[netcdfgroup] netCDF Operators NCO version 5.1.6 are coronated

• To: NCO Announce Mail List <nco-announce@xxxxxxxxxxxxxxxxxxxxx>, netCDF Mail List <netcdfgroup@xxxxxxxxxxxxxxxx>
• Subject: [netcdfgroup] netCDF Operators NCO version 5.1.6 are coronated
• From: Charlie Zender <zender@xxxxxxx>
• Date: Fri, 12 May 2023 16:46:06 -0700

netCDF Operators NCO version 5.1.6 are coronated

http://nco.sf.net (Homepage, Mailing lists, Help)
http://github.com/nc/ncoo (Source Code, Issues, Releases)

What's new?
Version 5.1.6 further polishes vertical interpolation, further
improves NCZarr safety, and fixes minor ncremap issues, improves EAMxx
support, adds basic support for regridding Coupler history files, and
employs CF Conventions, where possible, on all input files regardless
of whether they claim to be CF-compliant.
This release can be skipped if you would not use the NCZarr and
regridder improvements.

Work on NCO 5.1.7 has commenced and aims to add support for Zarr S3
stores and to polish support for new codecs.

Enjoy,
Charlie

NEW FEATURES (full details always in ChangeLog):

A. ncremap now diagnoses rather than prescribes the monotonicity
direction (increasing or decreasing in index space) and dimension
ordering of vertical grids prior to searching for maxima/minima
surfaces beyond which extrapolation is necessary. That's a mouthful.
Basically this feature fixes corner cases in which input or output 3D
vertical grids (e.g., hybrid sigma/pressure, or MPAS-Ocean-style
grids) with non-standard directionality (which way is up?) or
dimensionality could have caused previous versions of NCO to misjudge
the vertical domain of the grid, and thus prevented implementing
missing values beyond the valid domain. Also, missing values in the
grids are better handled when looking for vertical domain boundaries.
ncremap -P mpasocean --vrt_out=vrt.nc --map=map.nc in.nc out.nc
http://nco.sf.net/nco.html#ncremap
http://nco.sf.net/nco.html#vrt_out

B. ncclimo improves handling of output from the DOE EAMxx model
in two ways. ncclimo now understands the commonly used suffixes
for EAMxx monthly output files: "...YYYY-MM-01-00000.nc", and
allows these names to be used as the caseid argument for templating
filenames. Thanks to Chris Golaz (LLNL) prompting this feature.
caseid=really_long_string.0001-01-01-00000.nc
ncclimo -P eamxx -c ${caseid} -s 2000 -e 2019 -i $drc_in -o $drc_out
http://nco.sf.net/nco.html#ncclimo
http://nco.sf.net/nco.html#caseid

C. ncremap now supports non-spatial dimensions (temporal, spectral,
chemical) when vertically interpolating datasets. This now works for
all vertical grid types. Previously this only worked for hybrid/sigma
grids, and only then for temporal dimensions (this former limitation
never affected horizontal regridding). Now datasets with
non-spatio-temporal dimensions such as temperature(time, species,
wavelength, horizontal, vertical) should vertically regrid properly.
http://nco.sf.net/nco.html#ncremap

D. ncremap now supports vertical interpolation of timeseries data from
pure-pressure to pure-pressure grids
ncremap --vrt_out=ncep_L17.nc ncep_L10.nc out.nc
http://nco.sf.net/nco.html#vrt

BUG FIXES:

A. This version of NCO fixes and error with the overwrite/removal of
NCZarr stores implemented in 5.1.5.
in_hdf5="${HOME}//nco/data/in_zarr4.nc"
out_ncz="file://${HOME}/foo#mode=nczarr,file"
ncks -O -v one ${in_hdf5} ${out_ncz}
There is no workaround. The solution is to upgrade.

B. ncap2 previously ignored the --cmp option when writing
new variables defined in the script. Now ncap2 uses the
cmp_sng option, if any, to determine and set compression
levels for new variables. The workaround is to use the
(old) '-L 1' option for DEFLATE. The solution is to upgrade.
Thanks to Hugo Oliveira for reporting this issue.

Full release statement at http://nco.sf.net/ANNOUNCE
--
Charlie Zender, Earth System Sci. & Computer Sci.
University of California, Irvine 949-891-2429 )'(