rwgen.RainfallModel.postprocess

RainfallModel.postprocess(amax_durations=None, ddf_return_periods=None, amax_window_type='sliding', subset_length=200, output_filenames='default', calculate_statistics=True, simulation_format=None, start_year=None, timestep_length=None, calendar=None, simulation_subfolders=None, simulation_length=None, n_realisations=None)

Post-processing to calculate statistics from simulated point output.

Calculates statistics given in self.statistic_definitions by default, with options to extract annual maxima (AMAX) at multiple duration and to estimate depth-duration-frequency (DDF) statistics.

Parameters:

  • amax_durations (int or list of int) – Durations (in hours) for which annual maxima (AMAX) should be identified.

  • ddf_return_periods (int or list of int) – Return periods (in years) for which depths should be estimated, given the durations specified by amax_durations.

  • amax_window_type (str) – Use a 'sliding' (default) or 'fixed' window in AMAX extraction.

  • subset_length (int) – For splitting a realisation into subset_length years for calculating (seasonal) statistics (e.g. mean, variance, etc.). Does not affect AMAX extraction or DDF calculations.

  • output_filenames (str or dict) – Either key/value pairs indicating output file names, 'default' to use {'statistics': 'simulated_statistics.csv', 'amax': 'simulated_amax.csv', 'ddf': 'simulated_ddf.csv} or None to indicate that no output files should be written.

  • calculate_statistics (bool) – Indicates whether to calculate statistics (e.g. mean, variance, etc) for comparison with reference and/or fitted statistics. Default is True.

  • simulation_format (str) – Flag indicating point output file format (current option is txt).

  • start_year (int) – See self.simulate() arguments.

  • timestep_length (int) – See self.simulate() arguments.

  • calendar (str) – See self.simulate() arguments.

  • simulation_subfolders (dict) – See self.simulate() arguments (output_subfolders).

  • simulation_length (int) – See self.simulate() arguments.

  • n_realisations (int) – See self.simulate() arguments.

Notes

If the self.simulate() method has been run in the same session then the following arguments are not required: simulation_format, start_year, timestep_length, calendar, simulation_subfolders, simulation_length, n_realisations.