Joei:

I have processed the 2002 sy2u timegcm run (hourly for the entire
year). The files are in 

http://download.hao.ucar.edu/pub/foster/mendillo/timegcm1.2_sy2u

This run used constant gpi (i.e., not data), and ncep lower
boundaries. 

Next week I will do the sy2t case, which is constant gpi, and
no ncep boundaries, i.e., base case "climatological" run.
(the sy2y case, which I sent earlier, used both gpi and ncep data)

--Ben