[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20020612: gprcm




On Fri, 3 May 2002, Robert Mullenax wrote:

> Chiz,
>
> I just placed our modified gprcm (as gprcm.tar.Z) in the incoming directory
> in the gbuddy account.  It was done by one of our IT programmers
> that has been here for almost 20 years.  I included a FILTER_README.txt
> which describes how to specify the filtering.  You can do whatever
> you think best with it.. I use the centroid thinning a lot..but
> gave up on the tops and mesos because of the spurious data
> sent out in the RCMs.  I grid the NET data and then use gdplot
> to put on tops.
>
> Hope this contributes in some small way back to Unidata.
>
> Robert Mullenax
>


Robert,

I have packed up a tarfile of a new gprcm version that employs
filtering in the gbuddy ftp account:

~gbuddy/nawips-5.6/patches/gprcm_5.6.g.tar.Z


I looked at the FILTER in your contrib file, as well as the various
output files etc. I basically developed the filter based on the
sfmap idea of determining the bounding areas of text, markers and
barbs, and then using them to control overlap. This is different than
the maxtop filter you have which filters based on the top value.
The filter is based on the device normal coordinates, so zooming in
will maintain the degree of overlap allowed will gaining more data.

The filter can be specified for each of the centroid, tvs, meso, and
maxtop parameters using the same variables as before, just adding the ;filter
value (instead of creating a new variable). Filter can be Y, N, or
a real number as in other programs.

The main rewrite of the program is to use templates correctly.
I created a RCMDAT template in datatype.tbl like:
RCMDAT       $GEMDATA/../rcm           YYYYMMDDHH_rcm.dat        CAT_NIL
SCAT_NIL   72   4320     60

(excuse the line wrap)

Then the rcm invocation can look like:
 MAP       Map color/dash/width/filter flag  8
 GAREA     Graphics area                     us
 PROJ      Map projection/angles/margins|dr  STR/90.0;-105.0;0.0
 SATFIL    Satellite image filename(s)
 RADFIL    Radar image filename(s)
 LATLON    Line color/dash/width/freq/inc/l  0
 PANEL     Panel loc/color/dash/width/regn   0
 TITLE     Title color/line/title            1/-2/Radar Summary
 TEXT      Size/fnt/wdth/brdr/N-rot/just/hw  1/22/1/hw
 CLEAR     Clear screen flag                 YES
 DEVICE    Device|name|x size;y size|color   xw
 RCMFIL    RCM data file                     rcmdat
 RADTIM    Radar composite current/dattim    current
 RADDUR    Radar time window (minutes prior  90
 DITHER    dither level                      0
 MESO      meso color/marker/size/width/hw;  4/1/1/2/hw;.7
 TVS       tvs color/marker/size/width/hw;f  4/2/1/2/hw;n
 CNTR      Centroid barb color/size/width;f  1/.7/1;.6
 MAXTOP    Maxtop color;filter               6;1.3
 RADINFO   Radar status color                5
 ECHO      Plot radar echoes (0,1,2)         0
 CLRBAR    Color/ornt/anch/x;y/ln;wd/freq|t  5
 LUTFIL    Enhancement lookup table filenam
 IMBAR     Color/ornt/anch/x;y/ln;wd/freq|t
 OUTPUT    Output device/filename            t
 Parameters requested: MAP,GAREA,PROJ,SATFIL,RADFIL,LATLON,PANEL,TITLE,TEXT,
 CLEAR,DEVICE,RCMFIL,RADTIM,RADDUR,DITHER,MESO,TVS,CNTR,MAXTOP,RADINFO,ECHO,
 CLRBAR,LUTFIL,IMBAR,OUTPUT.
 GEMPAK-GPRCM>r


This saves the need to have a shell cat a bunch of files.
You can either use a template, or a list of files. At any rate,
its much faster than the old way of grabbing several hours in a file.


I also added the OUTPUT parameter which allows you to store the
information output by the program to a file, or supress it all
together (output=N). I saw that you had specific files being created-
but I didn't want to go that route to recreate the OUTPUT usage in GEMPAK.

Anyhow, you probably want to keep your version, but you might want to
try out the new version in a separate location- if at least for
the template usage.

Chiz