pycmt3d package

Submodules

pycmt3d.cmt3d module

class pycmt3d.cmt3d.Cmt3D(cmtsource, data_container, config)

Bases: object

Class that handles the solver part of source inversion

Parameters:

• cmtsource (pycmt3d.CMTSource) – earthquake source
• data_container – all data and window
• config (pycmt3d.Config) – configuration for source inversion

apply_station_correction(obsd, synt, istart, iend)

Apply station correction on windows based one cross-correlation time shift if config.station_correction

Parameters:

• obsd –
• synt –
• istart –
• iend –

Returns:

calculate_azimuth_bin()

Calculate the azimuth and sort them into bins

Returns:

calculate_criteria(obsd, synt, istart, iend)

Calculate the time shift, max cross-correlation value and energy differnce

Parameters:

• obsd (obspy.core.trace.Trace) – observed data trace
• synt (obspy.core.trace.Trace) – synthetic data trace
• istart (int) – start index of window
• iend – end index of window
• iend – int

Returns:

[number of shift points, max cc value, dlnA]

Return type:

[int, float, float]

calculate_dsyn(datalist)

Calculate dsyn matrix based on perturbed seismograms

Parameters:datalist – Returns:

calculate_kai_total_value()

Calculate the sum of kai value

Returns:

calculate_var_one_trace(obsd, synt, win_time)

Calculate the variance reduction on a pair of obsd and synt and windows

Parameters:

• obsd (obspy.core.trace.Trace) – observed data trace
• synt (obspy.core.trace.Trace) – synthetic data trace
• win_time (list or numpy.array) – [win_start, win_end]

Returns:

waveform misfit reduction and observed data energy [v1, d1]

Return type:

[float, float]

calculate_variance()

Calculate variance reduction based on old and new source solution

Returns:

compute_A_b(window, win_idx, dsyn)

Calculate the matrix A and vector b based on one pair of observed data and synthetic data on a given window.

Parameters:

• window (pycmt3d.Window) – data and window information
• win_idx (integer) – window index(a specific window)
• dsyn (numpy.array) – derivative synthetic data matrix

Returns:

compute_new_syn(datalist, dm)

Compute new synthetic data based on new CMTSOLUTION

Parameters:

• datalist – dictionary of all data
• dm – CMTSolution perterbation, i.e., (self.new_cmt_par-self.cmt_par)

Returns:

static construct_hanning_taper(npts)

Hanning taper construct

Parameters:npts – number of points Returns:

convert_new_cmt_par()

Convert self.new_cmt_par array to CMTSource instance

Returns:

static get_azimuth_bin_number(azimuth)

Calculate the bin number of a given azimuth

Parameters:azimuth – test test test Returns:

get_station_info(datalist)

Using the event location and station information to calculate azimuth and distance !!! Obsolete, not used any more !!!

Parameters:datalist – data dictionary(referred to pycmt3d.Window.datalist) Returns:

inversion_result_table()

Print out the inversion table

Returns:

invert_bootstrap()

It is used to evaluate the mean, standard deviation, and variance of new parameters

Returns:

invert_cmt()

ensemble all measurements together to form Matrix A and vector b to solve the A * (dm) = b A is the Hessian Matrix and b is the misfit

Returns:

invert_solver(A, b, print_mode=False)

Solver part. Hession matrix A and misfit vector b will be reconstructed here based on different constraints.

Parameters:

• A – basic Hessian matrix
• b – basic misfit vector
• print_mode – if True, then print out log information; if False, then no log information

Returns:

linear_solver(old_par, A, b, npar, na)

if invert for moment tensor with zero-trace constraints or no constraint

nonlinear_solver(old_par, A, b, npar, na)

if invert for moment tensor with double couple constraints setup starting solution, solve directly for moment instead of dm, exact implementation of (A16) logger.info(‘Non-linear Inversion’)

Returns:

normalize_weight()

Normalize the weighting and make the maximum to 1

Returns:

plot_stats_histogram(outputdir='.', format='png')

Plot inversion histogram

Parameters:outputdir – Returns:

plot_summary(outputdir='.', format='png')

Plot inversion summary

Parameters:outputdir – output directory Returns:

prepare_for_weighting()

Prepare necessary information for weighting, e.x., calculating azimuth, distance and energty of a window. Also, based on the tags, sort window into different categories.

Returns:

static print_cmtsource_summary(cmt)

Print CMTSolution source summary

Returns:

print_inversion_summary()

Print out the inversion summary

Returns:

setup_matrix()

Calculate A and b for all windows

Returns:

setup_weight(weight_mode='num_wins')

Use Window information to setup weight.

Returns:

setup_weight_for_category(window)

Setup weight for each category if config.normalize_category window_weight = window_weight / N_windows_in_category

Parameters:window – Returns:

setup_weight_for_location(window, naz_bin, naz_bin_all)

setup weight from location information, including distance, component and azimuth

Parameters:

• window –
• naz_bin –
• naz_bin_all –

Returns:

source_inversion()

the Source Inversion method :return:

write_new_cmtfile(outputdir='.')

Write new_cmtsource into a file

write_new_syn(outputdir='.', format='sac')

pycmt3d.config module

Configuration object for pycmt3d

class pycmt3d.config.Config(npar, dlocation=0.0, ddepth=0.0, dmoment=0.0, weight_data=True, weight_function=None, weight_azi_mode='num_files', normalize_window=True, norm_mode='data_only', normalize_category=False, station_correction=True, zero_trace=True, double_couple=False, lamda_damping=0.0, bootstrap=True, bootstrap_repeat=300)

Bases: object

Configuration for source inversion

Parameters:

• npar – number of parameters to be inverted
• dlocation – location perturbation when calculated perturbed synthetic data
• ddepth – depth perturbation
• dmoment – moment perturbation
• weight_data – bool value of weighting data
• weight_function – weighting function
• normalize_window – add window energy into the weighting term
• norm_mode – two modes: 1) “data_and_synt” 2) “data_only”
• station_correction – bool value of whether applies station correction
• zero_trace – bool value of whether applies zero-trace constraint
• double_couple – bool value of whether applied double-couple constraint
• lamda_damping – damping coefficient
• bootstrap – bool value of whether applied bootstrap method
• bootstrap_repeat – bootstrap iterations

print_summary()

Print function of configuration

Returns:

pycmt3d.const module

All the constants used in pycmt3d

pycmt3d.source module

Source and Receiver classes of Instaseis.

copyright:Lion Krischer (krischer@geophysik.uni-muenchen.de), 2014 Martin van Driel (Martin@vanDriel.de), 2014 license:GNU Lesser General Public License, Version 3 (http://www.gnu.org/copyleft/lgpl.html)

class pycmt3d.source.CMTSource(origin_time=UTCDateTime(1970, 1, 1, 0, 0), pde_latitude=0.0, pde_longitude=0.0, mb=0.0, ms=0.0, pde_depth_in_m=None, region_tag=None, eventname=None, cmt_time=0.0, half_duration=0.0, latitude=0.0, longitude=0.0, depth_in_m=None, m_rr=0.0, m_tt=0.0, m_pp=0.0, m_rt=0.0, m_rp=0.0, m_tp=0.0)

Bases: object

Class to handle a seismic moment tensor source including a source time function.

M0

Scalar Moment M0 in Nm

static adjust_eventname(eventname)

Remove the leading char :param eventname: :return:

classmethod from_CMTSOLUTION_file(filename)

Initialize a source object from a CMTSOLUTION file.

Parameters:filename – path to the CMTSOLUTION file

classmethod from_quakeml_file(filename)

Initizliaze a source object from a quakeml file :param filename: path to a quakeml file

moment_magnitude

Moment magnitude M_w

tensor

List of moment tensor components in r, theta, phi coordinates: [m_rr, m_tt, m_pp, m_rt, m_rp, m_tp]

tensor_voigt

List of moment tensor components in theta, phi, r coordinates in Voigt notation: [m_tt, m_pp, m_rr, m_rp, m_rt, m_tp]

time_shift

Time shift between cmtsolution and pdesolution

write_CMTSOLUTION_file(filename)

Initialize a source object from a CMTSOLUTION file.

Parameters:filename – path to the CMTSOLUTION file

pycmt3d.window module

class pycmt3d.window.DataContainer(par_list=[])

Bases: object

Class that contains methods that load data and window information

add_measurements_from_asdf(flexwinfile, asdf_file_dict, obsd_tag=None, synt_tag=None, stationfile=None, initial_weight=1.0)

Add measurments(window and data) from the given flexwinfile and the data format should be asdf. Usually, you can leave the obsd_tag=None and synt_tag=None unless if you have multiple tags in asdf file.

Parameters:

• flexwinfile –
• asdf_file_dict –
• tag –

Returns:

add_measurements_from_sac(flexwinfile, tag='untaged', initial_weight=1.0, load_mode='obsolute_time')

Add measurments(window and data) from the given flexwinfile and the data format should be sac

Parameters:flexwinfile – Returns:

check_and_load_asdf_file(asdf_file_dict)

get_stationxml_from_asdf(station_string, asdf_handle)

Used to extrace station location information from stationxml in asdf

get_trace_from_asdf(station_string, asdf_handle, tag)

Used to extract a specific trace out of an asdf file.

Parameters:

• station_string –
• asdf_handle –
• tag –

Returns:

load_data_from_asdf(win, asdf_ds, obsd_tag=None, synt_tag=None, station_dict=None)

load data from asdf file

Returns:

load_data_from_sac(win_obj, tag=None, mode=None)

Old way of loading obsd and synt data...

Parameters:win_obj – Returns:

static load_station_from_text(stationfile)

static load_winfile(flexwin_file, initial_weight=1.0)

old way of loading flexwin outputfile

print_summary()

Print summary of data container

Returns:

write_new_syn_file(format='sac', outputdir='.')

Write out new synthetic file based on new cmtsolution :return:

class pycmt3d.window.Window(station=None, network=None, location=None, component=None, num_wins=0, win_time=None, weight=None, obsd_fn=None, synt_fn=None, datalist=None, tag=None, source=None)

Bases: object

Obsd, synt, deriv synt trace and window information from one component of one station. Also, window weighting, station azimuth, distance is also included.

get_location_info(cmtsource)

calculating azimuth and distance, and then store it

Parameters:cmtsource – cmt source Returns:

win_energy(mode='data_and_synt')

Calculate energy inside the window

Parameters:mode – if mode == ‘data_and_synt’, then energy is calculated by sqrt(obsd_energy * synt_energy); if mode == ‘data_only’, then energy is calculated by obsd_energy only

Module contents