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.
Is this surprising? Didn’t everyone have a globe when they were a kid? I used to position it and marvel that the pacific was damned near half the world.
Turn it into an empirical model (just a best fit with a variable for time lag) and make a prediction for now-to-2018.
A full treatment would sequester 1/3 of the data, and both make a fit and determine error bars based solely off of the first 2/3 of the data. Then make a new fit with all of the data, yielding another set of errors as well. Then predict.
Interesting Graph. two things really stand out though.
1 The almost perfect agreement between the interrelationship
2 The apparent divergence at 1940 that indicates the Temperature portion is increasing
My theory is that the sun warms the ocean and – after an interval (a transient response) – the ocean warms the atmosphere after a 99-year time lag (152 years in the case of the Pacific, but the smaller oceans bring down the mean). If I’m right, the low level of solar activity in WWI and the 1920s are due to deliver us some bitter winters.
http://endisnighnot.blogspot.co.uk/2012/03/lets-get-sorted.html
I see several apparently instantaneous responses. For example, roughly, 1970, ’72 and ’74 have peaks in PDO that coincide with peaks in temperature. I don’t see a consistent lag overall, and only a year difference at the minimum of each trace at 1976-7ish. I do find the correlation disturbing though, it is cool enough thank you.
Nice, a model with a lag term would do it but I think some testing with AMO and some attempt at solar cycle variable construction from cycle normalization would also be interesting. The South Atlantic sea surface temperature data would also be worth looking at.
I believe I remember seeing on Dr. Spencer’s site that he’s a fan of the PDO for explaining climate change, although I can’t say whether or not a direct correlation like this is what he had in mind. Could be, for all I recall right now. I ought to go look.
So if this pattern holds, does that mean we can expect temperatures to drop for at least the next 5 years?
Yeah, Dr. Spencer lays out a case for PDO here:
http://www.drroyspencer.com/research-articles/global-warming-as-a-natural-response/
It makes perfect sense to me that measured air temps in the Northern Hemisphere would lag ocean temps.
Resourceguy suggest the South Atlantic sea surface temperature data would also be worth looking at. I’ll suggest separating the Southern Hemisphere from the Northern Hemisphere temperature data for the following reason. Remove the approximate 0.5 deg Anthony identified as due to positive biasing of the Northern Hemisphere temperature data from “corrections” and from the heat island effect. This may remove the “apparent divergence beginning at 1940 that indicates the Temperature portion is increasing” identified by Bryan A. Note that the Southern Hemisphere temperature data should not show the degree of temperature biasing shown by Anthony in the Northern Hemisphere.
PDO is temperature. predicting temperature from temperature tells you nothing. same with all the ocean patterns. OF COURSE they allow you to predict the global temps. they are part of the thing being predicted. Its like using CET to predict global temps. Of course you can.
Put another way. PDO is the effect
Science is the process of observation and analysis.
Nice observation, let’s analyse it
You are extrapolating existing graphs into the future. That is not prediction, except in the trivial sense that you are asking “Will the future resemble the past?”
For prediction, we need to know more about the PDO. It might consist of many other natural regularities. We need to know enough about some of them that we can create reasonably accurate descriptions of their characteristics. Those descriptions will take the form of general statements, something along the lines of Kepler’s First Law: All planetary orbits are ellipses with the Sun at one of the foci. From general statements and descriptions of existing conditions we can deduce and, thereby, predict descriptions of future conditions.
The PDO is worthy of serious study and experiment to learn its characteristics. Unfortunately, given the regime running climate science at this time, nothing is being done.
For the person who complains that we have only one PDO and, for that reason, cannot experiment on it, please note that experimentation should begin with passive observation of the natural regularities that make up the PDO.
I got interested in the PDO very early on – my recollection is somewhere around 2000 or 2001, but that may be in error. I think it is significant that the PDO was not even discovered by a climatologist or meteorologist, but by biologist Steven Hare in 1997, who was researching salmon catches in the Pacific Northwest. He found that when the northern Pacific water temps were in a warm regime, the salmon catches were low, and vice versa – and that the warm and cold regimes appeared to last decades before flipping to the other regime.
In 2005 I was emailing with then Oregon State Climatologist George Taylor. He wrote me the following (which I reproduce here without his permission and hope it is no big deal to him):
“I’ve known about PDO since the early 1990s, since before it was named (I missed my chance to name it!). Here’s a link to an article I wrote in 1996 and posted in 97…”
George was one of those in the climatology world who didn’t agree with the CAGW meme, and it was amazing in a state like Oregon that he was in his position. He got a LOT of flack within his state. Then he was let go as the state climatologist a few years ago. and the article was at some point removed from the state website.
I agree with Ashby Manson above, about how BIG the Pacific is. I looked it up once – 32% of the entire world’s surface area. To the eye it LOOKS like half the globe. To me whatever the Pacific does it does without the effect of humankind’s pitiful contribution. The Pacific is the elephant in the room. It seems pretty obvious to me that from the latency of its warm and cold regimes that nothing is going to push the Pacific around. It was said back then to be ten times as big as the El Niño.
No one is going around and arguing that humans are affecting ENSO. If we aren’t affecting ENSO, then what are the chances of us affecting the PDO? Pretty low.
And if we aren’t affecting the two big mamas of Pacific climate, then any pretense that we are affecting the rest of the globe’s climate are, without seriously incontrovertible evidence – what courtrooms call “stipulated evidence” – on very thin ice.
One more thing: The climate models throughout the 1990s were calaimed to be accurate and complete. Those were the ones upon which the IPCC based the projections of 5.8°C rise by 2100. Yet every one of those could not have included the PDO, which hadn’t been discovered yet. We were all to believe that without the single biggest factor in those model’s code, they were right. And then after the PDO entered the room, it took a good while before it was included. And then they acted as if they were right all along. How could they have been right both before and after, if the before was missing the elephant in the room?
Steve Garcia
Steven,
Oh, I dunno. Spencer says it might drive cloud changes. Of course, Trenberth will tell you clouds don’t control climate, so there you have it I guess. Except Spencer doesn’t agree.
Steven,
Wait a second here. How can it be the effect, if the effect is driven by CO2, yet the PDO changes phase every 20-30 years?
It does not make sense. Ladies and gentlemen of the alleged jury, If Chewbacca does not live on Endor you must acquit. Thank you.
The PDO could be regarded as the change in the ratio of ‘nino’ to ‘nina’ events , which seems to have a sixty year cyclicity comparable to the sixty year temperature cyclicity.
We’ve only been measuring the ENSO index for one cycle, but since temperature seems to be a good proxy for that cycle, it could be extended back into the 1800s.
But what, then, drives the PDO?
@Theo Goodwin June 5, 2013 at 3:44 pm:
“You are extrapolating existing graphs into the future. That is not prediction, except in the trivial sense that you are asking “Will the future resemble the past?”
???? You can’t be serious. ALL scientific predictions are extrapolations from existing evidence. All non-scientific predictions use crystal balls in place of existing evidence.
The ludicrous part of this assertion is that the climate models can NOT even make the future resemble the past. Yet all CAGW is based on assertions that they can extrapolate the future from and make it look like the past (even if they do it so utterly unsuccessfully).
EVERY linear trend line in all of climate science is an extrapolation. You argue against your own case.
(Personally I see the use of straight line linear regression as completely flawed, because nature doesn’t do straight lines in complex systems, not over any great length of time. However some linear regression is necessary – I just think the output should be some higher order curve..)
Steve Garcia
Some other correlations.
inverted SOI is leading indicator for global temps -7 Months
(“The Southern Oscillation Index, or SOI, gives an indication of the development and intensity of El Niño or La Niña events in the Pacific Ocean. The SOI is calculated using the pressure differences between Tahiti and Darwin. ” from BOM)
http://jonova.s3.amazonaws.com/guest/leyland/soi-global-temps-jan2011.pdf
Funnily SOI in turn seems to be correlated with the Ap index
see at 5:20
Perhaps someone has altered the data to match the PDO, but offset it by 5 years so is wasn’t so obvious? Just a thought, it seems to match too much, within nature that is so chaotic.
presented: temperature lags PDO by around 5 years.
Steven Mosher says: Put another way. PDO is the effect
reversing cause and effect, again?
Steven Mosher says:
June 5, 2013 at 3:40 pm
PDO is temperature. predicting temperature from temperature tells you nothing.
——————————————————
PDO going negative predicts about 30 years of lower trends or even falling temperatures.
I haven’t seen any such prediction from temperature.
Same for AMO.
#Steve Mosher
PDO = low frequency tail of ENSO accounting for an appreciable fraction of the warming trend
http://www.esrl.noaa.gov/psd/people/gilbert.p.compo/CompoSardeshmukh2008b.pdf
“Because its [ENSO’s] spectrum has a long low frequency tail, fluctuations in the timing, number and amplitude of individual El Nino and La Nina events, within, say, 50-yr intervals can give rise to substantial 50-yr trends…”
“…It [The Pacific decadal oscillation or the interdecadal Pacific oscillation] is strongly reminiscent of the low-frequency tail of ENSO and has, indeed been argued to be such in previous studies (e.g. Alexander et al 2002, Newman et al 2003, Schneider and Cornuelle 2005, Alexander et al 2008)…”
“…it also accountd for an appreciable fraction of the total warming trend…” (see figure 9b )
“…In this paper, we have argued that identifying and removing ENSO-related variations by performing regressions on any single ENSO index can be problematic. We stressed that ENSO is best viewed not as a number but as an evolving dynamical process for this purpose…”
I’m seeing the “PDO Effect” right out my window. Winter lasted until May, and spring is a weak, wet, cold and clammy thing, with little relief in sight. I have to do this for how many decades?