[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDV #ZZW-779500]: aggregation of NARR-A files in the IDV

> Not sure who sees this email, but a few days ago I sent a note to the
> idvusers mailing list about an error in the aggregation of multiple NARR-A
> using the "Data Source Type: Aggregate Grids by Time" in the Data Choosers
> window.
> 
> I have explored this problem further and discovered that in the metadata
> file, fields are assigned a time attribute of "time1=1" or "time=x" where
> x=the number of files being aggregated.  All of the fields shown in the
> metadata that are assigned "time=x" will display the appropriate number of
> valid times to select from... and will load the said grid in the Map Window
> for all the appropriate times.  On the other hand, many of the other fields
> (including ALL of the 3D fields) show a "time1=1" attribute, resulting in
> the listing of only ONE hour to load into the Map Window.
> 
> I am not sure if this is an IDV issue or an issue specifically with the
> NcML and the way it aggregates these NARR-A files and assigns these time
> attributes.

Hi Mike,
     I did follow the discussion in the idvuser list. Both Tom and Don provided 
the best solution to your problem available at this moment. Ideally, we would 
like to see the NcML can do the aggregation job for this kind of situation. But 
there are some restriction on the NcML side when the dataset has multiple time 
dimensions. I talked to the NcML developer, and he would consider to add 
support to multiple time dimension in the future release. I think this will 
take a few months to finish. In the meantime you can rely on the formula to do 
the job. If you save your setting in a bundle it can make your life easier. 
Sorry about the inconvenience.


Yuan
> 
> Below is a snippet showing the 3D Geopotential_height grid from an
> arbitrary aggregated NARR-A metadata file which DOES NOT properly display
> all the aggregated times:
> 
> float Geopotential_height(time1=1, isobaric=29, y=277, x=349);
> >      :_CoordinateAxes = "time1 isobaric y x ";
> >      :long_name = "Geopotential_height @ isobaric";
> >      :units = "gpm";
> >      :missing_value = -9999.0f; // float
> >      :grid_mapping = "Lambert_Conformal";
> >      :GRIB_param_name = "Geopotential_height";
> >      :GRIB_param_short_name = "HGT";
> >      :GRIB_center_id = 7; // int
> >      :GRIB_table_id = 131; // int
> >      :GRIB_param_number = 7; // int
> >      :GRIB_param_id = 1, 7, 131, 7; // int
> >      :GRIB_product_definition_type = "Forecast/Uninitialized
> > Analysis/Image Product";
> >      :GRIB_level_type = 100; // int
> >      :GRIB_VectorComponentFlag = "gridRelative";
> >
> 
> Below is a snippet showing the 2D Categorical_rain grid from the same
> arbitrary aggregated NARR-A metadata file which DOES properly display all
> the aggregated times:
> 
> float Categorical_rain(time=4, y=277, x=349);
> >      :_CoordinateAxes = "time y x ";
> >      :long_name = "Categorical_rain @ surface";
> >      :units = "yes=1;no=0";
> >      :missing_value = -9999.0f; // float
> >      :grid_mapping = "Lambert_Conformal";
> >      :GRIB_param_name = "Categorical_rain";
> >      :GRIB_param_short_name = "CRAIN";
> >      :GRIB_center_id = 7; // int
> >      :GRIB_table_id = 131; // int
> >      :GRIB_param_number = 140; // int
> >      :GRIB_param_id = 1, 7, 131, 140; // int
> >      :GRIB_product_definition_type = "Forecast/Uninitialized
> > Analysis/Image Product";
> >      :GRIB_level_type = 1; // int
> >      :GRIB_VectorComponentFlag = "gridRelative";
> >
> 
> 
> Thanks,
> Mike Umscheid
> 
> 


Ticket Details
===================
Ticket ID: ZZW-779500
Department: Support IDV
Priority: Normal
Status: Open