Climate Odds Look a Little Long

Although there is much debate about what is and is not science, there does seem to be agreement on the principle that any scientific method of inquiry must begin with a testable scientific hypothesis.

For example, the cake maker's recipe on the back of the box is his hypothesis;

"If you combine the listed cake variables in the proportions indicated and under the prescribed conditions, then the result will look like the picture on the front of the box."

Former Vice President Al Gore and his followers declare that just such a recipe which forecasts the climate future of the planet, complete with all the climate variables in their correct proportions, now exists and that unless man modifies the recipe, and soon, the world as we know it will no longer exist.

In the view of many, if not most, of these believers, the argument is over. The scientific questions have been asked, answered and verified. Not perfectly, perhaps, but close enough to know that unless meaningful action is undertaken now, disaster on a worldwide scale is inevitable.

One significant difference between the chef and the climate scientist is that the chef has cabinets to store his variables.

Scientific American says that there are "at least several hundred" climate variables, and certainly many of them can't be stored or even collected.

As a consequence, climate scientists rely heavily on mathematical models, and here is the real inconvenient truth - any science that depends upon a mathematical model to reach a result is beholden to the laws of mathematics.

This means, of course, that unless a global climate forecaster is 100 percent correct in his choice of variables and equations, his resulting forecast will be less than perfect. The only question is to what degree, and for that a (very) little review of probability theory may be in order.

On average, if I flip a coin, the coin will land "heads" one out of every two flips (1/2), and if I roll one die, I'll get a three one out of every six rolls (1/6).

But on average how many attempts will it take for me to achieve both a head on the flip and a three on the roll? Probability theory says that if you multiply the two "fractions of probability," you will have your answer. In this example, if you multiply 1/2 and 1/6, you get 1/12. This means that on average it will take 12 attempts to achieve the desired result.

Let's assume that the really excellent climate scientists have determined that every possible global climate affecting variable is now known and that these scientists also know with a 99 percent degree of accuracy how all of these variables interact and affect each other.

And since Freedictionary.com says several means more than two or three, let's assume that Scientific American's Estimate of "at least several hundred" climate variables means 400 total. According to probability theory, what then is the probability that these experts will be correct in their predictions?

Just use the same drill as the coin and die example above. Take the fraction 99/100 and multiply it times itself 400 times. The result is 0.0179505533, which converts to 1.8 percent.

This means that if we lived in a world in which virtually perfect humans could devise a virtually perfect climate model, there would be less than one chance in 50 that the forecast would be correct.

With the odds of a correct diagnosis at greater than 50 to 1 and the projected cure estimated in the hundreds of trillions of dollars, I believe a little skepticism is in order.