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

20030915: Couple of questions

>From: Gerry Creager N5JXS <address@hidden>
>Organization: Texas A&M University -- AATLT
>Keywords: 200309151437.h8FEbQLd000090

>Steve,
>
>Simple one first: Is there Gempak support for Level II radar data, 
>either now, or in the works?

Yes, see the 5.6.K release notes:
http://my.unidata.ucar.edu/content/software/gempak/GEMPAK5.6/whats_new.html#5.6K


>
>Now a little harder...
>
>I'm trying to use nex2img to generate a Texas-area mozaic, and a US 
>mosaic.  I've found that, after a couple of days, I seem unable to do 
>this.  I've got scoour.conf set to remove anything over 3 days, so I 
>guess the question is, am I already retaining too much?

I'd need to see more information from the program output, as to what
times are being found by the program. The system library for dirent 
is used to read the contents to the directory to find the required
data file.

Are you sure that your data is being written to disk within the RADDUR
windiw duration. If your pqact processing of FILE'ing the data to disk falls 
behind,
then the data file won't be found until its out of the specified window range.
That would happen if your system is experiencing high IO wait
(typically, you would see "Delay" messaged of increasing values
in the debug info from pqact (issue a kill -USR2 twice to the
pqact process to cycle into debug mode.





>
>It appears that nex2img looks at every directory in the NIDS area for 
>N0R data that falls within its area.  If this is correct, is there a way 
>I can speed up the Texas-area generation, and still maintain the US 
>composite capability?

Yes,
Create a station table from the default nexrad.tbl file of just your Texas 
region
stations and specify this in the STNFIL parameter. That way, you can eliminate 
the
mapping of radar's not in your region. Note that the calculation for 
image coordinates is done only once, so it will speed up on subsequent
iterations (eg the update frequency RADFRQ for continuous operation).
However, the disk IO on finding the file will still occur, so a paired down 
station
table would be advantageous.

Steve Chiswell


>
>Thanks,
>Gerry
>-- 
>Gerry Creager -- address@hidden
>Texas Mesonet -- AATLT, Texas A&M University   
>Cell: 979.229.5301 Office: 979.458.4020 FAX: 979.847.8578
>Page: 979.228.0173
>Office: 903A Eller Bldg, TAMU, College Station, TX 77843
>