Last week Kyoto came into effect. Apparently that was the signal for columns by a whole bunch of pundits who have two features in common: 1. they are manifestly ill-equipped to understand the science and 2. they are utterly certain that there is no such thing as global warming.
Our first pundit is Michael Duffy in the Daily Telegraph informs us:
The truth is we have no control over global warming, and in any case it's not a problem at all.
The myth holds that carbon dioxide in the upper atmosphere is increasing, due mainly to industrial activities, and this traps heat lower down, with the result that temperatures on the earth's surface rise.
The first problem with this is that the extra carbon dioxide we create is so minuscule in comparison with the amount of greenhouse gases in the atmosphere it's highly unlikely it could create this effect. Variations in the amount of heat the sun generates are a far more likely cause.
We have so far increased carbon dioxide by 30%, which is not minuscule. While there are other greenhouse gasses, carbon dioxide is an important one. Furthermore, the greenhouse effect is very large, keeping us 30°C warmer than we would be without an atmosphere, so even a relatively small increase in its strength produces significant warming. Variations in the sun do not explain the warming we have seen. The latest study on this found:
Along with his Scripps colleague, David Pierce, Barnett used a combination of computer models and hard, observed evidence to reach their conclusions. They determined that warming measured in the world's oceans closely matched the results predicted in computer models for warming caused by human activity.
When the models assessed whether the ocean warming could be caused by volcanic or solar activity, Barnett told reporters, the answer was stark: "Not a chance."
The second problem is that temperatures have not risen along with industrialisation over the past 200 years.
The graph on the right shows that temperatures have actually risen. You really have to work hard to remain as ignorant of this as Duffy is.
Our second pundit is Andrew Bolt (last seen arguing that a cold day in Melbourne was good evidence against global warming) claims
The truth is that despite the hype, not much about global warming is known for sure, not even how much the Earth has heated, and whether our carbon dioxide (CO2) caused it. So say even lead authors of the United Nation's Intergovernmental Panel on Climate Change, whose doctored "predictions" are most used to frighten us.
Doctored predictions? Odd, you would think that the hundreds of scientists whose work went into the IPCC would have noticed if their work had been doctored.
One of them, Professor John Christy of the University of Alabama in Huntsville, asks: "Will increases in CO2 affect the climate significantly? Are significant changes occurring now? Climate models suggest the answer is yes. Real data suggests otherwise."
Now this sounds like he is saying that the real data shows no warming and only the climate models show warming, but if you look at the context of his statement you'll find the real data shows warming at the surface (see graph above) and that Christy's calculations from satellite measurements also show warming, but not as much. Christy says this contradicts the climate models, but other researchers' calculations show more warming from the satellite data. And even if the climate models were wrong, it would not follow that we would expect no warming from increase CO2. All you would be able to say is that we don't know what the effects would be. No scientist has an explanation for the observed warming that does not involve CO2.
Adds another, Professor Richard Lindzen of the Massachusetts Institute of Technology: "The temperature is always changing for the earth, so it has only two choices---going up or going down. It has done both, and that doesn't say it's due to CO2; it doesn't say it's going to continue; it doesn't say anything beyond that."
In the preceding sentence Lindzen said:
For the last hundred years, I think there is a general agreement that there is something like a half-degree increase in temperature.
That's strange, Bolt claimed that Lindzen said that it was unknown "how much the Earth has heated" when in fact, Lindzen explicitly stated how much it had warmed.
And while Lindzen might express doubts about whether CO2 is causing the observed warming, the hundreds of other IPCC lead authors disagree with him. Not only that, Lindzen was one of the authors of the National Academy of Sciences report that concluded:
Greenhouse gases are accumulating in Earth's atmosphere as a result of human activities, causing surface air temperatures and subsurface ocean temperatures to rise. Temperatures are, in fact, rising. The changes observed over the last several decades are likely mostly due to human activities, but we cannot rule out that some significant part of these changes is also a reflection of natural variability. Human-induced warming and associated sea level rises are expected to continue through the 21st century.
Next, we have Melanie Philips, who is sure that global warming is a scam because (quoting McIntyre and McKitrick):
[Mann et al's method], when tested on persistent red noise, nearly always produces a hockey stick shaped first principal component (PC1) and overstates the first eigenvalue.
According to her biography Philips is a journalist with a degree in English. Back when I was an undergraduate learning about stuff like eigenvalues and mathematical physics, my friends studying English didn't learn about eigenvalues. Maybe it was different for Philips, or maybe she's done some post grad course in advanced statistical analysis, so I emailed her, asking her if she knew what red noise, principal components, or eigenvalues were. No reply. My guess is that she doesn't know what any of them are. (Oh, and M&M's "always produces a hockey stick" argument is a red herring.)
And to think the Hadley Centre was initially created to fabricate the illusion of global warming during Baroness Thatcher's premiership to diminish the power of the coal-miners union in the UK.
Curse you, Maggie Thatcher!
Heroically, I'll attempt to explain what an eigenvalue is to the liberal arts crowd (Politics, Philosophy & Economics, Oxford, 1991-4).
Basically, start off by thinking of a load of points in a three dimensional space. If you have a co-ordinate system, then you can refer to these points by their co-ordinates (x, y, z). Clearly, in this co-ordinate system, the axes going north-south, east-west and up-down are "special" vectors; they define the co-ordinate system and have all manner of special roles in describing the data.
Now, we generalise this model a little bit; assume that each of the points in the three-space has a numerical value associated with it. Given the "special" vectors (north-south, east-west and up-down), you can calculate all sorts of weighted sums; you can calculate the "centre of gravity" with respect to this co-ordinate system, how "big" the system is in a particular direction and so on.
Now, generalise the problem. Assume that you're now given a load of data of not necessarily known dimensionality. You have the job of deciding on an appropriate co-ordinate system for it.
It turns out that there is a mathematical procedure which will tell you what the right co-ordinate system is for your data, for some fairly sensible desiderata of a co-ordinate system. The "special" vectors of this co-ordinate system are called eigenvectors. Every eigenvector has an associated value which (roughly and heuristically) measures the "size" of the system in that direction. This is the eigenvalue.
When you do principal component analysis, you are basically carrying out this process, in a context which allows for the fact that your data are observed with error or are random variables. You're basically trying to describe a matrix (ie, several different data series, each series being a vector because it's a time series) with a set of "special" vectors (combinations of the different series) that forms the best available description. You often find that the vector with the biggest associated eigenvalue (the "heaviest") vector in that eigen-system represents a combination of the series which has an intuitively plausible interpretation (for example, MBH argue that a certain linear combination of tree ring sizes, ice core measurements, etc, is an important eigenvector in their dataset because all these things are being driven by temperature).
To be honest, I'm afraid it can't be made any simpler than that (as you can see, I've already massively distorted things with all that handwaving about eigenvalues measuring "size" or weight).
There's at least another error in Melanie Phillip's piece:
"Now the Wall Street Journal draws attention to the fact that one of these claims, which was absolutely crucial in informing the Kyoto protocol, has turned out to be not worth the paper it was written on. Michael Mann's 'hockey stick' curve showed continuous temperatures for about 700 years and then a sharp upward rise in the past 100 years."
The paper containing the 'hockey stick' was published in 1998. So how could it have "absolutely crucial in informing the Kyoto protocol", which was signed in 1997?
Hero you may be, dsquared, but my godlike ignorance will crush you. Whatcha saying about "deciding on an appropriate co-ordinate system for it"? Do you mean choosing appropriate XYZ factors that give a shape to the cloud of data points that makes sense in some way for the data?
The Eiger Sanction was a good read though.
I don't understand exactly. What's the linear transformation function going from and to? What the hell is red noise, and why is it relevent here? (Post script - red noise is "oceanic ambient noise (ie, noise distant from the sources) is often described as 'red' due to the selective absorption of higher frequencies." WTF?)
My linear algebra is 8 years out of date and the class I did singularly worst in, but I remember it, so don't skimp on the hard words and symbols, ifwhen you have the time.
Do you mean choosing appropriate XYZ factors that give a shape to the cloud of data points that makes sense in some way for the data?
Thank you for showing that some sanity still remains on this planet. Unfortunately the Mad Hatters run the asylum. The more I read about the state of this planet, the more depressed I get.
We have found out that western antarctica isn't as stable as we thought and could fall into the ocean raising sea levels 16 feet. Coral reefs around the world will soon be dead because of warming and acidification of the ocean (more than 25% of all marine life live on these reefs). The tundra is melting and huge quantities of frozen methane will be released. Forests are burning releasing more CO2. Peat bogs around the world are breaking down releasing even more green house gases. Glaciers are disappearing around the world. And this is just global warming. Nobody is talking about the plunder of life from the oceans, the disappearance of amphibians, the loss of top soil around the world. Major rivers no longer flow to the oceans. Rainforests are quickly disappearing. Every living thing on this planet now carries a toxic brew of chemicals courtesy of the chemical industry.
This could go on and on and our children are going to be stuck with this mess if they even survive. At some point in time people will realize our mistake but it will be too late.
I could just SCREAM!!!!!!
Sorry for the rant.
This is the sort of thing that makes me wary of the IPCC.
On another note... eigen, according to my German dictionary, means peculiar, in case anyone cares. We use eigenvalues and eigenvectors in a different context in engineering, usually to tell us about the dynamic modes of a linear system, and about system stability where it turns out the eigenvalues play a critical and interesting role. In graph theory, which is used to model networks, the eigenvalues of the graph laplacian (a certain matrix that describes the graph) tell you things about network conectedness and properties of information flow. Mathworld, as always, has a good explanation of eigenvalues for those who'd like to see the math.
Just pointing out that the usefullness or meaning of eigenvalues depends on the context, and they can be associated with data, linear systems of equations and other things. There's even a thing called an eigenfunction.
You've hit the nail on the proverbial head. Excellent post. My work as a population ecologist frequently leads me into discussions on these topics, and on the possible, if not likely outcomes. There are a number of major hurdles to our understanding of these issues: (1) scientists are not doing a good enough job of describing the seriousness of the current predicament; (2) the media, in an attempt to strike 'balance', gives equal time and space to opposing views which represent a small fraction (at most) of the scientific community; (3) powerful, vested interests that represent the establishment are investing huge sums of money to mislead the public and policymakers in an attempt to maintain the status quo; (4) people are desperate to hear good news and want to believe that consumption and economic growth will increase linearly far into the future without destroying (or even compromising at all) our ecological life-support systems; (5) many people are totally and utterly ignorant of the importance of natural systems in sustaining us and think, in part thanks to technology, that humans are somehow exempt from the laws of nature.
There is plenty of empirical evidence available - some of it mentioned in your post - that humans and the natural world are on a collision course. Moreover, important fundamental changes are desperately needed to avoid the consequences of this collision. But the elites of this world, who largely control the media and have a disproportionate influence on government, are investing heavily in a culture of denial.
On the bright side, the conservative Weekly Standard had a surprisingly sympathetic take on the Kyoto Treaty going into effect. Let's hope that's a sign that the intellectually honest on both sides are catching on.
Well, at least the US doesn't have a monopoly on idiots.
Lions and Tigers and Bears--and Chicken Littles! Tim I would say you have the best of the Clueless brigades posting on your site. Where is the "objective proof?"
Well to be honest John F, you are not really al that strong on the "objective proof" yourself. In response to a question you raised I posted two links to papers addressing the issue you questioned and you responded with a link to Junk Science.
Yelling, you posted a link which I could not read--only the abstract. In that abstract the term statistically significant was used. If the terms were more like a clear signal way above the noise I would take notice. As I stated earlier you did not respond to the lack of causality of the CO2 and temps from ice core data. The ppm level is not too far from today and if you take and scale the data and plot it, you find the CO2 responds to temperature and not the other way around. Actually you find 90% of the transistions in temp "LEAD" the delta CO2 response. There lies the real cause and effect. Now it appears the 1400's were as warm as today? Why? since we don;t know that answer to that we can't have chicken littles running around like in the 1970's claiming global cooling and the Earth is doomed!,Now later global warming and the Earth is doomed!
John F. - Yelling is right. While it is good of you to be concerned about the quality of data regarding CO2 and ice cores, your posts above are little more than cheap shots and I see nothing in them that is either correct, or properly documented.
1) "....you did not respond to the lack of causality of the CO2 and temps from ice core data."
What lack of causality? You've offered no documentation for this statement. Nor have you addressed any of the factors influencing how temperatures, CO2 levels, and trapped gases in ice cores are either correlated or uncorrelated. So far, you're just hand waving on this point.
2) "The ppm level is not too far from today and if you take and scale the data and plot it, you find the CO2 responds to temperature and not the other way around."
Your argument here seems a little puzzling to me, but if I understand it right, every word is incorrect. CO2 levels have changed quite significantly in the last century and in fact, are among the better documented datasets we have. This is well covered in the Chapter 3 of the IPCC Year 2001 Report. Data from both the Vostok and Taylor Dome ice cores (Siegenthaler et al., 1988; Neftel et al., 1994; Barnola et al., 1995; Etheridge et al., 1996) are shown alongside more modern independent determinations (Keeling and Whorf, 2000; Battle et al., 2000) here. Compared with the surface temperature record, it is evident that CO2 levels have been rising significantly since the early 19th century and lead temperature response, not the other way around. The most telling of these responses are the increases of roughly the last 40 years or so. Early 20th century temperature trends (your "90 percent" perhaps?) were driven by solar activity - which has waned ever since, and contrarian rants aside, has little to do with temperature trends today. The delay in the response is due to phase lags that are the result of the thermal capacity of the world's oceans - a fact well known to any undergraduate physics or engineering student, but that appears to be lost on most global warming skeptics and their front groups. More about CO2 and ice cores can be found at RealClimate here. More still on CO2 levels and ice cores can be found in Caillon et al., 2003. (and just in case I need to head this off at the pass, if you take issue with the IPCC Report cited above, spare us any rants about what politically motivated liberal &#$@*!'s they are or any other such thing - if you please, either refute their science properly or don't mention them).
3) "Now it appears the 1400's were as warm as today?
Nonsense. The Medieval Warm Period was warm compared to periods on either side of it, and approached (but did not exceed) levels of 20th century warming prior to 1990. They did not surpass anything since then. We are in fact, currently in the greatest period of warming since the dark ages. Recent attempts to prove otherwise (McIntyre and McKitrick, 2003; Soon et al., ) have been soundly refuted on a number of grounds, not the least of which was conflation of temperature and moisture proxies. For more on this, see this article from RealClimate and its citations.
4) "....we can't have chicken littles running around like in the 1970's claiming global cooling and the Earth is doomed!
Not only is this a junior high school cheap shot, it is one of the oldest and most stale pseudoscientific myths of the last 20 years. These claims originated (or at least spread most significantly) during the early 90's when Far-Right front groups discovered the Internet, but never bothered to actually read anything they cited. The source goes back to some papers from the mid 70's - most notably, Hays et al. (1976), and an an NAS Report from 1975. These addressed then current uncertainties about aerosol forcings and Milankovich Cycles - the latter of which operate mainly on millenial time scales that have nothing to do with anthropogenic greenhouse warming during the last, or the next century. More about this can be found here.
With all due respect John F., I appreciate your concerns. But all the cheap shots and CAPITAL LETTER WRITING on this earth are not going to make up for carelessness with facts. I suggest reading the actual science rather than extremist web sites and the like. If you have done so, I have yet to see evidence of it - please provide science rather than hysteria. Best.
According to her biography Philips is a journalist with a degree in English. Back when I was an undergraduate learning about stuff like eigenvalues and mathematical physics, my friends studying English didn't learn about eigenvalues.
I'm not sure it's right to assume that someone doesn't know about something because their degree is in another field. There is such a thing as an educated layperson. That being said, I don't know if Melanie Philips really knows what she's talking about; just that you can't use her english degree against her, any more than suggesting that if someone has a physics degree that means they can't know anything about literature.
Scott, get the Vostok data yourself and plot CO2 and Temp on the same plot with the peak-to-peak values scaled and tell me that the Temp is following CO2 or if it;'s the other way around. Don;t rely on your references you can read what someone else says.
I assume that the temperature data you used were corrected for deuterium-excess as per Cuffey.
However I think that the excellent paper that Scott links to makes an interesting case for CO2 rises taking place before de-glaciation in the Northern Hemisphere.
I am an engineer, but when it comes to choosing between thousands of years of history and 10 year old computer models which are still evolving, I am inclined to believe the history. Michael Mann accepts the reality of the Mediaeval Warm Period and Little Ice Age. The argument is about a few tenths of degrees, which in the absence of thermometers must be just speculation. Settlements and tree lines in some cases at higher altitudes and in others at higher latitudes which physical evidence indicates were successful over long periods are more convincing than proxy records in comparing past climate with this period.
David, settlements and tree lines ARE proxy records.
Saaaaay, John F:
I think something got deleted out of your post when you copied it over from Wurd. Surely, judging from the posting attitude you adopt of disdainful high intellect, you haven't fallen for the dupe-the-rube trick of pointing to a graph and stating "a-HA!!!!! Chicken Littles!!!!!"
So, having established your obvious high intelligence, do you "interpret" the "graph" as per Caillon et al. Pahnke et al. or Fischer et al. when you apprehend the data in this way? Or maybe Shackleton?Certainly the text that didn't transfer over from your paste from Wurd discusses, say, Fischer et al. and how they interpret recent anthro. CO2 increases. I'd be interested in hearing a smart guy discuss the recent recent anthro. CO2 increases WRT the interpretations of Vostok.