[ncl-talk] question regardig wrf_interp_2d_xy

[Bill Ladwig]

Hi Damjan,

You want to use wrf_user_intrp3d (
http://www.ncl.ucar.edu/Document/Functions/WRF_arw/wrf_user_intrp3d.shtml).
The wrf_interp_2d_xy is a lower level routine which is called by
wrf_user_interp3d.  To use this routine, you either set a start point and
end point, or an origin with an angle, to define the horizontal line.  With
either method, the points to define the horizontal line are grid indexes.

Also, the output vertical levels are chosen by the algorithm by taking
min/max for the vertical data and dividing it up in to ~100 increments.  It
will work with increasing or decreasing vertical coordinates, but
unfortunately the user cannot select the output vertical levels manually at
this time.  In general, the algorithm is expecting height or pressure
coordinates for the vertical.  Your mileage may vary if you choose another
type.

As a general warning, be careful when using non-WRF data with the 'wrf' NCL
routines, as nearly all of the WRF routines expect data without missing
values in the input arrays, despite many of the algorithms having missing
values in the output.  The algorithms will most likely crash if there are
missing values in the input.

Hope this helps,

Bill

On Mon, Jul 4, 2016 at 5:07 AM, Damjan Jelić <damjan.jelic at fmf.uni-lj.si>
wrote:

> Dear NCL support and users,
>
> I'm in need of vertical cross section data along some manually defined
> line (currently along New Zeland). Im using various datasets (currently
> reanalysis) in netcdf format. Is the function from title appropriate for
> such task? I did try to apply it but it seams its working only as
> latitudal or longitudal slicer and description is not detailed enough so
> Im sending this mail. Any how to about it would be great.
> On the other hand is there any other function which can do the cross
> section by interpolating neighboring points and not minding about type
> of vertical coordinate at hand or will i have to do it manualy?
>
> Sincerely,
> Damjan Jelic
> Uni. Lubljana
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20160705/4628dbcb/attachment.html