[ncl-talk] Question on interpolation

[Andrew Kren - NOAA Affiliate]

Dear ncl-talk,

I am working to interpolate a high-resolution model dataset (HWRF) to the
time and location of a dropsonde location. In the past for other datasets,
I have used linint1_n and linint2_points to interpolate in time and space
to the dropsonde position.

However, HWRF, which is 6-hour analysis output, has a varying grid every 6
hours as the domain "moves" with the tropical cyclone. Hence, the lat/lon
locations are not fixed in time. They have the same dimension sizes but the
points vary depending on the cyclone center. This led to an error in my
linint2_points function:

fatal:linint2_points: If xi is not one-dimensional, then it must have one
less dimension than fi



I know why this is happening. I read in two HWRF files around the dropsonde
time to do the interpolation. I used "join" to join the files and my
lat/lon arrays are (time, lat) and (time, lon). So I get the error.


But my question is how best to proceed? This is technically not an
unstructured grid, but could be considered one. Should I use ESMF?


Thanks,

-- 
Andrew Kren
Assistant Scientist
University of Miami CIMAS - NOAA/AOML
Global Observing Systems Analysis (GOSA) Group
NOAA/AOML Quantitative Observing System Assessment Program (QOSAP)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20190903/931d2f47/attachment.html>