Dear Jonathan,
thanks for your comments. The time axis and bounds are correctly described
(indeed this reminded me that it does make sense to include time_bounds in this
case which we often don't do). Automatic processing would indeed be useful
here. Further: since it is the maximum of an 8-hour mean, the cell_methods
should read "time: maximum (interval: 8 hours)", shouldn't it? In practice,
this might already suffice, unless I have a misconception what the interval
stands for. Reading the documentation (CF 1.5) again, I believe one can argue
that the "typical interval between the original data values to which the method
was applied" is indeed 8 hours in this case - the only ambiguity is that the
8-hour values were running means rather than discrete 8-hour intervals. If we
can live with such ambiguity: fine. Otherwise I would suggest to specify
"time: maximum (interval: 8 hours running mean)" or even "time: maximum
(interval: 8 hours running mean from 1 hour values)". If the parser st
ops after the standardized "8 hours", it will still get the main part. If it
reads on, it gets the full picture.
Cheers,
Martin
> Whether the CF convention needs to be extended for this depends on how
> much you need to be recorded in standardised metadata. I assume you have
> a time axis with times at daily intervals, whose bounds indicate periods of a
> day e.g. the time coordinates might be 2011-10-18 12:00, 2011-10-19 12:00
> with bounds [2011-10-18 0:00, 2011-10-19 0:00], [2011-10-19 0:00, 2011-10-20
> 0:00].
> If the time coordinate is "time", cell_methods can record a maximum made
> from hourly values with "time: maximum (interval: 1 hour)". This, however,
> does not record that the hourly value were running 8-hour means. That could
> done in an unstandardised way e.g. "time: maximum (interval: 1 hour
> comment: hourly values were running 8-hour means)". Is it sufficient to
> record the information in such a way, or do you need to be able to process it
> automatically in order to distinguish running 8-hour means from, say,
> instantaneous hourly values, or means for single hours?
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzender des Aufsichtsrats: MinDirig Dr. Karl Eugen Huthmacher
Geschaeftsfuehrung: Prof. Dr. Achim Bachem (Vorsitzender),
Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt,
Prof. Dr. Sebastian M. Schmidt
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
_______________________________________________
CF-metadata mailing list
[email protected]
http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
