I've got 96 hr WRF forecasts that focus on the Atlantic north of the
equator, the Caribbean and the Gulf of Mexico. THese are ~32GB. Would
that help?
gerry
Ben Domenico wrote:
Hi Peter,
One place you might look for a set of the most recent Global Forecast
Model runs from the US National Centers for Environmental Prediction on
our Undata server at:
http://motherlode.ucar.edu:8080/thredds/catalog/fmrc/NCEP/GFS/Global_0p5deg/files/catalog.html
These are the runs of the 0.5 degree resolution forecasts. Each file is
"only" about 3 GB, but you can combine a set of them to get to the size
you are looking for. If you need larger collections, I know that we
have about 20GB of ensemble forecast runs coming into UCAR/NCAR, but I
don't know if they are being made available publicly yet. Others on the
list may know.
In terms of interesting characteristics, you can drill down and look at
what's available in each forecast, e.g.,
http://motherlode.ucar.edu:8080/thredds/ncss/grid/fmrc/NCEP/GFS/Global_0p5deg/files/GFS_Global_0p5deg_20071218_0000.grib2/dataset.html
One thing you'll notice is the variety of vertical coordinates for
different parameters. In particular, the "geopotential height" is, in
effect, a dependent variable with the pressure level as the independent
coordinate variable. That's the sort of unusual characteristic where a
transformation might benefit some user communities that are not used to
working in pressure coordinates.
Hope this is enough to get you started.
-- Ben
