[netcdfgroup] netCDF Operators NCO version 5.3.8 mends fences

netCDF Operators NCO version 5.3.8 mends fences

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

What's new?
Version 5.3.8 contains fixes to ncclimo problems introduced with 5.3.7.
The ncclimo "flexible-months" feature suffered from incorrect behavior
for some inputs, and introduced a regression in seasonal filenames.
The only new feature is that ncclimo now supports generating climos
from high-frequency input data that are instantaneous (not mean)
values. Documentation now reflects the new CF Convention Homepage URL.

Users who upgraded to 5.3.7 should upgrade to 5.3.8, and 5.3.7 should
not be used anymore.

Enjoy,
Charlie

NEW FEATURES (full details always in ChangeLog):

A. ncclimo now supports generating climos from high-frequency input
data that are instantaneous (not mean) values. This mainly means that
that the time bounds and climatology bounds variables now use the
instantaneous time from the original time coordinate input, rather
than the average value of the time coordinate input.
Thanks to Jill Zhang (LLNL) for feature testing and feedback.
http://nco.sf.net/nco.html#ncclimo

B. All links to documentation now point to the newish CF Homepage
at https://cf-convention.github.io/cf-conventions.

BUG FIXES:

A. ncclimo flexible months mode introduced in 5.3.7 sometimes
generated read datasets in the wrong order.
Thanks to Jill Zhang (LLNL) for identifying this issue.
http://nco.sf.net/nco.html#ncclimo

B. ncclimo flexible months mode introduced in 5.3.7 sometimes
caused input years with leading zeros to be interpreted as octal
numbers and therefore generate inaccurate filenames.
Thanks to Xylar Asay-Davis (ANL) for reporting this issue.
http://nco.sf.net/nco.html#ncclimo

Full release statement at http://nco.sf.net/ANNOUNCE
--
Charlie Zender, Dept. of Earth System Science
University of California, Irvine 949-891-2429