Represent CDF Time Values

This example shows how to extract date and time information from CDF datasets. CDF has three data types that represent date and time values, each of which can be represented in MATLAB®:

• CDF_TIME_TT2000 data measures the number of nanoseconds elapsed since 01-Jan-2000 at 12:00:00 Terrestrial Time (TT). CDF_TIME_TT2000 data can be represented in MATLAB as datetime or int64 values.

• CDF_EPOCH data measures the number of milliseconds elapsed since 01-Jan-0000. CDF_EPOCH data can be represented in MATLAB as datetime or double values.

• CDF_EPOCH16 data measures the number of picoseconds elapsed since 01-Jan-0000. CDF_EPOCH16 data can be represented in MATLAB as double values.

Represent CDF Time Values as int64 or double Values

Read the tt2000 (of type CDF_TIME_TT2000) and Time (of type CDF_EPOCH) variables from the sample CDF file example_364.cdf. Set the DatetimeType name-value argument to "native". cdfread returns a cell array.

data_native = cdfread("example_364.cdf","Variables",{"tt2000","Time"}, ...
                      "DatetimeType","native");
whos data_native

  Name              Size            Bytes  Class    Attributes

  data_native      24x2              6144  cell               

Save the two CDF variables as MATLAB vectors and examine the output. The variable with CDF_TIME_TT2000 data has type int64, while the variable with CDF_EPOCH data has type double.

cdf_time_tt2000_data_native = [data_native{:,1}];
cdf_epoch_data_native = [data_native{:,2}];
whos cdf*_native

  Name                             Size            Bytes  Class     Attributes

  cdf_epoch_data_native            1x24              192  double              
  cdf_time_tt2000_data_native      1x24              192  int64               

cdf_time_tt2000_data_native(1:3)'

ans = 3x1 int64 column vector

   536500865284200300
   536500866284200300
   536500867284200300

cdf_epoch_data_native(1:3)'

ans = 3×1
1013 ×

    6.3146
    6.3146
    6.3146

Represent CDF Time Values as datetime Values

Read the same two CDF variables again. This time, set the DatetimeType name-value argument to "datetime".

data_datetime = cdfread("example_364.cdf","Variables",{"tt2000","Time"}, ...
                        "DatetimeType","datetime");
whos data_datetime

  Name                Size            Bytes  Class    Attributes

  data_datetime      24x2              9664  cell               

Again save the two CDF variables as MATLAB vectors and examine the output. Both variables have type datetime.

cdf_time_tt2000_data_datetime = [data_datetime{:,1}];
cdf_epoch_data_datetime = [data_datetime{:,2}];
whos cdf*_datetime

  Name                               Size            Bytes  Class       Attributes

  cdf_epoch_data_datetime            1x24              240  datetime              
  cdf_time_tt2000_data_datetime      1x24              480  datetime              

cdf_time_tt2000_data_datetime(1:3)'

ans = 3x1 datetime
   2016-12-31T23:59:57.100200300Z
   2016-12-31T23:59:58.100200300Z
   2016-12-31T23:59:59.100200300Z

cdf_epoch_data_datetime(1:3)'

ans = 3x1 datetime
   01-Jan-2001 00:00:00.000
   01-Jan-2001 01:00:00.000
   01-Jan-2001 02:00:00.000