The global warming alarmists would have us believe that mathematical climate models can accurately predict global temperatures in the year 2100. The several versions of GCMs (Global Climate Models) are predicting average temperature increases from about 2 to about 5 degrees Centigrade. Based on these predictions they want governments around the world to take drastic actions to decrease the production of greenhouse gases such as carbon dioxide. These drastic actions could have significant negative effects on the world economies. All this because they think the globe might warm by as little as 2 degrees?
Notice that their models disagree by more than the smallest predicted temperature increase. If the models they used show a spread from 2 to 5, what is the chance that additional models would widen that spread? Perhaps one might even predict a temperature decrease. Based on my own experience with mathematical models I have little confidence in any model’s ability to accurately predict 93 years into the future.
During most of my career I developed and used mathematical models to estimate weapon system performance for the Air Force. The models I worked with did not operate on a global scale but they were quite complex and took hours to run on high-speed computers. Nevertheless I, and my colleagues, never pretended that our models could accurately predict future outcomes. That is, although the models estimated future outcomes, we knew that the outcomes were based completely on the system and environmental data that is input to the model, and we knew that data was never perfect. We understood that there is no way that we could know either the precise environmental conditions at the time the weapon might be used in actual combat or the precise system parameters for the specific weapon fired in actual combat. The models were primarily used to perform system design trade-off studies and to produce probability of success charts to guide the pilots in employing the weapons.
No model of a physical process can reliably predict an outcome very far into the realm of the unknown. Models are constructed based on what we know today about how a physical process works. Models are run using data that we have collected to date — or data that someone makes up for the future. So models essentially only model the past or the present. When their users move beyond the present they are just guessing. The only way the GCMs can be used to predict 93 years into the future is to make up 93 years worth of input data, including the levels of carbon dioxide emitted into the atmosphere. The models then are subject to not only the errors in the physical model itself but also the errors in the input data.
If you are still not convinced that the glowarm alarmists can’t know, with any real confidence, that average global temperatures are going to be 2 to 5 degrees warmer in 2100, consider the five-day weather forecast. Although it’s called a five-day forecast it often changes every day — and I don’t mean by just adding a new fifth day. Today’s fourth day forecast is often different from yesterday’s fifth day forecast despite the fact that they represent the same day. This is because — with the passing of one day — the forecasters have new and better data that allows them to make a more accurate forecast. It will be interesting to see how much the predicted average global temperature for the year 2100 has changed next year and five years from now.
Understand that I’m not arguing that climatologists should not use the GCMs in their global climate research. I’m confident that they understand the limitations of their models better than I do. I’m trying to make an argument against putting too much stock in the crusades of devious egomaniacs like Al Gore, who are prone to misinterpret the model results.