[ferret_users] looking for an elegant way to regrid gridded netcdf with broken grid specs but with gridded lat/lon cell values

[Lev Tarasov <lev@xxxxxx>]

This is unfortunately a too frequent problem at least for data sources I'm 
using. The source files tend to be on curvilinear grids (eg DEM, or 
obscure GCM/RCM grid or some other curvilinear projection) and the 
internal net-cdf grid specs are broken so that simple regridding doesn't 
work. However, the dat set includes grid-center lat and lon for
each grid cell (so on the same grid). Now I could do a bunch of repeat
loops, list to file in 1D format and then use scat2grid* and repeat for 
all the 20 fields * 12 months  I want to extract, but I'm hoping
there's a much more elegant way to work directly with the netcdf file
(if not in ferret, then some other tool? (cdo remapbil doesn't work).


Eg, my current case:

yes? show dat
     currently SET data sets:
    1> ./MARv3.5.2-10km-monthly-ERA-Interim.yMonMeanClim19C.nc  (default)
 LAT      Latitude                         1:301     1:561     ...       ...       ...       ...
 LON      Longitude                        1:301     1:561     ...       ...       ...       ...
 TT       3m-Temperature                   1:301     1:561     ...       1:12      ...       ...
and 20 other fields


yes? show grid tt
    GRID GQA1
 name       axis              # pts   start                end
 X         LONGITUDE          301 r   5E                   65E(1505)
 Y         LATITUDE           561 r   5N                   2805N
 normal    Z
 TIME      TIME                12 i   15-JAN-2000 00:08    14-DEC-2000 19:32

Even though this is actually a curvilinear projection (from readme)
grid_mapping_name,stereographic "Stereographic Oblique Projection"
standard_name,stereographic "Stereographic Oblique Projection"
long_name,stereographic "Stereographic Oblique Projection"
longitude_of_projection_origin 40°W
latitude_of_projection_origin 70.5°N
false_easting -775 km
false_northing -1200 km
resolution 25km

so standard regridding:
shade TT[g=GRf525, l=6]
gives a messed up plot


yes? show grid GRf525
    GRID GRF525
 name       axis              # pts   start                end
 XLONGRF5  LONGITUDE          155mr   80.75W               3.75W
 YLATGRF25 LATITUDE           105 r   58.875N              84.875N


Cheers,

Lev


Lev Tarasov -   Dept of Physics and Physical Oceanography,
		Memorial University of Newfoundland.
                email: lev@xxxxxx
                http://www.physics.mun.ca/~lev/
		Tel (709)-864-2675
		Fax (709)-864-8739