From the University of Hawai’i’s International Pacific Research Center at Mānoa, comes this proxy reconstruction that does some similar things the infamous hockey stick reconstruction, but doesn’t need to pull any statistical “tricks” to make the case. Our old friend the Bristlecone pine in the southwest USA is the test subject again. Like Mann et al, they combined the recent instrumental record of ENSO variation with the proxy reconstruction record, except they didn’t need to delete any data, nor cover up any curves that didn’t behave as they wished. “Mike’s nature trick” wasn’t needed.
Why? Well it has to do with the Bristlecone pine being a better proxy for precipitation than temperature, and as we know, El Niño gives some significant precipitation impacts to the desert southwest. I’ve covered the issue of “treemometers” here and here previously, and they are worth a read as a refresher for this new paper.
They write:
During El Niño, the unusually warm surface temperatures in the eastern Pacific lead to changes in the atmospheric circulation, causing unusually wetter winters in the US Southwest, and thus wider tree rings; unusually cold eastern Pacific temperatures during La Niña lead to drought and narrower rings.
Liebig’s Law in a nutshell. In a desert, water is the limiting factor for growth, temperature, not so much. So I’m inclined to trust this reconstruction a bit more than I’d trust Mann’s. The test will be when somebody asks for the data for replication purposes.
Tree Rings Tell a 1100-Year History of El Niño
El Niño and its partner La Niña, the warm and cold phases in the eastern half of the tropical Pacific, play havoc with climate worldwide. Predicting El Niño events more than several months ahead is now routine, but predicting how it will change in a warming world has been hampered by the short instrumental record. An international team of climate scientists has now shown that annually resolved tree-ring records from North America, particularly from the US Southwest, give a continuous representation of the intensity of El Niño events over the past 1100 years and can be used to improve El Niño prediction in climate models. The study, spearheaded by Jinbao Li, International Pacific Research Center, University of Hawai’i at Manoa, is published in the May 6 issue of Nature Climate Change.
Tree rings in the US Southwest, the team found, agree well with the 150-year instrumental sea surface temperature records in the tropical Pacific. During El Niño, the unusually warm surface temperatures in the eastern Pacific lead to changes in the atmospheric circulation, causing unusually wetter winters in the US Southwest, and thus wider tree rings; unusually cold eastern Pacific temperatures during La Niña lead to drought and narrower rings. The tree-ring records, furthermore, match well existing reconstructions of the El Niño-Southern Oscillation and correlate highly, for instance, with δ18O isotope concentrations of both living corals and corals that lived hundreds of years ago around Palmyra in the central Pacific.
“Our work revealed that the towering trees on the mountain slopes of the US Southwest and the colorful corals in the tropical Pacific both listen to the music of El Niño, which shows its signature in their yearly growth rings,” explains Li. “The coral records, however, are brief, whereas the tree-ring records from North America supply us with a continuous El Niño record reaching back 1100 years.”
The tree rings reveal that the intensity of El Niño has been highly variable, with decades of strong El Niño events and decades of little activity. The weakest El Niño activity happened during the Medieval Climate Anomaly in the 11th century, whereas the strongest activity has been since the 18th century.
These different periods of El Niño activity are related to long-term changes in Pacific climate. Cores taken from lake sediments in the Galapagos Islands, northern Yucatan, and the Pacific Northwest reveal that the eastern–central tropical Pacific climate swings between warm and cool phases, each lasting from 50 to 90 years. During warm phases, El Niño and La Niña events were more intense than usual. During cool phases, they deviated little from the long-term average as, for instance, during the Medieval Climate Anomaly when the eastern tropical Pacific was cool.
“Since El Niño causes climate extremes around the world, it is important to know how it will change with global warming,” says co-author Shang-Ping Xie. “Current models diverge in their projections of its future behavior, with some showing an increase in amplitude, some no change, and some even a decrease. Our tree-ring data offer key observational benchmarks for evaluating and perfecting climate models and their predictions of the El Niño-Southern Oscillation under global warming.”
This research was funded by the National Science Foundation, National Oceanic and Atmospheric Administration, Japan Agency for Marine-Earth Science and Technology, National Basic Research Program of China, and the National Natural Science Foundation of China.
Citation: Jinbao Li, Shang-Ping Xie, Edward R. Cook, Gang Huang, Rosanne D’Arrigo, Fei Liu, Jian Ma, and Xiao-Tong Zheng, 2011: Interdecadal modulation of El Niño amplitude during the past millennium. Nature Climate Change.
Source here (PDF)
I hate to tell you guys but these people have no idea of what they are talking about. These peaks have nothing whatsoever to do with El Nino peaks which on the average are five years apart but may range from three to eight years due to variations in local conditions. The average distance between the peaks they show is about fifty or sixty years – ten times the spacing between ENSO peaks. But there is one phenomenon in the Pacific that supposedly does have a sixty year period, and that is the PDO (Pacific Decadal Oscillation). It is supposed to have warm phase and a cool phase, each thirty years, and the peaks they show are very likely the PDO and not ENSO. To me there was no convincing evidence up to now that such a thing as a PDO even existed but it looks like they’ve stumbled into away of proving that it does. Which brings into question the quality of that vaunted “peer review” system that allows such nonsense to be published. I have found that peer review is no longer true peer review I knew in the sixties but buddy review where journals actually encourage you to name the reviewer you would like to check out your own paper. Some of them will go so far as to demand that you name, for example, four reviewers, and if you don’t their software will not even let you complete your submission. That is the system that allowed the hockey stick to be published and now it brings us a completely misidentified trace of the PDO.
The cosmic rays ARE affected by the magnetic field of the sun and the magnetic field of the earth . Both fields fluctuate.
Fred says:
May 7, 2011 at 10:18 am
…but the idea that bristlecone pines might given information about global temperature is so outlandish that people looking into this are the worst scientists in the world? Just checking the logic here.
Yes, Fred, you are indeed correct that the people looking into BCP rings as GMT proxies – that is, those who have been criticized by the “skeptics” around here so far – couldn’t possibly be the “worst scientists in the world”, simply because, methodologically, those people were not even doing real science to begin with. The basic problem has always been that the “Climate Scientists” alleged to the world that they were in fact doing real science, when they simply weren’t.
Personally, yes, while BCP’s have been known to be good candidates as local precipitation proxies, I do think it’s ‘nuts’ to pursue the idea that wild BCP rings could yield a GMT signal [whatever on Earth a GMT is!] unless everything going on both above and below the ground in the case of each tree was known throughout their histories, in order to “zero” their ring data – in other words, except perhaps by pure chance with a very long confirmation testing period. But my above opinion is not science, either.
David Ball says:
May 7, 2011 at 8:00 pm
The cosmic rays ARE affected by the magnetic field of the sun and the magnetic field of the earth . Both fields fluctuate.
True, but the climate does not follow these fluctuations.
So there goes any remnants of the proposal that El Nino is the driving force behind Global warming.
Hah, that’s funny, Arno. I didn’t read the legends carefully, just looked at the curves, and thought it was the PDO cycles I was looking at.
=============
Mr. Eschenbach, “No, you’re not “checking the logic”. You’re just being a jerk, trying to sound precious and smart, and failing badly.”
Whatever happened to your post about debating the science instead of engaging in personal attacks?
Dear Bob and Leif,
I am puzzled by both your responses to my earlier comments because you both seem to make the same ‘error’.
I was careful to speak of averages over periods of time for both ENSO/PDO and solar data.
However each of you then rebutted my points about the implications arising from such averages by saying that there are ‘anomalous’ spikes within the period concerned and that those spikes invalidate the implications to be drawn from the averaging even though they are taken into account in arriving at the averaged figures.
Very odd.
Stephen Wilde says:
May 7, 2011 at 10:15 pm
However each of you then rebutted my points about the implications arising from such averages by saying that there are ‘anomalous’ spikes within the period
No, these were not ‘anomalous’, but quite normal for solar activity. Perhaps ‘anomalous’ if you have a preconceived fixation, otherwise not. If you want to take very long-term averages, such as: during the first 500 years both were low and during the last 500 years both were high [apart from even that being wrong], then the number of degrees of freedom is so low [like ‘one’] that there is no statistical significance left.
When A E Douglass founded dendrochronology he used it to track variations in precipitation on the West Coast of the USA over the last 2 millenium. He then went on to use it as a tool for dating ancient ruins in the nearby deserts. It is nice that this paper comes almost on the 100th anniversary of this early work to identify natural climate change during historical times…but it is a pity that I find no mention of this background in this press release (perhaps in the paper??).
See an early dendrochronology paper by Douglass in 1909:
http://journals.ametsoc.org/doi/abs/10.1175/1520-0493%281909%2937%5B225d:WCITGO%5D2.0.CO;2
And see also these results published by Huntington from 1914:
http://enthusiasmscepticismscience.wordpress.com/global-temperature-graphs/1914-huntington-tree-rings-precipitation-the-climatic-factor-as-illust-in-arid-america-p-153/
Leif Svalgaard says:
May 7, 2011 at 10:38 pm
I meant ‘anomalous’ as compared to the averaged figure (PDO/ENSO) or as compared to the background trend (solar) not anomalous as compared to the normal range of solar activity.
I didn’t actually average the level of solar activity. I merely pointed out that whilst the averaged level of El Nino was low for 500 years the level of solar activity was declining from the MWP to the LIA (Maunder Minimum) and that whilst the averaged level of El Nino was higher for the second period of 500 years the level of solar activity was increasing from Maunder Minimum (LIA) to date.
The existence of various peaks and troughs in solar activity during each of those 500 year periods does not detract from the background trends.
Furthermore your usual objections as to scale and proportion of solar changes do not apply because I am considering chemical effects and not radiative effects.
Stephen Wilde says:
May 7, 2011 at 10:59 pm
I didn’t actually average the level of solar activity. I merely pointed out that whilst the averaged level of El Nino was low for 500 years the level of solar activity was declining from the MWP to the LIA (Maunder Minimum) and that whilst the averaged level of El Nino was higher for the second period of 500 years the level of solar activity was increasing from Maunder Minimum (LIA) to date.
But that is not even the case. Solar activity in the 1780s was probably higher than in the latter part of the 20th century.
Furthermore your usual objections as to scale and proportion of solar changes do not apply because I am considering chemical effects and not radiative effects.
since what we get from the Sun is radiation and not chemicals, radiation is the relevant parameter.
Another interpretation is that the trend line shows the rate of growth of a tree from infant to adult with the flattening-off occurring at full maturity. Variations around the trend show the environmental effects on growth. As we know that environmental changes affect growth, this is a graph of changes in weather but not necessarily of climate. There may be a climate signal in there but it should not be assumed that there is.
The correlation of BCP with global temp in the last few decades of the instrumental record diverges badly, they don’t know why, they still used the series, and they covered up the divergence in their famous graph. Why don’t you check that logic?
######
as Mc points out the BCP have not been updated in recent years. you are confusing yamal with BCP.
I’ve been told that my use of electricity and gasoline causes climate extremes.
Have we all been misled?
Stephen Wilde says: “I am puzzled by both your responses to my earlier comments because you both seem to make the same ‘error’” And you continued, “I was careful to speak of averages over periods of time for both ENSO/PDO and solar data.”
And I provided you with graphs of average NINO3.4 SST anomalies over specific periods. Here’s one again:
http://i56.tinypic.com/zxmsg8.jpg
And here’s the other again:
http://i55.tinypic.com/33cwt4j.jpg
You continued, “However each of you then rebutted my points about the implications arising from such averages by saying that there are ‘anomalous’ spikes within the period concerned and that those spikes invalidate the implications to be drawn from the averaging even though they are taken into account in arriving at the averaged figures.”
Please document where on this thread I wrote of anomalous spikes. You wrote of averages and I provided you graphs of averages. I believe the ‘error’ is yours.
Stephen Wilde: Let me add to my May 8, 2011 at 1:19 am reply. You wrote, “I was careful to speak of averages over periods of time for both ENSO/PDO…”
You had not mentioned or implied the PDO until this comment.
“nother interpretation is that the trend line shows the rate of growth of a tree from infant to adult with the flattening-off occurring at full maturity. Variations around the trend show the environmental effects on growth”
not. tree rings are standardized to remove this effect. See CA.
“The test will be when somebody asks for the data for replication purposes.”
Shouldn’t “the data” and all other relevant information be released as a matter of corse?
Arno Arrak says: “I hate to tell you guys but these people have no idea of what they are talking about. These peaks have nothing whatsoever to do with El Nino peaks which on the average are five years apart but may range from three to eight years due to variations in local conditions…” You then went on to discuss the frequency of the PDO.
The text for the graph states, “PERIODS OF STRONG EL NINO ACTIVITY [my caps] are indicated by amplitudes above 1.0.” So the graph is not illustrating individual ENSO events. And if you refer to the Supplemental Materials for the paper…
http://www.nature.com/nclimate/journal/v1/n2/extref/nclimate1086-s1.pdf
…you will see that they’re also evaluating individual ENSO events.
So, maybe we do know what we’re talking about.
Bob Tisdale:
I was talking about this one from you:
http://i56.tinypic.com/2ag0u2u.jpg
Clearly the multidecadal shifts from your other graphs average out as virtually flat which implies that throughout the period the ‘excess’ of El Nino over La Nina was virtually constant.
As regards ‘spikes’ the record is made up of peaks (spikes) and troughs.
Leif Svalgaard says:
May 7, 2011 at 11:14 pm
“But that is not even the case. Solar activity in the 1780s was probably higher than in the latter part of the 20th century.”
But not for so long so the averaged slope continued upward until the peak of cycle 23.
“since what we get from the Sun is radiation and not chemicals, radiation is the relevant parameter.”
What we get is a variable mix of wavelengths and particles which seem to have a variable effect on atmospheric chemistry. The chemistry then changes the radiative flux so for climate change purposes chemistry is the relevant parameter.
Stephen Wilde says:
May 8, 2011 at 2:24 am
But not for so long so the averaged slope continued upward until the peak of cycle 23.
No, as the sunspot number is overestimated since 1945.
What we get is a variable mix of wavelengths and particles which seem to have a variable effect on atmospheric chemistry.
The particles [unvarying chemistry] follow the radiation except deliver six orders of magnitude less energy, so radiation is the driver, if anything.
Stephen Wilde replied, “I was talking about this one from you:
http://i56.tinypic.com/2ag0u2u.jpg”
Why are you ignoring the others I provided which clearly contradict you? They’re from the same dataset, but they’re presented differently. The data smoothed with a 121-month running-AVERAGE filter…
http://i43.tinypic.com/33agh3c.jpg
…clearly shows that “the size of the excess of El Nino over La Nina” does not remain “constant through the period.”
M White says: “Shouldn’t ‘the data’ and all other relevant information be released as a matter of corse?”
Some are.
http://www.ncdc.noaa.gov/paleo/recons.html#circulation
But it’s only been a few days since the paper was published.