[ncl-talk] Function rcm2rgrid producing "noisy" results

[Michael Toy]

Hello,

I am using the function “rcm2rgrid” to interpolate a 2D field onto a Cartesian grid from a smooth non-orthogonal curvilinear grid.  Note that I am not on the sphere, but on an idealized flat plane with doubly periodic boundary conditions.  When the field, which is relative vorticity (zeta), is plotted in the native curvilinear coordinate, the contours are smooth.  However, when the interpolated field is plotted, the contours appear very ragged.  I don’t know where the “noise” is coming from.  I’m wondering if “rcm2rgrid” is the appropriate interpolation function for what I’m trying to do.

I’ve attached the *.ncl file I’m using which reads in a netcdf file with the data in the native curvilinear coordinate (a 3.8MB file stored on Google Drive at the following link):
https://drive.google.com/file/d/0B5l8e_K0c8ymRnBqUm0xOGhSS0U/view?usp=sharing <https://drive.google.com/file/d/0B5l8e_K0c8ymRnBqUm0xOGhSS0U/view?usp=sharing>

The interpolated data is stored to a file named “zeta_xy.troubleshoot.nc”.  The first plot generated is zeta in the native coordinate, followed by the interpolated zeta, followed by a panel plot showing both fields.

The version of NCL I’m using is 6.3.0.  Any help with producing smoother interpolated fields would be appreciated.  Thanks!

— Mike



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20160322/71358ca8/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: zeta_xy_interpolate.troubleshoot.ncl
Type: application/octet-stream
Size: 3438 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20160322/71358ca8/attachment.obj 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20160322/71358ca8/attachment-0001.html