Updated 12:30 am Friday 25 June 2004
Added K-F2 PGI-MPI 36/12 graphics showing more precip than K-F2 using OpenMP.

36-km

Assessment: The really big differences in these plots is the convection near the Great Salt Lake.

12-km

Assessment: Here, we see the Utah differences and some differences around our area of interest. The Grell locations seem better though precip amounts in Washington are less. Conclusion 2: Not having a cumulus parameterization on for the 12-km domain kills the Washington convection. Conclusion 3: Note the differences at hour 21 of the KF2 run with OpenMP and run with PGI-MPI!

4-km

Assessment: We actually get convection in Washington with Grell!

To compare with satellites

Radar

24-hour observed precip in hundredths of an inch valid at 12Z on the 18th