Unidata 2006 Workshop - Lackmann Lab
Unidata 2006 Workshop - Lackmann Lab
Tuesday 11 July 2006
This web page houses graphics
needed to answer questions for the laboratory portion of this
presentation. If there is not enough time during the allocated
session, feel free to come back and visit this site at a later
time. Plots include animations from various experiments as well
as links to initial and boundary condition
files used to run the experiments.
Initial Environment and Initial Disturbance
The 10-km and 20-km scale initial disturbances appear to be larger than that,
due to the relatively coarse grid that was used to generate the
initial condition fields. The acutal model grid spacing for all
simulations was 20 km.
Initial Adjustment
WRF TC Experiments:
For each of the 5 main simulations, I've included a plot of
initial conditions and 2 animations: one of "model simulated
radar" and SLP, another of 10-m wind speed and SLP. Note that
because we are running a convective parameterization scheme in
these simulations, the model "radar" does not reflect all of the
model precipitation. However, a fairly large fraction of the
precipitation was of the grid-scale variety in all of the simulations.
For the wind speed plots (values in knots), the shading corresponds to
tropical-storm force winds (light blue), hurricane-force winds (first green)
and then to the different Saffir-Simpson categories.
10 km disturbance, 10C bubble
20 km disturbance, 10C bubble
100 km disturbance, 10C bubble
200 km disturbance, 10C bubble
1000 km disturbance, 10C bubble
Fun IDV Images!
Did you know that IDV can plot WRF output files directly? No need to run
any conversion programs for that! This is an easy and interesting way
to have a quick look at the model output. Just run IDV and point it
to the wrfout files. Try an isosurface of rain water mixing ratio,
or cloud water, etc. See some samples below.
The first image shows an isosurface of rainwater mixing ratio
superimposed on a plot of sea level pressure. The second and
third plots show this in addition to an isosurface of potential
temperature (an isentrope). The downward displacement of the
isentrope at the storm center is indicative of an excellent
warm-core structure in the model storm.
In the rainwater field, the extension of rain to higher altitudes
in the eyewall region is partly due to a higher freezing level
there associated with a warm core system, but it is also due to
strong upward motions, which loft liquid rain above the freezing level.
Summary Plot of Minimum SLP
One of my graduate students, Kevin Hill, gave me a nice utility program to
compute basic statistics for the storms in these idealized simulations.
Below is a graphic showing the evolution of the minimum central SLP for
each of the experimental runs.
WRF Model Files
In order to compile WRF you need a good FORTRAN compiler such as that
available from the Portland Group. The model runs shown here were generated
using WRF V2.1.2, which can be obtained from the first link below.
Once you have compiled WRF-ARW, you
simply need to copy the 3 files linked below into the WRFV2/test/em_real
directory, and type wrf.exe. These files are the namelist.input (which
specifies output frequency, run duration, etc.), wrfinput_d01 (the initial
condition file for the 200-km initial disturbance), and wrfbdy_d01 (the
lateral boundary condition file.
If the model runs properly, "wrfout###" files
will begin to appear in that directory. The easiest way to view the
output (perhaps) is to simply point the latest version of Unidata's IDV
to the wrfout files (but be sure to specify that they are NetCDF grid
files before adding the data source). IDV can also be downloaded from
the corresponding link below. Enjoy!
Note: I had to rename the wrfout file to replace the :s with -s in the
file name in order for it to behave properly on windows machines.
