Virtual science....

We live in an era of virtual science....where wishful thinking is more important than hard data...

The book is Science And Public Policy (Edward Elgar Publishing), and the author is Professor Aynsley Kellow, the head of the school of government at the University of Tasmania. Kellow believes that environmental science has often been corrupted by the good intentions of its practitioners, so that it consists of wishful thinking rather than facts and provable theories. Perhaps the first big case of this was the notorious Limits To Growth study published by the Club of Rome in 1972, based on computer modelling and subsequently disproved. One might expect the quality of models to improve, but since then they have been used for all sorts of predictions, and there is little evidence they have got much better.

Despite this, the predictions made by such models are now contained in scientific papers published in leading journals, which gives the status of science to what is often little more than wishful thinking. Kellow describes one paper published in the journal Nature in January 2004 that "warned of the loss of thousands of species with a relatively small warming over the next century. But just how virtual was this science is apparent when we consider that the estimates of species loss depended upon a mathematical model linking species and area; modelled changes in the … distributions of areas of habitat depended in turn upon the results of climate models tuned to reflect climate changes as a result of increasing greenhouse gases … these in turn were driven by scenarios of what [such] emissions might look like over the next century, driven in turn by economic models."

Kellow notes that a similar warming over the previous century had not left anything like the trail of species devastation being proposed in the paper, yet this observational data was considered irrelevant compared with the virtual world of the models.

The widespread concern over climate change is based substantially on calculations similar to the one just described. Is this a problem? Kellow thinks it is, because virtual science is ripe for manipulation, usually unconsciously, by virtuous scientists. Few people are aware of the large element of subjectivity, not only in the design of immensely complicated general circulation models, but in the data that goes into them. Even basic information such as contemporary temperatures is often incomplete or uncertain and tweaked by those who operate the models.

An interesting article on this appeared on the BBC website last month. The author is Dr John Christy, a professor of atmospheric science at the University of Alabama and a climate expert. He runs one of the two dozen or so general circulation models in the world. He says the only way to test models is to compare their predictions with outcomes "not known ahead of time", and when he has done that he has found "gross inconsistencies - hence I am sceptical of our ability to claim cause and effect about both past and future climate states".