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

Re: FMRC Best: possible problem with "coordinates" attributes

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: FMRC Best: possible problem with "coordinates" attributes
  • Date: Sat, 10 Jul 2010 05:52:15 -0600
Ethan Davis wrote:

All the data variables in the FMRC Best Time Series data I'm looking at
are shaped like "time*,z,y,x" and such as one would expect. However, all
the "coordinates" attributes still include time*_run. The time*_run
coordinate variables are still there as well. They just aren't
referenced in any of the arrays.

Ethan


i decided to leave the run time coordinate in the coordinates attribute, as a 
way of formally associating it with the data variable, eg:

 float Precipitable_water(time=8, y=103, x=108);
    :units = "kg m-2";
    :long_name = "Precipitable water @ entire_atmosphere";
    :missing_value = -9999.0f; // float
    :grid_mapping = "Lambert_Conformal";
    :coordinates = "time_run time y x ";

note that the runtime coordinate uses time as its dimension:

double time_run(time=8);
    :long_name = "run times for coordinate = time";
    :standard_name = "forecast_reference_time";
    :units = "hours since 2006-09-25T06:00:00Z";
    :missing_value = NaN; // double
    :_CoordinateAxisType = "RunTime";