20010406: stub to read .gem files?

Chris,

The email you enclosed was in regard to creating a grib
back out of a GEMPAK grid. 

You can do this with the gdgrib program in GEMPAK 5.6
(the code is in $GEMPAK/cource/contrib/hpc).

The best way to see how to extract data values otherwise would be to
look at gdlist which will access a grid, determine the points within
the projected area, and return the data values. 

Steve Chiswell
Unidata User Support



>From: address@hidden (Chris Herbster)
>Organization: UCAR/Unidata
>Keywords: 200104062248.f36MmPL02478

>(I am sending this once again as I discovered that my email client was 
>having a tough time with an imbedded email header.  Since I suspect that 
>others may also have this problem - thinking that the inserted text was 
>a new message - I think it's prudent to resend this.  Sorry for the 
>repeat if you have gotten this multiple times.)
>
>
>I had originally sent this message just to Steve, but then realized I should 
>have sent it to the generic "support" email.
>
>I've included a message I found in the support archive (between the "--->" 
>blocks).
>
>I was wondering if the stubbed code was ever generated - if so I would really 
>appreciate a copy as this is my first attempt to use the GEMPAK libraries myse
> lf
>to read a gempak grid file.
>
>Thanks for any and all help!
>
>Chris H.
>
>
>______________________________ Forward Header ________________________________
> __
>Subject: stub to read .gem files?
>Author:  Chris Herbster at BCL
>Date:    4/5/01 5:54 PM
>
>
>Steve,
>
>I found this in the gempak support archive:
>
>
>--->
>
>>From: Matt Fearon <address@hidden> 
>>Organization: Desert Research Institute 
>>Keywords: 199909202052.OAA24442
>
>>Steve,
>>
>>I am performing some calculations in gempak and then 
>>displaying in grads.
>>
>>After obtaining the eta grib data via ftp, I am converting the 
>>grib to gempak grid format. Then I am performing some 
>>calculations using gempak programs.
>>
>>After performing calculations and having a single gempak 
>>grid file, I was wondering if there is a way to re-convert 
>>the gempak grid file back to grib? Supposedly I can then
>>read the grib file into grads. Or a shorter way, if it is possilbe, 
>>may be to read the gempak grid file straight into grads w/o 
>>re-converting to grib?
>>
>>The reason I start in gempak is that I can perform certain 
>>calculations easier in gempak than in grads, but for the 
>>final product graphical display, in this case, I prefer grads. 
>>
>>If you have any ideas, please let me know. 
>>Thank you,
>>Matt Fearon
>>address@hidden
>>
>>
>>
>>
>
>Matt,
>I do not have any way of converting a Gempak file back to grib. 
>In general, you would need some information for generating the 
>grib product that is not stored in the Gempak file when it is 
>decoded.
>
>You could create a program to convert the Gempak file to the grads 
>format if you know the requirements there. Reading a gempak
>grid file is generally straight forward when using the gemlib.a 
>library routines-
>
>IN_BDTA  initiallizes the gempak interfaces 
>GD_OPNF  opens the gempak grid file
>GD_NGRD  determines the number of grids in the file
>GD_GGRD  returns a grid given its number in the file (eg from 1 to 
>number of grids)
>         returns the grid times, levels, vertical coordinate, parameter 
>name,
>         number of x,y gridpoints, the nagigation information and the 
>grid array.
>
>Once you have the gempak grid read in from the file, you can decide how 
>to
>output the grid into the grads file.
>
>If you need help creating the stub to read the gempak grid, I can 
>create that for you. Let me know.
>
>Steve Chiswell
>Unidata User support
>
>
>
>GD_GIDN  obtains the grid information (level, coordinate, parameter)
>
>--->
>
>I'm trying to write some code that will read data from either the RUC or 
>ETA grids to extract profiles that we're going to use to generate 
>synthetic satellite images.  (Pretty cool stuff.)
>
>I'm most interested in the last comment you make about creating the stub 
>to do the read.  Did you ever do this for Matt Fearon?  I'd *really* 
>appreciate a head start if you either have the code or could whip it 
>together ....
>
>Thanks for any help you can provide.
>
>Cheers,
>
>Chris H.
>
>