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

[IDV #WON-355400]: Problematic point netcdf file

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [IDV #WON-355400]: Problematic point netcdf file
  • Date: Thu, 19 Feb 2015 23:00:03 -0700
Hi David,

Oops, you've been trying to replicate the wrong file. The THREDDS page I 
pointed you to has 3 access methods: OPENDAP, NetcdfSubset, and HTTPServer. I 
meant for you to download from the HTTPServer service, not NetcdfSubset. My 
fault, I should have been more clear.

The original H.2.1.1.nc file, which HTTPServer will return to you unaltered, 
uses the Orthogonal multidimensional array representation of time series [1]. 
The NetcdfSubset service rewrote the original file to be in Indexed ragged 
array representation [2].

Orthogonal is what you want. For clarity, I'll go ahead and paste the ncdump of 
the correct file:

netcdf C:/Users/cwardgar/Desktop/H.2.1.1.nc {
  dimensions:
    station = 10;
    time = 5;
    name_strlen = 3;
  variables:
    float humidity(station=10, time=5);
      :standard_name = "specific_humidity";
      :coordinates = "time lat lon alt";

    double time(time=5);
      :standard_name = "time";
      :long_name = "time of measurement";
      :units = "days since 1970-01-01 00:00:00";

    float lon(station=10);
      :standard_name = "longitude";
      :long_name = "station longitude";
      :units = "degrees_east";

    float lat(station=10);
      :standard_name = "latitude";
      :long_name = "station latitude";
      :units = "degrees_north";

    float alt(station=10);
      :long_name = "vertical distance above the surface";
      :standard_name = "height";
      :units = "m";
      :positive = "up";
      :axis = "Z";

    char station_name(station=10, name_strlen=3);
      :long_name = "station name";
      :cf_role = "timeseries_id";

  // global attributes:
  :featureType = "timeSeries";
  :Conventions = "CF-1.6";
}

If you can get your dataset to look like that, you'll be good to go. You may 
wonder about the "station_name" variable and its "cf_role" attribute. From the 
CF docs [3]:

"It is strongly recommended that there should be a station variable (which may 
be of any type) with the attribute cf_role=”timeseries_id”, whose values 
uniquely identify the stations."

So, the values needn't be names or even strings, but they ought to be unique.

Cheers,
Christian

[1] 
http://cfconventions.org/Data/cf-conventions/cf-conventions-1.7/build/cf-conventions.html#idm43165546432
[2] 
http://cfconventions.org/Data/cf-conventions/cf-conventions-1.7/build/cf-conventions.html#idm43165513616
[3] 
http://cfconventions.org/Data/cf-conventions/cf-conventions-1.7/build/cf-conventions.html#time-series-data


Ticket Details
===================
Ticket ID: WON-355400
Department: Support netCDF Java
Priority: Normal
Status: Closed