Localize your weather stations

In looking for the monthly data for particular states, when I noticed that 8 out of 20 of you were from Michigan last week, I went to look for Michigan on the NDCD Geoportal. There was only one weather station for the GSN (monthly) data (several for the Global summary of the Day (GSOD)), but in looking at annual data over a number of years, you do not want to deal with all those daily values for 50 or 100 years. That weather station was Marquette, way up on the Upper Peninsula. Then I noticed there was usually only one weather station per state for the GSN data. It was often an atypical climate - more extreme in some way. So you can do Marquette for MI or pick maybe Indiana or Illinois's GSN data if you think it is closer to your own. Just a tip on dealing with that data. A very simple use of the chapter is to look at your annual average tempratures over as many years as you can get and add the trendline in excel - including it's equation so you can figure out the slope of the line. This allows you to computer the average rise in temperature over that 50 or 100 years. You could compare that to Key West or Barrow, for which you already have data in the chapter. So that is one suggestion.

Here is Marquette MI along with the data description file for 1900 to 2004. Create annual averages as described in the EET chapter. Graph year vs average temp and fit a trendline with equation in Excel.

Data description file - the column definitions and data units