[gdal-dev] "reboxing," or 3D regridding

[Tom Roche]

summary: I'd appreciate advice regarding tools and methods for
transforming values from an unprojected global 3D spatial grid to a
projected 3D grid with different horizontal and vertical resolution, or
pointers to other resources to consult.

details:

I have output from a global atmospheric model that I'd like to use as
initial and boundary conditions for a regional model. The global input
netCDF has dimensions=2.5° lon x 1.875° lat x 56 vertical levels. The
regional model runs over North America using a 12-km grid projected
Lambert Comformal Conic (LCC), with 34 vertical levels. Since its top
height is less than that of the global input, the extents of the output
domain are fully contained within the input domain.

Each box or voxel defined by the global input grid contains an estimate
for the N2O concentration for that volume. From those I want to compute
the concentrations for each output gridbox volume. I'd appreciate your
recommendations for tools that can handle this usecase. Particularly, is
this something that gdalwarp can do?

FWIW, the best tool I've seen so far is R package=gstat, but (IIUC)

- gstat expects projected input. I'm not sure if I can work around that
  for this usecase.

- as the name implies, 'gstat' is doing geostatistics, e.g., variogram-
  and covariance-based modeling. I'm not sure either how to setup the
  distance weighting for my scenario, and, frankly, I remain unconvinced
  that a statistical approach is necessary for this application (though
  it may be a sufficient or the best-available approach). This may be
  due to my statistical ignorance, however.

your assistance is appreciated, Tom Roche <Tom_Roche at pobox.com>