Re: How accurate are 10-day weather forecasts?
I actually took a course last semester about chaos and this was one of the subjects we talked about.
Weather forecasts past 3 or 4 days are pretty much useless. The reason for this, if you care, is that our measuring instruments aren't sensitive enough. We always have to drop decimal points. The reason this is a problem is that weather forecasts are calculated using compounding forumlas; that is, the use the results of the first calculation in the second, and so on. Therefore, the miniscule cutoffs in decimal points become enormous factors past 3 or 4 days. Also, even if we were to somehow develop perfect measuring devices, they would need to cover the entire atmosphere.
Bottom line is, weather forecasting can only be done with a useful degree of accuracy within 3 days, and it will never get much better than that.
|