TCS Daily

The Bad News Bearers

By Ross McKitrick - July 9, 2003 12:00 AM

There was news recently that some medical researchers in Australia discovered an effective treatment for melanoma, the deadliest form of skin cancer. If true, it would be a wonderful breakthrough. The article I read was properly cautious about the preliminary results, but allowed itself some delight in what is potentially an exciting result.

Medical science is conventional in this respect. A discovery that an experimental drug is effective, or that an old threat is receding, is treated as good news. These are happy things. When new diseases appear, or other health-related threats arise, it is considered bad news. Stories about SARS, for instance, are presented as if it is an unhappy occurrence, as one would expect.

Happy things: good news. Unhappy things: bad news. A normal formula, for a normal science.

Contrast that with the strange, masochistic world of global warming, where good news is at best ignored or, more often, resented and attacked. If the models show ruin and devastation ahead, the climate science crowd cannot hide its glee. Discover that things may not be so bad, however, and there are glum faces all around. Conversation gets defensive and hostile, and the expert's scowl deepens.

Actually the pattern of lamenting, resisting and denying good news is common across environmental issues. But it is nowhere more apparent than on global warming.

A few recent examples bear this out. First, witness the reaction to "On nonstationarity and antipersistency in global temperature series," by Olavi Kärner, in the Journal of Geophysical Research (v107 D20, pp ACL 1-1 to 1-11). Didn't hear about that one, did you? A deafening silence was all that greeted this astonishing study.

Here is what Kärner did. He noted that if the CO2-based mechanism proposed by the Intergovernmental Panel on Climate Change is the predominant force affecting the climate, there will be a cumulative positive feedback on atmospheric mean temperature in response to a forced change. This translates into a time series characteristic called persistency, which in turn implies that temperature will have something called a Hurst exponent with a value exceeding 0.5 (the technical details are in the paper). On the other hand the daily solar irradiance data from space-based radiometers displays a cumulative negative feedback, with a Hurst exponent of between 0.22 and 0.35 (depending on the estimation method), implying solar irradiance is antipersistent. So Kärner got the tropospheric temperature data and tested to see which it most resembles: the IPCC-hypothesized persistency mechanism or the observed antipersistent solar irradiance. The tropospheric temperature series turns out to be strongly antipersistent, with a Hurst exponent between 0.26 and 0.36, closely matching the solar data. He concludes:

"[The] solar forcing variability is actually the governing one among other existing (random or not) forcings in the Earth climate system...The revealed antipersistence in the lower tropospheric temperature increments does not support the science of global warming developed by IPCC"


"Dominating negative feedback also shows that the period for CO2 induced climate change has not started during the last 22 years. Increasing concentration of greenhouse gases in the Earth atmosphere appeared to produce too weak forcing to dominate in the Earth climate system."

This is great news! Like any other science result, it is tentative; it needs to be replicated and discussed and so forth, but surely it is good news, at least in principle. Even after all that CO2 we put into the air, the worst-case scenario mechanism of a positive cumulative feedback turns out to be unsupported by the data. Looks like we don't need to knock ourselves on the head with costly CO2 control policies. All that money could be used for more beneficial purposes. Happy things: good news.

Ah, but in the killjoy world of climate science, such news is bad. It is so bad it doesn't merit reporting at all. Even magazines like Scientific American that make a great show of highlighting the potential perils of global warming fall grimly silent when this sort of paper gets published, even in a leading journal. Maybe they only pretend to worry about warming: maybe they really do prefer bad news.

Sometimes SciAm manages to print snippets against the global warming doctrine. They did it in the June 2003 issue when they reported on (yet) another paper linking climate change to the sun. Richard C. Willson of Columbia University and his coauthor Alexander Mordvinov published a fresh batch of data showing that solar flux has been gradually intensifying over the past 24 years, suggesting by implication that CO2 is relatively less influential than hitherto believed. The lead author even told SciAm "In 100 years I think we'll find the sun is in control." Again, good news! At least in principle, at least to normal folk. The big crisis might not be such a problem after all.

But there was no cover article for this, no feature, no photos, no gladness, no joy in Mudville. They found a wee spot on page 28 to stick the article.

I'm picking on SciAm, whereas of course the pattern is the same in many other magazines. But SciAmhas the profile that it sets a tone others follow. SciAm reported on the Oregon Petition against Kyoto back in October 2001, but rather than being encouraged by the extent of professional opinion supporting an optimistic reading of the evidence, the publication sniffed through the names until the editors found six (unnamed) signatories who apparently have since changed their minds. This was offered as reassuring evidence of the mighty consensus behind the pessimistic IPCC Summary for Policy Makers. I would have liked to see them survey the chapter authors and reviewers of the full IPCC assessment report to see how many disagreed with the Summary for Policy Makers: but they might have unearthed some bad news doing so.

And of course there was the 11-page tantrum directed towards "The Skeptical Environmentalist" Bjorn Lomborg (January 2002). Most of us think that well-documented evidence of declining pollution, increasing life expectancy and increasing income is good news, but to SciAm it is apparently terrible news, sufficiently upsetting to send the magazine into a sputtering rage.

Now we have the newest snit, this one over a paper by TCS friends Willie Soon and Sallie Baliunas, which appeared in Climate Research (January 2003). Soon and Baliunas published a review essay discussing several hundred papers published in recent years, each of which tries to discern the clues about past climates left in tree rings, ice cores, ground layers, coral reefs and so forth. They find that there are common patterns across these site-specific "proxy" reconstructions. If you park your butt on any of these spots and go back in time you will encounter loads of natural variability, but generally you will find a stretch of weather consistent with a cold climate from about 200 to 700 years back, and a stretch of weather consistent with a warm climate for a further 500 years before that. The findings are neatly presented in tabular and graphical form, extensive caveats are made, primary references are given, and conclusions are stated.

To wit: pretty much anywhere you live you wouldn't find today's climate exceptional if you could experience the range of what happened in that place over the last 1000 years or so. While a famous paper by Mann, Bradley and Hughes a few years ago ran some temperature and proxy data through a statistical black box and came up with a hockey-stick shaped climate history for the Northern Hemisphere -- which seemed to demonstrate anthropogenic warming -- Soon and Baliunas don't presume to know the mechanism by which the actual climate integrates the temperature field across vast stretches of time and space into some fictive "Global Temperature"; and in particular they don't claim that this mechanism is well-represented by a principal components regression or any other linear stationary data compression routine. Maybe it is. No one knows. As Chris Essex and I argue in our book Taken By Storm, there is no such thing as a Global Temperature so there is no way to experimentally or theoretically verify any particular data averaging rule. Soon and Baliunas wisely choose to deal with the data as it stands, on a location-by-location basis, and let the reader decide how it might or might not aggregate up into a global climatology.

Any fair-minded reader will think it a meritorious piece of research presenting potentially reassuring news (even if she disagrees with the conclusions). But SciAm treats it as if it were bad news that needs to be resisted. They find people to rail against it and they probe around for some reason to explain its existence other than the simplest one that the authors wanted to present an argument for a position they believe to be true. It must not occur to SciAm that the Soon and Baliunas result is potentially good news, on a par with preliminary positive results from a drug trial, or word that a space mission is going well. Instead SciAm fusses and grumps and denigrates the research as politically-motivated and gratuitously controversial.

The Soon and Baliunas article is about as controversial as a phone book. Then again if you are committed to the belief that the phone system does not exist, a phone book would be quite a provocation. The SciAm article doesn't point out any flaws, only limitations and methodological specifics acknowledged within the article itself. Instead the SciAm article offers a disturbing sidebar attacking the journal that published it and insinuating misconduct on the part of the editor (Chris de Freitas of the University of Auckland) who supervised the review. That part of the tale has an interesting follow-up. Prof. Otto Kinne, the Director of Inter-Research (the publisher of Climate Research) personally reviewed the file, including the four referee reports and the process leading up to the publication decision. He dismissed the misconduct accusation, finding that the article was properly reviewed and that the editor, Prof. Chris de Freitas, did "a good and correct job as editor."

So I hereby nominate Chris de Freitas to be the next editor of Scientific American. Wouldn't it be great if it had someone capable of doing a good and correct job. Now that would be good news.

TCS Daily Archives