NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
Hi,we're used to distribute our NWP model-data using aggregation in thredds, either a simple 'best-estimate' aggregation or a FMRC aggregation. The model-output for one analysis-time is usually stored within one file which is then aggregated, e.g. http://thredds.met.no/thredds/catalog/metno/proff4km/default/
To have the forecast_reference_time stored within the files, we use a scalar variable as explained in http://cf-pcmdi.llnl.gov/documents/cf-conventions/1.6/cf-conventions.html#scalar-coordinate-variables
In above example the scalar variable is called 'runtime'.As soon as we use aggregation, this scalar 'runtime' variable gets only the value of the penultimate file, and this is wrong in nearly all cases.
In FMRC, it would probably be best to remove this variable for all aggregations except for the 'File Access', since in the other cases, the forecast_reference_time is added automatically, but I don't know if this is possible.
Is there a proposed way to use the forecast_reference_time together with aggregation? What are others using?
Sorry for cross-posting - in case of aggregation, I'm not sure if this is a thredds or netcdf-java issue.
Best regards, Heiko
netcdf-java archives: