next up previous contents index
Next: 21.2 AUTOPHASE, automatic phase Up: 21. Automatic routines in Previous: 21. Automatic routines in   Contents   Index

21.1 AUTOPIC and AUTO

AUTOPIC is a tool to automatically pick phases on events registered into the database. AUTOPICK can also be used for many events in connection with program AUTO, see below.

If an event file (S-file) has any readings, the AUTOPIC program will not reread in order to not destroy old picks. The automatic readings in the file are marked with an A after the weight column to indicate automatic pick. Each pick is evaluated by using the signal to noise ratio and an indication of the quality is given with the weight. The program will run on all waveform files given in an S-file. Each time the program runs, there is a file called autopic.out containing information about the run. If there are any 3-component stations, an azimuth will also be calculated, and the S-phase will be more reliable. The AUTOPIC program can also be used from EEV by typing Z (will run program AUTOPIC). When it is used from EEV, there is always an output in the S-file, which will be grouped at the bottom of the file, making it possible to compare manual and automatic readings. THE S-FILE MUST THEN BE EDITED MANUALLY IN ORDER TO REMOVE DOUBLE READINGS. The program requires an input parameter file in the working directory or DAT with the name AUTOPIC.INP. The program will first look in the working directory. The parameters in that file are explained below. NOTE: The file is formatted, data must be in columns exactly as shwown and no tabs must be used. The program uses a 4-pole filter running one way. This might result in phases being picked a bit late. However, it seems more accurate than the earlier version where the filter run both ways and picks were often far too early. The program is made mainly by Bent Ruud. For more information about how it works, see Ruud and Husebye [1992]; Ruud et al. [1988]. Description of parameters


                                                                   
%
% Input parameters common to all filters:
%
% LWIND  : used to define step length (DELTA=WINDOW/LWIND)
% ISHIFT : defines time shift between STA and LTA window (ISHIFT*DELTA)
%          Delay for LTA window (15 * 0.1) ==> 1.5 sec. after STA window.
% ISIGMA : defines fall off rate of LTA window (larger values   longer windows)
%          LTA(i) = (1   2^(isigma)) * LTA(i 1) + 2^( isigma) * STA(j)
% COHMIN : Polarization threshold.
%          Minimum coherence (see thresh_1 and thresh_2)
% NDMIN  : Mimimum number of consecutive triggered windows in a detection
% SVELO  : S wave velocity of the medium below the station (used for 3 comp)
% NFILT  : number of filters
% CRAT   : Ratio for calculation of coda duration ( range 1   4)
% LWIN   : Window used in coda duration routines (range 20   50 seconds)
% THRES  : Quality threshold (range 2   5). Used on the maximum to average
%          amplitude ratio in order to sort out the most noisy traces.
%
% Input parameters defined for each filter:
%
% WINDOW : length of the moving time window (sec)
% F1     : lower cutoff frequency (Hz) of band pass filter
% F2     : higher cutoff frequency (Hz) of band pass filter
% THRSH1 : STA/LTA threshold for polarized signals
% THRSH2 : STA/LTA threshold for unpolarized signals
%          If coherence > cohmin then detection is made on thresh_1
%          If coherence < cohmin then detection is made on  thresh_2
%
% Output parameters:
%
% D    : day of year
% H    : hour
% M    : minute
% SEC  : second
% DUR  : duration, i.e. time in detection state (sec)
% FRQ  : centre frequency of filter giving the best detection (Hz)
% SNR  : signal to noise ratio (SNR=STA/LTA)
% STA  : short time average (root mean square of amplitude)
% NT   : total number of triggered time windows in the detection
% NH   : number of windows with best SNR on one of the horizontal comp.
% NV   : number of windows with best SNR on the vertical comp.
% NC   : number of windows with acceptable polarization
% Q    : quality class, 1(best)   4(worst)
% PS   : P/S wave discriminator, 0(S)   10(P)
% AZI  : backazimuth in degrees measured from North through East
% DA   : variability in azimuth (deg)
% VEL  : apparent velocity (km/s)
% DV   : variability in apparent velocity (km/s)
%
% Note : azimuth and apparent velocity calculations are based on the
%        assumption of P wave, so that these variables should be
%        neglected for S waves.
%

Example of input file AUTOPIC.INP


% This is the parameterfile needed by program: --- PREPROCESS ---
%
% The following rules apply:
% 1. All lines with % in the first column are comment lines
% 2. Lines with a blank in column 1 are read for fixed parameters.
% 3. All lines starting with "filter_x", where x is a number, 
%    are read for filter variable parameters
% 4. All lines with * in the first column are read for stations to process
% 5. A breif explanation of all parameters is given in preprocess.inf
%
% FIXED PARAMETERS THAT ARE USED THROUGHOUT THE PROGRAM
% ----------------------------------------------------------------------
% Lwind Ishift Isigma Cohmin  Ndmin  Svelo  Nfilt   Crat   Lwin  Thres
%     !      !      !      !      !      !      !      !      !      !
% ----------------------------------------------------------------------
    4.0   30.0   06.0    0.1    3.0   2.75    4.0    1.6   30.0    3.0
%
%
% PARAMETERS THAT ARE FILTER DEPENDANT
% ----------------------------------------------------------------------
%Filter_nr    Window        F1        F2    Thrsh1    Thrsh2
%        !         !         !         !         !         !         
% ----------------------------------------------------------------------
filter_1         0.8       2.0       4.0      2.30      3.0
filter_2         0.6       5.0      10.0      2.30      3.00
filter_3         0.4       8.0      16.0      2.30      3.00
filter_4         2.0       0.5       2.0      4.0       5.0
%
%
% STATIONS TO USE IN THE PROCESSING
% ----------------------------------------------------------------------
*SUE  S  Z 3 component
*BER  S  Z
*HYA  S  Z
*KMY  S  Z 3 component
*ODD1 S  Z 3 component
*BLS5 S  Z 3 component
*ESG  S  Z
*EGD  S  Z
*KTK1 S  Z 3 component
*NSS  S  Z 3 component
*MOL  S  Z 3 component
*MOL  S  A 3 component
*JNW  S  Z 
*JNE  S  Z
*FRO  S  Z
*JMI  S  Z 3 component
*ASK1 S  Z 3 component
*ASK  S  Z 3 component
*MOR7 S  Z 3 component 
*MOR8 S  Z 3 component 
*LOF  S  Z 3 component
*LOF  A  Z 3 component
*OSG  S  Z 3 component
*TRO  S  Z 3 component
*FOO  S  Z 3 component
*ALVN S  Z 3 component
*UGA  S  Z 3 component
*ENT  A  Z 3 component 

AUTO

A Program for automatic processing: Phase picking, location, magnitude and fault plane solution determination.

This program run the picking program AUTOPHASE first (optionally AUTOPIC), locates events with or without outlier removal with HYP and then determines magnitudes with AUTOMAG. Automatic amplitudes for fault plane solutions can be done optionally with AUTORATIO. Fault plane solutions can also be done optionally with FPFIT or HASH. The default is to do AUTOPHASE, HYP with outlier removal and AUTOMAG using default parameters. Individual steps, like doing magnitudes, can optionally be deselected and some non default parameters for AUTOMAG can be selected. All changes are made using arguments, see list below. Thus the program can also be used to do any one of the operations. However to do only e.g. magnitudes, a more rational choice would be to use AUTOMAG only. Similarly for doing only locations, it is more logical to use HYP. The input can be all SEISAN types: a file, a data base or an index file. Outlier removal by HYP can be deselected. An alternative to outlier removal by HYP is to use the HYP option for residual weighting. For more detail of the individual programs, see program descriptions elsewhere in SEISAN manual. There are always two output files: auto.out: all events with the final results with updated values of location and magnitude. auto.log: a summary of what has been done to each event. If input is from a data base or index file, the results are also written back to the data base, overwriting what was there from before. The ID line is not updated by AUTO. When picking automatically, a few bad picks can throw the solution off so for a local event, the distance might wrongly be very large. It is then important that the range used for distance weighting in HYP is large so the initial wrong location can be made and the outlier rejection gets a chance to eliminate bad picks. Doing fault plane solution is only recommend with very good data. It is then important to use at least the default values set for both minimum number of polarities and maximum az gap. Note the gap is not the gap as in a hypocenter solutions but the gap when polarities are plotted on the projection on the focal sphere.

See list of argument below

 Examples of argument use:

auto          : pick phases, locate and do magnitudes
auto l m      : only pick phases with AUTOPHASE
auto j l m    : only pick phases with AUTOPIC
auto i l       : only do magnitudes
auto s 30     : all defaults except that spectral window for magnitude is 30 s
auto l m i f  : only do fault plane solutions with FPFIT
auto l m i ar h f: automatic amplitudes and fault plane solution with HASH

Writing auto help will give all the possible arguments:

c:\seismo\PRO>auto help
 Write auto help to get list of arguments. If no
 arguments, all defaults are used. This is autophase,
 location with outlier rejection and ML and spectral Mw

  s xxx : do spectrum for Mw, xxx is window length, default
          is 20 s. if 0, no spectrum. Default is spectrum

  w xxx : amplitude for Ml, xxx is window length
          if zero, no amplitude. Default is do amplitude
          default window length is 50 s

  n or e: use N or E component for magnitude, respectively.
          Default is Z

  p     : use P for Mw, default is S

  l     : do not locate, def. is to locate with rejection
          if magnitudes are done, location is also done
          after determination of amps and spectra in order
          to update magnitudes on header line

  r     : do not remove outliers when locating

  m     : no magnitude, default is to do magnitude

  i     : no phase pick, default is to do phase pick

  j     : if phase pick, use autopic, default is autophase

  ar    : do autoratio, default is not to do

  af  xx xx  : filter for autoratio, default is 2-4 Hz

  ad  xx   : max distance for autoratio, def 100km

  at  xx  : time window for autoratio, def 2 s

  ag  xx   : ground motion for autoratio, def 0 for none

  az    : if given, use z for s in autoratio

  f     : do fault plane solution with fpfit, def.  is not

  h     : do fault plane solution with hash, def.  is not

  n     : minimum number of polarities for fps, default 10

  g     : maximum gap for fps, default 180


next up previous contents index
Next: 21.2 AUTOPHASE, automatic phase Up: 21. Automatic routines in Previous: 21. Automatic routines in   Contents   Index
Peter Voss : Mon Feb 27 10:16:12 UTC 2017