Why individual weather patterns should not be used to discuss climate change

Spencer:

What a difference one year makes. With the anniversary of Hurricane Katrina's landfall (August 29, 2005) rapidly approaching, who would have predicted that we would now be in the middle of a near-normal Atlantic hurricane season? Weren't the global warming pundits' predictions for this hurricane season that it would be just as bad -- maybe even worse! -- than last year?

Yet, now at mid-August, we have had only three named tropical storms, compared to nine by this date last year. Normally, we would have had one hurricane by now, and we have not had any so far, so by that measure we are actually below normal.

That's not to say that climate change won't affect hurricanes, just don't make it easy for the dark side. Of course, his readers won't necessarily remember Roy's spin just a little less than a year ago :

There is some recent research that suggests that of all Atlantic and West Pacific tropical cyclones measured since the 1970's, a warming trend in sea surface temperatures has been accompanied by stronger and longer-lived storms. In fact, the increase in the total power generated by the storms that the study computed was actually much larger than could be accounted for by theory, suggesting changes in wind shear or other processes are operating in addition to just increased temperatures. (Unpublished results by the same researcher suggests, however, that this trend was not apparent in land falling hurricanes since the 1970's).

Last year it was a trend that could not possibly have been due to global warming, this year it's not a trend at all.