One of the issues raised by critics of global warming is scientists’ reliance on computer models of the Earth’s atmosphere. These constructions of mathematical algorithms and observational data are, say their detractors, unreliable, as they are based on an incomplete understanding of the atmospheric system, and include a number of questionable assumptions. The more disingenuous critics complain that meteorologists often get their weather forecasts wrong, so how can we trust long-term predictions of climate change?
Such comments conflate two completely different things: short-term and often chaotic weather fluctuations, and long-term changes which can be uncovered with the aid of statistical analysis. The critics also fail to understand the scientific method, and the integral part that models of reality play in developing and validating scientific theories.
Scientific theories, which provide our best understanding of the observed phenomena to which they relate, start life as hypotheses. These tentative explanations, which have not yet been fully tested, may be the result of individual creativity or collective brainstorming sessions, or simply hunches, born of gut feelings and intuition.
Transforming a hypothesis into a fully-fledged scientific theory requires comparing the core idea with hard data, and explaining any positive correlation found by means of a physical mechanism. It is not enough to simply state that X correlates with Y. Explaining the physics, chemistry or biology involves creating an abstract representation of reality in the form of a model which will necessarily be restricted in scope. That restriction allows us to tinker with this artificial world, and thus determine the relative importance of the variables contained within it.
For an explanation as to why climate models are a good thing, you would do well to read a short article posted in February of this year by Andrew Dessler on the environmental blog Grist. Dessler’s research into atmospheric water vapour and its influence on global warming is occasionally cited by climate sceptics in support of their prejudices, but the Texas A&M University scientist is no anthropogenic climate change denier. As well as the nuts and bolts of atmospheric physics, Dessler knows a thing or two about the misuse of science in the policy arena.
In his Grist article, Dessler touches on the problems involved in climate model verification. As he explains, we are not in a position to compare model outputs with 100+ years of observational data. Nor can we justify initialising the models with the atmosphere’s present state, in an attempt to correctly simulate 10-year timescales. What researchers do instead is validate individual processes in the models, and look at how these play with each other. This is scientific best practice, and all that can reasonably be achieved in the circumstances.
“If you can validate enough processes of the model (water vapor, clouds, ice, oceans, etc.) then you generate confidence that your model is probably making predictions that are at least in the ballpark,” says Dessler. “In addition, it gives you a good feeling of where the weak points in the model all are. For example, this type of analysis has demonstrated that most of the uncertainty in the model predictions arises due to clouds.”
Critics of climate models are demanding the impossible, and well they know it. It is therefore crucial that non-scientists understand what is involved in constructing and validating these virtual worlds, and their role in making predictions about the future state of our environment.