harpcollocate

The tool harpcollocate is used to derive a list of measurements that match in time, latitude and longitude for two sets of HARP compliant HDF4/HDF5/netCDF files.

Usage:
  harpcollocate sub-command [options]
    Determine the collocation filter for two sets of HARP files.

    Available sub-commands:
      matchup
      resample
      update

    Use 'harpcollocate <sub-command> --help' to get help on a specific
    sub-command.

  harpcollocate -h, --help
    Show help (this text).

  harpcollocate -v, --version
    Print the version number of HARP and exit.

Collocation

Collocation of two sets of files involves the following steps:

1. harpcollocate matchup is used to derive the collocation result file.
2. harpcollocate resample is optionally used to further process the collocation result file.
3. harpconvert -fca ... is used together with the collocation result file to select the matching measurements in each file of the primary dataset
4. harpconvert -fcb ... is used together with the collocation result file to select the matching measurements in each file of the secondary dataset

When needed, harpcollocate update can be used to refresh the collocation result file.

Obtaining collocation result file

The tool harpcollocate matchup is used to derive a list of measurement pairs that match in time, latitude and longitude.

Usage:
  harpcollocate matchup [options]
    Determine the collocation filter for two sets of HARP files,
    and optionally resample the collocation result

    Options:

      -h, --help
           Show matchup help (this text)
      -ia, --input-a <input>
           Specifies directory or names of input files of dataset A
      -ib, --input-b <input>
           Specifies directory or names of input files of dataset B
      -or, --output-result <output>
           Specifies collocation result file (comma separated values)

      Collocation options,
      set at least one of the following ([unit] is optional):
      -dt 'value [unit]'          : sets maximum allowed difference in time
      -dp 'value [unit]'          : sets maximum allowed point distance
      -dlat 'value [unit]'        : sets maximum allowed point difference
                                    in latitude
      -dlon 'value [unit]'        : sets maximum allowed point difference
                                    in longitude
      -da 'value [unit]'          : sets minimum allowed overlapping
                                    percentage of polygon areas
      -dsza 'value [unit]'        : sets allowed maximum difference
                                    in solar zenith angle
      -dsaa 'value [unit]'        : sets allowed maximum difference
                                    in solar azimuth angle
      -dvza 'value [unit]'        : sets allowed maximum difference
                                    in viewing zenith angle
      -dvaa 'value [unit]'        : sets allowed maximum difference
                                    in viewing azimuth angle
      -dtheta 'value [unit]'      : sets allowed maximum difference
                                    in scattering angle
      -overlap                    : sets that polygon areas must overlap
      -painab                     : sets that points of dataset A must fall
                                    in polygon areas of B
      -pbinaa                     : sets that points of dataset B must fall
                                    in polygon areas of A
      When '[unit]' is not specified, a default unit is used:
        Criteria; [default unit]
        -dt; [s]
        -dp; [m]
        -dlat; [degree_north]
        -dlon; [degree_east]
        -da; [%]
        -dsza, -dsaa, -dvza, -dvaa, -dvaa, -dtheta; [degree]

      Resampling options:
      -Rnna, --nearest-neighbour-a: keep only nearest neighbour,
                                    dataset A is the master dataset
      -Rnnb, --nearest-neighbour-b: keep only nearest neighbour,
                                    dataset B is the master dataset
      The nearest neighbour is the sample with which the squared sum
      of the weighted differences is minimal
      When resampling is set to 'Rnna' and/or 'Rnnb',
      the following parameters can be set:
      -wft 'value [unit]'         : sets the weighting factor for time
      -wfdp 'value [unit]'        : sets the weighting factor for
                                    point distance
      -wfa 'value [unit]'         : sets the weighting factor for
                                    overlapping percentage
      -wfsza 'value [unit]'       : sets the weighting factor
                                    for solar zenith angle
      -wfsaa 'value [unit]'       : sets the weighting factor
                                    for solar azimuth angle
      -wfvza 'value [unit]'       : sets the weighting factor
                                    for viewing zenith angle
      -wfvaa 'value [unit]'       : sets the weighting factor
                                    for viewing azimuth angle
      -wftheta 'value [unit]'     : sets the weighting factor
                                    for scattering angle
      When '[unit]' is not specified in the above, a default unit will be
      adopted:
        Weighting factors; [default unit]
        -wft; [1/s]
        -wfdp; [1/m]
        -wfa; [1/%]
        -wfsza, -wfsaa, -wfvza, -wfvaa, -wfvaa, -wftheta; [1/degree]
      When a weighting factor is not set, a default value of 1 and
      the default unit are adopted. Recommend value and unit for the
      weighting factors are the reciprocals of the corresponding
      collocation criteria value and unit that is used.

Resampling collocation result file

The command harpcollocate resample is used to apply resampling on the collocation result file. For example, to limit a series of matches to only the nearest neighbour.

Usage:
  harpcollocate resample [options]
    Resample an existing collocation result file

    Options:

      -h, --help
           Show resample help (this text)
      -ir, --input-result <input>
           Input collocation result file (comma separated values)
      -or, --output-result <output>
           Create a new file, and do not overwrite the input
           collocation result file

      Resampling options:
      -Rnna, --nearest-neighbour-a: keep only nearest neighbour,
                                    dataset A is the master dataset
      -Rnnb, --nearest-neighbour-b: keep only nearest neighbour,
                                    dataset B is the master dataset
      The nearest neighbour is the sample with which the squared sum
      of the weighted differences is minimal
      When resampling is set to 'Rnna' and/or 'Rnnb',
      the following parameters can be set:
      -wft 'value [unit]'         : sets the weighting factor for time
      -wfdp 'value [unit]'        : sets the weighting factor for
                                    point distance
      -wfa 'value [unit]'         : sets the weighting factor for
                                    overlapping percentage
      -wfsza 'value [unit]'       : sets the weighting factor
                                    for solar zenith angle
      -wfsaa 'value [unit]'       : sets the weighting factor
                                    for solar azimuth angle
      -wfvza 'value [unit]'       : sets the weighting factor
                                    for viewing zenith angle
      -wfvaa 'value [unit]'       : sets the weighting factor
                                    for viewing azimuth angle
      -wftheta 'value [unit]'     : sets the weighting factor
                                    for scattering angle
      When '[unit]' is not specified in the above, a default unit will be
      adopted:
        Weighting factors; [default unit]
        -wft; [1/s]
        -wfdp; [1/m]
        -wfa; [1/%]
        -wfsza, -wfsaa, -wfvza, -wfvaa, -wfvaa, -wftheta; [1/degree]
      When a weighting factor is not set, a default value of 1 and
      the default unit are adopted. Recommend value and unit for the
      weighting factors are the reciprocals of the corresponding
      collocation criteria value and unit that is used.

Updating collocation result file

The command harpcollocate update is used to:

1. Determine which files in the collocation result file are still in existance
2. Update the collocation result file accordingly

Usage:
  harpcollocate update [options]
    Update an existing collocation result file by checking
    the measurements in two sets of HARP files that still exist

    Options:
      -ia, --input-a <input>
           Specifies directory or names of input files of dataset A
      -ib, --input-b <input>
           Specifies directory or names of input files of dataset B
      -ir, --input-result <input>
           Input collocation result file (comma separated values)
      -or, --output-result <output>
           Create a new file, and do not overwrite the input
           collocation result file