Yesterday, WUWT covered the sharp drop in global temperatures that followed the peak of the 2015/16 El Niño. That caught the interest of John B from Toronto, and he writes in with a graph he has rendered that illustrates correlation of many climate metrics. While it is true that “correlation is not causation” it can also be said that correlation is well worth investigating further.
He writes:
Thought you might be interested in this plot I made. I’m not a climate scientist, but I have more than 15 years experience in signal analysis.
Coding is done by me using .NET with GDI graphics.
PS. I have done this in my spare time out of curiosity and to facilitate a personal understanding of whats going on. I have no interest in fame, fortune, politics or activism.
Click for a much larger image at 1920×1080 pixels. An even larger 3840×2160 (HD 4k) image is also available here. You really need to look at this graph in magnified form to get the details.
Note that the intensity of the red and green colors used to depict ENSO events (El Niño in red and La Niña in green) varies with the strength of the event. Note also that the Average Absolution Deviation of El Niño is +0.41 while La Niña is -0.27. This indicates warming events due to El Niño and in greater magnitude than cooling events due to La Niña, hence we have a cumulative heat growth from El Niño events during the period in the graph.
Links to data sources used in the graph:
Water Vapor: ftp://ftp.remss.com/vapor/
Temperature Lower Tropasphere: http://vortex.nsstc.uah.edu/public/msu/t2lt
ENSO: http://www.cpc.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml
ENSO: http://www.cpc.noaa.gov/products/analysis_monitoring/ensostuff/detrend.nino34.ascii.txt
Solar: http://www.sidc.be/silso/datafiles
CO2: ftp://aftp.cmdl.noaa.gov/products/trends/co2/co2_mm_gl.txt
Wind: ftp://ftp.cpc.ncep.noaa.gov/wd52dg/data/indices/zwnd200
Aerosol: http://data.giss.nasa.gov/modelforce/strataer/tau.line_2012.12.txt
Global Carbon Budget: http://cdiac.ornl.gov/GCP/
CO2ppm to CO2Gtonnes https://www.skepticalscience.com/print.php?r=45
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Beautiful. Perhaps it could be regularly updated as the datasets are updated?
Interesting, but some caution is in order. The graph maker had control over scale and offset so don’t put too much weight when you see graphs loosely correlate, your eye is ‘trying’ to find patterns, its just what it does.
For example the delta-CO2 to temperature looks loosely correlated, but perhaps not significantly so (statistically speaking).
Also warmer are can carry more moisture so perhaps it’s not surprising seeing the tight relationship between moisture and temperature, only by looking at the scale of the effect is it possible to find out if this is significant.
Great effort. I had no trouble, but a white background would help. My main hope is that John B from Toronto will update it at least every 6 months. The next two years may be fascinating.
I had some difficulty reading it until I copied the image into GIMP2(free photoediting software) and inverted the colours , giving a white background, then used the Levels in the Colour Tool to increase the readability .
No doubt the various Windows programs would do the same , I just wanted an excuse to play with the recently downloaded software.
It did appear however that the only occasions when large La Ninas follow El Ninos is when there is a near equatorial volcanic eruption coincidental with the El Nino as with El Chichon and Mt Pintaubo (17N and 15N respectively). It might suggest that the coming La Nina will be modest in global temperature amplitude.
This just leaves the question of the La Nina of 2008 . Previously when I have seen the global temperature graphs here , they have had a smoothing filter applied which gives the impression that the La Nina appeared from a flat background. However this author , as did others earlier this week , has supplied the monthly unfiltered data on his chart from which you can see that there was , to my untutored eye, a very short lived but quite strong preceeding El Nino.- too short lived perhaps to affect the smoothed result.
Hi Mike
Can you post it please, thanks
What is the point of the graph?
The text is nearly unreadable. It seems there are a large number of lines. Without some explanation, how is the viewer supposed to decipher the graph?
All we are told is:
“This indicates warming events due to El Niño …”
Well known to anyone paying attention to climate science.
“… cooling events due to La Niña …”
Well known to anyone paying attention to climate science.
” … cumulative heat growth …”
Well known to anyone not living in a cave.
So what is the point?
I’d ignore it and just not say another word if I felt like that.
Pretty presentation, but mostly unreadable on my elderly monitor.
Such is the progress of technology that I was able to look at it on my Win 8 Nokia phone – and expand it as I saw the need.
John B: I can buy into your hypothesis that the accumulated extra heat from all the El Ninos can show a net gain in time if the losses from La Ninas cannot cancel it out. It looks to me, from the chart that the next La Nina could well bring temps back to whatever is considered the norm. Which raises the question in my mind: whatever will the greenies make of a change in temp of 2 Deg C – down? Will they tell us how dangerous that is?
They did during the 70s.
skepticism of AGW requires that energy be created ex nihilo….
[No, it doesn’t, that’s just your distorted viewpoint – Anthony]
Mosh, with all due respect, you must learn to quit when you’re ahead. Snark ill becomes you.
All AGW requires is a variation in cloud cover (for one thing).
As I drive on the left, I guess I managed to miss your drive-by, Mosh. Not that it matters, it wouyldn’t have done much damage as it was typically cryptic – and worth the ‘nihilo’ – nothing.
Mosh
“skepticism of AGW requires that energy be created ex nihilo….”
“When the planet warms it will still exhibit patterns of warming and cooling”
Contradict your self much.
If it can have warming and cooling patterns when it warmers it can do so when it cools as well. Energy doesn’t have to be created either it can be stored and released as well. By the way your precious AGW forcing is essentially just a storing of energy as well not a magical forcing and it is minuscule in comparison to other natural storage phenomena.
AGW is an argument of post hoc ergo propter hoc! A logical fallacy and more Latin to boot!
Mosher, skepticism of the skepticism of AGW requires that the earth be in a stable equilibrium.
Your move.
So, Mosh, this is the next propaganda fib from the warmunistas, eh. Well good, because it leads you all into interesting territory…..
If temperature indeed follows water vapour content as the graph seems to show (look at the blue and yellow points in the graph, when the blue line goes up, a bit later the yellow line goes up), then the question remains: what is the cause of the quantity of water vapour in the atmosphere?
If CO2 should make sea temperatures rise, this initial rise of temperature is followed by extra evaporation (all other factors remaining the same) and this graph shows that in case of extra water content automatically COOLING will follow.
So the initial warming will not will be followed by an EXTRA rise of temperature, as assumed by IPCC and the model makers. But, the initial CO2 warming effect is DIMINISHED by water vapour.
And not enlarged.
A transmogrified version of John B’s graph is at the link below. This was a result of ‘messing about’ with an image editor and is more “different’ than ‘better’ but may be useful. Some notations have been expanded.
This is still fairly garish and hard to follow but some lines are easier to see and some will be harder. Depending what you want to see this one may work better.
The original graph appears to be of significant value, but ‘needs work’ in the presentation area to make it easier for people with eyes like mine [ 🙂 ] to follow it all easily.
Graph is at -> http://bit.ly/johnbensograph
Russell McMahon
Thanks 🙂
Here’s a thought…
* back in the 1970’s the “Ice-Age-Is-Coming” scare was fed in part by models showing that increasing aerosols in the atmosphere would reflect away more sunlight, causing a temperature drop
* El Chichon and Pinatubo dumped large amounts of aerosols into the atmosphere
* I remember Toronto regularly getting smog alerts during summers in the 1980’s and 1990’s
Then “stuff happened”…
* The EPA did its real job, and cut down particulate emission limits for cars and industries
* The US (and southwestern Ontario, too) “rust belt” happend as North American industrial production was outsourced to China, etc. This was not merely a transfer of pollution sources. Old more polluting factories in North America were replaced by newer factories in China. Yes, they still pollute, but not as much as the 50-year-old factories they replaced.
* With the downfall of the USSR, a lot of older, badly polluting factories were shut down. Crafty EU politicians ensured that the Kyoto Accord timing straddled that period. Thus the EU “greatly reduced it’s pollution” (both aerosols and CO2) by merely sitting on its hands while old polluting factories in the former East Germany were shut down.
* The economy cratered in 2008, and hasn’t fully recovered yet. More declines in manufacturing, industial activity, air and road travel, etc, etc. This was accompanied by a corresponding decline in aerosol emissions.
* Speaking of air travel, todays jets are much more fuel-efficient, and carry more passengers. Net result; much less emmisions per passenger-mile.
* No big equatorial volcanoes shooting off into the stratosphere recently.
Simple logic…
* more atmospheric aerosols == higher albedo == lower temperatures
* fewer atmospheric aerosols == lower albedo == higher temperatures
One vital thing I discovered from my GIS (Geographic Information Systems) classes and making maps, legends, charts, graphs, etc.
You must view your work in different media and on multiple computers/monitors.
Something can look great printed on paper, or on a computer monitor, or displayed through a projector, but it rarely looks good or is legible on all three.
I had some great looking work that ended up looking horrible when presented as a PowerPoint to a group of people using a projector.
John B,
I think that signal analysis types are best suited to rummage through the data. The don’t have any biases.
1) Temp and Precip correlate. OK Well understood.
2) [CO2} is rising with seasonal variation. OK
3) {CO2} increase anomaly seems to look like temp anomaly. Warmer temps evolve more CO2 from oceans so, OK.
4) El Nino rate > |La nina rate| so, Overall temp increase is cumulative due to El nino onset. Interesting.
Comments
1) Great work. (you must have no young children)
2) The predominance of the background red/green color imparts a cognitive bias in my mind.
3) Is there a lag between precip and temp?
4) Good on Anthony Watts for enabling this kind of contribution.
Nice work, but I think I see some potential problems: (1) Variables that autocorrelate (e.g., have similar time trends)? (2) Use of a derivative for one variable? (3) Possibly less than ideal data filters? This may be well-disguised wiggle-matching. I sure hope not. I appreciate the huge amount of work involved. The graphic legibility is not as bad as some allege here. Yes, it takes an effort to grasp it all, but it accomplishes its intended purpose.
“This indicates warming events due to El Niño and in greater magnitude than cooling events due to La Niña, hence we have a cumulative heat growth from El Niño events during the period in the graph.”
El Nino is a pattern of warming. It is not the cause of warming.
The planet does not warm on its own. Energy is not created out of nothing. The planet warms when there is an energy imbalance. The cause of the imbalance is forcing from GHGs. When the planet warms it will still exhibit patterns of warming and cooling. We name the patterns. They are the warming not the cause of the warming.
=========================================
[Gosh, this is what happens when Mr. Mosher tries to speak from authority.
NOAA’s Climate Prediction Center defines it as:
The California government website defines it as:
No mention of CO2 forcing there. It’s simply change in the wind pattern, and it’s been going on for millenia, well before CO2 forcing became an issue. it seems this paleo study of old piles of shells reveal ancient El Niño patterns: http://science.sciencemag.org/content/345/6200/1045.abstract?sid=3a410601-4f01-470b-b7ac-72d61ccfe5a0
Mosh, as they say on Mythbusters: BUSTED -Anthony]
Stephen Kosher
You’re talking through your hat! The measure of warming you are using is limited to near earth atmospheric temperatures. If you can’t imagine the quantity of heat held in atmospheric water vapour, the land masses and the oceans, then there is no point in your being here. Acting smug in the process just calls into question your motives and intellect..
Sometimes I actually like Mosh’s comments and contributions but we all have confirmation baises. On this day, he has clearly indicated his belief in the effects of GHG. That is good. We know which corner he is in and still I sometimes find gems in his comments and others of his belief. I am biased too. I studied engineering, geology, weather, history, astronomy, fluid dynamics and the water cycle along with a few other things like biology and microbiology (even one course under Suzuki no less) that made me skeptical from the start. That is my bias.
On another thread there was discussion of warming and cooling – perhaps Patrick Moore’s – that caused me to look this up on long term cycles:
http://www.ncdc.noaa.gov/paleo/abrupt/data2.html
Now I have to follow this: http://co2coalition.org/
Thanks A and WUWT. There is almost too much information here to digest, especially if you go read the links. Don’t know how some people do it. I try to keep and open mind but my guess is this issue won’t be resolved in my lifetime or even those of the youngest of readers here.
Nice graph from the perspective of there being many, many inputs to climate.
On a fine scale gifted by direct observation and unperturbed by ice core smoothing, we may be seeing the peak of the current interstadial. But what we don’t see if we could pan out is this:
http://nas-sites.org/americasclimatechoices/files/2012/10/Figure-14.png
https://nas-sites.org/americasclimatechoices/more-resources-on-climate-change/climate-change-lines-of-evidence-booklet/evidence-impacts-and-choices-figure-gallery/figure-14/
Even then, when panned out, we need to be aware of issues with the ice core data at the top:
http://digitalcommons.library.umaine.edu/cgi/viewcontent.cgi?article=1023&context=ers_facpub
At the middle:
https://www.researchgate.net/publication/268687179_A_review_of_the_brittle_ice_zone_in_polar_ice_cores
And at the bottom:
https://nsidc.org/data/docs/agdc/nsidc0202_wahlen/
Looking at the ice core data, I think it reasonable to suggest two conjectures: 1) That over the long term, increasing temperature is a prelude to decreasing temperature, and 2) The source of this long term increasing air temperature beyond the noise of weather must be coming from an entity that has hoarded it sometime in the past chilling the air, and is now releasing it warming the air, and does so an a long term pendulum scale.
My flight of fancy: That last conjecture leads me to peg the oceans as that entity, being large enough to hoard a whole lot of heat incrementally, and having the known capacity to release heat incrementally, warming the air and land near it.
I was thinking of something earlier today that is related. Isn’t any warming most likely a simple matter of having a greater level of positive ENSO region conditions versus negative ENSO regions over a 30 year cyclic pattern? That is what I see when viewing the MEI long term graph.
The positive ENSO conditions move masses of moist warmth air that are then carried to other parts of the globe. Any region which gets covered by the moving warm mass builds up heat during the day, as the blanket above limits the ability to shed the daily incoming solar energy. This is how heat builds up in the system. An example of what I am seeing would be the current extra warmth being experienced in a heat wave in the Southwest over the last week or two, and still ongoing. Although the current wind pattern is shifting slightly as compared to the last several weeks. Look at the surface wind patterns which are moving from warmer regions northward into the Southwest. Those surface winds are moist and warm, and a heat wave forms…https://earth.nullschool.net/#current/wind/surface/level/overlay=temp/orthographic=-105.39,42.10,819/loc=-122.896,40.434
It looks like the high pressure system responsible for the Arizona heat wave has moved a little west into the pacific, instead of moving east as the forecast predicted.
Now the jet stream is running along the U.S./Canadian border and bringing strong thunderstorms to the Great Lakes area. The center of the U.S. is starting to heat up and dry up.
https://earth.nullschool.net/#current/wind/isobaric/500hPa/orthographic=267.02,28.84,224
NOAA forecast that the central U.S. would be cooler than both the east and west coasts of the U.S. for the entire summer. I hope they are correct, but I don’t see it happening yet. Would love to see it. It was 105 heat index today while I was out mowing the lawn. I would prefer it a little cooler. 🙂
May and June have been very nice here in No California. Temps have held mainly average to below average. Plus there have been some decent rains. The surface winds off of the coast have been pushing south down the coastline for over a month. That leads to the night time temps dropping into the high 40s or low 50s. A hot day is easier to take when the night cools down like that…https://earth.nullschool.net/#current/wind/surface/level/orthographic=-107.15,42.03,819
Why is a chart like this inconsistent with the AGW theory? I am not an expert in this area. And I am not on one side or another — if anything I am a “skeptic”. But I am wondering what the rational response is to a hypothetical statement from an AGW advocate like this: “all this chart shows is generally rising temperatures, and generally intensifying El Ninos — in other words, a warming world, both in the air and in the oceans. Just what we’re talking about. And if the ‘deniers’ want to say that the bigger El Ninos are causing the warmer air, well, what is causing the bigger El Ninos?”
Dave,
“… I am wondering what the rational response is to a hypothetical statement from an AGW advocate like this: “all this chart shows is generally rising temperatures, and generally intensifying El Ninos …”
The chart is displaying a rather small temporal span during a generally warming period, for which there is information available to produce a chart like this. A similar chart depicting similar things during a cooling period, could render the same question in reverse, hypothetically, and there was a brief “global cooling” scare after such a spell . . right before the “global warming” scare began to heat up, so to speak ; )
This chart is displaying some details regarding the “process” of warming, and ought not be seen, I feel, as explanatory regarding “what is causing” the . . oceanic percolation if you will, to rise or fall over a relatively short span of time.
In broad terms, it’s the heat (light) from the sun of course, causing turbulence in the earths hydro/atmosphere (with some geological influences, like those two prominent volcanic events that seem (to me) to dampen the atmospheric temperature response to the concurrent El Nino . . “bubbles” . .
Good question. However, previous long term temperature reconstructions (that can be seen in the chart I displayed) do not indicate anything is amiss with the current warm period. And certainly cars cannot be blamed for previous epics.
The case for AGW is somewhat magical. Here is their case: Whatever caused the rise and fall of previous epics has been disrupted by the current mix of natural and anthropogenic CO2, which only the human-sourced CO2 is the culprit of that disruption. That seems implausible. To cause such a rise and fall over repeated epics seems to me to be something quite big. The small percentage of CO2 attributed to cars just seems, well, tiny in comparison to the graph.
+ many. Nice summary.
Neat! Thank you very much, John B.
2002 to 2009 is an interesting slice where the lack of correlation among pretty much all the the plots is fun to ponder. How many other slices (and how long would the slices be) that resemble 2002-2009 could we find with 1,000 years of data? Hard to say what even 100 years would show.
P.S. Yes, the image must be enlarged to start seeing all the goodies, but we were warned up front, so pony up and find a bigger screen, even if you have to beg, borrow, or steal. I’m going to see if I can get it on my HD TV screen. I have poor eyesight and I know I’ve missed stuff even on my decent-sized work monitor with my nose 3 inches (75mm) away.
I find myself using two sets of reading glasses for when I want to view fine detail. I suppose that I should get a nice magnifier on a head band to make life easier.
Another signal for John – ENSO appears to have a ~60 year periodicity. The same ~60 year cycle also appears in the AMO, PDO and HadCRUT 3v. I mention it because your of interest in ENSO signal processing, because in the shorter time frame of your data the ~60 year periodicity won’t show up.
The ~60 year cycle is of course well known, but despite being responsible for about 40% of temperature rise last century (because of the choice of start year and end year) it isn’t included in the IPCC ensemble models. Which is curious.
Something seems a bit off with the temperature which looks Karlised as others have noted. But I do like the way of plotting Ninos and Ninas using color bands which I think is the main point of this graph. It is also the reason for choosing a black background.
What I found interesting is that the El Ninos seem to lag temperature rises not lead them. And the nature of the temperature fall after an El Nino seems unrelated to whether or not there is a subsequent La Nina. To me the graph does not illustrate El Nino driven warming.
There is an obvious relationship between ENSO and temperature, but since ENSO is defined in terms of temperatures this is somewhat unsurprising. What it shows beyond that is unclear.
For those having trouble viewing this graph, I have posted a somewhat toned down version on my blog here: http://wp.me/p4MyHW-eo
Caution, it is a large file. I have changed nothing other than the brightness / contrast and the Hue / Saturation. It might be a little helpful for some to be able to view it a little easier.
John B, I hope you are not offended by my taking it on myself to do this. My eyes are more fuzzy than they used to be and adjusting it made it better on my monitor.
Thank you wyo, that’s much better.
That’s too short arecord to tell us anything about multidecadal, let alone millenial co-variation. For what it’s worth, cross-spectral analtyis of (UHI-uncorrupted)global temperature average and Maina Loa CO2 concentrations shows that CO2 variations either log those of emperature or are incoherent.
So much for the moshed-up notion that GHGs are the driver!
I agree that it is too short a record – anythnig shorter than about 100 years is too short, yer the alarmists seem fixated on temperatures since the 1970s. I was interested in your statement that :-
“cross-spectral analtyis of (UHI-uncorrupted) global temperature average and Ma[u]na Loa CO2 concentrations shows that CO2 variations either l[a]g those of [t]emperature or are incoherent. ”
Could you please tell me where I can find details of that ? Of course, Mauna Loa CO2 only goes back to 1959, but I find that Berkeley Earth global land & ocean temperatures (which are of course affected by UHI) and Mauna Loa CO2 from 1959-2015 are correlated (correlation coefficient = 0.924)
I did the cross-spectrum analysis for a client several years ago on a privileged basis. The results are not in the public domain.
Inasmuch as BEST’s temperature projections show a strong trend since 1959, they will exhibit strong correlation with any variable strongly trending at the same time (e.g., California real estate prices). Unlike frequency-dependent cross-spectral coherence and phase, such correlation can provide no unequivocal indication of physical relationship between the two variables.
Brilliant graph, John B from Toronto! Great work.
Another nail in the AGW coffin.
Kudos
ps_ White backgrounds are lame. Very nice graph.
I find white backgrounds with black text and black or coloured lines make the text and lines stand out more. They are also easier to print out without getting a lot of ink on the paper.