is most important question for climate science. Not “are we warming or cooling” – but understanding the conditions of rapid change. I don’t even hope for a prediction of rapid change, just a way to recognize it once it starts. The consequences for millions are profound and a few months can matter.
Yes, it does have predictive value and has already been done. See:
Easterbrook, D.J., 2011, Geologic evidence of recurring climate cycles and their
implications for the cause of global climate changes: the past is the key to the future: Elsevier.
D’Aleo, J., and Easterbrook, D.J., 2011, Relationship of multidecadal global temperatures to multidecadal oceanic oscillations: Elsevier.
The data certainly suggests some correlation but note that during the times when the PDO is declining the rate of fall in temperature is less than the rate of rise when the PDO is increasing. Suggests there is some underlying monotonic trend as well. This could be solar but it could conceivably also be CO2 – we dont know. What is obvious however is that that the monotonic component is FAR less than is claimed by warmists. Even if is were CO2 it might represent AGW certainly not CAGW.
Also, I note the continuous reference to very long time constants in the climate system. Usually from warmists claiming a lot of future heating is already locked in, we just haven’t seen it yet. Consider however the huge change in temperature both on land and sea between summer and winter. How can a system which has such a long time constant (I have seen predictions of 10’s of decades and even more) respond in a few months. Even more significantly, we can easily tell the seasonal time constant because we know the peak and trough of the insolation precisely – its the solstices. We also know the hottest and coldest times of the year. Here in Australia the hottest time is typically late January and the coldest is typically August. That translates to a lag of about 6-8 weeks. So how come the planet responds in 6-8 weeks to changes in solar energy input yet supposedly takes decades to respond to changes in energy input (or energy retention -same thing really) from other sources. If the time constant was really long and we are in fact seeing only a small portion of the potential seasonal change the lag would be 90degrees or 3 months not half that.
The only problem is that Hadcrut 4 is an obvious garbage obtained by “adjusting” drastically upwards the previous version of Hadcrut since 2001, which was already an upward adjustment of previous versions.,
http://www.woodfortrees.org/plot/hadcrut3vgl/from:2001/trend/plot/hadcrut4gl/from:2001/trend/plot/hadcrut3gl/from:2001/trend/plot/none
Actually they decreased the negative trend 2001-2013 from about 0.075 C to 0.025C.
This is the argument from false cause, or the post hoc ergo propter hoc fallacy. Whilst it may appear that the PDO has a predictive capacity to future global air temperatures, one might also remark that movements of the planet Mars appear to have a predictive capacity to the future course of planet Earth around the Sun.
Truly there are two sources of heat which ultimately affect the Earth’s “global” air temperature broadly speaking, and they are; 1. internal heat from the Earth’s molten core, manifested in volcanic activity, but also via heat transfer through the bedrock, and; 2. external heat from The Sun. The first process does affect the second process somewhat, by changing the transparency of the Earth’s atmosphere, and subsequent alterations to rainfall patterns, and so on.
Most significant however is The Sun, our very own Star, our own giant thermo-nuclear reactor which ultimately powers our planetary system. It is the magnetic field, as well as the heat, and Solar particles, which The Sun emits which affect the temperature and weather here on Earth more than anything else. The interaction of the Solar / Lunar magnetic fields, and to a lesser extent similar reactions between the Earth, Mars, Jupiter, and other cosmic events, such as Supernovae, which actually truly have the most significant effect upon the Earth’s stochastic climate system.
Man’s endeavours are a mere tickle from the leg of an insect, compared to the effects of the vast quantities of all wavelengths of energy that such solar-planetary systems are capable of transferring. The Sun puffs a solar flare of billions of tonnes of material, charged particles and radiation, and we should tremble, and the Earth’s atmosphere expands by three times the volume with the searing heat of the blast. There’s enough power absorbed to run New York for a thousand years. The Moon rises, and the land lifts by three feet, and giant ships float from the sea bed. This is the reality, see it and touch it.
Wake from your stupor, oh ye hapless victims of Hokum Pathos and False Logos. Remember lastly this, ye are all CARBON BASED Life-forms, whose very existence depends upon free Carbon Dioxide Gas in the Earth’s atmosphere. Pretend no more that ye control the Planet’s weather, for truly it controlleth ye ! Ask no more for ill gotten gains from honest labourers to fund your snake-oil scams, and bogus schemes.
Enough Already !
A signature of persistent natural thermohaline circulation cycles in
observed climate
Jeff R. Knight,1 Robert J. Allan,1 Chris K. Folland,1 Michael Vellinga,1
and Michael E. Mann2
Analyses of global climate from measurements dating
back to the nineteenth century show an ‘Atlantic
Multidecadal Oscillation’ (AMO) as a leading large-scale
pattern of multidecadal variability in surface temperature.
Yet it is not possible to determine whether these fluctuations
are genuinely oscillatory from the relatively short
observational record alone. Using a 1400 year climate
model calculation, we are able to simulate the observed
pattern and amplitude of the AMO. The results imply the
AMO is a genuine quasi-periodic cycle of internal climate
variability persisting for many centuries, and is related to
variability in the oceanic thermohaline circulation (THC).
This relationship suggests we can attempt to reconstruct
past THC changes, and we infer an increase in THC
strength over the last 25 years. Potential predictability
associated with the mode implies natural THC and
AMO decreases over the next few decades independent
of anthropogenic climate change.
http://www.meteo.psu.edu/holocene/public_html/shared/articles/KnightetalGRL05.pdf
A sensible question but not novel.
“Before it is safe to attribute a global warming or a global cooling effect to any other factor (CO2 in particular) it is necessary to disentangle the simultaneous overlapping positive and negative effects of solar variation, PDO/ENSO and the other oceanic cycles. Sometimes they work in unison, sometimes they work against each other and until a formula has been developed to work in a majority of situations all our guesses about climate change must come to nought.
So, to be able to monitor and predict changes in global temperature we need more than information about the past, current and expected future level of solar activity.
We also need to identify all the separate oceanic cycles around the globe and ascertain both the current state of their respective warming or cooling modes and, moreover, the intensity of each, both at the time of measurement and in the future.
Once we have a suitable formula I believe that changes in global temperature will no longer be a confusing phenomenon and we will be able to apportion the proper weight to other influencing factors such as the greenhouse effect of CO2.”
from here:
http://climaterealists.com/index.php?id=1302&linkbox=true&position=10
“The Real Link Between Solar Energy, Ocean Cycles and Global Temperature”
Wednesday, May 21st 2008, 8:20 AM EDT
The Pacific Multidecadal Oscillation (PMO – not PDO since the latter is only a measure of pressure variation between locations) is, I think, driven by solar variations altering global cloudiness and albedo in the manner I have suggested previously.
The change in the amount of solar energy entering the oceans is dependent on global cloudiness and albedo which is in turn dependent on the length of the lines of air mass mixing between climate zones.
Meridional jets produce longer such lines, more clouds and less energy into the oceans.
Zonal the opposite.
When the jets are meridional the reduction in energy into the oceans weakens El Nino events relative to La Nina events. That is what we have been seeing since 2000. It also matches the Maunder Minimum, the Dalton and all the other cooling spells.
When zonal, the opposite. That is what we were seeing pre 2000.and during the MWP.
The solar effect on the ENSO process takes about ten years to filter through to the Arctic Ocean so as to effect ice melt.
The 1998 El Ninos reached the Arctic around 2007 for record melt.
The 2012 high melt being due to El Ninos in 2002 /03
http://ggweather.com/enso/years.htm
So, with a suitable lag period we should be able to broadly anticipate Arctic ice melt from previous El Nino events
Theo Goodwin says:
June 5, 2013 at 7:48 pm
“Climate Science is in it’s infancy”
I’m not certain Climate Science is causing morning sickness yet.
I became interested in the PDO about 5 years ago, back then I eye-balled the correlation to just shy of 80%. Having just finished my Electrical Engineering degree and with the suggestion of Anthony W, will be beginning a degree in Applied Meteorology in August, so my skills in statistics are getting better every day. I now from the above graph see a correlation of mid 80’s, I still do not have the math skills or patience to mathematically prove my calibrated eye yet.
One of our previous commentors talked about a divergence between HADCRUT4 and JISAO, this should be expected. JISAO uses SST’s for PDO which is lower variability from land based temps. You would expect HADCRUT4 a combination of land based and SST to have a higher variation than a pure SST – JISAO data set. So I am sure the author of this article was not looking for an exact replica of rate of change.
One of the things we should look for is that the last ‘strong’ PDO (approx 1945-1977) caused a temp change of approx .25C, during a time when solar output was at its peak for the past century, cycle 19. So if there is not a double peak to cycle 24 and solar output continues to drop off for the next 7 years. What will this PDO bring us?
Second, thing I found quite interesting, is the correlation between strong ENSO’s and PDO, it is a negative correlation. The author of this paper, uses 1997 as the start point for this ‘stronger’ cycle of PDO for two reasons, the author of “A Pacific Interdecadel Climate Oscillation with Impacts on Salmon Production”, ended his study in 97, and said “1977 through (at least) the mid-1990’s” was a ‘weaker’ PDO(Mantua, 1997). Looking at the JISAO data, quite possibly, the weaker PDO of 1977 might have continued through 2005, and the reason we saw a drop in temps around ’98 was correlated to the El Nino of that time and not a phase shift in PDO. If the stronger PDO started in 05, we could be looking at 20 more years of stasis or cooling temps.
Thanks to Anthony W, for all the great work, I became a skeptic in mid Nov 06 as I could not justify Gavin Schmidt’s comments of “warmest Oct ever (06)”, that day I found a video “the great Global Warming Swindle”, and just days after that the alarmist had to retract their statement of warmest Oct ever, then again it was not their fault, it was faulty data, scientists have no responsibility to vet data, especially before running to the media. Thanks again for the great work here and suggestion of MSU’s program!
Well, if you look at the full record, the correlation is worse … but I don’t think you realize what you are doing.
The JISAO defines the PDO as the leading principle component of the sea surface temperature of the Pacific north of 20N … so it is nothing more than a measure of SST over the North Pacific.
As a result you are simply comparing the temperature of the whole world with the temperature of a bit less than a quarter of the world.
I would be extremely surprised of the temperature of maybe 20% of the planet did NOT correlate with the temperature of the whole planet … but as for the odds on the part (North Pacific) having predictive power for the whole planet, I’d be surprised.
w.
Precisely, and well analysed Willis Eschenbach !
Most of what is argued in here is simply trivial prattle.
Frankly it is irrelevant, even though it may be of some academic interest.
An interesting song from the seventies by the Brit-pop band, The Stranglers,
which was not thought to be so controversially radical at the time, now takes
on some new significance. Listen to the prophetic lyrics in this YouTube clip.
.http://www.youtube.com/watch?v=Um9kd-mhhK4
From Axel on June 5, 2013 at 10:44 pm:
Out in Wisconsin or Siberia, you get winter temps down to -30°F, but about ten to twelve feet down the temps will be in the low 50’s. Heat transfer “through the bedrock” is too slow to matter. Heat moves so slowly through the earth, there are geothermal heat pump systems that in summer will store the “waste” heat from cooling underground, where it is recovered in winter for heating.
The effects of volcanoes are variable, some yield warming, some cooling. Offhand the net effect on global temperature is nothing, the Earth tends to compensate quickly.
Your “first process” is irrelevant.
The lunar magnetic field is virtually non-existent, just crustal magnetization. There is no spinning dynamo core, no dipolar field. The “interaction of the Solar / Lunar magnetic fields” is such a tiny weak thing, so completely negligible compared to the vastly more immense magnetic field of Earth, it is hardly worth mentioning outside of a theoretical discussion.
The ocean water might move, the land much less so. As to the tides:
With everything lined up, on an idealized Earth, with Sun and Moon working together, you might get that three feet of lift from water. The land is much less flexible than water, nor does it flow as well. You will not get the Moon lifting the land by three feet.
Plus given the amount of sediment found deposited on old wrecked ships, leading to their being buried, it is quite obvious that giant ships DO NOT float from the sea bed.
Who are you to come here and leave such anti-scientific easily-disproved dreck littering the comments, your little pet theories, while rattling on like you’re one of us?
Hopefully Leif or Willis or one of the other main luminaries of the site will come by and put you out of your misery, by directing you towards real knowledge rather than what you have spewed.
Willis,
Here is a definition of PDO:
“A PDO definition
The Pacific Decadal Oscillation, or PDO, is often described as a long-lived El Niño-like pattern of Pacific climate variability (Zhang et al. 1997). As seen with the better-known El Niño/Southern Oscillation (ENSO), extremes in the PDO pattern are marked by widespread variations in Pacific Basin and North American climate. In parallel with the ENSO phenomenon, the extreme phases of the PDO have been classified as being either warm or cool, as defined by ocean temperature anomalies in the northeast and tropical Pacific Ocean.
Two main characteristics distinguish the PDO from ENSO. First, typical PDO “events” have shown remarkable persistence relative to that attributed to ENSO events – in this century, major PDO eras have persisted for 20 to 30 years (Mantua et al. 1997, Minobe 1997). Second, the climatic fingerprints of the PDO are most visible in the North Pacific/North American sector, while secondary signatures exist in the tropics – the opposite is true for ENSO. Several independent studies find evidence for just two full PDO cycles in the past century (e.g. Mantua et al. 1997, Minobe 1997): cool PDO regimes prevailed from 1890-1924 and again from 1947-1976, while warm PDO regimes dominated from 1925-1946 and from 1977 through (at least) the mid-1990’s. Recent changes in Pacific climate suggest a possible reversal to cool PDO conditions in 1998, an issue that is discussed in more detail at the end of this article.”
We all should have reason to question any SST older than 40 years ago as sampling is sparse, one of the big contributors to WUWT Bob Tisdale agrees with this as do many other scientist. For you to say “Well, if you look at the full record, the correlation is worse … but I don’t think you realize what you are doing.” We are for intents and purposes looking at most of the record. In a paper I wrote for school in ’10, I used the correlation of PDO back to 1880 to demonstrate trends in GMT, in my paper I used a lag of average 8 years, the author of this paper is using 5 years, albeit opposite direction as me. From 1880, till present, there is a high correlation of PDO to temp, like I said eye-balling it high 70’s.
Prior to reading this article, I had always felt that the phase change was within (after) 8 years of temps change, so that is where I ran with the correlation. Looking at this data and seeing the graphs now shows that it looks like, PDO changes and approx a year later temps follow, the change you do not see is the phase change around 1942 graphically PDO (1947 Mantua) and temps HADCRUT4 1945 was about 3 years.
So the question becomes could PDO be a climate driver? Some scientist believe that PDO is driven by ocean current oscillations and that a ‘strong’/ cool PDO is caused by upwelling of deeper colder pacific water. While, ‘weaker’/ warm PDO is less up-welling, similar to El Nino/ La Nina just on a longer time scale (months as compared to decades). We all know that a large body water can cause local weather to be enhanced by the temp of that water, growing up in Eastern Mass, we would head out to the beach when it was scorching in my backyard, about 10 miles from the water. The last turn to “Nantaskett’ beach you sometimes got a rude awakening, fog and temps dropping 20 degrees within a mile. So the question remains, can a body of water as large as the Pacific, change the weather/ climate of a large portion of the globe. Certainly Mantua believes so, http://www.atmos.washington.edu/~mantua/REPORTS/PDO/PDO_cs.htm.
I do agree with you, I have alot to learn, that might be why I am enrolled in a Applied Meteorology program (could not find an adequate Atmospheric sciences course 100% online). I do need more understanding of statistics, I wish I could work with data at the level you do, give me a couple years you will see a marked improvement on my posts. Thanks to you also for your informative posts, I enjoy them greatly.
Excellent. The pioneering work of Namais suggest this also. The PNA/PNO pattern seems to modulate ENSO, thereby driving natural climate cycles globally. The cooling of the 1960’s and 70’s…the warming of the 1980’s and 90’s. And lastly…the present climatic cooling.