Unstructured Grids SuperGrid Requirements

[tomgross]

The irregular grid specifications have brought up several spatial objects which are not precisely described by a single coordinate location. For instance the edge of a triangle, or the triangle itself. An edge must be described by two points and the triangle by three points. A variable such as heat flux or area are not, necessarily, located at a single point. For convenience we could define spatial locations, center of edge or center of triangle, which could serve as coordinate locations for these variables. As it will still be necessary to describe the edge or triangle, by calling out the locations of the vertices, these derived locations are redundant. These extra locations are very much like the gridspec supergrid points (Balaji 2006). They might be useful enough to become a requirement of the CF conventions.

Should there be a requirement for the inclusion of these auxiliary locations for the center points of higher dimensional objects, such as line segments or areas?

[tomgross]

Replying to tomgross: Given the fact that these auxliairy points are in fact defined points on some grids (for example, the velocity on the UnTRIM grid is specified at the cell center face, and the the depth is defined on the center point of an edge) I think there is utility in having them defined as a sub feature of a higher order geomtric feature like a line or an area.

Michael Piasecki

[taylor13]

Dear all,

Given that there is a small working group exploring options for representing unstructured grids, I am closing this ticket with the expectation that a new ticket will be opened when the working group comes up with one or more concrete proposals. [Thus, we are closing tickets 6 & 7 for the same reason.]

Best regards, Karl