pyrad.proc.process_phidp_kdp_lp

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

Estimates PhiDP and KDP using a linear programming algorithm.

This method only retrieves data in rain (i.e. below the melting layer).

The method has 3 steps: PhiDP unfolding (including clutter suppression), Processing PhiDP using the LP algorithm, Obtaining KDP by convoluting PhiDP with a Sobel window.

Required inputs for the LP algorithm are PsiDP and reflectivity.

RhoHV and SNR are used for the clutter suppression in the PhiDP unfolding step (note that SNR is used instead of Normalized Coherent Power used by the original algorithm). If they are not provided a gate_filter based on the values of reflectivity is used instead.

Freezing level height can be retrieved from iso-0 or temperature fields, from radio sounding or set by the user.

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,
      "PhiDP" or "PhiDPc" or "uPhiDP", and,
      "RhoHV" or "RhoHVc", (Optional, used when min_rhv is specified) and,
      "SNRh" (Optional, used when min_snr is specified), and
      "TEMP" or "H_ISO0" (Optional)
  fzl : float. Dataset keyword
      The freezing level height [m]. Default 2000.
  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 freezing_level is explicitely defined.
  ml_thickness : float. Dataset keyword
      The melting layer thickness in meters. Default 700.
  beamwidth : float. Dataset keyword
      the antenna beamwidth [deg]. If None that of the keys
      radar_beam_width_h or radar_beam_width_v in attribute
      instrument_parameters of the radar object will be used. If the key
      or the attribute are not present the beamwidth will be set to None
  LP_solver : str.
      The LP solver to use. Can be pyglpk, cvxopt, cylp, cylp_mp.
      Default cvxopt
  proc : int
      Number of worker processes. Only used when LP_solver is cylp_mp.
      Default 1
  min_snr, min_rhv : float
      Minimum SNR and RhoHV. Used to filter out non-meteorological
      echoes when performing the unfolding of the differential phase.
      Default 10 and 0.6
  sys_phase : float
      Default system differential phase offset in deg. Default 0.
  overide_sys_phase : bool
      If True the dynamic sys_phase not computed and the default
      sys_phase is used. If false a dynamic sys_phase is computed. If
      no dynamic value is found the default sys_phase is used.
      Default False
  first_gate_sysp : int
      First gate to use when determining the system differential phase
      offset. Default 0
  nowrap : int or None
      Gate number where to begin the phase unwrapping. None will unwrap
      from gate 0. Default None
  ncpts : int
      Minimum number of continuous valid PhiDP points. Segments below
      this number or starting at a gate below this number are going to
      be excluded from the unfolding. Default 2.
  z_bias : float
      reflectivity bias. Default 0 dBZ
  low_z, high_z : float
      mininum and maximum reflectivity values. Values beyond this range
      are going to be set to this range limits. The modified
      reflectivity is used in the LP algorithm. Default 10 and 53 dBZ
  min_phidp : float
      minimum differential phase. PhiDP values below this threshold are
      going to be set to the threshold values. The modified PhiDP is
      used in the LP algorithm. Default 0.1 deg
  doc : int
      Number of gates to doc at the end of the ray. Used in the LP
      algorithm. Default 0
  self_const : float
      selfconsistency factor. Used in the LP algorithm. Default 60000
  coef : float
      Exponent linking Z to KDP in selfconsistency. Used in the LP
      algorithm. kdp = (10**(0.1z))**coef. Default 0.914
  window_len : int
      Length of Sobel Window applied to PhiDP field before computing
      KDP. Default 35
  

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

Returns:

• new_dataset (dict) – dictionary containing the output field “PhiDPc” and “KDPc”

• ind_rad (int) – radar index