GRAPH Tutorial : Run
Edit g_plots.tbl
to select appropriate times, levels, and fields to plot.
Different MM5 system output files require g_plots.tbl be edited
slightly differently. For example:
- g_plots.tbl for Terrain
output
- g_plots.tbl for pressure-level
data (output from REGRID, RAWINS, LITTLE_R and INTERPB)
- g_plots.tbl for plotting
cross-sections and skew-T with pressure-level data
- g_plots.tbl for input
file to MM5
- g_plots.tbl for MM5 model
output files
- g_plots.tbl for plotting
sigma-level data on pressure and isentropic levels
- g_plots.tbl for plotting
cross-section and skew-T with sigma-level data
For complete reference to all plot variables, please see section
12.8 - 12.10 in Chapter
12 of the Tutorial Notes.
Create graphical output
Type the following command to run the Graph program:
graph.csh n m X_DOMAINx
where "n" in the command is the number of
plot files to split (rarely used, but may be used to split
plots into several pieces based on time periods), "m"
is the number of input files to be expected, and X_DOMAINx
is MM5 modeling system output file (e.g. MMOUT_DOMAIN1).
You can have more than one input file for Graph job, but
they must come from the same domain. So for a typical
Graph job where you don't want to split the plot file, and
you only have one input file, the command is:
graph.csh 1 1 X_DOMAINx
If you have two files as input to Graph (for example, one
from a MM5 run, and the other from the restart), you may type:
graph.csh 1 2 mm5-file1 mm5-file2
If you name the input files as mm5-fileA, mm5-fileB, mm5-fileC
etc., you can simply type the command using the 'root' name,
and the shell script knows to look for all relevant files:
graph.csh 1 3 mm5-file
A graphical output file called 'gmeta', is create, which
is an NCAR Graphics formatted plot file. To view the meta file
generated by GRAPH, one must use the NCAR Graphics utility 'idt'
(more information can be found on the man page: man idt),
idt gmeta&
You may also find another gmeta file in the directory: gmeta.split1.
This is the same as gmeta file. If you used n>1, you'd find gmeta.split1,
gmeta.split2, ..., gmeta.splitn in the directory. Each
of these split-ed files would contain a subset of the entire plot
file.
Sample plots from MM5 model output for SOC case:
|
