Hi Salah,
> My question:
>
> In general, I am wondering how curvilinear girds can be represented in
> NetCDF API such cylindrical boundaries "radar data", also is it
> possible to use it with irregular boundaries such as rivers banks,
> paths and littoral ocean environments or we have to use shape files
> because NetCDF API is not suitable for list data structure.
This kind of data can be represented using netCDF, although their are no
CF Conventions adopted yet for unstructured grids.
NetCDF CF conventions for curvilinear grids are described here:
http://cf-pcmdi.llnl.gov/documents/cf-conventions/1.4/cf-conventions.html#id2984605
For unstructured grids, a 2006 NOAA Unstructured Grids Workshop:
http://cf-pcmdi.llnl.gov/trac/wiki/UnstructuredGrids
resulted in discussion of several proposals and examples for
representing unstructured grids in netCDF:
http://cf-pcmdi.llnl.gov/trac/wiki/CdlDiscussion
Links from the latter page provide an example for unstructured grid cell
arrays and an example CDL for what's used in the FVCOM ocean model:
http://cf-pcmdi.llnl.gov/trac/wiki/FvCom
Some other examples are available from ELCIRC
http://sura-vims-pe6600-1.vims.edu/~drf/NETCDF/
and SELFE
http://www.stccmop.org/cyberinfrastructure/netcdf
Somewhat related to this is V. Balaji's GridSpec, which describes a
framework for representing many kinds of structured grids using the
netCDF-3 classic data model:
http://www.gfdl.noaa.gov/~vb/gridstd/gridstd.html
Software and utilities are available for creating the netCDF schema
for various kinds of grids described there, but no unstructured
triangular grids yet, as far as I know.
> My request:
>
> Also I will be very glade if some one provides me with curvilinear
> grids dataset.
There are links to such datasets in the examples above.
--Russ
