Source code for obspy.sac.sacio

#!/usr/bin/env python
# ------------------------------------------------------------------
# Filename: sacio.py
#  Purpose: Read & Write Seismograms, Format SAC.
#   Author: Yannik Behr, C. J. Ammon's, C. Satriano, L. Krischer
#    Email: yannik.behr@vuw.ac.nz
#
# Copyright (C) 2008-2015 Yannik Behr, C. J. Ammon's, C. Satriano,
#                         L. Krischer
# ------------------------------------------------------------------
"""
Low-level module internally used for handling SAC files

An object-oriented version of C. J. Ammon's SAC I/O module.

:copyright:
    The ObsPy Development Team (devs@obspy.org) & C. J. Ammon
:license:
    GNU Lesser General Public License, Version 3
    (http://www.gnu.org/copyleft/lesser.html)
"""

from __future__ import (absolute_import, division, print_function,
                        unicode_literals)
from future.builtins import *  # NOQA
from future.utils import native_str

import os
import time
import warnings

import numpy as np

from obspy import Trace, UTCDateTime
from obspy.core.compatibility import frombuffer
from obspy.core.util import AttribDict, gps2DistAzimuth


# we put here everything but the time, they are going to stats.starttime
# left SAC attributes, right trace attributes, see also
# http://www.iris.edu/KB/questions/13/SAC+file+format
convert_dict = {'npts': 'npts',
                'delta': 'delta',
                'kcmpnm': 'channel',
                'kstnm': 'station',
                'scale': 'calib',
                'knetwk': 'network',
                'khole': 'location'}

# all the sac specific extras, the SAC reference time specific headers are
# handled separately and are directly controlled by trace.stats.starttime.
SAC_EXTRA = ('depmin', 'depmax', 'odelta', 'o', 'a', 't0', 't1', 't2', 't3',
             't4', 't5', 't6', 't7', 't8', 't9', 'f', 'stla', 'stlo', 'stel',
             'stdp', 'evla', 'evlo', 'evdp', 'mag', 'user0', 'user1', 'user2',
             'user3', 'user4', 'user5', 'user6', 'user7', 'user8', 'user9',
             'dist', 'az', 'baz', 'gcarc', 'depmen', 'cmpaz', 'cmpinc',
             'nvhdr', 'norid', 'nevid', 'nwfid', 'iftype', 'idep', 'iztype',
             'iinst', 'istreg', 'ievreg', 'ievtype', 'iqual', 'isynth',
             'imagtyp', 'imagsrc', 'leven', 'lpspol', 'lovrok', 'lcalda',
             'kevnm', 'ko', 'ka', 'kt0', 'kt1', 'kt2', 'kt3', 'kt4', 'kt5',
             'kt6', 'kt7', 'kt8', 'kt9', 'kf', 'kuser0', 'kuser1', 'kuser2',
             'kdatrd', 'kinst', 'cmpinc', 'xminimum', 'xmaximum', 'yminimum',
             'ymaximum', 'unused6', 'unused7', 'unused8', 'unused9',
             'unused10', 'unused11', 'unused12', 'unused13')

FDICT = {'delta': 0, 'depmin': 1, 'depmax': 2, 'scale': 3,
         'odelta': 4, 'b': 5, 'e': 6, 'o': 7, 'a': 8, 'int1': 9,
         't0': 10, 't1': 11, 't2': 12, 't3': 13, 't4': 14,
         't5': 15, 't6': 16, 't7': 17, 't8': 18, 't9': 19,
         'f': 20, 'stla': 31, 'stlo': 32, 'stel': 33, 'stdp': 34,
         'evla': 35, 'evlo': 36, 'evdp': 38, 'mag': 39,
         'user0': 40, 'user1': 41, 'user2': 42, 'user3': 43,
         'user4': 44, 'user5': 45, 'user6': 46, 'user7': 47,
         'user8': 48, 'user9': 49, 'dist': 50, 'az': 51,
         'baz': 52, 'gcarc': 53, 'depmen': 56, 'cmpaz': 57,
         'cmpinc': 58, 'xminimum': 59, 'xmaximum': 60,
         'yminimum': 61, 'ymaximum': 62, 'unused6': 63,
         'unused7': 64, 'unused8': 65, 'unused9': 66,
         'unused10': 67, 'unused11': 68, 'unused12': 69}

IDICT = {'nzyear': 0, 'nzjday': 1, 'nzhour': 2, 'nzmin': 3,
         'nzsec': 4, 'nzmsec': 5, 'nvhdr': 6, 'norid': 7,
         'nevid': 8, 'npts': 9, 'nwfid': 11,
         'iftype': 15, 'idep': 16, 'iztype': 17, 'iinst': 19,
         'istreg': 20, 'ievreg': 21, 'ievtype': 22, 'iqual': 23,
         'isynth': 24, 'imagtyp': 25, 'imagsrc': 26,
         'leven': 35, 'lpspol': 36, 'lovrok': 37,
         'lcalda': 38, 'unused13': 39}

SDICT = {'kstnm': 0, 'kevnm': 1, 'khole': 2, 'ko': 3, 'ka': 4,
         'kt0': 5, 'kt1': 6, 'kt2': 7, 'kt3': 8, 'kt4': 9,
         'kt5': 10, 'kt6': 11, 'kt7': 12, 'kt8': 13,
         'kt9': 14, 'kf': 15, 'kuser0': 16, 'kuser1': 17,
         'kuser2': 18, 'kcmpnm': 19, 'knetwk': 20,
         'kdatrd': 21, 'kinst': 22}

TWO_DIGIT_YEAR_MSG = ("SAC file with 2-digit year header field encountered. "
                      "This is not supported by the SAC file format standard. "
                      "Prepending '19'.")


[docs]class SacError(Exception): """ Raised if the SAC file is corrupt or if necessary information in the SAC file is missing. """ pass
[docs]class SacIOError(Exception): """ Raised if the given SAC file can't be read. """ pass
[docs]class SacIO(object): """ Class for SAC file IO. Functions are given below, attributes/header fields (described below) can be directly accessed (via the :meth:`~obspy.sac.sacio.SacIO.__getattr__` method, see the link for an example). .. rubric::Description of attributes/header fields (based on SacIris_). .. _SacIris: http://www.iris.edu/manuals/sac/SAC_Manuals/FileFormatPt2.html ============ ==== ========================================================= Field Name Type Description ============ ==== ========================================================= npts N Number of points per data component. [required] nvhdr N Header version number. Current value is the integer 6. Older version data (NVHDR < 6) are automatically updated when read into sac. [required] b F Beginning value of the independent variable. [required] e F Ending value of the independent variable. [required] iftype I Type of file [required]: * ITIME {Time series file} * IRLIM {Spectral file---real and imaginary} * IAMPH {Spectral file---amplitude and phase} * IXY {General x versus y data} * IXYZ {General XYZ (3-D) file} leven L TRUE if data is evenly spaced. [required] delta F Increment between evenly spaced samples (nominal value). [required] odelta F Observed increment if different from nominal value. idep I Type of dependent variable: * IUNKN (Unknown) * IDISP (Displacement in nm) * IVEL (Velocity in nm/sec) * IVOLTS (Velocity in volts) * IACC (Acceleration in nm/sec/sec) scale F Multiplying scale factor for dependent variable [not currently used] depmin F Minimum value of dependent variable. depmax F Maximum value of dependent variable. depmen F Mean value of dependent variable. nzyear N GMT year corresponding to reference (zero) time in file. nyjday N GMT julian day. nyhour N GMT hour. nzmin N GMT minute. nzsec N GMT second. nzmsec N GMT millisecond. iztype I Reference time equivalence: * IUNKN (5): Unknown * IB (9): Begin time * IDAY (10): Midnight of reference GMT day * IO (11): Event origin time * IA (12): First arrival time * ITn (13-22): User defined time pick n, n=0,9 o F Event origin time (seconds relative to reference time.) a F First arrival time (seconds relative to reference time.) ka K First arrival time identification. f F Fini or end of event time (seconds relative to reference time.) tn F User defined time {ai n}=0,9 (seconds picks or markers relative to reference time). kt{ai n} K A User defined time {ai n}=0,9. pick identifications kinst K Generic name of recording instrument iinst I Type of recording instrument. [currently not used] knetwk K Name of seismic network. kstnm K Station name. istreg I Station geographic region. [not currently used] stla F Station latitude (degrees, north positive) stlo F Station longitude (degrees, east positive). stel F Station elevation (meters). [not currently used] stdp F Station depth below surface (meters). [not currently used] cmpaz F Component azimuth (degrees, clockwise from north). cmpinc F Component incident angle (degrees, from vertical). kcmpnm K Component name. lpspol L TRUE if station components have a positive polarity (left-hand rule). kevnm K Event name. ievreg I Event geographic region. [not currently used] evla F Event latitude (degrees north positive). evlo F Event longitude (degrees east positive). evel F Event elevation (meters). [not currently used] evdp F Event depth below surface (meters). [not currently used] mag F Event magnitude. imagtyp I Magnitude type: * IMB (Bodywave Magnitude) * IMS (Surfacewave Magnitude) * IML (Local Magnitude) * IMW (Moment Magnitude) * IMD (Duration Magnitude) * IMX (User Defined Magnitude) imagsrc I Source of magnitude information: * INEIC (National Earthquake Information Center) * IPDE (Preliminary Determination of Epicenter) * IISC (International Seismological Centre) * IREB (Reviewed Event Bulletin) * IUSGS (US Geological Survey) * IBRK (UC Berkeley) * ICALTECH (California Institute of Technology) * ILLNL (Lawrence Livermore National Laboratory) * IEVLOC (Event Location (computer program) ) * IJSOP (Joint Seismic Observation Program) * IUSER (The individual using SAC2000) * IUNKNOWN (unknown) ievtyp I Type of event: * IUNKN (Unknown) * INUCL (Nuclear event) * IPREN (Nuclear pre-shot event) * IPOSTN (Nuclear post-shot event) * IQUAKE (Earthquake) * IPREQ (Foreshock) * IPOSTQ (Aftershock) * ICHEM (Chemical explosion) * IQB (Quarry or mine blast confirmed by quarry) * IQB1 (Quarry/mine blast with designed shot info-ripple fired) * IQB2 (Quarry/mine blast with observed shot info-ripple fired) * IQMT (Quarry/mining-induced events: tremors and rockbursts) * IEQ (Earthquake) * IEQ1 (Earthquakes in a swarm or aftershock sequence) * IEQ2 (Felt earthquake) * IME (Marine explosion) * IEX (Other explosion) * INU (Nuclear explosion) * INC (Nuclear cavity collapse) * IO\_ (Other source of known origin) * IR (Regional event of unknown origin) * IT (Teleseismic event of unknown origin) * IU (Undetermined or conflicting information) * IOTHER (Other) nevid N Event ID (CSS 3.0) norid N Origin ID (CSS 3.0) nwfid N Waveform ID (CSS 3.0) khole k Hole identification if nuclear event. dist F Station to event distance (km). az F Event to station azimuth (degrees). baz F Station to event azimuth (degrees). gcarc F Station to event great circle arc length (degrees). lcalda L TRUE if DIST AZ BAZ and GCARC are to be calculated from st event coordinates. iqual I Quality of data [not currently used]: * IGOOD (Good data) * IGLCH (Glitches) * IDROP (Dropouts) * ILOWSN (Low signal to noise ratio) * IOTHER (Other) isynth I Synthetic data flag [not currently used]: * IRLDTA (Real data) * ????? (Flags for various synthetic seismogram codes) user{ai n} F User defined variable storage area {ai n}=0,9. kuser{ai n} K User defined variable storage area {ai n}=0,2. lovrok L TRUE if it is okay to overwrite this file on disk. ============ ==== ========================================================= """
[docs] def __init__(self, filen=False, headonly=False, alpha=False, debug_headers=False): self.byteorder = 'little' self.InitArrays() self.debug_headers = debug_headers if filen is False: return # parse Trace object if we get one if isinstance(filen, Trace): self.readTrace(filen) return if alpha: if headonly: self.ReadSacXYHeader(filen) else: self.ReadSacXY(filen) elif headonly: self.ReadSacHeader(filen) else: self.ReadSacFile(filen)
[docs] def InitArrays(self): """ Function to initialize the floating, character and integer header arrays (self.hf, self.hs, self.hi) with dummy values. This function is useful for writing SAC files from artificial data, thus the header arrays are not filled by a read method beforehand :return: Nothing """ # The SAC header has 70 floats, then 40 integers, then 192 bytes # in strings. Store them in array (and convert the char to a # list). That's a total of 632 bytes. # # allocate the array for header floats self.hf = np.ndarray(70, dtype=native_str('<f4')) self.hf[:] = -12345.0 # # allocate the array for header integers self.hi = np.ndarray(40, dtype=native_str('<i4')) self.hi[:] = -12345 # # allocate the array for header characters self.hs = np.ndarray(24, dtype=native_str('|S8')) self.hs[:] = b'-12345 ' # setting default value # allocate the array for the points self.seis = np.ndarray([], dtype=native_str('<f4'))
[docs] def fromarray(self, trace, begin=0.0, delta=1.0, distkm=0, starttime=UTCDateTime("1970-01-01T00:00:00.000000")): """ Create a SAC file from an numpy.ndarray instance >>> t = SacIO() >>> b = np.arange(10) >>> t.fromarray(b) >>> t.GetHvalue('npts') 10 """ if not isinstance(trace, np.ndarray): raise SacError("input needs to be of instance numpy.ndarray") else: # Only copy the data if they are not of the required type self.seis = np.require(trace, native_str('<f4')) # convert start time to sac reference time, if it is not default if begin == -12345: reftime = starttime else: reftime = starttime - begin # if there are any micro-seconds, use begin to store them # integer arithmetic millisecond = reftime.microsecond // 1000 # integer arithmetic microsecond = (reftime.microsecond - millisecond * 1000) if microsecond != 0: begin += microsecond * 1e-6 # set a few values that are required to create a valid SAC-file self.SetHvalue('int1', 2) self.SetHvalue('cmpaz', 0) self.SetHvalue('cmpinc', 0) self.SetHvalue('nvhdr', 6) self.SetHvalue('leven', 1) self.SetHvalue('lpspol', 1) self.SetHvalue('lcalda', 0) self.SetHvalue('lovrok', 1) self.SetHvalue('nzyear', reftime.year) self.SetHvalue('nzjday', reftime.strftime("%j")) self.SetHvalue('nzhour', reftime.hour) self.SetHvalue('nzmin', reftime.minute) self.SetHvalue('nzsec', reftime.second) self.SetHvalue('nzmsec', millisecond) self.SetHvalue('kcmpnm', 'Z') self.SetHvalue('evla', 0) self.SetHvalue('evlo', 0) self.SetHvalue('iftype', 1) self.SetHvalue('npts', len(trace)) self.SetHvalue('delta', delta) self.SetHvalue('b', begin) self.SetHvalue('e', begin + (len(trace) - 1) * delta) self.SetHvalue('iztype', 9) self.SetHvalue('dist', distkm)
[docs] def GetHvalue(self, item): """ Read SAC-header variable. :param item: header variable name (e.g. 'npts' or 'delta') >>> from obspy.sac import SacIO # doctest: +SKIP >>> tr = SacIO('test.sac') # doctest: +SKIP >>> tr.GetHvalue('npts') # doctest: +SKIP 100 This is equivalent to: >>> SacIO().GetHvalueFromFile('test.sac','npts') # doctest: +SKIP 100 Or: >>> tr = SacIO('test.sac') # doctest: +SKIP >>> tr.npts # doctest: +SKIP 100 """ key = item.lower() # convert the item to lower case if key in FDICT: index = FDICT[key] return(self.hf[index]) elif key in IDICT: index = IDICT[key] return(self.hi[index]) elif key in SDICT: index = SDICT[key] if index == 0: myarray = self.hs[0].decode() elif index == 1: myarray = self.hs[1].decode() + self.hs[2].decode() else: myarray = self.hs[index + 1].decode() # extra 1 from item #2 return myarray else: raise SacError("Cannot find header entry for: " + item)
[docs] def SetHvalue(self, item, value): """ Assign new value to SAC-header variable. :param item: SAC-header variable name :param value: numeric or string value to be assigned to header variable. >>> from obspy.sac import SacIO >>> tr = SacIO() >>> print(tr.GetHvalue('kstnm').strip()) -12345 >>> tr.SetHvalue('kstnm', 'STA_NEW') >>> print(tr.GetHvalue('kstnm').strip()) STA_NEW """ key = item.lower() # convert the item to lower case # if key in FDICT: index = FDICT[key] self.hf[index] = float(value) elif key in IDICT: index = IDICT[key] self.hi[index] = int(value) elif key in SDICT: index = SDICT[key] if value: value = '%-8s' % value else: value = '-12345 ' if index == 0: self.hs[0] = value.encode('ascii', 'strict') elif index == 1: value1 = '%-8s' % value[0:8] value2 = '%-8s' % value[8:16] self.hs[1] = value1.encode('ascii', 'strict') self.hs[2] = value2.encode('ascii', 'strict') else: self.hs[index + 1] = value.encode('ascii', 'strict') else: raise SacError("Cannot find header entry for: " + item)
[docs] def IsSACfile(self, fh, fsize=True, lenchk=False): """ Test for a valid SAC file using arrays. """ cur_pos = fh.tell() fh.seek(0, os.SEEK_END) length = fh.tell() fh.seek(cur_pos, os.SEEK_SET) try: npts = self.GetHvalue('npts') except: raise SacError("Unable to read number of points from header") if lenchk and npts != len(self.seis): raise SacError("Number of points in header and " + "length of trace inconsistent!") if fsize: sizecheck = length - (632 + 4 * int(npts)) # size check info if sizecheck != 0: msg = "File-size and theoretical size are inconsistent.\n" \ "Check that headers are consistent with time series." raise SacError(msg) # get the SAC file version number version = self.GetHvalue('nvhdr') if version < 0 or version > 20: raise SacError("Unknown header version!") if self.GetHvalue('delta') <= 0: raise SacError("Delta < 0 is not a valid header entry!")
[docs] def ReadSacHeader(self, fh): """ Reads only the header portion of a binary SAC-file. :param fh: file or file-like object. >>> from obspy.sac import SacIO # doctest: +SKIP >>> tr = SacIO() # doctest: +SKIP >>> with open('test.sac', 'rb') as fh: ... tr.ReadSacHeader(fh) # doctest: +SKIP This is equivalent to: >>> with open('test.sac', 'rb') as fh: ... tr = SacIO(fh, headonly=True) # doctest: +SKIP """ # -------------------------------------------------------------- # parse the header # # The sac header has 70 floats, 40 integers, then 192 bytes # in strings. Store them in array (and convert the char to a # list). That's a total of 632 bytes. # -------------------------------------------------------------- self.hf = frombuffer(fh.read(4 * 70), dtype=native_str('<f4')) self.hi = frombuffer(fh.read(4 * 40), dtype=native_str('<i4')) # read in the char values self.hs = frombuffer(fh.read(24 * 8), dtype=native_str('|S8')) if len(self.hf) != 70 or len(self.hi) != 40 or len(self.hs) != 24: self.hf = self.hi = self.hs = None raise SacIOError("Cannot read all header values") try: self.IsSACfile(fh) except SacError as e: try: # if it is not a valid SAC-file try with big endian # byte order fh.seek(0, os.SEEK_SET) self.hf = frombuffer(fh.read(4 * 70), dtype=native_str('>f4')) self.hi = frombuffer(fh.read(4 * 40), dtype=native_str('>i4')) # read in the char values self.hs = frombuffer(fh.read(24 * 8), dtype=native_str('|S8')) self.IsSACfile(fh) self.byteorder = 'big' except SacError as e: self.hf = self.hi = self.hs = None raise SacError(e) try: self._get_date() except SacError: warnings.warn('Cannot determine date') if self.GetHvalue('lcalda'): try: self._get_dist() except SacError: pass
[docs] def WriteSacHeader(self, fh): """ Writes an updated header to an existing binary SAC-file. :param fh: open file or file-like buffer >>> from obspy.sac import SacIO # doctest: +SKIP >>> with open('test.sac', 'rb') as fh: ... tr = SacIO(fh) # doctest: +SKIP >>> with open('test2.sac', 'wb') as fh: ... tr.WriteSacBinary(fh) # doctest: +SKIP >>> with open('test2.sac', 'rb') as fh: ... u = SacIO(fh) # doctest: +SKIP >>> u.SetHvalue('kevnm','hullahulla') # doctest: +SKIP >>> with open('test2.sac', 'rb+') as fh: ... u.WriteSacHeader(fh) # doctest: +SKIP >>> u.GetHvalueFromFile('test2.sac',"kevnm") # doctest: +SKIP 'hullahulla ' """ fh.seek(0, os.SEEK_SET) try: # write the header fh.write(self.hf.data) fh.write(self.hi.data) fh.write(self.hs.data) except Exception: raise SacError("Cannot write header.")
[docs] def ReadSacFile(self, fh, fsize=True): """ Read read in the header and data in a SAC file The header is split into three arrays - floats, ints, and strings and the data points are returned in the array seis :param fh: file or file-like object. >>> from obspy.sac import SacIO # doctest: +SKIP >>> tr = SacIO() # doctest: +SKIP >>> with open('test.sac') as fh: ... tr.ReadSacFile(fh) # doctest: +SKIP This is equivalent to: >>> with open('test.sac') as fh: ... tr = SacIO(fh) # doctest: +SKIP """ # -------------------------------------------------------------- # parse the header # # The sac header has 70 floats, 40 integers, then 192 bytes # in strings. Store them in array (and convert the char to a # list). That's a total of 632 bytes. # -------------------------------------------------------------- self.hf = frombuffer(fh.read(4 * 70), dtype=native_str('<f4')) self.hi = frombuffer(fh.read(4 * 40), dtype=native_str('<i4')) # read in the char values self.hs = frombuffer(fh.read(24 * 8), dtype=native_str('|S8')) if len(self.hf) != 70 or len(self.hi) != 40 or len(self.hs) != 24: self.hf = self.hi = self.hs = None fh.close() raise SacIOError("Cannot read all header values") # only continue if it is a SAC file try: self.IsSACfile(fh, fsize) except SacError: try: # if it is not a valid SAC-file try with big endian # byte order fh.seek(0, os.SEEK_SET) self.hf = frombuffer(fh.read(4 * 70), dtype=native_str('>f4')) self.hi = frombuffer(fh.read(4 * 40), dtype=native_str('>i4')) # read in the char values self.hs = frombuffer(fh.read(24 * 8), dtype=native_str('|S8')) self.IsSACfile(fh, fsize) self.byteorder = 'big' except SacError as e: raise SacError(e) # -------------------------------------------------------------- # read in the seismogram points # -------------------------------------------------------------- # you just have to know it's in the 10th place # actually, it's in the SAC manual npts = int(self.hi[9]) if self.byteorder == 'big': self.seis = frombuffer(fh.read(npts * 4), dtype=native_str('>f4')) else: self.seis = frombuffer(fh.read(npts * 4), dtype=native_str('<f4')) if len(self.seis) != npts: self.hf = self.hi = self.hs = self.seis = None raise SacIOError("Cannot read all data points") try: self._get_date() except SacError: warnings.warn('Cannot determine date') if self.GetHvalue('lcalda'): try: self._get_dist() except SacError: pass
[docs] def ReadSacXY(self, fh): """ Read SAC XY files (ascii) :param f: File or file-like object. >>> from obspy.sac import SacIO # doctest: +SKIP >>> tr = SacIO() # doctest: +SKIP >>> tr.ReadSacXY('testxy.sac') # doctest: +SKIP >>> tr.GetHvalue('npts') # doctest: +SKIP 100 This is equivalent to: >>> tr = SacIO('testxy.sac',alpha=True) # doctest: +SKIP Reading only the header portion of alphanumeric SAC-files is currently not supported. """ data = bytes(fh.read()) data = [_i.rstrip(b"\n\r") for _i in data.splitlines(True)] if len(data) < 14 + 8 + 8: raise SacIOError("Not a valid SAC file.") # -------------------------------------------------------------- # parse the header # # The sac header has 70 floats, 40 integers, then 192 bytes # in strings. Store them in array (and convert the char to a # list). That's a total of 632 bytes. # -------------------------------------------------------------- # read in the float values self.hf = np.array([i.split() for i in data[:14]], dtype=native_str('<f4')).ravel() # read in the int values self.hi = np.array([i.split() for i in data[14: 14 + 8]], dtype=native_str('<i4')).ravel() # reading in the string part is a bit more complicated # because every string field has to be 8 characters long # apart from the second field which is 16 characters long # resulting in a total length of 192 characters for i, j in enumerate(range(0, 24, 3)): line = data[14 + 8 + i] self.hs[j:j + 3] = np.fromstring(line, dtype=native_str('|S8'), count=3) # -------------------------------------------------------------- # read in the seismogram points # -------------------------------------------------------------- self.seis = np.array([i.split() for i in data[30:]], dtype=native_str('<f4')).ravel() try: self._get_date() except SacError: warnings.warn('Cannot determine date') if self.GetHvalue('lcalda'): try: self._get_dist() except SacError: pass
[docs] def ReadSacXYHeader(self, fh): """ Read SAC XY files (ascii) :param fh: open file or file like object. >>> from obspy.sac import SacIO # doctest: +SKIP >>> tr = SacIO() # doctest: +SKIP >>> tr.ReadSacXY('testxy.sac') # doctest: +SKIP >>> tr.GetHvalue('npts') # doctest: +SKIP 100 This is equivalent to: >>> tr = SacIO('testxy.sac',alpha=True) # doctest: +SKIP Reading only the header portion of alphanumeric SAC-files is currently not supported. """ data = bytes(fh.read()) data = [_i.rstrip(b"\n\r") for _i in data.splitlines(True)] if len(data) < 14 + 8 + 8: raise SacIOError("Not a valid SAC file") # -------------------------------------------------------------- # parse the header # # The sac header has 70 floats, 40 integers, then 192 bytes # in strings. Store them in array (and convert the char to a # list). That's a total of 632 bytes. # -------------------------------------------------------------- # read in the float values self.hf = np.array([i.split() for i in data[:14]], dtype=native_str('<f4')).ravel() # read in the int values self.hi = np.array([i.split() for i in data[14: 14 + 8]], dtype=native_str('<i4')).ravel() # reading in the string part is a bit more complicated # because every string field has to be 8 characters long # apart from the second field which is 16 characters long # resulting in a total length of 192 characters for i, j in enumerate(range(0, 24, 3)): line = data[14 + 8 + i] self.hs[j:j + 3] = np.fromstring(line, dtype=native_str('|S8'), count=3) try: self.IsSACfile(fh, fsize=False) except SacError as e: raise SacError(e) try: self._get_date() except SacError: warnings.warn('Cannot determine date') if self.GetHvalue('lcalda'): try: self._get_dist() except SacError: pass
[docs] def readTrace(self, trace): """ Fill in SacIO object with data from obspy trace. Warning: Currently only the case of a previously empty SacIO object is safe! """ # extracting relative SAC time as specified with b try: b = float(trace.stats['sac']['b']) except KeyError: b = 0.0 # filling in SAC/sacio specific defaults self.fromarray(trace.data, begin=b, delta=trace.stats.delta, starttime=trace.stats.starttime) # overwriting with ObsPy defaults for _j, _k in convert_dict.items(): self.SetHvalue(_j, trace.stats[_k]) # overwriting up SAC specific values # note that the SAC reference time values (including B and E) are # not used in here any more, they are already set by t.fromarray # and directly deduce from tr.starttime for _i in SAC_EXTRA: try: self.SetHvalue(_i, trace.stats.sac[_i]) except (AttributeError, KeyError): pass return
[docs] def WriteSacXY(self, fh): """ Write SAC XY file (ascii) :param fh: open file or file-like object. >>> from obspy.sac import SacIO # doctest: +SKIP >>> with open('test.sac', 'rb') as fh: ... tr = SacIO(fh) # doctest: +SKIP >>> with open('test2.sac', 'wb') as fh: ... tr.WriteSacXY(fh) # doctest: +SKIP >>> with open('test2.sac', 'wb') as fh: ... tr.IsValidXYSacFile(fh) # doctest: +SKIP True """ # header try: np.savetxt(fh, np.reshape(self.hf, (14, 5)), fmt=native_str("%#15.7g%#15.7g%#15.7g%#15.7g%#15.7g")) np.savetxt(fh, np.reshape(self.hi, (8, 5)), fmt=native_str("%10d%10d%10d%10d%10d")) for i in range(0, 24, 3): # numpy 1.6 compatibility fh.write(self.hs[i:i + 3].tostring()) fh.write(b'\n') except Exception as e: raise SacIOError("Cannot write header values.", e) # traces npts = self.GetHvalue('npts') if npts == -12345 or npts == 0: return try: rows = npts // 5 np.savetxt(fh, np.reshape(self.seis[0:5 * rows], (rows, 5)), fmt=native_str("%#15.7g%#15.7g%#15.7g%#15.7g%#15.7g")) np.savetxt(fh, self.seis[5 * rows:], delimiter=b'\t') except Exception as e: raise SacIOError("Cannot write trace values.", e)
[docs] def WriteSacBinary(self, fh): """ Write a SAC binary file using the head arrays and array seis. :param fh: open file or file-like object. >>> from obspy.sac import SacIO # doctest: +SKIP >>> with open('test.sac', 'rb') as fh: ... tr = SacIO(fh) # doctest: +SKIP >>> with open('test2.sac', 'wb') as fh: ... tr.WriteSacBinary(fh) # doctest: +SKIP >>> import os >>> os.stat('test2.sac')[6] == os.stat('test.sac')[6] # doctest: +SKIP True """ try: self._chck_header() # numpy 1.6 compatibility fh.write(self.hf.tostring()) fh.write(self.hi.tostring()) fh.write(self.hs.tostring()) fh.write(self.seis.tostring()) except Exception as e: msg = "Cannot write SAC-buffer to file: " raise SacIOError(msg, e)
[docs] def PrintIValue(self, label='=', value=-12345): """ Convenience function for printing undefined integer header values. """ if value != -12345: print(label, value)
[docs] def PrintFValue(self, label='=', value=-12345.0): """ Convenience function for printing undefined float header values. """ if value != -12345.0: print('%s %.8g' % (label, value))
[docs] def PrintSValue(self, label='=', value='-12345'): """ Convenience function for printing undefined string header values. """ if value.find('-12345') == -1: print(label, value)
[docs] def ListStdValues(self): # h is a header list, s is a float list """ Convenience function for printing common header values. :param: None :return: None >>> from obspy.sac import SacIO # doctest: +SKIP >>> t = SacIO('test.sac') # doctest: +SKIP >>> t.ListStdValues() # doctest: +SKIP +NORMALIZE_WHITESPACE <BLANKLINE> Reference Time = 07/18/1978 (199) 8:0:0.0 Npts = 100 Delta = 1 Begin = 10 End = 109 Min = -1 Mean = 8.7539462e-08 Max = 1 Header Version = 6 Station = STA Channel = Q Event = FUNCGEN: SINE If no header values are defined (i.e. all are equal 12345) than this function won't do anything. """ # # Seismogram Info: # try: nzyear = self.GetHvalue('nzyear') nzjday = self.GetHvalue('nzjday') month = time.strptime(repr(nzyear) + " " + repr(nzjday), "%Y %j").tm_mon date = time.strptime(repr(nzyear) + " " + repr(nzjday), "%Y %j").tm_mday pattern = '\nReference Time = %2.2d/%2.2d/%d (%d) %d:%d:%d.%d' print(pattern % (month, date, self.GetHvalue('nzyear'), self.GetHvalue('nzjday'), self.GetHvalue('nzhour'), self.GetHvalue('nzmin'), self.GetHvalue('nzsec'), self.GetHvalue('nzmsec'))) except ValueError: pass self.PrintIValue('Npts = ', self.GetHvalue('npts')) self.PrintFValue('Delta = ', self.GetHvalue('delta')) self.PrintFValue('Begin = ', self.GetHvalue('b')) self.PrintFValue('End = ', self.GetHvalue('e')) self.PrintFValue('Min = ', self.GetHvalue('depmin')) self.PrintFValue('Mean = ', self.GetHvalue('depmen')) self.PrintFValue('Max = ', self.GetHvalue('depmax')) # self.PrintIValue('Header Version = ', self.GetHvalue('nvhdr')) # # station Info: # self.PrintSValue('Station = ', self.GetHvalue('kstnm')) self.PrintSValue('Channel = ', self.GetHvalue('kcmpnm')) self.PrintFValue('Station Lat = ', self.GetHvalue('stla')) self.PrintFValue('Station Lon = ', self.GetHvalue('stlo')) self.PrintFValue('Station Elev = ', self.GetHvalue('stel')) # # Event Info: # self.PrintSValue('Event = ', self.GetHvalue('kevnm')) self.PrintFValue('Event Lat = ', self.GetHvalue('evla')) self.PrintFValue('Event Lon = ', self.GetHvalue('evlo')) self.PrintFValue('Event Depth = ', self.GetHvalue('evdp')) self.PrintFValue('Origin Time = ', self.GetHvalue('o')) # self.PrintFValue('Azimuth = ', self.GetHvalue('az')) self.PrintFValue('Back Azimuth = ', self.GetHvalue('baz')) self.PrintFValue('Distance (km) = ', self.GetHvalue('dist')) self.PrintFValue('Distance (deg) = ', self.GetHvalue('gcarc'))
[docs] def GetHvalueFromFile(self, thePath, theItem): """ Quick access to a specific header item in specified file. :param f: filename (SAC binary) :type hn: str :param hn: header variable name >>> from obspy.sac import SacIO # doctest: +SKIP >>> t = SacIO() # doctest: +SKIP >>> t.GetHvalueFromFile('test.sac','kcmpnm').rstrip() # doctest: +SKIP 'Q' String header values have a fixed length of 8 or 16 characters. This can lead to errors for example if you concatenate strings and forget to strip off the trailing whitespace. """ # # Read in the Header # with open(thePath, "rb") as fh: self.ReadSacHeader(fh) return(self.GetHvalue(theItem))
[docs] def SetHvalueInFile(self, thePath, theItem, theValue): """ Quick access to change a specific header item in a specified file. :param f: filename (SAC binary) :type hn: str :param hn: header variable name :type hv: str, float or int :param hv: header variable value (numeric or string value to be assigned to hn) :return: None >>> from obspy.sac import SacIO # doctest: +SKIP >>> t = SacIO() # doctest: +SKIP >>> t.GetHvalueFromFile('test.sac','kstnm').rstrip() # doctest: +SKIP 'STA' >>> t.SetHvalueInFile('test.sac','kstnm','blub') # doctest: +SKIP >>> t.GetHvalueFromFile('test.sac','kstnm').rstrip() # doctest: +SKIP 'blub' """ # Read the header. with open(thePath, "rb") as fh: self.ReadSacHeader(fh) # Modify it. self.SetHvalue(theItem, theValue) # Write it. with open(thePath, "rb+") as fh: self.WriteSacHeader(fh)
[docs] def IsValidSacFile(self, thePath): """ Quick test for a valid SAC binary file (wraps 'IsSACfile'). :param f: filename (SAC binary) :rtype: boolean (True or False) >>> from obspy.sac import SacIO # doctest: +SKIP >>> SacIO().IsValidSacFile('test.sac') # doctest: +SKIP True >>> SacIO().IsValidSacFile('testxy.sac') # doctest: +SKIP False """ # # Read in the Header # try: self.ReadSacHeader(thePath) except SacError: return False except SacIOError: return False else: return True
[docs] def IsValidXYSacFile(self, fh): """ Quick test for a valid SAC ascii file. :param file: file or file-like object. :rtype: boolean (True or False) >>> from obspy.sac import SacIO # doctest: +SKIP >>> SacIO().IsValidXYSacFile('testxy.sac') # doctest: +SKIP True >>> SacIO().IsValidXYSacFile('test.sac') # doctest: +SKIP False """ try: self.ReadSacXY(fh) return True except: return False
[docs] def _get_date(self): """ If date header values are set calculate date in julian seconds >>> t = SacIO() >>> t.fromarray(np.random.randn(100), delta=1.0, ... starttime=UTCDateTime(1970,1,1)) >>> t._get_date() >>> t.reftime.timestamp 0.0 >>> t.endtime.timestamp - t.reftime.timestamp 100.0 """ # if any of the time-header values are still set to -12345 then # UTCDateTime raises an exception and reftime is set to 0.0 try: ms = self.GetHvalue('nzmsec') * 1000 yr = self.GetHvalue('nzyear') if 0 <= yr <= 99: warnings.warn(TWO_DIGIT_YEAR_MSG) yr += 1900 self.reftime = UTCDateTime(year=yr, julday=self.GetHvalue('nzjday'), hour=self.GetHvalue('nzhour'), minute=self.GetHvalue('nzmin'), second=self.GetHvalue('nzsec'), microsecond=ms) b = float(self.GetHvalue('b')) if b != -12345.0: self.starttime = self.reftime + b else: self.starttime = self.reftime self.endtime = self.starttime + \ self.GetHvalue('npts') * float(self.GetHvalue('delta')) except: try: self.reftime = UTCDateTime(0.0) b = float(self.GetHvalue('b')) if b != -12345.0: self.starttime = self.reftime + b else: self.starttime = self.reftime self.endtime = self.reftime + \ self.GetHvalue('npts') * float(self.GetHvalue('delta')) except: raise SacError("Cannot calculate date")
[docs] def _chck_header(self): """ If trace changed since read, adapt header values """ if self.byteorder == 'big': self.seis = np.require(self.seis, native_str('>f4')) else: self.seis = np.require(self.seis, native_str('<f4')) self.SetHvalue('npts', self.seis.size) if self.seis.size == 0: return self.SetHvalue('depmin', self.seis.min()) self.SetHvalue('depmax', self.seis.max()) self.SetHvalue('depmen', self.seis.mean())
[docs] def _get_dist(self): """ calculate distance from station and event coordinates >>> t = SacIO() >>> t.SetHvalue('evla',48.15) >>> t.SetHvalue('evlo',11.58333) >>> t.SetHvalue('stla',-41.2869) >>> t.SetHvalue('stlo',174.7746) >>> t._get_dist() >>> print('%.2f' % t.GetHvalue('dist')) 18486.53 >>> print('%.5f' % t.GetHvalue('az')) 65.65415 >>> print('%.4f' % t.GetHvalue('baz')) 305.9755 The original SAC-program calculates the distance assuming a average radius of 6371 km. Therefore, our routine should be more accurate. """ eqlat = self.GetHvalue('evla') eqlon = self.GetHvalue('evlo') stlat = self.GetHvalue('stla') stlon = self.GetHvalue('stlo') d = self.GetHvalue('dist') if eqlat == -12345.0 or eqlon == -12345.0 or \ stlat == -12345.0 or stlon == -12345.0: raise SacError('Insufficient information to calculate distance.') if d != -12345.0: raise SacError('Distance is already set.') dist, az, baz = gps2DistAzimuth(eqlat, eqlon, stlat, stlon) self.SetHvalue('dist', dist / 1000.) self.SetHvalue('az', az) self.SetHvalue('baz', baz)
[docs] def swap_byte_order(self): """ Swap byte order of SAC-file in memory. :param: None :return: None >>> from obspy.sac import SacIO # doctest: +SKIP >>> t = SacIO('test.sac') # doctest: +SKIP >>> t.swap_byte_order() # doctest: +SKIP """ if self.byteorder == 'big': bs = '<' self.byteorder = 'little' elif self.byteorder == 'little': bs = '>' self.byteorder = 'big' self.seis = self.seis.byteswap(True).newbyteorder(bs) self.hf = self.hf.byteswap(True).newbyteorder(bs) self.hi = self.hi.byteswap(True).newbyteorder(bs)
[docs] def __getattr__(self, hname): """ convenience function to access header values :param hname: header variable name >>> tr = SacIO() >>> tr.fromarray(np.random.randn(100)) >>> tr.npts == tr.GetHvalue('npts') # doctest: +SKIP True """ return self.GetHvalue(hname)
[docs] def get_obspy_header(self): """ Return a dictionary that can be used as a header in creating a new :class:`~obspy.core.trace.Trace` object. Currently most likely an Exception will be raised if no SAC file was read beforehand! """ header = {} # convert common header types of the ObsPy trace object for i, j in convert_dict.items(): value = self.GetHvalue(i) if isinstance(value, (str, native_str)): null_term = value.find('\x00') if null_term >= 0: value = value[:null_term] value = value.strip() if value == '-12345': value = '' # fix for issue #156 if i == 'delta': header['sampling_rate'] = \ np.float32(1.0) / np.float32(self.hf[0]) else: header[j] = value if header['calib'] == -12345.0: header['calib'] = 1.0 # assign extra header types of SAC header['sac'] = {} for i in SAC_EXTRA: header['sac'][i] = self.GetHvalue(i) # convert time to UTCDateTime header['starttime'] = self.starttime # always add the begin time (if it's defined) to get the given # SAC reference time, no matter which iztype is given # note that the B and E times should not be in the SAC_EXTRA # dictionary, as they would overwrite the self.fromarray which sets # them according to the starttime, npts and delta. header['sac']['b'] = float(self.GetHvalue('b')) header['sac']['e'] = float(self.GetHvalue('e')) # ticket #390 if self.debug_headers: for i in ['nzyear', 'nzjday', 'nzhour', 'nzmin', 'nzsec', 'nzmsec', 'delta', 'scale', 'npts', 'knetwk', 'kstnm', 'kcmpnm']: header['sac'][i] = self.GetHvalue(i) return header # UTILITIES
[docs]def attach_paz(tr, paz_file, todisp=False, tovel=False, torad=False, tohz=False): ''' Attach tr.stats.paz AttribDict to trace from SAC paz_file This is experimental code, taken from obspy.gse2.libgse2.attach_paz and adapted to the SAC-pole-zero conventions. Especially the conversion from velocity to displacement and vice versa is still under construction. It works but I cannot guarantee that the values are correct. For more information on the SAC-pole-zero format see: http://www.iris.edu/files/sac-manual/commands/transfer.html. For a useful discussion on polezero files and transfer functions in general see: http://seis-uk.le.ac.uk/equipment/downloads/data_management/\ seisuk_instrument_resp_removal.pdf Also bear in mind that according to the SAC convention for pole-zero files CONSTANT is defined as: digitizer_gain*seismometer_gain*A0. This means that it does not have explicit information on the digitizer gain and seismometer gain which we therefore set to 1.0. Attaches to a trace a paz AttribDict containing poles zeros and gain. :param tr: An ObsPy :class:`~obspy.core.trace.Trace` object :param paz_file: path to pazfile or file pointer :param todisp: change a velocity transfer function to a displacement transfer function by adding another zero :param tovel: change a displacement transfer function to a velocity transfer function by removing one 0,0j zero :param torad: change to radians :param tohz: change to Hertz >>> from obspy import Trace >>> import io >>> tr = Trace() >>> f = io.StringIO("""ZEROS 3 ... -5.032 0.0 ... POLES 6 ... -0.02365 0.02365 ... -0.02365 -0.02365 ... -39.3011 0. ... -7.74904 0. ... -53.5979 21.7494 ... -53.5979 -21.7494 ... CONSTANT 2.16e18""") >>> attach_paz(tr, f,torad=True) >>> for z in tr.stats.paz['zeros']: ... print("%.2f %.2f" % (z.real, z.imag)) -31.62 0.00 0.00 0.00 0.00 0.00 ''' poles = [] zeros = [] if isinstance(paz_file, (str, native_str)): paz_file = open(paz_file, 'r') while True: line = paz_file.readline() if not line: break # lines starting with * are comments if line.startswith('*'): continue if line.find('ZEROS') != -1: a = line.split() noz = int(a[1]) for _k in range(noz): line = paz_file.readline() a = line.split() if line.find('POLES') != -1 or line.find('CONSTANT') != -1 or \ line.startswith('*') or not line: while len(zeros) < noz: zeros.append(complex(0, 0j)) break else: zeros.append(complex(float(a[0]), float(a[1]))) if line.find('POLES') != -1: a = line.split() nop = int(a[1]) for _k in range(nop): line = paz_file.readline() a = line.split() if line.find('CONSTANT') != -1 or line.find('ZEROS') != -1 or \ line.startswith('*') or not line: while len(poles) < nop: poles.append(complex(0, 0j)) break else: poles.append(complex(float(a[0]), float(a[1]))) if line.find('CONSTANT') != -1: a = line.split() # in the observatory this is the seismometer gain [muVolt/nm/s] # the A0_normalization_factor is hardcoded to 1.0 constant = float(a[1]) paz_file.close() # To convert the velocity response to the displacement response, # multiplication with jw is used. This is equivalent to one more # zero in the pole-zero representation if todisp: zeros.append(complex(0, 0j)) # To convert the displacement response to the velocity response, # division with jw is used. This is equivalent to one less zero # in the pole-zero representation if tovel: for i, zero in enumerate(list(zeros)): if zero == complex(0, 0j): zeros.pop(i) break else: raise Exception("Could not remove (0,0j) zero to change \ displacement response to velocity response") # convert poles, zeros and gain in Hertz to radians if torad: tmp = [z * 2. * np.pi for z in zeros] zeros = tmp tmp = [p * 2. * np.pi for p in poles] poles = tmp # When extracting RESP files and SAC_PZ files # from a dataless SEED using the rdseed program # where the former is in Hz and the latter in radians, # there gains seem to be unaffected by this. # According to this document: # http://www.le.ac.uk/ # seis-uk/downloads/seisuk_instrument_resp_removal.pdf # the gain should also be converted when changing from # hertz to radians or vice versa. However, the rdseed programs # does not do this. I'm not entirely sure at this stage which one is # correct or if I have missed something. I've therefore decided # to leave it out for now, in order to stay compatible with the # rdseed program and the SAC program. # constant *= (2. * np.pi) ** 3 # convert poles, zeros and gain in radian to Hertz if tohz: for i, z in enumerate(zeros): if abs(z) > 0.0: zeros[i] /= 2 * np.pi for i, p in enumerate(poles): if abs(p) > 0.0: poles[i] /= 2 * np.pi # constant /= (2. * np.pi) ** 3 # fill up ObsPy Poles and Zeros AttribDict # In SAC pole-zero files CONSTANT is defined as: # digitizer_gain*seismometer_gain*A0 tr.stats.paz = AttribDict() tr.stats.paz.seismometer_gain = 1.0 tr.stats.paz.digitizer_gain = 1.0 tr.stats.paz.poles = poles tr.stats.paz.zeros = zeros # taken from obspy.gse2.paz:145 tr.stats.paz.sensitivity = tr.stats.paz.digitizer_gain * \ tr.stats.paz.seismometer_gain tr.stats.paz.gain = constant
[docs]def attach_resp(tr, resp_file, todisp=False, tovel=False, torad=False, tohz=False): """ Extract key instrument response information from a RESP file, which can be extracted from a dataless SEED volume by, for example, using the script obspy-dataless2resp or the rdseed program. At the moment, you have to determine yourself if the given response is for velocity or displacement and if the values are given in rad or Hz. This is still experimental code (see also documentation for :func:`obspy.sac.sacio.attach_paz`). Attaches to a trace a paz AttribDict containing poles, zeros, and gain. :param tr: An ObsPy :class:`~obspy.core.trace.Trace` object :param resp_file: path to RESP-file or file pointer :param todisp: change a velocity transfer function to a displacement transfer function by adding another zero :param tovel: change a displacement transfer function to a velocity transfer function by removing one 0,0j zero :param torad: change to radians :param tohz: change to Hertz >>> from obspy import Trace >>> import os >>> tr = Trace() >>> respfile = os.path.join(os.path.dirname(__file__), 'tests', 'data', ... 'RESP.NZ.CRLZ.10.HHZ') >>> attach_resp(tr, respfile, torad=True, todisp=False) >>> for k in sorted(tr.stats.paz): # doctest: +NORMALIZE_WHITESPACE ... print(k) digitizer_gain gain poles seismometer_gain sensitivity t_shift zeros >>> print(tr.stats.paz.poles) # doctest: +SKIP [(-0.15931644664884559+0.15931644664884559j), (-0.15931644664884559-0.15931644664884559j), (-314.15926535897933+202.31856689118268j), (-314.15926535897933-202.31856689118268j)] """ if not hasattr(resp_file, 'write'): resp_filep = open(resp_file, 'r') else: resp_filep = resp_file zeros_pat = r'B053F10-13' poles_pat = r'B053F15-18' a0_pat = r'B053F07' sens_pat = r'B058F04' t_shift_pat = r'B057F08' t_shift = 0.0 poles = [] zeros = [] while True: line = resp_filep.readline() if not line: break if line.startswith(a0_pat): a0 = float(line.split(':')[1]) if line.startswith(sens_pat): sens = float(line.split(':')[1]) if line.startswith(poles_pat): tmp = line.split() poles.append(complex(float(tmp[2]), float(tmp[3]))) if line.startswith(zeros_pat): tmp = line.split() zeros.append(complex(float(tmp[2]), float(tmp[3]))) if line.startswith(t_shift_pat): t_shift += float(line.split(':')[1]) constant = a0 * sens if not hasattr(resp_file, 'write'): resp_filep.close() if torad: tmp = [z * 2. * np.pi for z in zeros] zeros = tmp tmp = [p * 2. * np.pi for p in poles] poles = tmp if todisp: zeros.append(complex(0, 0j)) # To convert the displacement response to the velocity response, # division with jw is used. This is equivalent to one less zero # in the pole-zero representation if tovel: for i, zero in enumerate(list(zeros)): if zero == complex(0, 0j): zeros.pop(i) break else: raise Exception("Could not remove (0,0j) zero to change \ displacement response to velocity response") # convert poles, zeros and gain in radian to Hertz if tohz: for i, z in enumerate(zeros): if abs(z) > 0.0: zeros[i] /= 2 * np.pi for i, p in enumerate(poles): if abs(p) > 0.0: poles[i] /= 2 * np.pi constant /= (2. * np.pi) ** 3 # fill up ObsPy Poles and Zeros AttribDict # In SAC pole-zero files CONSTANT is defined as: # digitizer_gain*seismometer_gain*A0 tr.stats.paz = AttribDict() tr.stats.paz.seismometer_gain = sens tr.stats.paz.digitizer_gain = 1.0 tr.stats.paz.poles = poles tr.stats.paz.zeros = zeros # taken from obspy.gse2.paz:145 tr.stats.paz.sensitivity = tr.stats.paz.digitizer_gain * \ tr.stats.paz.seismometer_gain tr.stats.paz.gain = constant tr.stats.paz.t_shift = t_shift
if __name__ == "__main__": import doctest doctest.testmod()