Dear Jonathan,
> standard_error here means statistic uncertainty or systematic (measurement)
> uncertainty. Does that meet the need?
>
Yes.
> If the purpose of the variable is to indicate missing or non-missing, why do
> you need to store it in a separate variable, rather than indicating it with
> missing data values in the data variables?
>
For the missing values, we are using the _FillValue. The purpose of this
variable is to flag the bad values.
But we have only one flag for several values.
> That doesn't answer your question, though. I understand there might be a need
> to share ancillary data among several data variables with different standard
> names. I suppose that there is nothing to prevent a variable with standard_name
> "X" having an ancillary_variables attribute which points to a variable with
> standard name "Y status_flag", for instance. Perhaps we should review this
> convention.
>
So we will have for example :
float ewct(time, latitude, longitude) ;
ewct:standard_name = "eastward_sea_water_velocity";
ewct:ancillary_variables = "ct_flag";
float nsct(time, latitude, longitude) ;
nsct:standard_name = "northward_sea_water_velocity";
nsct:ancillary_variables = "ct_flag";
byte ct_flag(time, latitude, longitude) ;
ct_flag:standard_name = "eastward_sea_water_velocity status_flag";
That's it ? Both ewct and nsct variables have the same ancillary variable.
But it's not obvious when you look at ct_flag that it applies to two
variables !
Best regards,
St?phane Tarot
> Best wishes
>
> Jonathan
> _______________________________________________
> CF-metadata mailing list
> CF-metadata at cgd.ucar.edu
> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
>
>
Received on Thu Oct 08 2009 - 09:33:41 BST
