Unidata Support wrote:
To: support-idv@xxxxxxxxxxxxxxxx
From: "Stuart Wier" <wier@xxxxxxxxxx>
Subject: IDV - NetCDF data files with curvilinear coordinates / CF convention
Organization: UCAR/Unidata
Keywords: 200512191651.jBJGpYJU012442
Institution: unavco
Package Version: 1.2
Operating System: linux
Hardware Information: the ususal
Inquiry: I have a GEON IDV user who wants to see seismic tomography (seismic
velocity values in 3D) from a 3D volume data grid. His data file is a
collection of what are unique longitude-latitude pairs for each grid point at
each depth:
long lat depth Pv
138.717 37.1541 9 5.8325
138.773 37.1542 9 5.9639
138.829 37.1543 9 6.1133
138.886 37.1543 9 6.1892
138.942 37.1543 9 6.0828
138.998 37.1543 9 5.9052
139.054 37.1543 9 5.8052
139.111 37.1543 9 5.8607
139.167 37.1542 9 5.9771
139.223 37.1541 9 6.0249
(there are 17 megabytes of this, spanning about 100 depth levels. The depths
are regular: identical 2D grids for each depth.)
I suspect that this is a regular rectangular x-y grid mapped onto longitude-latitude locations.
I recall this can be handled by the CF convention if enough is known about the
projection, right? What exactly is required in the NetCDF file, and is there
an example of a NetCDF file to use as a guide?
Yes, he can use the CF conventions. Just have a 2D lat and a 2D lon variable,
see section 5.6 of CF manual.
Basically you have:
float dataVariable(x,y,z);
:coordinates="lat lon";
float lat(x,y);
float lon(x,y);
float z(z);
If its really from a projection (and its one we can deal with) then thats a
more efficient way to go.
Stu
--
NOTE: All email exchanges with Unidata User Support are recorded in the
Unidata inquiry tracking system and then made publicly available
through the web. If you do not want to have your interactions made
available in this way, you must let us know in each email you send to us.
