Observations for comparison to
Deep Thunder (RAMS) results during Supercomputing'99
Results from RAMS on the previous page showed
predictions of weather for November 16 and 18, 1999 for the Northwestern
Unites States. For comparison purposes, consider satellite observations
from GOES-9 showing cloud measurements on for November 18.
A
more detailed view from the outer nest of the RAMS runs corresponding to
this period is shown below. Qualitatively, the correspondence
between the observations and the predictions is quite good. Of course,
the simulated "clouds" are multiple isosurfaces of total cloud water density,
and not the same as cloud imagery.
Now consider radar observations for approximately the same period of
time. The animation sequence is pseudo-colored by precipitation type
and density.
Since
this presentation is not calibrated in terms of actual observable units,
it is difficult to compare to the simulation results. In an attempt
to do so, the following animation shows maximum predicted reflectivity
for the outer nest. One mm contours of accumulated hourly precipitation
are overlaid to enable some comparison to the radar results. Again,
there is rough correspondence between these visualizations of radar observations
and the forecast simulation/
lloydt@watson.ibm.com
January 4, 2000
