Wednesday, July 25, 2012 ... Français/Deutsch/Español/Česky/Japanese/Related posts from blogosphere

CO2 may lag temperature just by 400 years or so

Phys.ORG is among the most science-oriented outlets that inform about the new paper

Tightened constraints on the time-lag between Antarctic temperature and CO2 during the last deglaciation (full text PDF)
by Pedro, Rasmussen, and Ommen published in Climate of the Past. The content of the paper is simple.




They looked at the Antarctic data between 17,000 BC and 9,000 BC (to help delayed readers of this blog in the year 3000 AD, I express the timing relatively to Jesus Christ instead of our humble present). The CO2 concentration increases or decreases after the temperature does the same thing but the newly determined lag is just about 400 years instead of the figure 800 years we loved to repeat.

It's still safely positive so that we know which change is the cause (temperature change, caused mostly by the Milankovitch cycles) and which change is the consequence (changes of CO2 due to the temperature-dependent ability of ocean water to store trace gases).

However, it's also much shorter than we used to say. It means that the excess CO2 we are adding into the atmosphere disappears much earlier than those 800 years we used to promote. Based on the currently observed CO2 budget, it seems obvious to me that the excess CO2 in the atmosphere would be reduced e-times within 50 years or so, much shorter even than their estimate of 400 years. But the motion of CO2 could have been slower when (and where) the Earth was cooler and less hospitable for life.

(Again, it's easy to calculate the 50-year figure. The amount of CO2 we are emitting would increase the CO2 concentration by 4 ppm a year. The actual observed increase is about 1.9 ppm in average. So it's clear that Nature – oceans and the biosphere – are absorbing 2.1 ppm out of the excess CO2 every year. This absorption only depends on the elevated concentration because that's the only thing that the oceans and plants "feel". So it would continue even if we stopped all CO2 emissions. Consequently, CO2 would start to drop by 2.1 ppm a year although the decrease would be gradually slowing down. The current excess CO2, 393-280 ppm = 113 ppm or so, would be largely undone in 50 years or so. The CO2 concentration would quickly return to the level compatible with the equilibrium given the current interglacial temperatures – and it makes no difference whether you add the tiny man-made contribution to the interglacial temperatures or not. The modest warming arguably caused by CO2 would be undone in those 50 years, too.)

Of course, no outlets actually explain that this finding makes the hypothetical "CO2 problem" twice (or more) less long-lasting a problem. The environmentalist journalists only inform about bad news and true or hypothetical yet untrue effects of the industrial activity. In fact, if you look at the titles of articles that inform about the paper by Pedro et al., all of them effectively want the reader to believe – or continue to believe – that the research suggests that CO2 causes changes in the temperature even though the truth is the other way around.

There are many journalists who spin everything to help criminals such as Al Gore to remain at large.



Prague as Joseph of Canada remembers it

A new Greenland panic

That's also the case of the newest hysteria about the Greenland. It's been warm in the Greenland in recent weeks so a thin surface layer across the 97% of the are of the island melted a bit. The same thing occurred a century ago and possibly much more frequently than that. But we're almost told about an apocalypse by many journalists.

Needless to say, there's nothing remarkable about such a percentage. First of all, it's just a thin layer, so much much more is needed to transform the Greenland into a green land again. Also, the percentage of the Greenland experiencing surface melting at a given July day (throughout the history) is (50 ± 20)% so the excursion towards 97% is a 2+ sigma effect that is guaranteed to occasionally happen by the laws of statistics. When forced to say the truth, even those folks admit that the same thing happened in 1889 when the industrial activity was negligible and given the expected frequency, this event is right on time.



This is my copy of the temperature map. It will not evolve with time. Click to zoom in.

The degree of cherry-picking and dishonesty in this story-writing about the Greenland has breathtaking proportions in the eyes of those who are following the current temperature anomalies across the world (see TRF for more current maps). More precisely, the link in the previous sentence shows the temperature anomaly averaged over the next 8 days. It's a prediction but the meteorological models at this timescale and this spatial resolution work almost flawlessly.

You see that the bulk of the Greenland is yellow, kind of +4 °C warmer than the long-term average for this date. You find a similar, even warmer, region in central Russian Asia. However, what the journalists don't say is that Eastern Canada, Australia, and Chukotka are among the regions that are symmetrically –4 °C cooler than the normal temperatures right now. The spatial variability of the temperature anomaly is normal, too; 250 years ago, you would get qualitatively similar maps.

And the focus on the Greenland is even more remarkable if you look at the Antarctica – which is 6 times larger a territory than the Greenland. The prevailing green color shows that the freezing continent is actually about –7 °C cooler than the normal and if you look carefully, especially at the violet colors near 70°E, you will realize that there are regions of the Antarctica that currently experience temperatures more than –20 °C cooler than the normal temperatures for this date! (Please don't double count the signs. I wanted the adjectives to be "natural" and the figures to show the sign of the anomaly, too.)

This is five times larger an anomaly than the anomaly in the Greenland and the continent experiencing this freezing cold is six times larger than the Greenland but... the temperature anomaly has the politically incorrect sign so no environmentalist journalistic crook writes about it. Regardless of their gender, the journalists who keep on promoting the climate alarm are mother-sexual-intercourse-havers.

Richard Lindzen and climate models

Phys.ORG was balanced enough to at least offer a summary of MIT's Richard Lindzen's talk,
Predictions by climate models are flawed, says invited speaker at Sandia
Via Marc Morano.

Richard presented quite some arguments why the predictions of the models are untrustworthy, wrong, and why it's preposterous to trust them given the available data. I can't resist one comment about the first remark by a passionate climate panic mujahideen called Voleure under the article:
"He showed a graph that demonstrated that more energy consumption leads to higher literacy rate, lower infant mortality and a lower number of children per woman."

Lindzen should be more careful when throwing stones at the entire scientific community to keep his sponsors cards closer to his chest and not reach beyond the data for his ideology.
Richard doesn't really depend on some inflows of money from the "sponsors" who appreciate his work but an untold secret is that the MIT actually does. Richard is bringing a disproportionately high amount of funding to the institute as important sponsors began to realize how unhinged and politically motivated the rest of the scholars in similar disciplines have become – and the climate alarmists are among the top examples of this deterioration. Those sponsors are often told by the MIT officials "But we also have Richard" and MIT keeps on going.

So the right conclusion is that unless MIT wants to decline, it should more carefully protect its financially key scholars such as Richard Lindzen against garbage of the Voleure type.

PSU students watching Gore's movie in class

Xanthippa (via Tom Nelson) has found a syllabus of Michael Mann's course at Penn State University. Together with Tom, I must say "Wow". Just "Wow".

The first detail that will make you breathless is the name of the course: "Gaia – the Earth System". It sounds exactly like a parody – someone trying to humiliate Mann and similar folks – but the syllabus is real. It's enough to see that the crook can do literally anything at PSU. But that's just the beginning.

Two classes taught by the notorious researcher who just hired a Big Tobacco lawyer to protect himself against some true statements in the media (you can't make it up) were dedicated to watching of Al Gore's movie, An Inconvenient Truth. It's hyenous to brainwash schoolkids at the basic school by this movie but at least, the intellectual level of this movie shot by a politician who doesn't have the slightest idea about science is appropriate for basic schools. But a college? Moreover, these are not generic college kids; they're potential future climate scientists. Are climate scientists really learning from Al Gore these days? Michael Mann has also forced the students to buy his and Kump's idiotic book about "dire predictions".

The intellectual atmosphere at that college must be absolutely shocking if such things are considered tolerable. And this Gore-level insanity must be added to PSU's being a hotbed of child abuse and even loop quantum gravity.

Add to del.icio.us Digg this Add to reddit

snail feedback (9) :


reader Gene Day said...

The Penn State situation is shocking but recall that Paul Erlich has been on the Stanford faculty for decades without seriously damaging that institution. No doubt Erlich has dented Stanford's reputation a bit but our universities have always been tolerant of fear mongers who have tenure. I guess that's the price for academic freedom.


The Sandusky affair and coverup has likely done far more damage to Penn State because of lost revenues from football and alumni (and other) donations.


reader Luboš Motl said...

True, Gene, I don't even think of Stanford when you say "Ehrlich". Stanford may be a bigger name that may afford such things.¨


Even the fanciest apartment house in central Pilsen, Palace Ehrlich


http://www.palacehrlich.cz/video/



which is under construction hasn't been hurt by the name so far. ;-) PSU may still be a smaller name than Stanford, having fewer super positive famous players in it, so a Mann or two may make difference.


reader AJ said...

Assume a cyclical forcing and a constant exponential decay response. As the lag becomes a lesser fraction of the period the value of tau will approach the value of the lag itself.


Here's the formula for tau that I use:


tau = period*tan(2pi*lag/period)/2pi


Here's the calculated value for tau given a 100,000 yr cycle and the stated lag:


lag tau


050 050.00
200 200.01
400 400.08


That said, I find tau=50 reasonable given the emissions and accumulated growth rates you mention. The paper's PDF's don't exclude this value and there may very well be a bias in the measurements. Then again, maybe constant exponential decay isn't the proper model?


Also, assuming a linearly increasing growth rate in emissions and a constant exponential decay of the excess, then the excess will at first grow quadratically and then over time transition to linear growth.


reader Robert Sykes said...

Pedro et al go out of their way to imply that CO2 drives temperature in the records they examine. Is that the only way they could get published?


reader Frederick Bloggs said...

Penn State sounds like a prison and the intellectual level must be similar.

Interesting to note that there is no session on introductory statistics in Mann's course. Indeed the word statistics is not even mentioned.


reader Frederick Bloggs said...

Mann problem sets
Q1: You are asked for your data. How do you respond? Do you respond ?
Q2: A sceptic submits what looks like a ground-breaking paper showing significant negative feedback effects and you are asked to be a reviewer. How do you respond?
Q3: Despite your best efforts the paper in Q2 is published. How do you respond ? This is a group question.
Q4: Regress this time series against proxy measurements. Make sure you get a hockey stick.
Q5: Explain why R-squared is a rubbish measure of statistical regression and create your own measure with some complicated name.
Q6:Contrary to expectations the results of Q3 show a decline in temperatures post 1960. What do you do to correct what has clearly been a failure of the model to detect the truth [tip:look closely at my paper in Nature] ?
Q7 ...
ad nauseam ...


reader AJ said...

The Pedro paper makes me wonder if their "temperature" proxy is actually a proxy for energy in the climate system. The reason I wonder this is because of the Calder/Roe/Milankovitch observation that the rate of change in ice volume has a zero lagged anti-correlation with the solar insolation forcing. I reasoned that the rate of change was a proxy for the the specific air temperature. Others believe that ice volume is the proper proxy. This doesn't make sense to me as the rate of change would be zero when temperatures were both at their coldest and warmest. I view ice volume as a proxy for energy in the climate system and that it is a cumulative function of temperature. Given a sinusoidal cycle and a one-box temperature model, energy would lag temperature by 1/4 the period.


The Pedro paper states "Water stable isotope ratios (d18Oice and dDice) from the ice are proxies for temperature above the inversion layer at the time of snow formation (Jouzel et al., 1997), while CO2 is preserved in air bubbles in the ice.". My question is, not necessarily directed at you, do water stable isotope ratios correlate with ice volume or the rate of change thereof?


Eyeballing the Pedro paper, there appears to be two cycles in their ~10,000yr sample, giving a cyclical period of ~5000yrs. Could it be that co2 is lagging temperature by 1250+400 yrs? On a 100,000yr cycle, could it be that co2 lags temperature by 25,000+400yrs :) I must be missing something.


reader JimNtexas said...

I'm confused by the dates on the chart, 25 July 2012 - 02 Aug 2012. Is this a forcaste, observed data, or some combination thereof?


reader Luboš Motl said...

Dear Jim, I was confused, too, but I got de-confused and answered it. It is a forecast but it is a damn accurate forecast because at this level of regional resolution and accuracy and 8-day averaging, the forecasting models simply work almost flawlessly. Free feel to check it on August 2nd.