[ncl-talk] regridding using ESMF weights

[Denise Worthen - NOAA Affiliate]

All,

I've read the information on using ESMF_regrid and weights generation when
the mask varies across levels (say for ocean depths from tripole to
rectilinear). I'd like to be sure I understand the "trick" which is
described for situations where the mask for missing values would change
with level.

Is this the correct understanding:

For each vertical level, the same weights file can be used if first you do
an interpolation from a variable field where the missing values have been
set to 1 and valid values set to 0. Using the resulting interpolated field,
define another mask where values >0 would show where the missing values
'bled' into the field. That mask can then be applied to the interpolated
fields. This trick removes the necessity of generating a weights field for
each vertical level.

Thanks
Denise
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20190320/0bfb0c25/attachment.html>