[Paraview] Point data on/in sphere - converting to structured grid

Lester Anderson lester_anderson1963 at hotmail.com
Wed Aug 11 04:15:08 EDT 2010


Hi ken

Thanks for the pointers. If it is possible to load in a data file with Longitude, Latitude, and Radius (sphere) and convert the input grid to a x, y, z that would be great.

I had a look at the calculator filter but could not see what to use. Would the result be a spherical representation of the data?

I do have the data in netCDF and CSV files. I did try converting all the data to spherical coordinates for X, Y, Z but Paraview only worked to do Table to Points.

Thanks for the guidance

Lester

From: kmorel at sandia.gov
To: lester_anderson1963 at hotmail.com; paraview at paraview.org
Date: Tue, 10 Aug 2010 14:14:23 -0600
Subject: Re: [Paraview] Point data on/in sphere - converting to structured grid





Message body


I’m confused as to why Table to Structured Grid won’t work.  A structured grid has no problem representing data on regular longitude, latitude, and radius.  If the coordinates are given in lon,lat,r, just use the calculator filter to convert them to x, y, z.



If you were to ask what the ideal format for the data would be, I would suggest creating netCDF files using the COARDS or CF convention.  It’s a much more efficient format than CSV and was specifically designed with geospatial data in mind.  That said, this only makes sense if you have control over how your data is generated in the first place.  I know of no easy way to convert CSV to netCDF.



-Ken





On 8/10/10 8:06 AM, "Lester Anderson" <lester_anderson1963 at hotmail.com> wrote:



Hello all



I have made a start with building the global mantle model. Because the data are referenced to the Sphere, it is not possible

to do the import of CSV using Table_to_Structured Grid and so I had to use Table_to_Points (see attached image).



The plot shows point data 50 depths 50-1600 km (final layers extend to 2800 km). The spatial distribution of data is 1 x 1 degree, and each layer increment is 50 km (eg 50, 100, 150 ... 2800) - data converted to X,Y,Z coordinates. As such it would seem that one could describe the data as being a structured grid with a curvilinear geometry.



Is there a way to convert or reformat the data (in Paraview .pvd format) to a cell-based geometry given that one has the values and coordinates of each "cell" from the point data?



I did have a quick try of Delauny3D but it crashed Paraview (3.8.0)



If there is a more efficient way of dealing with the CSV file that would be useful to know.



Thanks



Lester       





   ****      Kenneth Moreland

    ***      Sandia National Laboratories

***********  

*** *** ***  email: kmorel at sandia.gov

**  ***  **  phone: (505) 844-8919

    ***      web:   http://www.cs.unm.edu/~kmorel



 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20100811/f014e8be/attachment.htm>


More information about the ParaView mailing list