The KMA completed the
new version (WRFV2.2) for DFI on CrayX1E.
l
There are some
changes for optimization, but itÕs not a big deal.
l
On CrayX1 system,
numtiles should set to 4.
l
IÕm rerunning the
KMA cycle test. Experiments are COL, CYC, DFL, DFI, and TDFI.
l
The design of
execution is back.exe & ddfi.exe (DFI filtering=>WRFV2.2), wrf.exe (forecast=>WRFV2.1.2).
Almost done except for
a few figures:
l
Calculating the
RMSE to compare to the increment between analysis and initialization.
l
Calculating the
upper air verification scores.
l Continue to retest for the cycle, drawing the rest figures.
l Test some experiments for extended domain:
n 24hr forecast=>75hr forecast
n
10km
domain=>the same domain with 30km.