[Dart-users] group question for obs_diag changes

Hacker, Joshua (Josh) (CIV) jphacker at nps.edu
Wed Nov 14 11:05:58 MST 2012


Hi Tim,

Does this eliminate the ability to inventory an obs sequence other than obs_seq.final?

Josh

--
Joshua Hacker
Naval Postgraduate School
Department of Meteorology
589 Dyer Rd, Root Hall 254
Monterey, CA 93943
email: jphacker at nps.edu
phone: 831-656-2722
fax: 831-656-3061

From: Tim Hoar <thoar at ucar.edu<mailto:thoar at ucar.edu>>
Date: Wednesday, November 14, 2012 10:00 AM
To: "dart-users at mailman.ucar.edu<mailto:dart-users at mailman.ucar.edu>" <dart-users at mailman.ucar.edu<mailto:dart-users at mailman.ucar.edu>>
Subject: [Dart-users] group question for obs_diag changes

obs_diag is very complicated and difficult to maintain.
I'd like to simplify it.

Since we can get the location information from ANY observation sequence (obs_seq.in, obs_seq.out, obs_seq.final) by converting the observation sequences to a netCDF file (with obs_seq_to_netcdf) - I would like to remove the 'plot_locations' functionality from obs_diag.

How negatively would that impact people?

Speak up and tell me what you think!

Long version: If I can require that the [prior/posterior]_ensemble_mean and
[prior/posterior]_ensemble_spread are present in the file, I can remove a lot
of the checking. This means only "obs_seq.final" files would be supported in obs_diag.

Tim Hoar, Associate Scientist
National Center for Atmospheric Research
thoar at ucar.edu<mailto:thoar at ucar.edu>
303.497.1708

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/dart-users/attachments/20121114/e87ec8ee/attachment.html 


More information about the Dart-users mailing list