I've been having a little fun with climate data. Mr Motl has been posting some basic analyses of data released as HadCRUT3. I've done something similar, but focused on the United States. Here is a graph of mean temperature by month and year:
There really seems to be very little trend there to me. Maybe 1/2 of a degree in the past century or so. Nothing to get hot under the collar about?
What is interesting to me is that the summer months seem much less volatile year to year than the winter months. Why would that be? I don't know.
I've been coding in Matlab, which is my wont. I have used the code I posted a little over a year ago. I think it works very well.
Note that the data set I am using is one upon which the IPCC is based, and was produced by the Hadley / Met Office. They are the "climate-gate" folks, as I recall.
Ann says: if I had to guess about the greater variability in winter, I would guess it had something to do with the pacific oscillations and el nino events. Warming and cooling cycles in the oceans might change winter weather more than summer.