Conditional calculations for arbitrary time windows in massive data set

[m.a.a.psu]

Hello,

I have a very large data set comprised of hourly temperature, degree days (an integrated measurement of time and temperature) and solar radiation measurements at a site in Greenland from August 1995 to the end of 2009. During the summers of each year from 1996 to 2009, biological sampling events occurred approximately 7 days apart (sometimes, 5,6,8 days apart) from June until the snows come in August or September. For each year, for each interval preceding each sampling event, I need to calculate the sum and average of degree days, average temperature, temperature variance, average solar radiation, and solar radiation variance. After stabbing at it for a while, I have no idea how to calculate those statistics for the variables in question apart from doing each stat for each year for each interval individually, which I hope not to have to do. I thought there may be a way to simply add a column for each of the stats in question and write a lengthy conditional that would spit out the appropriate values for intervals bounded by certain stipulated dates, but am now not so sure. Any ideas? I am attaching a very small sample from the datasheet so that you might see its layout. Air_temperature_200cm, TDH (thawing degree hours), and In_sw_rad (solar radiation) are the three variables of interest. Thanks for any help you might be able to provide