Massive extinction of logic

• Article
• Comments ()
• Click-2-Listen
• Videos

By

Much has been made of a paper published on Jan. 8 in the journal Nature by Chris Thomas and 18 co-authors, claiming global warming will cause a massive extinction of the Earth's biota. Mr. Thomas told The Washington Post: "We're talking about 1.25 million species. It's a massive number."

It turns out that there is a massive number of glaring problems with their study that clearly eluded the peer review process. This is evinced by the rapid turnaround for the manuscript, with acceptance in final form a mere five weeks after original submission. No one can clear revisions through 19 authors in that time unless there weren't many revisions suggested, or, if there were, they were ignored by the journal's editors in a rush to publication.

In fact, acrimonious debates about what should or should not be published about global warming are the rule rather than the exception, simply because papers are being published -- on many sides of the issue -- that can be shredded after only a cursory review. Unfortunately, the debate may have started with Nature itself.

In 1996, conveniently a day before the U.N. conference that gave birth to the Kyoto Protocol, Nature published a paper purporting to match observed temperature with computer models of disastrous warming. It used weather balloon data from 1963 through 1987. The actual record, however, extended (then) from 1958 through 1995, and, when all the data were used, the troubling numbers disappeared. Since that famous incident, people have been very leery of what major scientific journals publish on global warming. The Thomas extinction paper only throws more fuel on an already roaring inferno.

The work of Mr. Thomas et al. is an interesting exercise in computer modeling showing again that what comes out of a computer is a product of the assumptions that go in. The scientists examined the distribution of more than 1,000 plants and animal species, calculated their current climatic range, and then used a climate model to determine whether the amount of land the species could occupy in the future would shrink or expand. If there was a likely shrinkage, the researchers expected an increased chance of extinction.

Fair enough. But this assumes climate change is the sole driver of changes in biodiversity, which is hardly true. Consider the effects on an ecosystem of the mutation of some previously harmless bacterium, a clearly nonclimatic cause of extinction. The plethora of factors that influence ecosystems, besides climate, determine the composition of the community. In fact, placing all the onus for extinction on climate also calls the entire dramatic result into question.

Their lowest scenario for warming is bounded at 0.8 degree Celsius in the next 50 years, and produces an extinction of roughly 20 percent of the sampled species.

There's a convenient reality check available. That's because surface temperatures indeed have risen this amount in the last 100 years. But there is absolutely no evidence for massive climate-related extinctions. (One would think the reviewers at Nature would have picked that up.)

There are several other major problems:

• Global climate models, in general, predict a warmer surface and an increased rate of rainfall. As long as there is adequate moisture, the most diverse ecosystems on Earth are in the warmest regions, the tropical rainforest being the prime example. Consequently, the general character of future climate is one which is more, not less hospitable for biodiversity.

* Temperatures have been bouncing up and down a lot more than 0.8 degrees Celsius during the past several hundred thousand years. The published methodology implies there are large extinctions for each and every increment of equivalent change, whether the temperature goes up or down. Applying this method to all those changes would make extinct just about every species on Earth.

• Species often thrive well outside their gross climatic "envelope." The U.S. Agriculture Department has mapped the distribution of all major tree species in North America. For almost every species, there are separate "disjunct" populations far away from the main climatic distribution. A fine example is the Balsam fir, Abies balsamea, whose main distribution is across Canada. But there is a tiny fir forest, a relic of the Ice Age, still standing in eastern Iowa, hundreds of miles south (and about 10 degrees warmer) than the climatic "envelope" people normally assume to circumscribe its distribution. These disjuncts are the rule, not the exception, in nature, and are one reason why the most diverse ecosystem on Earth -- the tropical rainforest -- managed to survive the Ice Age.

* Perhaps most egregious, this work makes what the famed agronomist Paul Waggoner has called the "dumb people" assumption: that people won't adapt to changing conditions. In fact, we have been preserving diversity artificially, in parks and zoos, for centuries. In addition, the amount of "artificial" genetic diversity is rising dramatically with the technology of modern genetics. It is difficult to imagine, decades from now, that these technologies would not be applied to ameliorate a prospective massive extinction.

Obviously, there is a lot to criticize in this paper. What is surprising is that something with so many inconsistencies and unrealistic assumptions made it unscathed through the review process in such a prestigious journal as Nature. The politicization of scientific papers on global warming and the tendency of science journals to rush to judgment have to end.

Patrick J. Michaels is senior fellow in environmental studies at the Cato Institute.