[Nereus]

Hi there:

I have been given a fine mesh with transport parameters (i.e., velocities, turbulent mixing coefficients) defined at the mesh nodes. I'm constructing a coarse mesh and determining the transport parameters at my new nodes by means of interpolation using Akima's algorithms. Is this the best approach, however? My concern is that, depending where the new nodes fall, this may result in very different values for my parameters. For example, imagine a region with very sharp gradients in parameter values. The exact placement of a new node in such a region could have a big impact on the parameter values estimated using interpolation. Would some sort of averaging approach over the region make more sense? Is this a standard problem and does it have a standard solution?

Thanks in advance!