[ncl-talk] data interpolation problem

[WIND]

Thanks Karin,
 With your help to use the "function" ESMF_regrid, I had regrided my data. (I only changed example's DstGridType to "0.45deg", DstLLCorner to (/30.d,   120.d /) and DstURCorner to (/66.d, 156.d /) ) But I found that the plot outcoming is quite different, the value at the same point is unmatched. My former grid interval is about 0.44 degree, I changed to regrid to 0.5 or other degree, they all failed all the same. The output pic's link is here https://pan.baidu.com/s/1ObFbK6lTDpzyyYgTpZp64w. Did that mean the data is unsuitable to regrid in this way?
 Really looking forward to your advice.
 Sincerely,
 Yan
  ------------------ Original ------------------
Message: 8
Date: Wed, 27 Nov 2019 14:38:20 +0100
From: Karin Meier-Fleischer <meier-fleischer at dkrz.de>
To: ncl-talk at ucar.edu
Subject: Re: [ncl-talk] data interpolation
Message-ID: <517bd9c2-644b-412a-57eb-43fa715103d4 at dkrz.de>
Content-Type: text/plain; charset="windows-1252"; Format="flowed"
Hi Yan,
if you have lat(y,x) and lon(y,x), which is a curvilinear grid, you can 
use ESMF to regrid it to a regular latlon grid.
Have a look at ESMF example 7 
http://ncl.ucar.edu/Applications/Scripts/ESMF_regrid_7.ncl.
-Karin
Am 27.11.19 um 13:18 schrieb WIND via ncl-talk:
> Dear all,
> Thanks for reading, I'm dealing with a Two-dimensional rotated pole 
> grid data, the coordinate changing format is different at each grid 
> (for eg. the latitude variable has two dimensions). I want to 
> interpolate the data to Regular grid format (so the dimension of 
> latitude and longitude is one). I had tried the function "linint2", 
> but it didn't work out.?So I wonder are there any useful function to 
> solve this problem?
> Any help would be appreciated!
> Sincerely,
> Yan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20191128/42433480/attachment.html>