Thanks for the prompt reply, Bill.
> It sounds like your original temperatures are sampled on
> a square grid with NaNs at any grid points outside the
> circular plate, and that you then reample to a more
> circular Irregular2DSet. If so, then the problem is that
> you don't have temperature values at the edge of the plate,
> and any procedure for creating them is just guessing. So
> the problem is with the original data, and you'll have to
> solve it there.
We use another disc(same size with the plate) with an array of
temperature sensors to measure the temperature of the target plate and
there are no sensors at the very edge of the disc. So it is very
likely that we can not solve the problem by measuring the edge region
of the target plate. Maybe I can try setting the NaNs to be weighted
average of several adjacent points, but this will use more CPU time.
Do you have any suggestion?