pyrad.proc.process_hydroclass

pyrad.proc.process_hydroclass(procstatus, dscfg, radar_list=None)

Classifies precipitation echoes

Parameters:

• procstatus (int) – Processing status: 0 initializing, 1 processing volume, 2 post-processing

• dscfg (dictionary of dictionaries) –

  data set configuration. Accepted Configuration Keywords:

  datatype : list of string. Dataset keyword
      The input data types, must contain
      "dBZ" or "dBZc", and,
      "ZDR" or "ZDRc", and,
      "RhoHV", or "uRhoHV", or "RhoHVc", and,
      "KDP", or "KDPc", and,
      "TEMP" or "H_ISO0" (optional)
  HYDRO_METHOD : string. Dataset keyword
      The hydrometeor classification method. One of the following:
      SEMISUPERVISED, UKMO
  centroids_file : string or None. Dataset keyword
      Used with HYDRO_METHOD SEMISUPERVISED. The name of the .csv file
      that stores the centroids. The path is given by
      [configpath]/centroids_hydroclass/
      If None is provided default centroids are going to be used
  compute_entropy : bool. Dataset keyword
      Used with HYDRO_METHOD SEMISUPERVISED. If true the entropy is
      computed and the field hydroclass_entropy is output
  output_distances : bool. Dataset keyword
      Used with HYDRO_METHOD SEMISUPERVISED. If true the de-mixing
      algorithm based on the distances to the centroids is computed and
      the field proportions of each hydrometeor in the radar range gate
      is output
  vectorize : bool. Dataset keyword
      Used with HYDRO_METHOD SEMISUPERVISED. If true a vectorized
      version of the algorithm is used
  weights : array of floats. Dataset keyword
      Used with HYDRO_METHOD SEMISUPERVISED. The list of weights given
      to each variable
  hydropath : string. Dataset keyword
      Used with HYDRO_METHOD UKMO. Directory of the UK MetOffice
      hydrometeor classification code
  mf_dir : string. Dataset keyword
      Used with HYDRO_METHOD UKMO. Directory where the UK MetOffice
      hydrometeor classification membership functions are stored
  ml_depth: float. Dataset keyword
      Used with HYDRO_METHOD UKMO. Depth of the melting layer [km].
      Default 500.
  perturb_ml_depth: float. Dataset keyword
      Used with HYDRO_METHOD UKMO. if specified, the depth of the
      melting layer can be varied by +/- this value [km], allowing a
      less-rigidly defined melting layer. Default 0.
  fzl: float or None. Dataset keyword
      If desired, a single freezing level
      height can be specified for the entire PPI domain. This will be used
      only if no temperature field is available.
  sounding : str. Dataset keyword
      The nearest radiosounding WMO code (5 int digits). It will be used to
      compute the freezing level, if no temperature field name is specified,
      if the temperature field is not in the radar object or if no
      fzl is explicitely defined.
  use_dualpol: Bool. Dataset keyword
      Used with HYDRO_METHOD UKMO. If false no radar data is used and
      the classification is performed using temperature information
      only. Default True
  use_temperature: Bool. Dataset keyword
      Used with HYDRO_METHOD UKMO. If false no temperature information
      is used and the classification is performed using radar data only.
      Default True
  use_interpolation: Bool. Dataset keyword
      Used with HYDRO_METHOD UKMO. If True gaps in the classification
      are filled using a nearest-neighbour interpolation. Default False
  map_to_semisupervised: Bool. Dataset keyword
      Used with HYDRO_METHOD UKMO. If True the output is map to the same
      categories as the semi-supervised classification. Default True
  append_all_fields: Bool. Dataset keyword
      Used with HYDRO_METHOD UKMO. If True auxiliary fields such as
      confidence and probability for each class are going to be added to
      the output
  

• radar_list (list of Radar objects) – Optional. list of radar objects

Returns:

• new_dataset (dict) – dictionary containing the output fields “hydro”, “entropy” (if compute_entropy is 1), and “propAG”, “propCR”, “propLR”, “propRP”, “propRN”, “propVI”, “propWS”, “propMH”, “propIH” (if output_distances is 1)

• ind_rad (int) – radar index