Misunderstandings about the Pacific Decadal Oscillation
Guest post by Bob Tisdale
INITIAL NOTE
The first version of this post (The Common Misunderstanding About The PDO dated June 26, 2008) incorrectly described the method for calculating the Atlantic Multidecadal Oscillation. I originally intended to do a quick correction in agreement with my post The Atlantic Multidecadal Oscillation – Correcting My Mistake, but then I decided to expand this post.
INTRODUCTION
Many climate change bloggers often note that global temperatures rise when the Pacific Decadal Oscillation (PDO) is positive and drop when the PDO is negative. They then make the assumption that it’s the PDO that causes global temperature to vary. To dispel this, let’s first examine what the PDO is.
THE PACIFIC DECADAL OSCILLATION
The Pacific Decadal Oscillation (PDO), Figure 1, is “derived as the leading PC of monthly SST anomalies in the North Pacific Ocean, poleward of 20N. The monthly mean global average SST anomalies are removed to separate this pattern of variability from any ‘global warming’ signal that may be present in the data.” The quote is from the JISAO website: http://jisao.washington.edu/pdo/PDO.latest
The main JISAO PDO webpage is here:
http://jisao.washington.edu/pdo/
http://i41.tinypic.com/vrq7uq.jpg
Figure 1
The semi-periodic variation in the PDO can be better seen when the data is smoothed with a 121-month running-average filter, Figure 2.
http://i39.tinypic.com/20svqsx.jpg
Figure 2
THE METHOD USED TO CALCULATE THE PDO
Nathan Mantua of the University of Washington and JISAO, in an email, described the process used to calculate the PDO. And it is a process:
“The full method for computing the PDO index came from Zhang, Y., J.M. Wallace, D.S. Battisti, 1997: ENSO-like interdecadal variability: 1900-93. J. Climate, 10, 1004-1020.
“They labeled this same time series “the NP index” (see their figs 5 and 6). The steps are listed below, and files described below can be found at: ftp://ftp.atmos.washington.edu/mantua/pdofiles/
“Data used:
* monthly 5×5 Hadley Center SST 1900-93
“Method:
1. create monthly anomaly fields for all grid points
2. create a monthly mean global SST anomaly time series for all months, 1900-93, using gridpoints specified in file grid.temp.glob_ocean.977
3. create a “residual SST anomaly” field for the North Pacific by subtracting out the global mean anomaly from each North Pacific grid point in file grid.N_Pac_SST.resi.172 (20N-65N, only in Pacific Basin) for all months and locations
np_resi(mo,loc)= np_ssta(mo,loc) – global_mean(mo)
4. compute the EOFs of the North Pacific residual SST anomaly fields, and ignore all missing data point (set them to zeros)
5. the PDO index is the leading PC from the above analysis
6. for PDO index values post 1993, project observed ‘North Pacific residual SST anomalies’ onto the leading eigenvector (what we call the ‘PDO pattern’ of ssts) from the EOF analysis done in step 4. We now do this with the Reynold’s and Smith Optimally Interpolated SST (version 2) data.”
###
A link to the referenced Zhang et al (1997) paper is here:
http://www.atmos.washington.edu/~david/zwb1997.pdf
The point of listing that multistep process was to show that the PDO is a statistically created dataset. Let’s look at what the PDO does not represent.
THE PDO DOES NOT REPRESENT NORTH PACIFIC SST ANOMALIES
SST anomalies for the North Pacific Ocean (20N-65N) and scaled PDO data are illustrated in Figure 3. The PDO does not represent SST anomalies for the North Pacific.
http://i43.tinypic.com/29fp8ad.jpg
Figure 3
THE PDO DOES NOT REPRESENT DETRENDED NORTH PACIFIC SST ANOMALIES
The PDO is not calculated in the same fashion as the Atlantic Multidecadal Oscillation (AMO). NOAA ESRL calculates the AMO by detrending SST anomalies for the North Atlantic. Refer to The ESRL AMO webpage:
http://www.cdc.noaa.gov/data/timeseries/AMO/
In Figure 4, the PDO (scaled) is compared to detrended North Pacific (North of 20N) SST anomalies (calculated the same as the AMO). While there are semi-periodic variations in detrended North Pacific SST anomalies, the PDO does not represent them.
http://i42.tinypic.com/17pev8.jpg
Figure 4
THE PDO DOES NOT REPRESENT VARIATIONS IN THE DELTA T BETWEEN NORTH PACIFIC SST AND GLOBAL TEMPERATURES
Let’s subtract Global temperature anomalies (LST & SST) from North Pacific SST anomalies to see what that curve looks like. Refer to Figure 5. The PDO does not represent the difference between global temperature anomalies and North Pacific SST anomalies.
http://i42.tinypic.com/345kgsk.jpg
Figure 5
SO WHAT DOES THE PDO DESCRIBE?
The PDO represents a pattern of SST anomalies in the North Pacific. The operative word in that sentence is PATTERN. Figure 6 (from the JISAO PDO webpage) illustrates the warm and cool phases of the PDO. When the PDO is positive, SSTs in the eastern North Pacific are warmer than in the central and western North Pacific, and when the PDO is negative, the reverse is true.
http://i39.tinypic.com/20v1934.jpg
Figure 6
Keep in mind, though, that the PDO data itself represents only the North Pacific, north of 20N, which I’ve blocked off in Figure 7. Figure 7 is a map of SST anomalies from April 14-21, 2008 that shows a negative PDO pattern. It’s from the NASA Earth Observatory webpage here:
http://earthobservatory.nasa.gov/IOTD/view.php?id=8703
Specifically, this linked page:
http://earthobservatory.nasa.gov/images/imagerecords/8000/8703/sst_anomaly_AMSRE_2008105_lrg.jpg
http://i39.tinypic.com/262prfa.jpg
Figure 7
PDO VERSUS ENSO
There is also a popular belief that the sign of the PDO dictates whether El Nino or La Nina events dominate. There is, however, an analysis that contradicts that belief. Refer to:
http://www.cdc.noaa.gov/people/gilbert.p.compo/Newmanetal2003.pdf
And for those who enjoy PowerPoint presentations for the visuals:
http://www.cpc.noaa.gov/products/outreach/proceedings/cdw28_proceedings/mnewman_2003.ppt
In “ENSO-Forced Variability of the Pacific Decadal Oscillation”, Newman et al state in the conclusions, “The PDO is dependent upon ENSO on all timescales. To first order, the PDO can be considered the reddened response to both atmospheric noise and ENSO, resulting in more decadal variability than either. This null hypothesis needs to be considered when diagnosing and modeling ‘internal’ decadal variability in the North Pacific. For example, the observed spatial pattern of Pacific SST decadal variability, with relatively higher amplitude in the extratropics than in the Tropics, should be at least partly a consequence of a reddened ENSO response.”
In the introduction, Newman et al explain, “Anomalous tropical convection induced by ENSO influences global atmospheric circulation and hence alters surface fluxes over the North Pacific, forcing SST anomalies that peak a few months after the ENSO maximum in tropical east Pacific SSTs (Trenberth and Hurrell 1994; Alexander et al. 2002). This ‘atmospheric bridge’ explains as much as half of the variance of January-March seasonal mean anomalies of SST in the central North Pacific (Alexander et al. 2002). Furthermore, North Pacific SSTs have a multiyear memory during the cold season. Deep oceanic mixed layer temperature anomalies from one winter become decoupled from the surface during summer and then ‘reemerge’ through entrainment into the mixed layer as it deepens the following winter (Alexander et al. 1999). Thus, over the course of years, at least during winter and spring, the North Pacific integrates the effects of ENSO.” [Emphasis added]
They continue, “The prevailing null hypothesis of mid latitude SST variability posits that the ocean integrates forcing by unpredictable and unrelated weather, approximated as white noise, resulting in ‘reddened’ noise with increased power at low frequencies and decreased power at high frequencies (e.g., Frankignoul and Hasselmann 1977). In this paper, we propose an expanded null hypothesis for the PDO: variability in North Pacific SST on seasonal to decadal timescales results not only from red noise but also from reddening of the ENSO signal.”
Figures 8 and 9 are comparative graphs of the PDO and NINO3.4 SST anomalies, smoothed with 12-month and 121-month filters.
http://i41.tinypic.com/fd4vgz.jpg
Figure 8
##########
http://i41.tinypic.com/n14010.jpg
Figure 9
CLOSING
As discussed and illustrated, the PDO cannot directly explain global temperature variations because it represents a pattern of SST variability, not SST. And the Newman et al paper explains why the low frequency variations of the PDO are greater than ENSO. They write in their abstract, “Variability of the Pacific decadal oscillation (PDO), on both interannual and decadal timescales, is well modeled as the sum of direct forcing by El Nino-Southern Oscillation (ENSO), the ‘reemergence’ of North Pacific sea surface temperature anomalies in subsequent winters, and white noise atmospheric forcing.” [Emphasis added]
Do other areas of the Global oceans integrate the effects of ENSO like the North Pacific?
SOURCES
The links for the PDO data are included in the text of the post. HADISST NINO 3.4 SST anomaly data, HADISST North Pacific SST anomaly data, and the combined CRUTEM3+HadSST2 global temperature anomaly data are available through the KNMI Climate Explorer website:
http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere
Good post Bob,
Having worked with these ocean indices, I can say the PDO does not provide very good correlations and the ENSO, by itself, provides a much better reflection of the impacts this region of the planet has on global temperatures.
It would be relatively straight-forward to create a PDO index similar to the method used for the AMO. But one problem is, it would be covering a very large part of the globe and, hence, would already be a large part of the global temperature anomaly itself – ie. not a sufficiently independent variable.
Bob,
Can you address whether the PDO is a short term, or long term phenomenon? You seem to be saying that it is simply a particular manifestation of ENSO. But ENSO varies on much shorter time scales than has traditionally been attributed to the PDO. Can you clarify, or elaborate, on that?
Basil
Don Easterbrook is an expert on the PDO.
He seems to have a completely different understanding of the phenomenon, as he:
1. Sees the PDO signal in the climate record (glacial advance and retreat, specifically) going back for (at least) centuries
2. Does regard the PDO as a driver
and
3. Predicts 20-30 years of mild cooling as we move forward, specifically due to the changeover of the PDO to its negative phase
One thing that puzzles me in all of this. So-called “global temperature” as represented by GISStemp and the like; being just some sort of average of something (not knowing the details of Hansen’s AlGorythm) cannot explain any sort of global patterns or circulations, since it requires temperature differences to cause energy flows in any direction (other than radiation of course; which however is not in any particulaqr direction).
So I don’t see how “global warming” effects as predicted by green house gas theories of the MMGWCC kind, can cause any sort of oscillation such as the PDO or ENSO or anything else; after all they are global effects not local and it requires local temperature variations to create a drive for energy flows around the world; simply jackling all temperatures up by 0.2 deg C is not going to cause thermal energy to flow anywhere it wasn’t already going.
So for me, I can’t see any human influence appearing in PDOs or ENSOs, which presumably are perfectly normal occasional natural climate changes.
So how much is known about just what physical processes cause PDO/ENSO/La Nina events ?
I am clearly one of the bloggers who are being contradicted by the author above. Sorry, after having read the text, I am entirely unconvinced by any of the “negative” statements in this article.
As far as I can see, all widely read climate bloggers realize that PDO is about a pattern – let’s say that the PDO index is the difference of temperature anomalies between the U.S. West Coast and (minus) the central Northern Pacific. Also, most of us are convinced that the pattern helps to drive global temperatures, probably via the influence on ENSO.
I don’t see any rational counter-evidence to this statement written in the article. The observed correlation between the time-derivative of the global mean temperature and the PDO index seems to be powerful.
Also, the correlation between the relative frequency of El Nino and La Nina episodes on one side, and the PDO index on the other side, seems to be strong, and seems to be eplainable by a natural mechanism. During negative PDO phases, the warmer central Northern Pacific Ocean is able to “suck” potential El Nino conditions, bringing La Nina to the equatorial Pacific Ocean instead.
The statement that “PDO is influenced by ENSO” contradicts very basic physical principles. PDO is the slower process among the two, so it can’t possibly be measurably influenced by faster processes such as ENSO.
ENSO is faster which means that the PDO index acts as one of the external parameters determinining ENSO dynamics – at least this influence is much more important and logical to organize the events than the opposite influence. Everyone who knows what Renormalization Group or Born-Oppenheimer approximation mean will surely know where I am coming from.
Basil: You asked, “Can you address whether the PDO is a short term, or long term phenomenon?”
If we look at the last two graphs, the PDO appears to have a long term “cycle” that’s better defined than ENSO. If Newman is correct (and if I’m interpreting that paper correctly), then it’s the reemergence of anomalies (created by ENSO) that causes the PDO to persist longer and at greater amplitudes than ENSO.
Lubos Motl : ‘The statement that “PDO is influenced by ENSO” contradicts very basic physical principles. PDO is the slower process among the two, so it can’t possibly be measurably influenced by faster processes such as ENSO. ‘
That is a very good point there! unless there is any evidence of long term cycles of ENSO which could act as drivers?
PDO driving ENSO makes sense because as can be seen in the diagrams, anomalies are overall negative at the equator during cold phase, which would surely indicate a tendency towards La Nina? just an observation, I am no expert but generally all blogs/articles I have read through do put forward the idea of PDO driving ENSO, this article was quite a surprise but interesting nevertheless. However PDO graphs that are commonly used never go further back than 1900 and this is suspicious, especially since there is less oscillation in the visible data at +-1900
Per Harold Ambler’s ref to Easterbrook, see:
Don Easterbrook
Global climate change, global warming: Recent global cooling — news articles and interviews
‘During negative PDO phases, the warmer central Northern Pacific Ocean is able to “suck” potential El Nino conditions, bringing La Nina to the equatorial Pacific Ocean instead.’
It appears in the diagrams according to the arrows that during cool phase PDO, warmer waters shift away from the equator towards the central Northern and Southern Pacific Oceans and cold waters upwell at the equator from beneath hence La Nina domination.
And during warm phase, warm water ‘accumulates’ at the equator hence El Nino, whilst cold upwelling occurs at north and south.
Seems obvious, but it appears that PDO could logically be the driving force behind ENSO events, or at least it decides what type of events will occur.
Thanks, Bob.
By the way, here is another timely PDO article, although some of the statements in it may not agree with what you have posted.
The Pacific Ocean’s Influence on Climate Change: How Low Will the PDO Go?
http://www.stormx.com/agriculture/severe-weather/2009/04/the-pacific-oceans-influence-on-climate-change-how-low-will-the-pdo-go/
The PDO is also found to be a global driver of temperature, partly because of its intimate link to the El Niño Southern Oscillation cycle. La Niña is more likely during periods of negative PDO, while periods of positive PDO yield more frequent and intense El Niño’s. Thus the tropical temperature anomalies are highly correlated with the PDO phase and in turn help produce quantifiable fluctuations in global average temperatures. From the warm to the cool PDO phase, global temperatures drop by around 0.3-0.35°F. Some may contend that these observed global temperature differences are explained by the global warming trend, as the PDO cool phase was centered in the 1950s and the PDO warm phase in the 1990s. However, it remains plausible and perhaps likely that the PDO may be an important cause of the global change and observed temperature trends during the latter half of the 20th century. Scientifically, an improved understanding of global climate change may be on the horizon, as another cool PDO phase has begun and its impacts on climate will be studied closely. If global warming ceases or cooling occurs during the next 20 years, the scientific community may come to realize that decadal fluctuations within the oceans may play a larger role in climate than is presently acknowledged. The first signs of the PDO’s impacts may already be occurring as satellite measurements of lower tropospheric temperatures derived by the University of Alabama in Huntsville (UAH) and Remote Sensing System (RSS) show no temperature increases since 2002, and in fact show a slight cooling trend. With solar activity at a 100 year minimum and the PDO expected to remain negative for several decades, the cooling trend may persist. If it does, an inconvenient truth about climate change may emerge.
Bob,
Very interesting information. Your comment “And it is a process” summed up my reaction when I first looked at the JISAO material. As all this is supposed to produce a “pattern”, and it seems to, they must have stayed up nights to think this through. It violates the “keep it simple stupid” (KISS) rule.
One of the problems of visualizing these things is the issue of map projections. We get a false impression of the areas on many of these maps. Those used here, Figure 7, for example, exaggerate the area in the far north and south. Most apparent here if one looks at Antarctic – on a globe Australia looks to be much the same size, not so here. Likewise, Greenland and Mexico are similar in size. Again, not so here.
I make an issue of this projection problem because people forget that there is substantially more of Earth’s surface area and more ocean between the Tropics and more insolation. Thus, it should not be a surprise that ENSO exerts the sorts of influences it does.
We also tend to forget that Earth is rotating and most have not looked closely at the many ocean currents. There is a country song all should listen to a few times and get this idea in their heads:
chorus
cause and effect, chain of events all of the chaos makes perfect sense
when your spinning round, things come undone
welcome to earth third rock from the sun
Joe Diffe, “Third Rock From the Sun”.
Lubos Motl: You wrote, “The statement that ‘PDO is influenced by ENSO’ contradicts very basic physical principles. PDO is the slower process among the two, so it can’t possibly be measurably influenced by faster processes such as ENSO. ”
Lubos, have a look at the monthly PDO versus NINO3.4 SST anomalies. There’s nothing slow about the PDO. Both signals are so noisy, I had to divide the data into four graphs.
NINO3.4 vs PDO from 1900 to 1925:
http://i30.tinypic.com/30adrph.jpg
NINO3.4 vs PDO from 1925 to 1950:
http://i32.tinypic.com/2dt2hbq.jpg
NINO3.4 vs PDO from 1950 to 1975:
http://i25.tinypic.com/dyx7bl.jpg
NINO3.4 vs PDO from 1975 to 2008:
http://i27.tinypic.com/2zf4taw.jpg
They’re from my post “The Chicken or Egg: PDO or ENSO?”
http://bobtisdale.blogspot.com/2008/06/chicken-or-egg-pdo-or-enso.html
Regards
There isn’t really any generation or removal of heat from “the system” with any of these processes, is there? If not, then there’s no “global mean temperature” involved, it’s just a moving around of heat to different areas. Unless new heat is being added to the system, the totally meaningless “global mean temperature” should not change. Of course that assumes complete global coverage and no garbage temperature monitoring.
Mr. Tisdale,
From reading what you have written, it is unclear: are you saying you misunderstand the PDO, or that those who see a massive long term shift in the Pacific as a strong influence on climate fail to understand the PDO?
Your clarification is appreciated.
George E. Smith (09:52:27)
“…simply jackling all temperatures up by 0.2 deg C is not going to cause thermal energy to flow anywhere it wasn’t already going.”
The idea is that a generally warmer temperature will push the big weather makers poleward a little so that, say the boundary between summer high pressure along North America’s west coast and the Westerlies, slides northward X miles. Precipitation would decline in W. Washington and S. British Columbia. Energy would also be redistributed in these situations. There are probably better examples.
jackling? He typed that, not me. John
“As discussed and illustrated, the PDO cannot directly explain global temperature variations because it represents a pattern of SST variability, not SST.”
That looks like a non sequitur. Can the author explain further?
A pattern in SST can explain changes in total SST. It could do this by representing changed transport of heat towards the poles, or transport of heat to areas where loss by radiation was more or less efficient, or due to heat having different effects on wind, cloudiness, or precipitation depending on where it is. It’s like the way blood circulation changes make your skin go white when it’s cold and red when it’s hot. It’s a change in the pattern of where the hot blood is, but it affects temperature and heat flow too.
The correlation is saying that a large part of the change in temperature may be related to non-linear effects of the distribution of heat around the globe. For example, it could be that without any change in input, something has pushed the heat into places from where it doesn’t convect and radiate to space so easily, and so warmed the Earth. The atmosphere is not one-dimensional, as a lot of the simplistic explanations would have it, and a change in output does not necessarily imply a change in input.
Is there anything in the above analysis to show that nothing like this is happening?
Lubos: Your quote, “PDO is influenced by ENSO”, which everyone is repeating, is curious. It appears nowhere in my post. The only place the word influence was used was in the quote from Newman et al. They wrote, “Anomalous tropical convection induced by ENSO influences global atmospheric circulation and hence alters surface fluxes over the North Pacific, forcing SST anomalies that peak a few months after the ENSO maximum in tropical east Pacific SSTs (Trenberth and Hurrell 1994; Alexander et al. 2002).”
Newman et al go to great detail to explain their conclusion. If you disagree with the specifics of the paper, please detail them.
Thanks
hunter: You wrote. “From reading what you have written, it is unclear: are you saying you misunderstand the PDO, or that those who see a massive long term shift in the Pacific as a strong influence on climate fail to understand the PDO?
Your clarification is appreciated.”
I don’t misunderstand the PDO. I’ve illustrated how it’s calculated, what it is, what it is not, and I’ve quoted from a paper that appears to be controversial.
Regarding long-term shifts, note that the major shifts in NINO3.4 and PDO occur at the same time in Figure 9.
Please read the Newman at al paper. Thanks.
Dear Alex, exactly! The general principle is that slower processes effectively behave as “constants” that determine the external conditions for faster processes.
Whenever we focus on a typical time scale (over which we average things etc.), the processes with the same typical duration or periodicity (in the case of periodic processes) are the most important ones. The slower processes act as external parameters (which can skew the balance of positive/negative events, like La Nina vs El Nino) while the faster processes tend to be averaged out.
Dear Bob, we must be looking at the same graphs and numbers in a very different way.
PDO is significantly slower than ENSO. This statement is obvious even from the very graphs that you have included. With the 121-month filter, it is very clear that the warm and/or cool phases of PDO last 30+ years or so, agreeing with the general “binary” classification of the warm and cool stages, see e.g.
http://en.wikipedia.org/wiki/Pacific_decadal_oscillation#Regime_shifts
You wouldn’t get any similar 30-year long eras out of ENSO. Of course that there is noise over there. Noise is everywhere. The difference is that also 30-year or 50-year signals can be isolated from PDO indices but they can’t be isolated from ENSO/ONI indices.
The most extreme PDO variations are very slow – multidecadal, therefore the name (!). On the other hand, La Ninas and El Ninos demonstrably take 1-3 years in average, so they’re about 10 times faster than PDO phases. This statement can also be derived theoretically because ENSO primarily depends on some atmospheric/ocean coupled behavior in a smaller region – the near-equatorial Pacific – than PDO (PDO influences 1/3 of the world’s greatest ocean) which also means that it takes a shorter time for it to switch from the negative to the positive phase.
I don’t believe that you actually disagree with any of these things because they’re completely manifest.
The statement by Newman et al. you ended up with is just a postmodern cliche of the type “everything influences everything else”. Well, yes, no, what of it? Of course that at some vague level, it’s right that anything related to ENSO influences anything else, and so on. But these things surely can’t be used to “debunk” the influence of PDO on ENSO.
What is more important than “everything impacts everything” philosophy is that one must choose a specific time scale or otherwise defined realm of phenomena and the slower processes can be approximated by constants while the faster processes may be assumed to average out.
PDO is slower than ENSO. If Mr or Ms Newman disagrees with that, and I don’t think that you have shown any evidence even for this modest statement, it simply means that Mr or Ms Newman has no idea about the ocean cycles.
I’m curious what people think about this presentation I was recently required to attend for a class: http://oregonstate.edu/groups/hydro/Seminars/spring09/docs/042209_Mote.pdf
There is video of the presentation on the same website.
I just wanted to note that the actual PDO index has much more variability than the smoothed graphics you might have seen on the Internet.
Here is what the PDO Index looks like with thinner lines so you can see the additional variability. In my mind, this chart looks very, very different than the charts usually shown on the Internet.
http://img168.imageshack.us/img168/8917/pdoindex.png
[This is why I also do not favour using smoothing unless absolutely necessary and why I keep my line thickness down to the bare minimum so that this kind of additional detail shows up.]
Curious: You wrote, “That looks like a non sequitur. Can the author explain further?”
The PDO is independent of global temperature. It is independent of the temperature difference between the North Pacific SST anomalies and global temperature. It explains whether the anomalies in the eastern part of the North Pacific are warmer or cooler than the western and central parts. It has been correlated to other effects, such as western North American temperature, precipitation, and snowpack.
http://www.atmos.washington.edu/~mantua/REPORTS/PDO/PDO_egec.htm
Can someone explain how the following is true:
The statement that “PDO is influenced by ENSO” contradicts very basic physical principles. PDO is the slower process among the two, so it can’t possibly be measurably influenced by faster processes such as ENSO.
To me this seems counterintuitive. What does the speed of a process have to do with what sort of force or affect it can exert on another process?
Jeff,
The question is, how much heat do the oceans exhaust or absorb over time? ENSO is important because a great deal of heat energy is exhausted into the atmosphere. The PDO is important because there appears to be a correlation between PDO phases and ENSO.
One could use the classic biological phases of North Pacific fish migration to chart the PDO. For it was fishermen who first noted that every 2-3 decades there was an obvious migration.
Lubos: “You wouldn’t get any similar 30-year long eras out of ENSO.”
Figure 9 illustrates that there are 30-year-long eras of ENSO. It’s the frequency and magnitude of El Nino events versus La Nina events that dictate whether the 121-month smoothed curve of ENSO is positive or negative. During the periods when El Ninos dominate, global temperatures rise, and when La Ninas are dominant, temperatures drop.
So, accordimg to picture 6, typhoons are to be expected in Japan, while just a few showers in the tropics..(if any). That will be a big deception for Him ( the bearer of all future distress), provided that red spot in the gulf does not mean hot waters..:), because la Nina affects that region also:
http://www.osdpd.noaa.gov/PSB/EPS/SST/data/anomw.4.27.2009.gif
bob
You and I have blogged on this topic before. If
“The PDO represents a pattern of SST anomalies in the North Pacific” and if this pattern of anomalies are indicators of certain weather and climate conditions or patterns especially in the Northern hemisphere including land temperatures in North America which repeat over a period of time , then various levels of PDO can be used as predicters or indicators of past and future of climate patterns .It may not be a direct indicator of Pacific Ocean SST but it does reflect a multiple of things which originate from SST, does it not.?
You explain well, Bob, what PDO is not, but do not explain well what it is, what it can be used for and why it was developed in the first place. The definition in the CLOSING is far too confusing for any blogger
I think it would help me to see, patternwise, further discussion of any ocean circulation and jet stream changes rolled into the presentation.
I admit to a provincial interest in NA weather in particular as determined by PDO and don’t care so much about the chimera, average global temp.
At any rate, ENSO as climate master doesn’t particularly alarm.
I am conscious that ENSO in the form of El Nino and La Nina events seems to be wind driven to some extent but I am nevertheless unclear as to whether the associated changes in SSTs cause further changes in the air as a positive feedback or whether changes in the air are actually the drivers of the ENSO phenomenon.
I am also conscious that the PDO phase shifts are much larger events over longer multidecadal time scales.
Importantly there is no need for the two processes to be linked. It is quite possible that the underlying ocean cycles such as PDO, AMO et. al. could have a different causation from more temporary shifts in surface SSTs.
What matters for global air temperatures is the netted out energy characteristics of ALL the ocean cycles and various SSTs at any one time and as I have said elsewhere there are times when they work together and times when they offset one another and then one also has to add in solar variability over as many as 6 11year solar cycles or 3 22 year solar cycles.
Given that a positive PDO enhances El Nino and suppresses La Nina (or appears to) and that a negative PDO suppresses El Nino and enhances La Nina it is quite likely both that ocean cycles are seperately caused by influences other than those that cause shorter term SST variations and that those deeper longer cycles are driven more by solar variations affecting the thermohaline crculation than by changes in the air.
I don’t see the point of asserting a null effect from PDO variations by simply asserting that they represent merely a changed distribution of SSTs.
The fact is that global air changes are seen to occur after SST changes and given the basic physical differences between water than air that must be expected.
As another poster said the change in position of warm and cool areas of SST is itself quite sufficient to change the air circulation patterns and therefore the rate of energy transmission from surface to space.
Furthermore it must be perverse to suggest that changes in the SSTs such as those observed can have a zero effect on the temperature of the air globally.
This seems to be an attempt to deny the possible influence of ocean multidecadal cycles just as they are beginning to be seen to be highly significant out in the real world.
I’m not sure whether Bob supports the Newman analysis or is just putting it up as a discussion point.
Bob,
Thanks. But that doesn’t clear things up much.
You say “The PDO is independent of global temperature.” Are you saying that the PDO is not statistically correlated with the rate of change of global temperature anomaly? Or are you saying that the mechanisms causing them both are such that the PDO cannot cause or influence the changes in global temperature’s rise and fall? i.e. a physical independence? If the latter, what is the argument for this?
People have looked at graphs and thought there was a correlation – that global mean temperature anomaly rose in periods of positive PDO. How has this happened, if they are independent? Eyeball failure? Or smoothing/processing errors?
If you’re saying that because PDO is not temperature, so it isn’t directly contributing to temperature in the sort of arithmetical way that both sea temperature and land temperature contribute to global temperature, then I’d agree. But I wasn’t aware that anyone had suggested such a thing. I thought the popular theory was that unknown circulation and heatflow relationships caused the low frequency component of PDO to oscillate, and that the changed patterns in heat and heat flow distribution changed the efficiency with which the Earth absorbed or radiated heat by unknown mechanisms, and that when integrated over time this caused long-term rises and falls in temperature. You wouldn’t expect PDO to equal temperature on this basis.
Are you simply correcting misunderstandings in somebody else’s theory?
Thanks.
Its true that Don Easterbrook foresees just a mild cooling due to the PDO.
But..
As far as i know, he is using a GISS trend to foresee the future. (!!)
So the mild GISS cooling from 1940-77 is what he expects.
Any GISS based anything, i will not have too much faith in:
http://www.klimadebat.dk/forum/vedhaeftninger/gissdivergence.gif
Dear Bob,
Figure 9, the very Figure 9 that you included, shows very clearly that the typical timescale over which PDO changes the sign is comparable to 30 years while the typical timescale at which ENSO changes the sign is 1-4 years.
The former is shown on the figure because the blue line has huge variations that change per 30 years. The latter is shown by seeing that there are almost no variations left after the averaging over 100+ months. If you averaged over 1-2 years, the ENSO index would give you an equally pronounced sin-like signal like PDO gives at the multidecadal scale.
The ENSO index almost certainly influences the temperature but the ENSO index itself – the relative frequency of El Ninos vs La Ninas during a decade or two – is influenced by PDO. The data clearly show this much and there are good theoretical reasons to be sure that this influence exists, too.
Best wishes
Lubos
Ah, think I see the problem.
Newman’s analysis and by implication that of Bob is that the PDO has no independent existence because it is merely comprised of data derived from the varying SST conditions involved in the ENSO phenomenon.
My comment on that, if true, is that some unrecognised underlying ocean cycle is causing that variable data to change at approximately 30 year intervals and we may as well call it the PDO.
Bob says this above:
“Figure 9 illustrates that there are 30-year-long eras of ENSO. It’s the frequency and magnitude of El Nino events versus La Nina events that dictate whether the 121-month smoothed curve of ENSO is positive or negative. During the periods when El Ninos dominate, global temperatures rise, and when La Ninas are dominant, temperatures drop.”
i.e. The 30 year variability is a function of ENSO not any underlying cycle so PDO does not exist independently.
Well in my opinion something external to ENSO is causing those phase shifts and imposing them onto the ENSO cycle. That is what I have always meant by the PDO. Perhaps we should rename it ?
Why, otherwise would the relative dominance of El Nino and La Nina change at all ?
And the same thing is going on in every ocean it seems.
I recently wrote a research article on the PDO that quantifies the impacts on climate. While we can debate as to whether the PDO and global temperatures connections are cause or effect, the raw numbers from each phase indicate that during cool phases global temperatures decrease slightly. IMO this is mostly caused by cooling in tropical Pacific, presumably from more frequent La Nina events during negative PDO.
I strongly believe that our understanding of the negative PDO’s impacts on climate will be better understood as we seek to quantify them over the next decade or so. Please check out the PDO research note here:
http://www.stormx.com/agriculture/severe-weather/2009/04/the-pacific-oceans-influence-on-climate-change-how-low-will-the-pdo-go/
BOB
I also found this helpful.
http://www.atmos.washington.edu/~mantua/REPORTS/PDO/PDO_cs.htm
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.
PDO cycle and US temperature record are in excellent agreement. Global temperature datasets play with PDO quite well as well – warming 1907-1940, cooling 1940-1977, warming 1977-2005. Europe is also affected by AMO, which has switched to cold recently.
Variating Solar baseline with PDO/AMO variations on the top can describe whole 20th century record very well.
@Lubos, pozdrav z Bratislavy.
All I get from this is that the PDO is not the be-all and the end-all of temperature changes globally, which makes sense to me. It does influence, however, North American temperatures – and ENSO does so especially, which with our 24 hour media coverage gets overemphasized. Weather is not climate, but it seems neither side can resist stories that illustrate their particular point of view. This is, I suppose, understandable. There are a lot more factors to global temperature than just the PDO and/or ENSO, certainly, but the world’s largest major body of water has to have at least some influence on global temperatures. If it influences cloud cover or airflow in its various temperature variations (and in NA at least this is shown to be true) then it has global reach by effect though that effect may be weak at best and overridden by stronger regional or local influences. The Earth is a closed system, therefore finite, and therefore (like your bank account) when something is spent in one place it is lacking in another. How large the effect is and how tight the correlation from place to place seems to be the key question regarding PDO and/or ENSO. That is at least my amateurish take.
First I would like to thank Bob Tisdale for his article and input. It is good in general to challenge conventional thoughts and views. Science cannot advance unless it is challenged and debated.
For now, I must remain critical of the concept however. Long term events like the PDO would be expected to have embedded short term oscillations. Likewise, the NINO cycle would as illustrated have some long term trend constructions possible. That is true of any variable data.
1998 was a very warm El Nino event. I don’t think anyone would argue that. However, I submit that the event could not have occurred without the PDO also being very warm at the time. What I make note of is figure 9. The amplitude of change in the PDO is far greater than that of the NINO.
To me it seems that the state of the ENSO can augment or mitigate the effects from the PDO but the PDO has to be the greater driver of overall temps. Along that line; referring to figure 8 and specifically the period from 1960 – 1980 it appears that the NINO was, for the most part, divergent from observed global temperature conditions while the PDO was consistent with those same conditions. Thus it would appear again that the PDO was the greater factor.
Re: Stephen Wilde (13:10:46)
After I read the Newman et al. (2003) paper awhile back I had some similar thoughts – chicken-egg thing – is it a deck? or is it 52 cards? (Some say ENSO drives LOD; some say vice versa…) Research funding, semantics, & politics….
The Newman et al. (2003) paper was a valuable contribution.
I read a paper that the PDO had significant impact on US 20th century temperatures.
Given that the anomaly is close to the US west coast, this would make some sort of sense.
Is that correlation also disputed?
Thanks for the article Bob. A couple of questions if you have the time.
From fig. 7 they only use north of 20N. What percentage of the whole Pacific would this be?
From fig. 6 the max is +0.8 and the min is-.06. Why would the scale not be symmetrical? Thanks.
(Should it really be called the Far North PDO?)
BOBBY LANE
Don’t underestimate the impact of the pattern of SST anomalies that the PDO reflects as shown below. There is a correlation.
LEAST SQUARE TREND LINES PER DECADE FOR VARIOUS PAST WARM AND COOL PERIODS
[Per HADCRUT3vgl]
1900-1926 0.048 COOL [AMO –VE, PDO –VE &+VE]
1926-1944 0.187 WARM [AMO & PDO POSITIVE]
1964-1976 0.108 COOL [AMO& PDO NEGATIVE]
1994 -2008 0.187 WARM [AMO & PDO POSITIVE]
1900 -2009 0.073 PAST CENTURY [equivalent of 0.73/century]
2002 -2009 -0.195 LATEST COOL [PDO –VE SEPT/07, AMO –VE JAN/09]
OTHER DECADAL TRENDS
1976 -1994 0.138 C [ PRIOR TO 1994]
1994 – 2008 0.187 C [ KEY GLOBAL WARMING]
1976 – 2008 0.178 C
Notice that the period 1926-1944 had the same rate of warming as 1994-2008. Periods of global warming existed well before 1976-2008.
What made the most recent warming period [1994-2008] more significant was that some of the AMO and PDO levels were higher than usual [ AMO levels were third[1998],fourth[2003] and fifth[2005] highest ever after 1878 and 1937]
Another observation is that most of the recent warming period was really in the period 1994- 2008 and not 1976-2008. So the real warming was a decade plus three to four years only, a very short period indeed and not a climate trend or long term trend at all. It is amazing how 13-14 years got blown out of all proportions by the AGW science and misrepresented as an alarming and an unprecedented climate trend when it was really another warm hiccup of this planet where regular alternating cool and warm hiccups are par for the planet.
The AGW supporters have commented that the recent seven years of global cooling[since 2002] is too short a period to be considered a climate trend, yet the last global warming period lasted only 13-14 years, equally not a climate trend.
Rabidkangaroo (11:46:12) About the presentation
This is standard material based on a false premise. Thus, like all the others of its ilk, each of the points can be refuted. I suggest you look at a few of the State reports and see what was done. Which was: a consulting group sold this batch of stuff to many states, controlled the process, and thus the outcomes. They are all pretty much alike.
Go to this site: http://scienceandpublicpolicy.org/scarewatch/
On the right side, under Reports, click on SPPI State Climate Profiles.
Pick a couple of the states and see what they say. When you are only allowed to consider a warming based on GHGs – and nothing else – the reports and their critiques are repetitive.
People seem to like the concept of the PDO so I hesitate to write this.
But it seems to me that the PDO is just a reflection of the accumulated impact of the last few El Ninos or La Ninas over the last year or two.
It is the ENSO which drives the PDO, not the other way around, or it might be better to think of them as part of the same system.
http://upload.wikimedia.org/wikipedia/commons/c/cc/Ocean_surface_currents.jpg
The ENSO migrates across the Pacific at the equator from the Nina 1,2 region to Nina 3 to the Nina 4 region, from the South American coast to the Gilbert Islands/New Guinea driven by the Trade Winds.
At this point, the surface ocean currents turn to the North to form part of the Kurushio Current (the Gulf Stream of the Pacific) and the North Equatorial Counter-Current. The currents are mostly blocked to the South by shallow ocean and islands and the southery flow is weak.
The Kurushio Current then flows across the Pacific in a gyre and the ocean temperatures in the North Equatorial Counter-Current filter off further northward and, eventually, these two currents influence ocean temperatures across the entire North Pacific.
As well, at the end of the Nino 4 region at the Gilbert Islands/New Guinea, the prevailing winds shift to the NorthEast toward Alaska and California.
If the ocean surface temperatures conditions are cool at this location, the prevailing winds migrate those cooler conditions NorthEast across the Pacific (fairly rapidly) in conjunction with the ocean currents and this where we see the common wedge pattern of the PDO anomaly.
It is evident in this animation. Let it load and then speed it up as fast as your computer will allow and notice how the cool surface spreads across the Pacific from the end of the Nino 4 region.
http://www.osdpd.noaa.gov/PSB/EPS/SST/anom_anim.html
Through these two impacts, ocean currents and prevailing winds/poleward transport of temperatures, the ENSO drives the North Pacific PDO. It also explains why the ENSO has very limited effect on southern hemisphere temperatures beyond those directly in the Tropics. Most of its energy is filtered northward.
The PDO can flow back into the ENSO region through the California Current although this current is not strong. The majority of the ocean current flowing into the ENSO region is through the Southern Pacific Peruvian/Humbolt Current and through upwelling from the deep ocean immediately below the Nina 1 and 3 regions.
Right now, despite the very negative PDO conditions, it appears that an El Nino is building. The Southern Oscillation Index has suddenly shifted to signal El Nino, the Trade Winds have slowed, Atmospheric Angular Momentum is now signaling El Nino and the Upper Ocean Heat Content is showing much warmer ocean temperature water ready to surface.
http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ocean/anim/wkxzteq_anm.gif
One only needs to llok at a current SST anomoly map
http://weather.unisys.com/surface/sst_anom.html
and see why the PDO has an effect on Global Temps as the strong
cold current that flows equatorward off the western USA coast ends up in the Tropical Pacific Ocean making the current attempt by the western South American current to form an El Nino with warm water broken down by the PDO cold current mixing into and overriding it…That is why we have less El-Ninos and weaker ones generally during a cold PDO phase and of course cooler Tropical Pacific Ocean area is strongly linked to Global Temperatures…My research suggests strong links between AMO and PDO phases and Global temperatures, and in the PDO case it is not so much the measuring area of the PDO, but the effect it has on the large Tropical Pacific Ocean Area, and the El-Nino-La Nina phenomenon, which then in turn effects global temperature strongly. I cannot see how this does not make good common science sense.
My understanding, after reading this article, is that the PDO represents the SST anomaly for the North Pacific after subtraction of the global SST anomalies (please correct me if I am wrong). A positive anomaly therefore indicates the North Pacific is warmer than expected given the ‘current’ global SSTs. Does this then imply that a ‘local’ (i.e. Pacific ocean) input of energy has produced the relative increase in SSTs (e.g. volcanic, see surge tectonic theory), or can it be explained by a re-distribution of energy?
It also appears to me (due to the way the PDO is calculated) that the relationship between the phase of the PDO index and specific patterns in SST anomalies (relative difference in anomalies between east and west) are only associations (not necessarily stable).
That leads me to another aspect of the SST patterns associated with the PDO phase. I would like to know how closely the PDO anomalies correspond to SST anomalies for the Eastern Pacific. Are the PDO anomalies in the East simply an artifact of the analysis, due to the dominance of the Western and Central pacific on the calculation of the PDO index, and the influence of the Western Pacific on world climate (which influences Global SST anomalies and the PDO index) ? Or in another way, do the actual SST anomalies vary as much in the East as they do in the West?
Bob,
In the past, I have removed the *immediate* ENSO signal from the PDO. Is this curve of any use in showing the underlying variation of the PDO and a possible link to ENSO?
A commentary on PDO from Jeremy Ross:
http://www.stormx.com/agriculture/severe-weather/2009/04/the-pacific-oceans-influence-on-climate-change-how-low-will-the-pdo-go/
OT… couldn’t find a place to put this comment about pan evaporation rates.
Changes in Australian pan evaporation from 1970 to 2002
Michael L. Roderick, Graham D. Farquhar *
Cooperative Research Centre for Greenhouse Accounting, Research School of Biological Sciences, Institute of Advanced Studies, The Australian National University, Canberra, ACT 0200, Australia
email: Graham D. Farquhar (farquhar@rsbs.anu.edu.au)
*Correspondence to Graham D. Farquhar, Cooperative Research Centre for Greenhouse Accounting, Research School of Biological Sciences, Institute of Advanced Studies, The Australian National University, Canberra, ACT 0200, Australia
Abstract
Contrary to expectations, measurements of pan evaporation show decreases in many parts of the Northern Hemisphere over the last 50 years. When combined with rainfall measurements, these data show that much of the Northern Hemisphere’s terrestrial surface has become less arid over the last 50 years. However, whether the decrease in pan evaporation is a phenomenon limited to the Northern Hemisphere has until now been unknown because there have been no reports from the Southern Hemisphere. Here, we report a decrease in pan evaporation rate over the last 30 years across Australia of the same magnitude as the Northern Hemisphere trends (approximately -4 mm a-2). The results show that the terrestrial surface in Australia has, on average, become less arid over the recent past, just like much of the Northern Hemisphere. Copyright © 2004 Royal Meteorological Society.
It seems as though Australia has actually been getting wetter along with the Northern Hemisphere… Another inconvenient truth. Some are blaming it on Solar Dimming…
The PDO governs the temperatures in western North America – especially in the western U.S. — A graph in this document: http://www.appinsys.com/GlobalWarming/PDO_AMO.htm shows a very strong correlation between PDO and Washington/Oregon temperatures.
The shift between warm/cold phases may be related to the 22-year solar Hale cycle – see figure at the end of the PDO section in the above link.
Maybe I’m reading this wrong but my best assessment is that the author is arguing that the PDO cannot be responsible for warming or cooling as it is not a direct indicator of SST anomalies in the north pacific or difference between SST and LST anomalies. If that is correct then I can’t really say I’m impressed. I would have thought it was obvious that any link between the PDO and global temperatures would most likely lie in the changes to ocean and air circulation patterns that the PDO might result in, not based on the method we use to measure it. I really don’t get the argument.
Well, here’s my take….
(1) The PDO is a descriptor of an underlying phenomenon that affects global temperature but is also an influence (below). But for the purposes of the climate models, it really doesn’t matter because the models don’t consider anything with a period of less than a century because the assumption is that these shorter periods will merely cancel out over the long term so can be ignored. Other temperature-affecting phenomena such as aerosols and volcanoes are not periodic and therefore are thrown into the mix by the models at random (volcanoes) or ad hoc (aerosols).
The problem here is that though the PDO has a ‘period’ of about thirty years and thus supposedly cancels out after about sixty years, that leaves room for an odd number of oscillations in a century. So the 20th century had two warmish phases but only one coolish phase thus the PDO didn’t have a chance to cancel out as ‘weather noise’.
The ‘weather noise’ will not cancel out for another twenty years or so, therefore the global temp rise so far must have part of its attribution handed over to the ocean which the models have not done.
(2) When I view the ‘overhead’ shots of negative PDO and positive PDO and where the warm SST’s vs the cooler SST’s in the northern pacific are, it reminds one more of a huge low vs a huge high with their counterclockwise vs clockwise energy flows.
ie, when the PDO is negative, the warm waters are in the middle (as would be the case under a gigantic high pressure system) and the cooler waters surround. the ‘flow’ would be clockwise and the ‘bottom’ of the ‘high’ would push equatorial waters westward and result in la nina conditions (cooler waters upwell to replace the displaced warmer water). When the PDO is positive that westward flow reverses or halts, allowing warmer water temperatures to build in the enso3.4 region giving el nino conditions.
But what would cause the PDO to ‘flip’ from positive to negative or vice versa? Surface winds and currents must play a roll here, no? Both from the ENSO region and from the periphery, especially the west coast of No Amer. These ‘highs’ and ‘lows’ drift and wobble and effect the flow of heat in the waters below (hence the noisy anomalies). ‘Drifts’ over areas of strong wind and current that are in place for other reasons may break up the flow around the ‘high’ or ‘low’ and wear it down and eventually there’s more warm/cold around the periphery than before and the pattern of wind between hot and cold, low and high pressure, eventually ‘congeals’ into the opposite pattern: the shift in the PDO.
So, basically, though ENSO does not directly dictate the PDO, it certainly influences it back. I’d say the PDO is the driver of ENSO here, but the feedback from prolonged ENSO trends, chaotic nudges, and the winds going along for the ride, gives us the flips and the flops in an amazingly consistent timeframe.
That’s my story and I’m sticking to it.
Stephen Wilde (12:31:32) : “I am conscious that ENSO in the form of El Nino and La Nina events seems to be wind driven to some extent but I am nevertheless unclear as to {1} whether the associated changes in SSTs cause further changes in the air as a positive feedback or {2} whether changes in the air are actually the drivers of the ENSO phenomenon.”
Good comment. I believe the former {1} is the case. My take on it is this: “normal or La Nina conditions” result in Earth-rotation-winds displacing a 1/2 meter high mound of water from the East and mid-Pacific (EP and MP) into the West Pacific (WP). This is obviously not a stable situation. Eventually (especially after a La Nina), the system pressure causes increasing amounts of cold water to well up in the EP. This lowers SST’s in the EP, which in turn cause two phenomena: higher seawater viscosity (see Note 2) and lower air T and humidity in the EP.
The air is thus denser and more difficult for the trade winds to move; the water is “stiffer” and also harder for the trade winds to move. This is effectively the same as an apparent decrease in wind velocities. The system can then reverse, spreading the heated water from the WP back across the MP and EP, into full El Nino conditions.
Note 1: radiation losses for highly differentiated EP/WP temperatures are theoretically higher than for more isothermal (average) conditions. However, I’m unsure at the moment whether the T^4 weighted average SST for a La Nina is greater or less than that for El Nino conditions.
Note 2: seawater viscosity is 100 times more sensitive than seawater density to temperature changes.
Good post.
But what would cause the PDO to ‘flip’ from positive to negative or vice versa? Surface winds and currents must play a roll here, no?
It follows the Southern Oscillation, an atmospheric oscillation (not dissimilar to the NAO) which is measured by a Tahiti/Darwin pressure differential.
The PDO may be said to be a result of the SO rather than a cause, in and of itself. One may even say it is a result of magnified Nino or Nina phases of the SO.
SO went positive in 1977 along with PDO. Don’t know where it is now, other than I guess it has probably recent gone cold.
P.S., I have just read two mentions that the AMO has gone into cold phase. And even some evil rumors going ’round about the the AO and NOA. Watts Up With That?
“Right now, despite the very negative PDO conditions, it appears that an El Nino is building. The Southern Oscillation Index has suddenly shifted to signal El Nino, the Trade Winds have slowed, Atmospheric Angular Momentum is now signaling El Nino and the Upper Ocean Heat Content is showing much warmer ocean temperature water ready to surface.”
Breathing in, breathing out. The oceans are ready to exhale.
Thar she blows! Or will it be merely a sigh.
So, what causes those 30 year phase changes ?
There is no mechanism whereby the much more rapid El Nino/ La Nina changes can cause such large background changes on that time scale.
We are now told that we cannot allocate the term ‘PDO’ to those phase changes because the PDO is simply an artifact arising from the SO and ENSO signals with no independent existence of it’s own.
I for one as well as many others have been using the term ‘PDO’ to refer to those 30 year phase changes in the Pacific so I think the best thing to do is either rename the phenomenon or change the definition of PDO.
There seems to be similar academic and media confusion between the more rapid SST changes and underlying multidecadal cycles in every ocean.
There are multidecadal phase changes in every ocean.
They have profound effects on global air temperatures.
They interact with each other and solar variations.
They seem to occur independently of changes in the air alone.
They render the effect of CO2, and other changes in the air alone, insignificant.
I nominate the term ‘Wildean Ocean Cycles’.
All: Excuse me for disappearing yesterday, but something unexpected came up and took me away from home until late last night.
TO THOSE WHO CONTINUE TO INSIST THAT THE PDO DRIVES ENSO: Let me refer you to the linked Zhang et al “ENSO-like Interdecadal Variability: 1900–93”. As noted above, the calculation of the PDO is based on the methods used in that paper. They found that the PDO lags ENSO by about a season and concluded that spatial pattern of the PDO is a response to ENSO, not vice versa. And they also found that the interdecadal variability of the PDO was a lagged response to ENSO. Refer to:
http://www.atmos.washington.edu/~david/zwb1997.pdf
Zhang et al refer to the PDO as “NP”, and they use the Cold Tongue Index (CT), which are SST Anomalies of 6S-6N, 180-90W, in place of NINO3.4 SST Anomalies of 5S-5N, 170W-120W. In Figure 7, they illustrate the cross-correlation functions between the Cold Tongue and the other time series they examined. Note how in the bottom cell NP (PDO) lags (CT) ENSO. Their Figure 7 is copied here:
http://i39.tinypic.com/14o3beb.jpg
They wrote on page 1011 (pdf page 8), “Figure 7 shows the cross-correlation function between CT and each of the other time series in Fig. 5. The lag is barely perceptible for TP and G and it increases to about a season for G – TP and NP, confirming that on the interannual timescale the remote features in THE PATTERNS SHOWN IN Fig. 6 ARE OCCURRING IN RESPONSE TO THE ENSO CYCLE RATHER THAN AS AN INTEGRAL PART OF IT, consistent with the conclusions of Alexander (1992a,b) and Yulaeva and Wallace (1994).” [Emphasis added]
Their Figure 6 shows the spatial pattern for the North Pacific associated with the PDO:
http://i44.tinypic.com/112h3k8.jpg
They also observed the interdecadal variability of the PDO (NP), but did not appear to feel it conflicted with the above findings that the PDO occurs in response to ENSO. On page 1012 (pdf page 9) they wrote, “In summary, of the time series in Fig. 8, CT is most strongly dominated by the interannual variability associated with the ENSO cycle, while G – TP and NP exhibit the clearest evidence of interdecadal variability. This distinction is also evident in the autocorrelation functions shown in Fig. 9: CT’s negative sidelobe reflects the ENSO cycle, WHILE NP’S POSITIVE VALUES OUT TO LAGS OF 5 yr AND BEYOND REFLECT THE GREATER PROMINENCE OF INTERDECADAL VARIABILITY.” [Emphasis added]
Their Figure 9 is here:
http://i41.tinypic.com/200zfk0.jpg
Regards
matt v: You wrote, “You and I have blogged on this topic before. If “The PDO represents a pattern of SST anomalies in the North Pacific” and if this pattern of anomalies are indicators of certain weather and climate conditions or patterns especially in the Northern hemisphere including land temperatures in North America which repeat over a period of time , then various levels of PDO can be used as predicters or indicators of past and future of climate patterns .It may not be a direct indicator of Pacific Ocean SST but it does reflect a multiple of things which originate from SST, does it not.?”
It does. In his paper “The Pacific Decadal Oscillation and Climate Forecasting for North America”, Mantua did list the temperature, precipitation, etc. correlations with the PDO, which would then aid in prediction. Refer to:
http://www.atmos.washington.edu/~mantua/REPORTS/PDO/PDO_cs.htm
You wrote, “You explain well, Bob, what PDO is not, but do not explain well what it is, what it can be used for and why it was developed in the first place. The definition in the CLOSING is far too confusing for any blogger.”
Sorry, but there are just so many topics one can cover in a post. Hopefully the above link will help explain what it can be used for. The method of calculating the PDO was developed to explain the “ENSO-like Interdecadal Variability: 1900-93” in the Global Oceans, the North Pacific and others areas. Refer to the Zhang et al paper:
http://www.atmos.washington.edu/~david/zwb1997.pdf
With respect to the closing discussion, does my above comment (April 29 @ 03:42:15) help explain it better? That the spatial pattern and interdecadal variations of the PDO are lagged responses to ENSO?
Regards
Stephen Wilde: You wrote, “I am conscious that ENSO in the form of El Nino and La Nina events seems to be wind driven to some extent but I am nevertheless unclear as to whether the associated changes in SSTs cause further changes in the air as a positive feedback or whether changes in the air are actually the drivers of the ENSO phenomenon.”
Refer to Bill Kessler’s discussion on ENSO. Trade winds, SST gradients across the tropical Pacific, and ENSO depend on one another.
http://faculty.washington.edu/kessler/occasionally-asked-questions.html#q1
You also wrote, “I am also conscious that the PDO phase shifts are much larger events over longer multidecadal time scales,” and “Importantly there is no need for the two processes to be linked.”
As noted in my April 29 @ 03:42:15 comment, above, Zhang et al found the PDO pattern and the decadal variability to be lagged responses to ENSO.
“Recent studies suggest that ENSO teleconnections with North American climate are strongly dependent on the phase of the PDO, such that the “canonical” El Niño and La Niña patterns are only valid during years in which ENSO and PDO extremes are “in phase” (i.e. with warm PDO+El Niño, and cool PDO+La Niña, but not with other combinations) (Gershunov and Barnett 1999, Gershunov et al. 1999, McCabe and Dettinger 1999).”
IMHO Money quote from Bob’s Mantua link.
Curious: You wrote, “You say ‘The PDO is independent of global temperature.’”
If you were to replace the North Pacific SST anomalies in Figure 3 with Global Temperature, and have global temperatures compared to the PDO, the same PDO value and pattern have occurred at various global temperatures. And I also showed in Figure 5 that the Delta T between global temperature and the North Pacific SST do not correlate. (Before you reply, continue to read this comment.)
You asked, “Are you saying that the PDO is not statistically correlated with the rate of change of global temperature anomaly…”
As noted in a number of posts at my blog, a scaled running total of Hadley Centre’s NINO3.4 SST anomalies will reproduce the global temperature anomaly curve.
http://i39.tinypic.com/2w2213k.jpg
The graph is from my post “Reproducing Global Temperature Anomalies With Natural Forcings”
http://bobtisdale.blogspot.com/2009/01/reproducing-global-temperature.html
The curve of a running total (not scaled) of PDO reflects the variations, but not the Global Temperature curve.
http://i39.tinypic.com/6iy3ro.jpg
Even if I could coax the global temperature anomaly curve from the PDO, it’s immaterial, because as I noted in my April 29 @ 03:42:15 comment above, Zhang et al found the PDO pattern and the decadal variability to be lagged responses to ENSO, So any global temperature response to changes in the PDO are ultimately responses to ENSO.
Bob,
I’ve looked at the Zhang study and found the following towards the end:
“It is difficult to compare the present study with those
that have attempted to recover separate ‘‘modes’’ of
variability using techniques such as conventional EOF
analysis (Parker and Folland 1991; Miller et al. 1994;
Deser and Blackmon 1995), rotated EOF analysis (Kawamura
1994), or Principal Oscillation Pattern analysis
(Latif et al. 1997) because the interdecadal variability
that we have identified exhibits a spatial signature that
is similar, in so many respects, to that associated with
the ENSO cycle. The confusion is exacerbated by the
general upward trend in global-mean SST over the century,
which renders the results of conventional EOF
analysis difficult to interpret. In view of these problems
and the limited length of the historical record, we are
not convinced that a formal modal separation involving
interdecadal variability is meaningful.”
They are clearly hedging their bets as to whether there is a seperate mode of variability which operates independently of the ENSO influence. They recognise that others do think that other modes exist, as do I.
I find it difficult to exclude the possibility that varying solar input to the oceans over several solar cycles has an effect on the energy characteristics within the oceans which in turn affects movement of large bodies of water within the oceans.
If we describe ALL the observed oceanic effects on the air as consequences of processes in the air driving the oceans (as you seem to suggest) then that presupposes a static oceanic background despite an ever changing solar effect on ocean dynamics over multiple decades or even centuries.
I think that treatment of the main body (as opposed to the surface) of the oceans as an unchanging monolith for energy absorption and release purposes is a major defect in AGW theory and the main defect in the point of view put forward at the start of this thread.
PDO, AMO, NAO et.al. might all be statistical artifacts derived from observational SST data but to treat them as such does in my view deter one from considering modes of variation within the main body of the oceans themselves.
We should seperate PDO, AMO, NAO et.al. from their SST counterparts and look into the possibility that they are indeed independently driven and so affect the net balance between the different SST modes and thereby drive global air temperatures as much as is necessary to explain observed temperature changes without needing to attribute a large forcing to CO2.
We could also group them together under the generic name that I have suggested.
Bob,
Serious scientists working in the field know that the PDO is just an index
that represents a PATTERN in the sea surface temperature anomalies in the Nth central Pacific over time. Thanks however, for taking the time to point this out to the wider public.
You seem to be totally convinced by Newman et al. arguements even though it is obvious from the data and timing arguments that Newman et al. cannot be true.
Newman et al. basically argue that the PDO is a sea surface annomaly
pattern that is produced by the long term integration by the Nth. Pacific Ocean of the ENSO signal + white noise.
However, Newman et al. cannot explain the following sequence of events
that can be ascertained from Gergis & Fowler (2006)
See:
http://www.springerlink.com/content/2242tp4610562j55/
http://www.soest.hawaii.edu/~timm/files/workshop2007/gerigs.ppt
1) The PDO flips from negative to postive
Bob, you know that these switches (as indicated by your own graphs) are abrupt climate events like the great PDO shift in the mid 70’s. Normally
the flip in the PDO index from one extreme to the other only takes a couple of years.
2) Historical records (both instrumental and PROXY) show that the
PDO stays in a positive phase for any where from 10 – 40 years.
3) At the start of the postive PDO period, the strength of El Nino events
are typically weak.
4) As the postive PDO persists, the average strength of the EL Nino events
get’s strong and stronger.
5) The PDO eventually flips from positive to negative
– again abruptly over a period of a few years.
6) The average strength of the El Nino events start to decrease and
continue to decrease while the PDO remains negative.
This is NOT, repeat NOT, the time sequence that you would expect to observe if the PDO was just a long term intergration of the ENSO signal.
In fact if Newman et . al.’s model was correct you would expect the
gradual 10 – 40 year intensifications of the El Ninos to lead to slow
buildup in the PDO strength. This does NOT take place.
Hence, I believe that Newman et. al. ‘s model is directly contradicetd by the observations and so is false.
Sorry, but I cannot agree with Newman et al.
Bob,
I’m still trying to understand your main point. Here’s an abstract of another paper. Is this consistent with what you are trying to say?
The Pacific decadal oscillation (PDO), defined as the leading empirical orthogonal function of North Pacific sea surface temperature anomalies, is a widely used index for decadal variability. It is shown that the PDO can be recovered from a reconstruction of North Pacific sea surface temperature anomalies based on a first-order autoregressive model and forcing by variability of the Aleutian low, El Niño–Southern Oscillation (ENSO), and oceanic zonal advection anomalies in the Kuroshio–Oyashio Extension. The latter results from oceanic Rossby waves that are forced by North Pacific Ekman pumping. The SST response patterns to these processes are not orthogonal, and they determine the spatial characteristics of the PDO. The importance of the different forcing processes is frequency dependent. At interannual time scales, forcing from ENSO and the Aleutian low determines the response in equal parts. At decadal time scales, zonal advection in the Kuroshio–Oyashio Extension, ENSO, and anomalies of the Aleutian low each account for similar amounts of the PDO variance. These results support the hypothesis that the PDO is not a dynamical mode, but arises from the superposition of sea surface temperature fluctuations with different dynamical origins.
http://ams.allenpress.com/perlserv/?request=get-abstract&doi=10.1175%2FJCLI3527.1
Let’s suppose that this is correct, that the PDO is not a “dynamical mode.” To borrow terms from my discipline, I think that is simply saying that PDO is endogenous, not exogenous, i.e. it is the result of other forces, not a external force itself. It doesn’t change the fact that the PDO has characteristic modes, or phases, i.e. cool phases and warm phases, of decadal or even multi-decadal time scale.
Moreover, if, as the abstract I cite, and the Newman, et al paper you cite (“The PDO is dependent upon ENSO on all timescales.”) are correct that ENSO plays a part in the decadal pattern of variation in the PDO, how does that constitute a disproof of the proposition that during a PDO cool phase La Nina’s dominate, while during a PDO warm phase El Nino’s dominate?
And what is the bottom line here? Are you denying that the apparent recent shift of the PDO into a cool phase may be indicative of a climate shift that is likely to persist for some period of time?
Regardless of whether it is endogenous, or exogenous, there is plenty of evidence for a multi-decadal variation in climate here, isn’t there?
Basil
Whoops. Remove NAO from previous post. That is primarily atmospheric rather than oceanic.
Stephen,
I have recently attempted to assess your comment on solar impacts:
“I find it difficult to exclude the possibility that varying solar input to the oceans over several solar cycles has an effect on the energy characteristics within the oceans which in turn affects movement of large bodies of water within the oceans.”
Please see the bottom graph on this page, which shows that ENSO is controlled somewhat by high and low sunspot periods.
http://www.stormx.com/agriculture/weather-risk/2009/04/solar-activity-lowest-in-almost-100-years-implications-for-climate-potentially-significant/
I live in the high mountainous plains of NE Oregon. Understanding the PDO is essential for farmers, especially for those who grow less than robust crops. The lows, highs, and precip can be forecasted based on what the PDO is doing. That information tells farmers what to plant and when. Unless they are just plain dumb and plant spring wheat every year hoping this is the year it won’t freeze the tender spring shoots. Gardens are just now being plowed and no garden veggie seeds are in the ground yet that I can see (unless you are just plain dumb…see above). For those who understand this, seeds are being germinated in damp paper towels, etc. And recipes are being dusted off for green tomato salsa. Even peas will be a bit late this year but with the cool temps, we should have another bumper crop. Fast growing corn should be planted instead of the 10ft tall variety. I would skip potatoes. The tubers will freeze in the ground this fall before they are ready to be dug up. I would also put in the short season stubby carrot, not the long variety.
I could go on. You can see that this is just the tip of the iceberg when it comes to understanding the PDO and local weather pattern variation. It is information that makes or breaks the backs of farmers. I wish weather reports and services would go back to the days of reporting this kind of information instead of “climate change” warnings.
Bob Tisdale
Thanks for the answers. You said ” As discussed and illustrated, the PDO cannot directly explain global temperatures variations because it represents a pattern of SST variability, not SST” I think that the pattern that PDO index represents does reflect a changing SST pattern in the Pacific which is a forecaster of certain weather patterns and tempertaures for North America and hence the northern Hemisphere and the globe . In this context the pdo can explain global temperatures , in my opinion. I still disagree with your statement. It may not be a direct explanation but it does still explain global temperatures in my opinion.
Ian Holton
I agree with your oservations about El Nino’s . Here is my count of El Nino’s and La Nina’s during positive and negative phases of PDO.
DURING NEGATIVE OR COOL PDO AND AMO PHASE [1944-1976]
Number of La Nina’s
STRONG 2
MODERATE 4
WEAK 2
————————
Total 8
Number of El Nino’s
STRONG 1
MODERATE 3
WEAK 4
———————-
Total 8
DURING POSITIVE OR WARM PDO AND AMO PHASE [1977—2007]
Number of La Nina’s
STRONG 2
MODERATE 2
WEAK 3
————————-
Total 7
Number of El Nino’s
STRONG 3
MODERATE 2
WEAK 8
————————
Total 13
Lubos: You wrote, “If you averaged over 1-2 years, the ENSO index would give you an equally pronounced sin-like signal like PDO gives at the multidecadal scale.”
For you I’ve generated two more graphs. Here’s the PDO and ENSO 3.4 SST anomalies smoothed with a 25-month filter. The NINO3.4 data produces a signal that is more equally balanced than the PDO, but both have underlying variations as well.
http://i39.tinypic.com/23t5pi9.jpg
The second is a curve of the PDO MINUS NINO3.4 SST anomalies, as suggested by Carl Wolk. I will agree that the additional variations in the North Pacific have to influence the tropical Pacific, but…as noted in Zhang et al, the PDO pattern and the decadal variability are lagged responses to ENSO. Then the additional long-term variability in the North Pacific (referred to as the PDO) would be a decadal component of ENSO, providing feedback to the basic ENSO signal.
http://i40.tinypic.com/2mop4dj.jpg
Regards
Stephen Wilde: You wrote, “Well in my opinion something external to ENSO is causing those phase shifts and imposing them onto the ENSO cycle.”
Bratcher and Giese believe the shifts originated in the subsurface South Pacific. Refer to “Revisiting Bratcher and Giese (2002)”:
http://bobtisdale.blogspot.com/2009/04/revisiting-bratcher-and-giese-2002.html
The WUWT post of it is here:
http://wattsupwiththat.com/2009/04/10/revisiting-bratcher-and-giese-2002/
Also:
The long-term variations in the detrended North Pacific SST anomalies and the AMO (detrended North Atlantic SST anomalies) agree quite well. Are they climatically coupled?
http://i43.tinypic.com/2sbqtrk.jpg
And ENSO influences the AMO as discussed in my post “There Are ENSO Induced Step Changes In The North Atlantic”:
http://bobtisdale.blogspot.com/2009/02/there-are-also-el-nino-induced-step.html
And ENSO Influences AMOC. The following graph was taken from my post “Atlantic Meridional Overturning Circulation Data”:
http://i33.tinypic.com/5cyglz.jpg
Then there are what appear to be 100-year cycles in ~80% of the Southern Ocean (green curve).
http://i41.tinypic.com/qsjwwp.jpg
That’s from my post “A Closer Look At The ERSST.v3b Southern Ocean Data”:
http://bobtisdale.blogspot.com/2009/04/closer-look-at-ersstv3b-southern-ocean.html
Basil: You asked, “Moreover, if, as the abstract I cite, and the Newman, et al paper you cite (“The PDO is dependent upon ENSO on all timescales.”) are correct that ENSO plays a part in the decadal pattern of variation in the PDO, how does that constitute a disproof of the proposition that during a PDO cool phase La Nina’s dominate, while during a PDO warm phase El Nino’s dominate?”
It doesn’t disprove it at all. BUT, why not think of the domination of La Nina or El Nino events as dictating the sign of PDO? That would be more consistent with those papers and with the Zhang et al.
I keep seeing three solar cycles per PDO phase, six per cycle. All you need is something alternating each solar cycle, to put two of one type of solar cycle and one of the other into each phase of the PDO, thus giving alternating warming and cooling phases. Well, the shape of cosmic ray peaks alternate each cycle from pointed to broadened. And cosmic rays are tentatively linked to clouds and albedo. It isn’t enough to satisfy Leif, but the bare bones of a mechanism are there.
===========================================
Jeremy Ross, PhD: Thanks for the link. As soon as I catch up with replying to comments, I’ll enjoy reading it.
Let me pose something for you and for others to ponder. First, a scaled running total of Hadley Centre NINO3.4 SST anomalies reproduces the global temperature anomaly curve. Refer to:
http://bobtisdale.blogspot.com/2009/01/reproducing-global-temperature.html
Second, Newman et al wrote, “Thus, over the course of years, at least during winter and spring, the North Pacific integrates the effects of ENSO.”
Could the similarities of the scaled running total and global temperature anomalies indicate that additional portions of the global ocean also integrate the effects of ENSO?
I think this is still skating round the issue.
That issue is whether the ENSO cycle is sufficient on it’s own to account for observations without requiring any seperate input from independent oceanic variations.
From the links provided I get the impression that there is doubt and that there are differences of opinion on the matter.
For my part I cannot envisage there being no independent input from ocean variability.
Nor can I see how ENSO variability alone could give rise to phase shifts at 30 year intervals with a complete cycle of twice that.
It seems to me that the logical solution is that ENSO contributes air induced variations over short periods of time, seperate oceanic cycles contribute their own independent variations over longer periods of time and the general underlying trend is dependent on slow long term changes in solar output over many solar cycles.
On that basis one can explain the short term climate consequences of individual El Nino and La Nina events, the 30/60 year longer term variations in the dominance of El Nino and La Nina and also the slow rise in global air temperatures since the little ice age.
Indeed, one can also account for the stepped upward temperature movement when each positive longer term oscillation leaves the air temperatures a little higher than the one before as during the 20th Century.
Now if the sun were to go into a long term cooling trend (albeit very slow) then I would expect to see stepped downward temperature movements develop but it would take two more 30 year phase shifts for that to become apparent.
In the meantime a coincidence of weaker solar cycles and negative oceanic phases (especially if in all the oceans at the same time) would be capable of giving us a large a downward shift in temperatures as we saw a large upward shift in temperatures during the recent warming when we had both a more active sun and positive oceanic phases.
‘Large’ being a relative term of course. I’m quite sure that natural forcings can do far more than our records to date have ever noted.
Bob Tisdale (08:50:12) : It doesn’t disprove it at all. BUT, why not think of the domination of La Nina or El Nino events as dictating the sign of PDO? That would be more consistent with those papers and with the Zhang et al.
Okay, I must admit I misunderstood your:
There is also a popular belief that the sign of the PDO dictates whether El Nino or La Nina events dominate. There is, however, an analysis that contradicts that belief.
I now read you to be saying that the order of “causation” is reversed. However, without going back and looking at the Newman paper, or the Zhang paper, the paper I cited allows for decadal or multi-decadal influences on the PDO besides ENSO. In fact, if you are saying that “The PDO is dependent ONLY upon ENSO on all timescales” — then a disagreement remains. I would, for instance, leave room for various other influences, such as the lunar nodal cycle, solar, and atmospheric changes related to changing LOD influencing the multi-decadal climate regimes indicated by the PDO (which, as you know, is not limited to the instrumental period, but is indicated in proxy data).
Is anybody actually denying an ENSO influence on the PDO? I think the real debate, if there is one, is whether ENSO dominates the PDO to the point that we can ignore the concept of regime shift and long periods where climate regimes so dominate global temperatures that we cannot isolate the AGW influence from climate regime trends. I.e., if PDO can be reduced to ENSO, and if ENSO is basically an intradecadal phenomenon (say a 5-6 year cycle), then the multidecadal trend that began in the mid 1970’s must owe to AGW influences, and cannot be attributed in any significant measure to climate regimes.
Frankly, showing a relationship between PDO and ENSO at the intradecadal scale is trivial. In spectral terms, Nino 3.4 and PDO both show a strong peak at ~5.6 years. But the PDO shows multi-decadal trends that are not evident in the Nino 3.4 series. So the PDO must be something more than just ENSO.
Do you agree with that, or not?
Thanks, by the way. Regardless of where you come out on this, it has been a stimulating discussion of the kind that has helped me clarify some things about this issue.
Basil
Frankly this seems very circular and dissembling.
To seek to dismiss the PDO, even as it ahs been driving the temps down and changing rain patterns in a very predictible patter seems to be motivated by something other than seeking answers. It seems much more like seeking to squelch things that invalidate AGW.
hunter,
You may say that but I couldn’t possibly do so.
Sorry for posting this on this thread but I would like to hear some discussion on a research project done on African droughts . Here is a quote from the National Geographic article:”More and more, it’s starting to look like the AMO is a big player affecting climate change around the Northern Hemisphere, including drought variability over Western Africa and western North America,” Overpeck said.”
http://blogs.nationalgeographic.com/blogs/news/chiefeditor/2009/04/extreme-droughts-west-africa.html
Another article, which I can not find at the moment, also connected this event to the Little Ice Age.
Correction:
hunter,
You may say that but I couldn’t possibly comment.
alf,
When the air around the globe is warming the air circulations move poleward.
When that air is cooling the air circulations move equatorward.
The desert areas shift polewad or equatorward accordingly.
Currently the air is cooling so areas south of the Sahara will get drier after a period of alleviation whilst the air was warmimg.
It’s not just the AMO or PDO. What matters is the net warming or cooling effect of all the oceans combined as regards their effect on the air. That dictates the latitudinal positions of all the weather systems in so far as they move beyond normal seasonal variability.
Willis Eschenbach provided an interesting graph on CA back in Sept of 2007.
UAH Trop temps with and without El Nino effect.
This is his image…
http://homepage.mac.com/williseschenbach/MSU_and_el_nino.jpg
This is the page on CA….
Anthony Watts at UCAR
http://www.climateaudit.org/?p=1999
This is the reply / comment containing the image…
http://www.climateaudit.org/?p=1999#comment-133427
This is his post image comment….
As you can see, eliminating the El Nino effect removes some of the swings, but does not affect the trend.
——————————-
Hence, it would appear that if the general trend still exists after removing the El Nino effect then that general trend comes not from the ENSO but rather from another source.
On the other hand. If I look at the PDO by itself the only potentially significant real agreement (even after adjustment for volcanic activity) I see occurs in 1999 with the ‘great shift’. While investigating the divergences I compared NINO3.4 0l.v2 SST with UAH lower trop temps and found a surprising level of agreement between the two.
At first I thought the was a good deal of disparity but then it became clear that what I was seeing was an apparent lag between the Nino3.4 and the global trop temps which, to me, is logical.
Prior to the shift in 1999 there appears to be very close agreement, after the shift the only real disparity was a greater drop in temps in the Nino than in the Lower trop. That again makes logical sense since the rest of the oceans did not have the same drop the Nino did and the Lower trop is effected by ocean temps as a whole; global SST and UAH lower trop temps in close agreement.
Though I am not totally ready to file for divorce from my thoughts on the PDO this blog entry by Bob Tisdale has invoked a further review into the facts and I thank him for that.
Tony Hansen: You asked, “What percentage of the whole Pacific would this be?”
The area of the Pacific North of 20N should be approximately 22 to 25% of the Pacific Ocean. If you want to check my math, the Latitude/Longitude Distance Calculator helps:
http://www.nhc.noaa.gov/gccalc.shtml
You asked, “Why would the scale not be symmetrical?” Good question. The years JISAO used as “typical” years and the base years of the anomalies would dictate the scale. Maybe it’s like my Figure 5, where the data fit better with a scale that ran from -0.3 to 0.4 deg C.
Thanks Bob
As luck would have it. The new paper out by Douglass and Christy ( Limits on CO2 Climate Forcing from Recent Temperature Data of Earth ) has a great graph of the NINO / UAH comparison. Had I seen it earlier I would have spared myself a bit of work this morning. It shows the lag I was talking about in my previous post.
Anyway.. . it appears on page 20 of the PDF found here…
http://www.pas.rochester.edu/~douglass/papers/E&E%20douglass_christy-color.pdf
Steven Garland: “My understanding, after reading this article, is that the PDO represents the SST anomaly for the North Pacific after subtraction of the global SST anomalies (please correct me if I am wrong).”
You asked for a correction, so here it is. If global SST anomalies are subtracted from the SST anomalies of the North Pacific (north of 20N), creating a residual, the resulting curve is very similar to “North Pacific SST Anomalies MINUS Global Temperature Anomalies” curve in Figure 5. A North Pacific Residual (North Pacific SST Anomalies MINUS Global SST Anomalies) versus PDO comparison is here:
http://i40.tinypic.com/w9d4k6.jpg
The calculation of the PDO involves many more steps than a simple residual, as noted above in the post.
Thanks Bob Tisdale for all your efforts to help us understand that the PDO is not a climate driver. I am going to try this in everyday language, so the science will probably get botched. The PDO is not a driver of anything in and of itself; it tells us what has happened re changing ocean temperatures (in the North Pacific) because of certain changes in the global climate system, most specifically through ENSO. It tells us this because scientists have developed the statistics so it can do so. It “represents a pattern of SST anomalies”, of “SST variabilites”. Now, If ENSO “provides” the main temperature changes that instruments record, moving the North Pacific temperature from warm to cool in short bursts that eventually add up to approximately 30 year periods of more warm vs more cool we know as the PDO, then I think you are directing our attention to what drives or causes ENSO variability.
Your graph, Figure 4, in “Revisiting Bratcher and Giese (2002)” amazingly (to my eyes) shows the step change caused by NINO3 SST anomaly from an average of 0.003 Deg C from 1950-1975 ratcheting up to an average of 0.045 Deg C from 1978-Present. Therefore, ENSO initiated the warm PDO. In your other work, I can see the influence of ENSO on the Indian Ocean.
It seems that you are pointing us to one of the most exciting areas of research. ENSO appears to be extremely influential. It certainly doesn’t have to “do everything”, but what makes it so important? It appears to be a collection of “small” TSI variabilities (without cloud cover) affecting at the very least SSTs, input from the Humbolt Current, changing pressure gradients, winds, mixing of deep oceanic layers with other layers and the surface, and, I imagine, the influence of the topography and geothermal heating — in the largest global ocean and the greatest ocean extent of the tropics. What drives ENSO seems to be an important place to begin with this dynamic system.
About the AMO. I don’t yet understand its difference from the PDO, but where you graph them, I notice the AMO leads the PDO most of the time. I think you have suggested that ENSO also significantly influences the AMO.
Stimulating research. You certainly get an interesting back and forth with other knowledgable researchers. I hope you generate a intense new focus of research energies — but how to get the grants for something important rather than see them go to line the pockets of the corporate-marxist-university-government-publication elites.
AlexB: You wrote, “Maybe I’m reading this wrong but my best assessment is that the author is arguing that the PDO cannot be responsible for warming or cooling as it is not a direct indicator of SST anomalies in the north pacific or difference between SST and LST anomalies.”
You missed a few points. I tried to provide basic information for those new to the PDO as well as advanced discussion material for those who are well-versed. The intent of the post was to define what the PDO is, illustrate how it is calculated, illustrate what it is not, and present the findings of the Newman et al study, which indicated that the PDO is dependent on ENSO. I expected Newman et al to be controversial, and it was. I also linked the Zhang et al study, which was the basis for the PDO. Then in follow-up comments, I noted how the Zhang study reinforces Newman inasmuch as it concludes the spatial pattern and multidecadal nature of the PDO are lagged responses to ENSO.
Regards.
Syl: You wrote, “The problem here is that though the PDO has a ‘period’ of about thirty years and thus supposedly cancels out after about sixty years, that leaves room for an odd number of oscillations in a century.”
As you’ll note in Figure 9, the NINO3.4 data also has periods, during which El Nino events or La Nina events dominate. The two referenced papers indicate the PDO is an aftereffect of ENSO, so I’ll address the ENSO portion of Figure 9. The “cancels out” portion of your comment only works if the frequency and amplitude of the positive events are equal to the frequency and magnitude of the negative events. They clearly are not. El Nino (positive) events are clearly dominant. During periods when El Nino events dominate, global temperatures rise, and global temperatures decline when La Nina events dominate. The only period when that doesn’t work is the 1960s+, but during the 60s, volcanic aerosols overrode ENSO.
“Lubos Motl (09:54:48) :
The statement that “PDO is influenced by ENSO” contradicts very basic physical principles. PDO is the slower process among the two, so it can’t possibly be measurably influenced by faster processes such as ENSO.”
This assertion is false. If I hit one of those heavy boxing bags in rapid succession in a pattern, first on one side, then on the other, that patterned sequence of rapid blows will cause a slower-moving oscillation in the punching bag. This is true whethor or not the punching bag is already oscillating due to someone else hitting it, too. When the earth revolves around the sun once a year, the gravitational force between the two causes an oscillation of the sun (albeit a lot smaller magnitude) with a much slower cycle. The fact that the PDO oscillates on a much slower cycle than ENSO does not disprove the proposition that the PDO is a causal function of ENSO.
gary gulrud: You quoted Nate Mantua and wrote, “IMHO Money quote from Bob’s Mantua link.”
Thanks for emphasizing that. I believe it might also help explain another phenomenon. If you were to compare Global Temp and NINO3.4 from 1978 to 2007, you’ll note that the global temperature response to El Ninos was much greater than it was to La Ninas. Has that swapped now that the PDO is in the negative mode? Global temperature definitely responded to the 2007/08 La Nina. It’ll be interesting to see how global temps respond to an El Nino now.
Stephen Wilde: You quoted Zhang et al, then commented, “They are clearly hedging their bets as to whether there is a seperate mode of variability which operates independently of the ENSO influence. They recognise that others do think that other modes exist, as do I.”
I don’t interpret the quote as Zhang et al hedging their bets. To me they’re downplaying the referenced studies. In fact, their final sentence emphasizes that. They wrote, “In view of these problems and the limited length of the historical record, we are not convinced that a formal modal separation involving interdecadal variability is meaningful.” In other words, the other studies tried to separate the modes and they see no reason for doing so.
Basil: You wrote, “Is anybody actually denying an ENSO influence on the PDO?”
Based on the comments on past posts here at WUWT and at other bogs, there are many who don’t even consider it. The standard comment goes something to the effect, “Well now that the PDO has shifted global temperatures will drop.” The PDO is the only driver of climate in their eyes. Maybe I need to introduce the “G” Time Series data to those who don’t know it exists. It’s also from the Zhang et al paper.
http://jisao.washington.edu/data_sets/G/
You wrote, “So the PDO must be something more than just ENSO,” then asked, “Do you agree with that, or not?”
I don’t disagree with it. Newman et al and Zhang et al both fail to address the impact of Meridional Overturning Circulation in the North Pacific. I seem also to recall reading about teleconnections between the North Atlantic and North Pacific. There are a multitude of other factors that can influence both ENSO and the PDO on yearly, decadal, and multidecadal bases.
hunter: You wrote, “Frankly this seems very circular and dissembling.
To seek to dismiss the PDO, even as it ahs been driving the temps down and changing rain patterns in a very predictible patter seems to be motivated by something other than seeking answers. It seems much more like seeking to squelch things that invalidate AGW.”
I am in NO way attempting to squelch things that invalidate AGW. If you were to read my other posts here at WUWT or at my blog, you’d note that my primary empahsis is to illustrate phenomenon that downplay AGW. My hope with this post was to redirect the understanding of primary driver from the PDO to ENSO.
Bob,
Good answers but for the moment I’ll continue my view that there is something more than ENSO variability involved in the PDO, probably independent variation within the oceans.
I have no problem with you emphasising the ENSO influence on short term variability within the background PDO cycle.
Ninderthana: If I understand your argument, referring to Figure 9, you would have expected the PDO from the mid-70s to early 2000s to have reacted more like the PDO did from the 1920s to the 1940s. From the 1920s to the 1940s, the PDO rose gradually until it peaked in the late 1930s. But the period from the mid-70s to the early 2000s had the sharp rise at the beginning, which you believe contradicts the Newman hypothesis. Did I get that right?
My first thought: There were differences between the two periods. The latter period had two major El Nino events, where the early period had none. The latter period also had two significant volcanic eruptions, where the early period had none.
In the following graph, for the period of January 1975 to February 2009, I subtracted NINO3.4 SST anomalies from the PDO and compared it to NINO3.4 SST anomalies (as a reference for timing only) and comapred them to Northern Hemisphere Sato Index data (used also only for timing so the Sato Index scaling is unimportant). Then I smoothed the data with a 12-month filter.
http://i40.tinypic.com/otd739.jpg
In looking at the graph, two anomalous periods stand out, but there may be more than two. The first occurred after the 1982/83 El Nino and the 1982 eruption of El Chichon. The difference between the PDO and NINO3.4 SST anomalies rose in an atypical fashion. Did the El Chichon eruption alter the normal effects of ENSO on the North Pacific and, if so, what changed? The second anomalous period occurred after the 1997/98 El Nino. Did the 1997/98 El Nino shift enough heat to the North Pacific to upset the normal SST anomaly patterns of the PDO?
Since I can’t answer those questions, I can’t dismiss the Newman et al hypothesis (assuming I’ve correctly interpreted your argument).
Hmmn. I’m not sure whether I’m one of the “those” bloggers, but assuming I am, my argument is a little different. It is simply that long term climate oscillations modulate the warming trend. Currently, we seem to be in a cooling phase – that suppresses the indicators of warming, but doesn’t change the build up of heat. When we switch back into a warming phase, global temperatures will increase at double the average rate for the century.
So is the PDO cause or effect? I’m not too concerned. But I do think it is an indicator of the global climate oscillation. That oscillation is not a driver of temperatures, but a modulator.
30 years of cooling? Yay! I’m alright. Let the kids fry.
bob
I also thank you Bob for bringing up this track. It helps all of us to focus on a very vital topic. The one area that I still have probem is your statement ” PDO cannot directly explain global temperature variations because it represents a pattern of SST variability, not SST.” What is a “pattern of SST anomlies’, what is its make up and why do these anomalies do not affect temperatures ? To me, PDO patterns explains local , hemispheric and ultimately global tempertures as well as the global warm and cool periods very clearly . It may not be the prime or first level driver but whether ENSO events are the prime driver is also up for debate as we really do not understand what triggers them . We do not even understand what triggers or moves the deep ocean currents .
I would love to hear your comments on the make up of AMO, Bob
vibenna: You wrote, “Currently, we seem to be in a cooling phase – that suppresses the indicators of warming, but doesn’t change the build up of heat. When we switch back into a warming phase, global temperatures will increase at double the average rate for the century.”
Reallly???!!!
Where is the build up of heat? Temperatures are dropping in the lower troposphere. Sea surface temperatures are dropping. Ocean heat content is either flat or dropping, depending on the dataset. So where exactly is this heat being stored for future release?
Bob Tisdale, please forgive my earlier post. Delete it if you can, at least from your mind. I still fell into the same trap as many others. I have some questions that were struggling to be formed in that garbled attempt, in part because I follow your research, and in part because the oceans seem to me to be the most significant storage, heat-cold mixing, heat-releasing, and heat-circulating source on earth. I will try again later. Thanks for you unwavering attempts to clarify the PDO. You better be careful; soon you will gain a reputation something like Leif’s.
matt v.: You asked, “What is a “pattern of SST anomlies’, what is its make up and why do these anomalies do not affect temperatures?
In the following SST anomaly map, I’ve cropped Figure 7 from the post so that it only shows the North Pacific, North of 20N. That is the only part of the Pacific Ocean expressed by the PDO. Nothing else. (Before you comment on that, read the rest of this reply, please.)
http://i39.tinypic.com/4r5oxx.jpg
The illustration shows SST anomalies in a cool phase, which means the SST anomalies in the eastern North Pacific are cool while the SST anomalies in the central and western portions are warm. But note that the warm area is significantly larger than the cool area in the east. The average SST anomalies for the North Pacific north of 20N in that case are probably warm even though the PDO is in the cool phase. And if the average SST anomaly is positive, it is contributing more positive anomalies to the global average than “normal”.
Now scroll up to Figure 5. It illustrates the PDO, and it also shows the temperature difference between the SST anomalies of the North Pacific north of 20N (the same area as the PDO) and Global Temperature anomalies. That second dataset is calculated as North Pacific SST anomalies minus Global Temperature anomalies. Note that between the early 1940s and the late 1970s, the PDO was below zero (in the cool phase) for the most part. But during that same period, North Pacific SST anomalies were greater than Global temperature anomalies, so that part of the Pacific Ocean was actually contributing positive anomalies to the global average—in other words, it was heating.
I do understand that while the PDO is in the cool phase, other parts of the Pacific are NORMALLY cooler than normal, like the eastern tropical Pacific, like the equatorial Pacific (the NINO areas), etc. And someone could try to argue that fact. The point is, the PDO only deals with the area highlighted in Figure 7. Nothing else. There is another dataset to express the pattern of variability in the entire Pacific basin, and it’s called the Interdecadal Pacific Oscillation or IPO. And there’s another dataset for discussions of the pattern of variability for the global ocean called the “G” Time Series. If someone wants to discuss the eastern equatorial Pacific SST anomalies, there are the NINO indices and the Cold Tongue Index (CTI).
You wrote, “I would love to hear your comments on the make up of AMO, Bob”
I covered the AMO in a recent post here:
http://bobtisdale.blogspot.com/2009/04/atlantic-multidecadal-oscillation.html
Based on the way it is calculated, the AMO is in no way related to the PDO.
Bob,
I have some results you might find interesting, showing maybe even more clearly the correlation/correspondence between ENSO and PDO.
If you are interested, email me at blcjr2 at gmail dot com and I’ll send it to you.
Basil
bob
Your comments seem at odds with the quote from this web page ? Especially the last paragraph.
My own research of warm and cool temperature anomalies along the west coast of Canada and the western half of Canada is that they correlate with warm and cool phases of PDO. How do you explain that? Warm and cool PDO levels correlate and thus explain local temperatures which become part of the make up of global temperaure anomalies
“The PDO is often quantified by the use of an index, referred to as the PDO Index. The PDO Index is calculated by spatially averaging the monthly sea surface temperature (SST) of the Pacific Ocean north of 20°N. The global average anomaly is then subtracted to account for global warming (Mantua, 2000). Normally only October to March values are used in calculating the PDO index because year-to-year fluctuations are most apparent during the winter months (Mantua, 2001).
When the PDO Index is positive, waters in the north central Pacific Ocean tend to be cool, and waters along the west coast of North America tend to be warm. The opposite is true when the PDO Index is negative (Null, 2002).
The effects of the PDO are most drastic in the Pacific Northwest. In this region, a positive, or warm phase PDO, generally correlates with lower than average rainfall and higher than average air temperatures. Likewise, a negative, or cool phase PDO, correlates with relatively high precipitation rates and low air temperatures (Null, 2002).
http://ffden-2.phys.uaf.edu/645fall2003_web.dir/Jason_Amundson/pdoindex.htm
Vibenna – and anyone else concerned that once the PDO turns positive again, global temperatures will accelerate – as Bob pointed out, there is no extra heat built up since 2002, though the oceans still hold 84% of the past century global warming signal (IPCC), but you need to check out a few related things:
1. the carbon dioxide model uses an unproven multiplier (x3) in relation to water vapour, and on its own, the gas even when doubled in concentration can warm things only a little (between 0.5 and 1.0 C);
2. the global rise between 1980-2000 was taken as proof of the multiplier, but…
3. in 2007/8 re-analysis of upper ocean heat content revised the heat stored in that period downward by x2;
4. the computer models ignored the PDO (it wasn’t invented when they were created and they still have not incorporated these long term cycles), but assumed the global fall from 1945-1980 was due to anthropogenic sulphur – in 2005 new analysis of satellite data showed this was a false assumption (admitted also by the IPCC in 2007) and that the human effects were localised – the global effect was therefore natural and due to cloud and aerosols (which i assume are linked to ocean cycles).
Finally, this demonstrates that the current suite of computer models upon which projections of future warming are based are seriously flawed – this is not yet admitted and efforts are being made to incorporate ocean cycles.
And for the link to global temperatures – do not underestimate the spatial patterning around Alaska – this has a knock-on effect, delayed by a few years, on the Beaufort Gyre and the dynamics of the Arctic Basin – the cooling of SSTs off Alaskan pacific coasts in late 2006 and the 2007/2008 cooling of the Alaskan shelf, these will reverse the gyre and expel warm water from underneath the sea-ice, and also lead to less cloud (Arctic cloud increased above the ice from 1982-2000 by 14%). Although GHG theory expects a stronger effect in the water-vapourless Arctic, the clouds and warm seawater are the dominant factors – and these have now reversed their trend, hence in 2008, there was 9% more sea-ice in the summer.
It will be interesting to see whether an El Nino will drive world temperatures up – this was predicted for 2007 by Hadley – and they got it spectacularly wrong – that prediction was based on an incipient El Nino signal at the end of 2006.
The following areas correlate with PDO
http://ffden-2.phys.uaf.edu/645fall2003_web.dir/Jason_Amundson/trends.htm
matt v.: You wrote, “Your comments seem at odds with the quote from this web page ? Especially the last paragraph.”
The final paragraph agrees with what I’ve written. The “Pacific Northwest” Coast of North America, which is land, lies adjacent to the eastern North Pacific Ocean. During the cool phase of the PDO the eastern North Pacific (ocean) is cooler, so the Pacific Northwest Coast of North America (land) would in turn be cooler. But again, the western and central portion of the ocean surface area included in the PDO are warmer when the eastern part of the North Pacific Ocean is cool. The area included in the PDO is only sea surface, not the adjoining landmass.
bob
You are correct. I mixed my east and west and forgot that our west coast[land] is Pacific ocean’s east side[water]. It is the effect of this eastern half of the Pacific ocean that correlates with our air temperatures in the western half of our continent as the previous post showed.Since the jet streams go west to east , it impacts the weather all the way to the Great Lakes and many times to our Atlantic coast. These are significant weather fronts .
Thanks again. I have no further comments.
matt v.: Regarding you last comment, I agree that the Eastern North Pacific SST anomalies of the area included in the PDO impact Western North America Land Surface Temperature anomalies. No doubt about it. In fact, here’s a graph to confirm that fact:
http://i41.tinypic.com/2ee8uj7.jpg
HOWEVER
On the other side of the Pacific, the Western North Pacific SST anomalies have an impact on Eastern Asian Land Surface Temperatures, and here’s that graph.
http://i42.tinypic.com/20ppslw.jpg
Looks like a wash then for the direct impacts of the PDO on land surface temp anomalies. Why? During the extremes of the PDO, if the Eastern Pacific is warm or cool, the Western Pacific has changed in the opposite direction.
Again, the effects on global temperatures of the Pacific are dictated by the entire Pacific basin, not only the area of the North Pacific included in the PDO.
Also, here’s a map of the areas included in those graphs.
http://i43.tinypic.com/2w3z0o4.jpg
Regards
BOB
Great graphs. They show clearly how ocean SST affect climate in the Pacific region. What would CRUTEM 3 GLOBAL and HADISST Western Pacific look like . Perhaps HADISST Western Pacific is even a better indicator of global temperatures because it is more vast?
The traditional 2 cool and 2 warm periods are clearly better apparent.
Bob
HADSST 2NH for NORTHERN HEMISPHERE OCEAN SURFACE TEMPERATURE ANOMALIES and HADCRUT 3GGL for GLOBAL SURFACE TEMPERATURE ANOMALIES have a super fit . This is like AMO[ATLANTIC SST ]and PACIFIC STT
combined which match global surface temperatures.
http://www.woodfortrees.org/plot/hadsst2nh/from:1976/to:2009/plot/hadcrut3gl/from:1976/to:2009
And the reason is, the trade wind blows all that hot surface water up against landforms to the East of us. While we freeze in uncovered and upwelled cold water, Japan melts in the warm waters we blew over their way.
matt v, your graph begs the question why we continue to discuss climate change, CO2, and the Sun, when we should be discussing weather pattern variation. What’s that line from the Wizard of Oz? Something about it’s right in your own backyard?
matt v., regarding your graph: Land surface temperatures follow sea surface temperatures. Always have, alway will. LST exaggerates SST a little and there’s some other noise, but LST is pretty much just along for the ride.
Regarding the AMO: It’s calculated (method used by NOAA ESRL) by detrending North Atlantic SST anomalies. They aren’t the same, but are based on the same data.
Someone on topic: I just posted the April 2009 SST anomalies:
http://bobtisdale.blogspot.com/2009/05/april-2009-sst-anomaly-update.html
Regards
Bob / Pamela
Thanks for the data. There is a world out there that still thinks Co2 is responsible for climate warming and climate change . If you tell a lie often enough the myth becomes stronger than the truth.
Bob, thanks for the reply.
You responded “A North Pacific Residual (North Pacific SST Anomalies MINUS Global SST Anomalies) versus PDO comparison is here:
http://i40.tinypic.com/w9d4k6.jpg”
“The calculation of the PDO involves many more steps than a simple residual, as noted above in the post”
Bob, if you inverted one of the graphs they would look a lot more similar!
Steven Garland: You wrote, “Bob, if you inverted one of the graphs they would look a lot more similar!”
Here’s the residual graph again, so you don’t have to scroll up.
http://i40.tinypic.com/w9d4k6.jpg
But in this correct form, it shows that during the cool phase of the PDO (between the early 1940s and late 1970s) the SST anomalies for the North Pacific were greater than Global SST anomalies. This means that the North Pacific was actually adding to the global SST anomalies during this period; i.e., it was warming.
Remember, it’s the small Eastern portion of the North Pacific being cool that’s indicative of the cool mode. The anomalies of the much larger central and western portion of the North Pacific are elevated and they represent a much greater area.
It’s the typical “cool” pattern over the rest of the Pacific that dictates whether global temperatures fall, and that pattern is a product of ENSO. (It’s also dominated by ENSO.)
The reverse holds true when the PDO is in the warm mode.
Regards