August 17, 2004

California Nightmare

It’s increasingly difficult to keep a straight face while reading any global warming paper in a major scientific journal. Even by this standard, a recent article on deaths in California and destruction of its wine industry (of course, because of dreaded global warming) is a true belly-slapper.

The fact that it appeared in the Proceedings of the National Academy of Sciences (PNAS) is a lot less funny. What on earth is happening to the peer-review process in science, and how are papers this bad getting through that process?

How bad is it? This paper uses a computer model that can’t predict U.S. temperatures (much less California’s) and relies on research on heat-related deaths that is 15 years out-of-date.

Here is the cookbook for modern climate “science”:

1) Run a “General Circulation Model” (GCM). These are large computer simulations designed to calculate global temperature changes as atmospheric chemistry is altered by human activity.

2) “Downscale” that model to a region or a state. In this case, a fine-scale grid laid across California.

3) Input the output of that model to some regional process, like viticulture or urban death.

4) Submit the result to a major journal for sure publication and a cursory peer review.

Later on, we’ll discuss why (4) is inevitable.

But let’s start at the top. One of the two main models used in this paper, from the British Meteorological Office, “resolves” the earth’s climate at 2.5 degrees latitude by 3.75 degrees longitude, or about 36,000 square miles, at California’s latitude.

A similar version of this model was used by the Clinton Administration to estimate warming in the lower 48 states, in its 2000 “National Assessment” of global warming. It didn’t work. It performed worse than a table of random numbers when applied to decadal temperature changes that occurred as the greenhouse effect has increased.

This criticism was first expressed by a reviewer of that report. The feds were so concerned that they replicated the reviewer’s experiment and found it to indeed be true. (The Administration went forward with the “Assessment” anyway.)

If you can’t simulate the climate of the United States as a whole, you obviously can’t do California. But, nonetheless, the PNAS paper drops the initial 36,000 square-mile resolution (which didn’t work over the United States) down to a remarkable 56 square miles, which is guaranteed not to work!

It then applies this result to, among other things, California viticulture and heat-related death rates in California cities.

A model that cannot correctly simulate surface temperature must be similarly incapable of properly forecasting precipitation. That’s because the relationship of surface temperature to what occurs aloft determines whether or not it rains.

Nonetheless, the PNAS report notes that its GCM predicts reduced rain, which will be bad for California wine.

Is that so? In fact, the warmer and drier things are, in general, the better the wine. Can anyone remember the abysmal vintage of 1998, coming off the wet El Nino year? Does anyone realize that futures on the 2003 Bordeaux are through the roof, thanks to the excessive heat and dryness of that European summer?

Then there’s the problem of urban heat-related deaths. One of the two citations, published in 1989, contains Robert E. Davis as an author. Davis and his colleagues looked at mortality data and concluded that there was a “threshold” temperature at which mortality begins to increase. Ever since, people who should know better have simply extrapolated those results using ever increasing urban temperatures and the same threshold for mortality.

They are using what is known colloquially as the “dumb people scenario.” It assumes that, as our cities warm (which is inevitable: Cities always warm compared with the surrounding countryside), the inhabitants will simply elect to slowly fry and die rather than adapt to changing conditions.

The same Robert Davis (along with Patrick J. Michaels, chief editor of World Climate Report) hypothesized that this was a little demeaning of human nature and creativity. We examined whether, over time, the temperature “threshold” for mortality becomes higher and higher; in other words, whether people adapt to warming conditions.

Indeed they do. We found that in many of our urban cores, heat-related death, statistically speaking, has been virtually engineered away—by air-conditioning, better emergency care, and people taking sensible precautions.

This work was recently awarded the climate science “Paper of the Year” by the Association of American Geographers. But it is nowhere to be found in the PNAS paper.

Which leads us to wonder many whys.

Why could a paper be published in such a prestigious journal that is based upon a model that did not work? Why weren’t the authors cognizant of the Paper of the Year, which just happened to be on urban warming and death statistics? How could the reviewers miss this?

Patrick Michaels has a new book coming out on this issue, called Meltdown: The Predictable Distortion of Global Warming by Scientists, Politicians, and the Media. It describes dozens of papers in the scientific literature that are as flawed as this one.

Why do they continue to appear? The reason is obvious. The scientific community is supported by gloom-and-doom, which gets us money from our single sugar daddy, the Federal Government. No one ever leveraged billions out of or Nation’s Capital (the current annual outlay for “global change” research is 4 billion) unless they threaten the worst. Then the political process takes credit for saving us from certain destruction and gets itself re-elected.

And don’t expect scientific peer-review to stop this process. Those who sit in judgment of science are those who receive the same largesse. Who would rationally derail this gravy train? That is why papers using models that don’t work, or that ignore critical citations that might spoil an otherwise perfectly apocalyptic hypotheses, are the rule rather than the exception. Which is what just happened in the Proceedings of the National Academy of Sciences.

Davis, R.E., Knappenberger, P.C., Novicoff, W.M., and P.J. Michaels, 2003. Decadal Changes in Summer Mortality in U.S. Cities, International Journal of Biometeorology, 47, 166-175.

Hayhoe, K., et al., 2004. Emissions pathways, climate change, and impacts on California. Proceedings of the National Academy of Sciences, 101, 12422-12427.

Michaels, P.J., 2004. Meltdown: The Predictable Distortion of Global Warming by Scientists, Politicians, and the Media. Available on September 27, 2004, The Cato Institute.

No Comments

No comments yet.

RSS feed for comments on this post. TrackBack URI

Sorry, the comment form is closed at this time.

Powered by WordPress