NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
Pedro, I have something similar to what you're asking for set up using gdcntr to graphically display the difference between a model's initial (i.e. f000) run and a previous run's forecast verifying at the same time (i.e. the f012 forecast of the model run 12 hours prior). For mine, I basically put the call to gdcntr and set all the variables within a shell script, and in the gdfile I used the "+" operator to specify that the data was coming from two separate model runs, for example: GDFILE = $direc/na_${newdir}.gem+$GEMDATA/model/gfs_save/${newdir}f000_gfs.gem This line is taken directly from my script, the $newdir variable is actually the name of a directory in YYYYMMDDHH format which contains the data for that specific model run, so as you can see just changing that variable allows you to compare any model runs you want. Then for the gdpfun line, I have something like this: GFUNC = sub[mag[vecr(urel^F${fcast},vrel^F$fcast)],mag[vecr(urel^F000@10+2,vrel^F000@10+2)]] Which in this case is just subtracting the wind speed from the F0000 run of the current model from the $fcast run of the older model, so just change $fcast to whatever forecast time you want that matches the time of the F000 run. So I guess in a nutshell what I'm suggesting is to use the "+" operator in the GDFILE variable to specify two different model runs, and then use variables within a shell script to denote the various model runs and forecast times if you want to automatically produce output comparing several runs or forecast times. If you'd like a copy of my entire script let me know, I'll be happy to send it along to you. Best regards, Jon Oxtoby
-----Original Message----- From: gembud-bounces@xxxxxxxxxxxxxxxx [mailto:gembud-bounces@xxxxxxxxxxxxxxxx] On Behalf Of WAMM Sent: Monday, September 10, 2007 4:17 PM To: chiz@xxxxxxxx; gembud@xxxxxxxxxxxxxxxx Subject: [gembud] Recent model run vs. previous one All: I'm trying to write a script that makes a comparison of the most current model run, say 24h forecast, versus the previous cycle's (say it is 6h earlier) 30h forecast, effectively valid at the same time. I'd like to just be able to see changes in the model solution from previous runs. I have something like: gdpfun=3D sub(hght^1200f012, hght^0600f018) which actually works, but I'd like to be able to do something like this with specifying the times, but a more general script in which it compares the last run versus the previous one, something in the form of: gdpfun=3D sub(hght^last, hght^'(last-1)') I'd appreciate any suggestions as soon as possible. Thanks, Pedro
gembud
archives: