Re: Converting GMT files to the netCDF format

• To: Roy Mendelssohn <rmendels@xxxxxxxxxxxxx>
• Subject: Re: Converting GMT files to the netCDF format
• From: "Mr. Chad W. Saxon" <csaxon@xxxxxxxxxxxxxxxxx>
• Date: Thu, 8 May 2003 14:43:08 -0500 (CDT)

Roy,


 i converted a GMT grid (small.grd for example) to a .cdf file as 
explained in the GMT documentation using a script which uses the grdinfo 
and grd2xyz commands from the gmt library. i now have small.grd, 
small.cdl, and small.cdf


  i was looking through the code and ncdump does work.. actually when i 
do a :

  ncdump -h small.cdf i got the results
                                   small
                                    {

                                    }

  ..there is not header numbers for the rows, columns, latitude/longitude.

  i tried an ncdump on a netCDF file that i knew was in the right format 
and i get something like this..

                               file {
                                     rows: 300
                                     columns: 100
                                     etc etc
                                    }

                                     all the (x,y,z) coordinates


i have looked deeper into the two files to see whats going on.
typing the command at the prompt: od -c small.cdl | more, i look at the 
first line of the header and i see a \r \n (carriage return & a line 
feed) ..??? 

i do the same for a .cdf file that i know that works and there is only a 
\n (line feed) on the first line. so this is what i do. i take the 
small.cdl file from above and at the prompt type dos2unix small.cdl... 
and guess what, the carriage return is gone and i successfully am able to 
convert it to a .cdf file(using ncgen utility) and thus i can run the 
data and get the visualization that i need. The \r i am seeing is a dos/windows 
thing and not a unix, which explains my dos2unix solution.  

I am on a linux box, which explains why i didnt get any output before on 
the ncdump since the \r is dos/windows. i know this is a lot, but these 
are the initial results and findings if you could email me back. it is 
not 2:45 pm central time and will not be back until Tuesday 13th, so if 
you respond after 4:30 central time you will know why i dont respond 
until then. i hope this helps more.....

Chad 

 On Thu, 8 May 2003, Roy 
Mendelssohn wrote:

> Hi Chad:
> 
> GMT does do netCDF but it follows a specific convention.  What 
> visualization program are you trying to read it in?  Even more 
> importantly, for the GMT netcdf file, can you do the following 
> command (assuming you have the utility and the file is foo.nc):
> 
> ncdump -h foo.nc
> 
> and send the results.  The problem usually resides with the 
> dimensioning of the variables.
> 
> -Roy M.
> 
> 
> At 1:29 PM -0500 5/8/03, Mr. Chad W. Saxon wrote:
> >i'm sorry if i was unclear before.. i am using netCDF. The GMT file is
> >for the most part the same as the netCDF file but there is one or two
> >slight variations. this is the problem i am having when figuring out how
> >my program can read both formats.
> >
> >chad
> >
> >On Thu, 8 May 2003, John Caron wrote:
> >
> >>  Mr. Chad W. Saxon wrote:
> >>
> >>  >hi there everyone. i am new to the list and have an interesting question
> >>  >regarding netCDF. I have two data formats, one in GMT and one in CDF. i
> >>  >have a computer program that uses CDF as a way to read data and
> >>  >visualize it, however the program does not read the GMT format. Is there
> >>  >a simple script or solution to convert the GMT format to CDF? i was
> >>  >thinking there was a GMT command to convert it to CDF format but am
> >>  >uncertain as to what that is. I know there is not much difference between
> >>  >the two, but its enough to cause my program not to run correctly. i hope
> >>  >i have been clear in my problem, and i would appreciate your help. thanks
> >>  >a lot.
> >>  >
> >>  >Chad
> >>  > 
> >>  >
> >>  By "CDF" do you mean "netCDF" ?
> >>
> >>  A "GMT file" might be a netCDF file already, as far as I can tell from:
> >>
> >  > http://gmt.soest.hawaii.edu/gmt/doc/html/GMT_Docs/node108.html
> >  >
> >  > do you know what you have?
> >  >
> 
> 
> -- 
> Roy Mendelssohn
> Pacific Fisheries Environmental Laboratory
> 1352 Lighthouse Avenue
> Pacific Grove, CA 93950-2097
> 
> e-mail: rmendels@xxxxxxxxxxxxx
> voice: (831)-648-9029
> fax: (831)-648-8440
> www: http://www.pfeg.noaa.gov/
> 
> "Old age and treachery will overcome youth and skill."
> 

• References:
  • Re: Converting GMT files to the netCDF format
    • From: Roy Mendelssohn