Wednesday, August 15, 2012

Special: Why Long-Range Climate Models May Not Be Correct

By now, I am sure you have seen the climate models spelling an extremely warm world in the next few decades. I was thinking about this prospect and realized that these models may have forgotten a few crucial pieces to this puzzle.

DISCLAIMER: Because I did not make or run any climate models, I cannot say for sure if the factors listed below were taken into account. For informative purposes, I will assume that the following factors were not taken into account.

Changes in CO2 Emissions to Come
I have a feeling that the long range climate models were run with no change to future CO2 productions. While it is plausible that CO2 emissions will increase in the future as third-world countries gain further access to technology, the first-world countries have been, and will continue, to enhance recycling efforts and more fuel-efficient cars. Effectively, should these CO2-cutting proposals work, it may come down to the global temperature not rising as fast. If we're lucky, it may not rise at all for a while.

Sun Hibernation
As I have discussed recently, the sun is forecast to move into a period of hibernation over the next several decades. As illustrated in the chart to the right, the latest sunspot cycle maxes out at a weak 50 or so before going back down into an even weaker state. This cycle, known as Cycle 24, appears to be the last solar cycle before we dip into an extended solar hibernation, during which the Earth may very well cool. It is unknown what exact effects this will have on weather, but the idea of a general cool down appears likely.

Long Range
One must remember that these long range models are the same as everyday models, like the GFS. The only difference between the two is that the climate model is fitted for only a few specific parameters, and it goes out several decades.
That does not eliminate the error potential that regular forecasting models have. Who says that these climate models are perfect and accurate 40 years away from now? One must remember- that is 40 years. Not a few days- 40 years. If this were run on a regular model, you might as well trash that model.
What i'm saying is that just because climate models are long range does not, in any way, exclude them from a wide error margin.