[Thread Prev][Thread Next][Index]

binary data


Thanks to Mark Verschell for answering my question about
reading 3D vector binary data.

I am trying to define a strategy for Ferret-visualization of
FORTRAN generated model output that includes several variables
some of them scalar, some vector, some 2D, some 3D.
Example of FORTRAN generated binary output:

      DIMENSION U(360,180,33),V(360,180,33), UA(360,180), VA(360,180),
      DIMENSION T(360,180,33), S(360,180,33), MLD(360,180)
      WRITE(80) U,V,UA,VA,T,S,MLD

where U,V is horizontal velocity vector; UA, VA is vertical
average of U, V;  T,S are temperature and salinity, MLD is
mixed layer depth. There are more variables of this type.

The task is to minimize combined complexity of formats on
FORTRAN side and on Ferret side. In the example above 'WRITE'
command works as is - no need to run DO loops and specify indexes.
In the past I used graphical language UNIRAS in which it was
possible to read the data with FORTRAN commands, so I would just run

     READ(...) U,V,UA,VA,T,S,MLD

In Mark Verschell's example of reading such data in Ferret
there is need to specify length of records on Ferret side
like (360*180*33 + 1) et al.

Is this the least complex way? Did anyone try, direct access binary format,
ASCII format? How does intermediate conversion from binary or ASCII
to NetCDF format sounds in this context.

Any advice is greatly appreciated.

Grigory Monterey
1352 Lighthouse Avenue
Pacific Grove, CA 93950-2097

Phone: (408) 648-0623
FAX:   (408) 648-8440
email: gmonterey@pfeg.noaa.gov

         \\ //
         (o o)

[Thread Prev][Thread Next][Index]

Dept of Commerce / NOAA / OAR / PMEL / TMAP

Contact Us | Privacy Policy | Disclaimer | Accessibility Statement