open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,
I was wondering: how are the algorithms programmed in the 2D interpolator of BlueKenue. The following methods are possible:
* linear
* inverse distance
* nearest neighbour

For the first two interpolation methods, how many points are used, and is it based on specific zones (like quadrants for example)?
(There are also some special cases with a constraint outline, a lack of points, the use of maximum search distance... but they are not my main focus).
I already searched in the tutorial and the documentation unsucessfully.

PS: I am not dealing with interpolation in a triangular mesh (like this topic).

Thanks in advance,
Luc

Hi Luc,

In all cases we reduce geometry from all source objects (linesets, grids, meshes) to points. Lines are resampled to a "small" interval automatically computed from the geometry of the target object.

1. Linear option:
- triangulate all source points (via a Delaunay triangulator)
- locate the triangle under the target point
- perform a linear interpolation on the triangle.

2. Inverse Distance option:
- locate the closest source point in each quadrant
- perform distance weighted interpolation from these 4 points

3. Nearest Neighbour:
- simple assignment from the closest source point.

Maximum search distance is only for Inverse Distance interpolation.
- only points within this radius are used.

Constraint outline allows the user to tell the the Map Object function to only interpolate values for locations inside this polygon.

Hope this helps... Martin

Thank you Martin for your quick and precise answer!