pysiral.retracker.tfmra

Attributes

__author__
CYTFMRA_OK

Classes

cTFMRA	The default TFMRA retracker implementation using cythonized
TFMRAMultiThresholdFreeboards	Level-2 processor step to compute elevations from a range of TFMRA thresholds.

Functions

bnsmooth(x, window)

Bottleneck implementation of the IDL SMOOTH function

Module Contents

pysiral.retracker.tfmra.__author__ = 'Stefan Hendricks <stefan.hendricks@awi.de>'

pysiral.retracker.tfmra.CYTFMRA_OK = True

class pysiral.retracker.tfmra.cTFMRA

Bases: pysiral.retracker.BaseRetracker

The default TFMRA retracker implementation using cythonized functions for numerical perfomrance

DOCSTR = 'Threshold first maximum retracker (TFMRA)'

set_default_options(specified_options=None)

Set the default options for the cTFMRA retracker. This can be modified by the specified_options keyword that will overwrite the values in the default options dictionary :param specified_options: (dict or dict-like) specific options :return:

property default_options_dict

create_retracker_properties(n_records)

Mandatory method, but unused :param n_records: :return:

l2_retrack(rng, wfm, indices, radar_mode, is_valid)

API Calling method for retrackers.

Parameters:

• rng – (np.array, dim:(n_records, n_bins)

• wfm

• indices

• radar_mode

• is_valid

Returns:

get_tfmra_threshold(indices)

Compute the TFMRA threshold for each waveform with several options. :param indices: A list of array indices for which to compute thresholds (e.g. for sea ice waveforms) :return: An array of thresholds with the dimensions of the l2 data object, but only values for <indices>

get_preprocessed_wfm(rng, wfm, radar_mode, is_valid)

Returns the intermediate product (oversampled range bins, oversampled and filtered waveforms, indices of first maxima and peak power norm for custom applications

get_thresholds_distance(rng, wfm, fmi, t0, t1, return_all_values=False)

Return the distance between two thresholds t0 < t1

static get_filtered_wfm(rng, wfm, oversampling_factor, window_size)

Return a filtered version of the waveform. This process inclused oversampling, smoothing and normalization to the first maximum power. :param rng: (np.array, dim:n_records) window delay for each range bin :param wfm: (np.array, dim:n_records) the power for each range bin with

sensor dependent units

Parameters:

• oversampling_factor – (int) The waveform oversamling factor

• window_size – (int) The filter size of the box filter

Returns:

static get_first_maximum_index(wfm, peak_minimum_power, first_valid_idx=0)

Return the index of the first peak (first maximum) on the leading edge before the absolute power maximum. The first peak is only valid if its power exceeds a certain threshold :param wfm: (np.array, dim=(n_range_bins)): Normalied waveform power :param peak_minimum_power: (float) threshold for normalized power that

a peak must surpass to be regarded as a first maximum candidate

Parameters:

first_valid_idx – (int):

Returns:

static get_threshold_range(rng: numpy.ndarray, wfm: numpy.ndarray, first_maximum_index: int, threshold: float, first_valid_idx: int = 0) → Tuple[float, float, int]

Return the range value and the power of the retrack point at a given threshold of the firsts maximum power

Parameters:

• rng – (np.array, dim=(n_range_bins) Window delay in meters

• wfm – (np.array, dim=(n_range_bins) Waveform power in normalized units

• first_maximum_index – (int) Index of first maximum

• threshold – (float) Power threshold

• first_valid_idx – (int) First valid index for first maximum / leading edge

Returns:

tfmra range (float), tfmra power (float)

class pysiral.retracker.tfmra.TFMRAMultiThresholdFreeboards(*args, **kwargs)

Bases: pysiral.l2proc.procsteps.Level2ProcessorStep

Level-2 processor step to compute elevations from a range of TFMRA thresholds. NOTE: Computational expensive, should be handled with care

execute_procstep(l1b, l2)

Computes the elevation of a range of defined of TFMRA retrackers thresholds to all lead or ice elevations. :param l1b: :param l2: :return: error_status_flag

property l2_input_vars

property l2_output_vars

property auxid_fmt

property error_bit

pysiral.retracker.tfmra.bnsmooth(x, window)

Bottleneck implementation of the IDL SMOOTH function