Saturday, December 5, 2009

Science with its pants down

Because the consequences of a human contribution to climate change are huge, scientific research that could shed light on this question is very important, and has risen to high visibility all over the world. Thus it has come to pass that a scandal of sorts in the world of climate science, referred to by some as “climategate”, has drawn a lot of attention.

I need not rehearse here the saga of the past couple of decades of intense debate over the question of whether humans are causing climate change. The implications for the workings of modern society are enormous. If it were the case that fossil fuel emissions are causing an increase in the temperature of the planet, and if that increase has the potential to cause disruption of society at many levels, a world-wide effort to mitigate that increase would be called for. We have had a succession of reports from the Intergovernmental Panel on Climate Change (IPCC), each presumably updating and improving upon its predecessor, that point to a likely increase in global temperature over the next 50 to 100 years. The predictions point to alarming changes in weather pattern s, and a potentially disastrous rise in sea level, along with a host of other changes that would require great adjustments in human society. So we are coming up to the Copenhagan conference, at which the nations of the world are once again under great moral pressure to respond to this threat. The costs of effective mitigation will run into the trillions of dollars.

On November 20 some files and e-mails originating in the Climate Research Unit of East Anglia University were made available publicly, by whom or by what means I don’t know. For those who have read the files and dug into some of the background, what they reveal is not a pretty story. Meagan McCardle summarized some of her conclusions and reactions in Atlantic magazine on December 1. What emerges from the discussions of this episode is that influential climate scientists at this very prominent voice of expertise on climate change appear to have behaved badly in several respects. They exercised undue power over the peer review process on papers dealing with climate change, and thus were able to stifle work that did not meet their well baked-in ideas of what is happening in the field. Here is a quote from one of the e-mails uncovered:
“I can't see either of these papers being in the next IPCC report. Kevin and I will keep them out somehow - even if we have to redefine what the peer-review literature is !”

An even more serious problem has to do with the quality of much of the historical data on which the modeling studies depend. Climate models are tested in part by their capacity to produce temperature and other trends that match the historical data. It now emerges that the original data sets are nowhere to be found! The problem is that many of the older data have been adjusted for one reason or another over the years, by persons unknown, and for reasons that were not well documented. Much of the original computer code is formatted in computer languages no longer in use, and backing out the original data is maddeningly difficult if not impossible. These data are the source of what eventually becomes an estimate of the global temperature. So we have the situation that the historical record of the planet’s temperature is now in disarray, and may not even exist!

This episode is very distressing to me as it is to all who want to see science respected as a reliable and truthful source of knowledge of the world. I’ve written in Imperfect Oracle of the ingredients that make for science’s epistemic authority. I won’t recite all that here, but clearly truthfulness and a disinterested approach to one’s work are key ingredients. However, aside from what appear to be serious lapses from professional ethical standards, something else about this case strikes me as especially interesting. The reactions to this story would have one imagine that that original historical record of the global temperature is somehow sort of gold-plated scientific data, the true story of the planet’s temperature over the past 150 years. I believe that to be false. At best, much of that record is quite unreliable and subject to uncertainties that are much larger than the variations that are being talked about.

I wrote about the concept of global temperature in a book published in 2003, Making Truth: Metaphor in Science, (see pages 163-165). As I pointed out, there is no physical thing that corresponds to the global surface temperature, or at least there has not been to this point. Measuring the earth’s temperature is not like sticking a thermometer under the tongue and getting a measurement that is satisfactorily representative of the temperature regime throughout the entire body. When the tongue thermometer registers a change of a degree or two from the normal, the fact of that temperature change is evident in the way the person feels: feverish. There is no analog to the under-the-tongue thermometer in measuring the surface temperature of the planet. Until the advent of satellite measurements which only now are becoming well enough standardized to serve as a reliable measure, climatologists relied upon a patchwork of measurements non-uniformly covering the planet’s surface, many of them individually unreliable, which was then somehow put together to give a number we call the global temperature. In truth, it is a sad fact that this idea of a reliable global temperature has been sold by climatologists with scarcely any acknowledgement of how sketchy it really is.

It is important to note that the absence of a reliable historical record of the planet’s surface temperature does not render impotent the idea that the planet is warming. Suppose you had an unreliable thermometer for measuring your body temperature. Even if the device did not give you a reliable measurement, you would know if you had a significant fever, right? In the same way, it is evident from what we see occurring in nature that the planet is growing warmer: disappearing arctic ice, glacial melting, shifts in weather patterns that clearly betoken warming in both hemispheres. The big issue, however, is whether and to what degree this warming is due to human activity. This is where climate modeling comes in. If they are sufficiently complete and self-consistent, the very large, complex computational programs that the climatologists have produced should be able to tell us how much change in surface temperature could be accounted for by greenhouse gases added by human activity. But are the models sufficiently realistic ? One way to test this is to see whether they reproduce historical climate change. Also, though I am unsure about this point, the programs may need the historical data for some aspects of the computations. But if we don’t have a reliable, sufficiently complete historical record…well, climate science has a credibility problem.

This episode has dealt a blow to science’s epistemic authority, and to its moral authority as well.

No comments:

Post a Comment