The problem you outline involves vector data.
Applying interpolation schemes designed for scalar quantities
(eg, temperature) to vector components could lead to
problems, particularly near the poles. Values derived from
the interpolated components should, as Erik noted, be interpreted
with caution.
For the specific problem you outlined, I have added two new
NCL functions [uv2vr_cfd, uv2dv_cfd] described at:
http://ngwww.ucar.edu/ngdoc/ng/ref/ncl/functions/vrdvcfd.html
The grid may contain missing data and the grids can be
be over a limited domain. These will be included in
version 4.2.0.a012 available [30 October 2000]. It may be
be downloaded free of charge. Contact Mary Haley [haley@ucar.edu]
The new functions will not do calculations where missing data
are present. However, if you want 'complete' grids [ie, no
missing values present], you can then apply standard scalar
interpolation schemes to the derived relative vorticity and
divergence fields since these are *scalar* quantities.
good luck
Dennis Shea
This archive was generated by hypermail 2b29 : Thu Oct 26 2000 - 14:21:17 MDT