Fully unstructured 3D mesh and netcdf

4 posts / 0 new
Last post
markus.luisser_...
Fully unstructured 3D mesh and netcdf
Hi! I'm interested in using VAPOR to visualize transient CFD simulation results of unstructured meshes. The simulation tool is OpenFOAM and the result files are available as vtk data files.

From what I have seen so far, I need to generate netcfd files from my vtk files in order to use them in VAPOR (via ncdf2vdf).

What I have not found out so far, is if VAPOR can deal with unstructured data at all and if it can, what kind of data is expected in the netcdf file.

[ With 'unstructured' I mean what is described here under 'Fully 3D unstructured': https://github.com/ugrid-conventions/ugrid-conventions/blob/master/src/i... ]

clyne

Hi,

 

Sorry, but vapor does not support unstuctured data at this time. We are working on support for data following the UGRID conventions and expect to have a release in late 2017 or early 2018.

 

Best,

 

jc

markus.luisser_...
Thank you for the clarification, clyne! Is there some more information about the mesh requirements? I'm thinking about things like: point data or cell data? What kind of structured cells are supported? Is it possible to use data from refined subgrids etc? I'm able to re-grid my data (though it requires some effort) but without a specification (and no familiarity with the supported meteorological grid formats) I feel a bit lost.
clyne

We intend to support meshes descriable by the UGRID conventions. However, layered 2d grids will be our priority as these are most commonly used in the geosciences.

 

Regarding re-gridding: This may be your best option for the short term. There are numerous 3rd party tools available for re-gridding to a lat-lon grid. See the ESMF re-gridding software or NCO for example.