On Aug 23, 2012, at 2:37 PM, Randy Horne <[email protected]> wrote:
> > What is the “generally accepted” approach for how a specific quality flag > value should be assigned for a corresponding data value (in the corresponding > data variable) that has a _FillValue ? > > Is it sufficient that, because the data variable value is _FillValue, the > corresponding quality flag value can be undefined ? I do want to mention one variant to consider: if the quality of a particular point is the *reason* that the data value is _FillValue, then it is useful for that corresponding quality flag to retain its value. We use this approach in our Data Quality Screening Service, which allows users to filter (some of) our satellite data based on the quality flags. In the output product, we replace data values with _FillValues if the quality flag indicates it does not meet specified quality criteria, but then we retain the quality flag so that the user will know WHY that particular point has a _FillValue (vs. not being able to get a retrieval at all). -- Dr. Christopher Lynnes, NASA/GSFC, Code 610.2, 301-614-5185 _______________________________________________ CF-metadata mailing list [email protected] http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
