Dear Brian,
I assume you therefore stand by your previous example of the
"forecast_reference_time" used as below:
variables:
double atime(atime)
atime:standard_name = "forecast_reference_time" ;
atime:units = "hours since 1999-01-01 00:00" ;
double time(time);
time:standard_name = "time" ;
time:units = "hours since 1999-01-01 00:00" ;
float temp(time,atime,level,lat,lon);
temp:long_name = "Air temperature on model levels" ;
temp:standard_name = "air_temperature" ;
temp:units = "K" ;
data:
time = 24 ;
a_time = 0 ;
I can see the advantage of this being that any non-CF compliant software
still gets to read the data following the standard time dimension, even if
it cannot understand the concept of a forecast. I also take your point that
using 'forecast_validity_time' does not add any new information. As long as
both time axes are fully defined with a standard units attribute I am happy
with the description.
Kind regards,
Ag
> -----Original Message-----
> From: Brian Eaton [mailto:eaton at ucar.edu]
> Sent: 16 July 2003 18:20
> To: cf-metadata at cgd.ucar.edu
> Subject: [CF-metadata] Example of forecast data
>
>
> Hi Jonathan and Ag,
>
> I'm opposed to adding forecast_validity_time to the standard
> name table for
> the following reasons:
>
> The purpose of standard names is to identify quantities and
> to help users
> of data from different sources decide whether or not certain variables
> contain data that is comparable. The name
> forecast_validity_time does not
> help in this regard. forecast_validity_time means exactly
> the same thing
> as "time", so adding that name makes it possible to say the same thing
> using more than one standard name. This makes things more complicated
> without adding information.
>
> That data may have been produced by a forecast is a property
> of the data
> and not of the time coordinate. Same is true of an analysis or
> observational data. The time at which a forecast, analysis,
> or observation
> is valid should be indicated by an axis with the standard name "time".
>
> The standard name table currently contains only one other
> name which refers
> to an instant in time, i.e., "forecast_reference_time". It
> is sufficient
> to have these two names to distinquish between valid and
> reference times of
> a data variable.
>
> Brian
> _______________________________________________
> CF-metadata mailing list
> CF-metadata at cgd.ucar.edu
> http://www.cgd.ucar.edu/mailman/listinfo/cf-metadata
>
Received on Thu Jul 17 2003 - 01:09:25 BST
