HASH Manual V1.2

HASH_manual_v1.2

User Manual:

Open the PDF directly: View PDF .
Page Count: 17

Download
Open PDF In Browser	View PDF

HASH: A FORTRAN Program for Computing Earthquake FirstMotion Focal Mechanisms – v1.2 – January 31, 2008
Jeanne L. Hardebeck, US Geological Survey, Menlo Park, CA, jhardebeck@usgs.gov.
Peter M. Shearer, IGPP, Scripps Institution of Oceanography, La Jolla, CA,
shearer@igpp.ucsd.edu.
INTRODUCTION:
Focal mechanisms of small earthquakes are typically determined from P-wave firstmotion polarities. These fault plane solutions are notoriously sensitive to various sources
of error, including imperfect knowledge of the seismic velocity structure. To address this
problem, we have developed a new method (called HASH, for HArdebeck & SHearer)
for producing more stable focal mechanisms. This method generates a set of acceptable
mechanisms for each event given the various sources of uncertainty, and returns the most
likely mechanism. Mechanism quality is assigned based on the solution stability with
respect to model uncertainty, represented by the spread of the acceptable mechanisms.
This technique has been shown to produce more accurate focal mechanisms than prior
methods for cases in which we believe the correct mechanisms are known. For example,
the mechanisms for clusters of events with similar seismic waveforms in the Northridge,
California, aftershock sequence are more similar (Hardebeck & Shearer, 2002; Shearer et
al., 2003), and the mechanisms of events along the San Andreas, Calaveras and Sergeant
Faults in northern California are more consistent with the fault orientations delineated by
earthquake locations (Kilb & Hardebeck, 2006). We have also expanded this technique
to include S-wave to P-wave amplitude ratios.
This manual is intended to help researchers run HASH on their own data sets. The source
code may be obtained at: http://quake.wr.usgs.gov/research/software/#HASH. The
methodology is discussed in more detail in our publications. Please cite these papers if
you use our codes. Thank you!


Hardebeck, Jeanne L. and Peter M. Shearer, A new method for determining firstmotion focal mechanisms, Bulletin of the Seismological Society of America, 92,
2264-2276, 2002.



Hardebeck, Jeanne L. and Peter M. Shearer, Using S/P Amplitude Ratios to
Constrain the Focal Mechanisms of Small Earthquakes, Bulletin of the
Seismological Society of America, 93, 2434-2444, 2003.

As described in our papers, this technique uses a grid-search to determine P-wave
polarity first-motion (or P-polarity and S/P amplitude ratio) focal mechanisms. For each
earthquake, a set of acceptable mechanisms is found. The spread of the acceptable
mechanisms determines the uncertainty and the assigned solution quality. The set of
acceptable mechanisms takes into account the uncertainty in polarity measurements,
1

event location, and takeoff angle (velocity model.) Therefore, you will need an estimate
of the rate of polarity errors in your data, an estimate of the location uncertainty, and an
estimate of the takeoff angle uncertainty (which can be represented as a set of possible
1D velocity models for your region.)
We've tried to make the code as input-format independent as possible. The idea is to
have a main driver code that does the I/O and gets the input data into the internal arrays,
and subroutines that do the actual computation of the focal mechanisms and uncertainties
and deal with the station locations and the seismic velocity models. To use for different
networks and data formats, you should only have to edit the main driver code and the
station subroutines.
The code was developed while we were primarily using data from the Southern
California Seismic Network (SCSN – also sometimes called TriNet), obtained through
the Southern California Earthquake Data Center (SCEDC). Therefore the example
formats are similar to the standard distribution formats of SCEDC phase data, and we
occasionally refer to the SCSN or the SCEDC. We also refer to FPFIT, which is
currently the most widely used focal mechanism program, see Reasenberg &
Oppenheimer (1985). The new example 4, added for the release of HASH 1.2, uses the
current SCEDC phase and station formats as of January 2008, including 5-character
station names.
The programs are in FORTRAN 77, and have been tested by the authors on Sun
workstations of various configurations, and a Mac G4 running OSX (FORTRAN 77
compiler available from FINK: http://fink.sourceforge.net/). Others have used the codes
on Linux. Please contact the authors if any changes need to be made to run on other
platforms. And of course please let us know if you discover any bugs.
OVERVIEW:
The following files should appear in the HASH directory.
Source Code:
 hash_driver1.f
\
 hash_driver2.f
- example main driver programs
 hash_driver3.f
/
 hash_driver4.f - NEW EXAMPLE, updated SCEDC formats, 5 character stations
 hash_driver5.f - NEW EXAMPLE, SIMULPS format for 3D ray tracing
 fmamp_subs.f
- subroutines for computing focal mechanisms, P and S/P
 fmech_subs.f
- subroutines for computing focal mechanisms, P only
 pol_subs.f
- polarity distribution/misfit routines
 station_subs.f
- station location/polarity reversal routines
 station_subs_5char.f - NEW, 5 character station location/polarity routines
 uncert_subs.f
- subroutines to compute mechanism uncertainty
 util_subs.f
- utility subroutines
 vel_subs.f
- subroutines for seismic velocity tables

2

Include Files:
 param.inc
- some parameters to set array sizes
 rot.inc
- determine the grid spacing for focal mechanism search
 vel.inc
- velocity table parameters
Makefile:
 Makefile
Example Control Files (correspond to example driver programs):
 example1.inp: Example with P polarity data only, take-off angle uncertainties are
specified in the input.
 example2.inp: Example with P polarity data only, take-off angle uncertainties are
represented by a suite of 1D velocity models.
 example3.inp: Example with P polarity and S/P amplitude data, take-off angle
uncertainties are represented by a suite of 1D velocity models.
 example4.inp: NEW EXAMPLE, like example2.inp, but with updated SCEDC
formats, 5 character stations.
 example5.inp: NEW EXAMPLE, like example2.inp, but using azimuth and takeoff angles from a SIMULPS-like format file, for 3D ray-tracing
Example Data Files:
 north1.phase
- P polarity phase file, for example 1
 north2.phase
- P polarity phase file, for examples 2 & 3
 north3.amp
- P and S amplitude file, for example 3
 north3.statcor
- station S/P ratio correction, for example 3
 north4.phase - NEW EXAMPLE, P polarity phase file in updated SCEDC format
 north5.simul - NEW EXAMPLE, azimuth and take-off angles from SIMULPS
 scsn.stations
- station locations, for examples 2 & 3
 scsn.stations_5char - NEW 5 character station locations, for example 4
 scsn.reverse
- station polarity reversals, all examples
 vz.socal, etc
- 1D velocity models, for examples 2 & 3
Example Output Files:
 example1.out
- preferred mechanisms, example 1
 example1.out2
- set of acceptable mechanisms, example 1
 example2.out
- preferred mechanisms, example 2
 example2.out2
- set of acceptable mechanisms, example 2
 example3.out
- preferred mechanisms, example 3
 example4.out
- NEW, preferred mechanisms, example 4
 example4.out2
- NEW, set of acceptable mechanisms, example 4
 example5.out
- NEW, preferred mechanisms, example 5
 example5.out2
- NEW, set of acceptable mechanisms, example 5
To Compile, type "make hash_driverX", where “X” is the example number, 1, 2, 3, 4 or
5. To Run, type "hash_driverX" and it runs interactively, or type "hash_driverX <
exampleX.inp" to run using the given control file. If the codes have compiled and are
running correctly, the generated output files should closely match the provided examples.
NOTE: The generated output may not exactly match the samples output. The input for
each Monte Carlo trial is selected randomly, so if there is a difference in random number
generation, this can lead to a small difference in output. Because of this, sometimes the

3

solutions are slightly different by a few degrees (very small compared to the uncertainty),
or the other nodal plane is listed, so the files aren't strictly identical.
RUNNING THE CODE:
Computing Focal Mechanisms:
The driver, or main, program is primarily for input and output. The actual computation
of the focal mechanism for each event is done in three subroutine calls from the main
program. You should edit the main driver routines to most efficiently get your data
formats into the arrays passed to these subroutines.
1. Computing the set of acceptable mechanisms.
Separate but similar subroutines are used depending on whether or not you wish to use
S/P amplitude ratios in addition to P-wave polarities. The inputs to both subroutines
include the P-wave polarities (and S/P amplitude ratios) for the event at a set of stations,
and the azimuth and takeoff angle to each station. An estimate of uncertainty in these
inputs is also required, in order to test the stability of the focal mechanism solutions.
The uncertainty in the azimuth and takeoff angle is represented by multiple sets of
reasonable values of these angles. It’s assumed that these values come from repeated
trials, for instance each trial for a different event depth and/or velocity model. For
example, for trial number 5, the event location and velocity model are held fixed, and the
resulting azimuth and takeoff angles for the 7th station are stored in p_azi_mc(7,5) and
p_the_mc(7,5), for the 8th station are stored in p_azi_mc(8,5) and p_the_mc(8,5),
etc. The polarity, pick quality, and S/P ratios are the same for each trial, and for the 7th
station they are given in p_pol(7), p_qual(7), and sp_amp(7), etc. The number of
trials is given by nmc, and the number of station observations is given by npsta. A set of
acceptable mechanisms will be constructed for each trial, and these will be combined to
make the set of acceptable mechanisms for this event.
The uncertainty in the P-wave polarity picks is accounted for by specifying how many
polarity errors are acceptable. The returned set of acceptable mechanisms will include
mechanisms with up to this many misfit polarities. The number of allowed polarity
misfits is defined as the number of misfit polarities for the best-fit solution plus an
additional nextra. If this total number of allowed misfits is less than ntotal, then
ntotal is used as the number of allowed misfits. We usually use a value of ntotal
equal to the total number of polarity observations times the known rate of polarity errors
for the network, therefore allowing for at least the expected number of polarity misfits.
We usually use a nextra half of ntotal. Similarly, for the S/P amplitude ratios, the
allowed log10(S/P) misfit is defined as the best-fit solution misfit plus qextra, or qtotal
if this is greater. We usually used qtotal equal to the total number of S/P observations
times the estimated average uncertainty, and qextra half of qtotal.

4

There are two additional control parameters: dang, which specifies the resolution of the
focal mechanism grid search, and maxout, which gives the maximum desired number of
acceptable focal mechanisms output.
The total number of acceptable mechanisms found is given by nf (but a maximum of
maxout are returned.) For each acceptable mechanism, the mechanism parameters and
the normal vectors to the two nodal planes are returned.
For P-wave polarity data only:
subroutine FOCALMC(p_azi_mc,p_the_mc,p_pol,p_qual,npsta,nmc,
dang,maxout,nextra,ntotal,nf,strike,dip,rake,faults,slips)

Inputs:
 p_azi_mc(npsta,nmc)
 p_the_mc(npsta,nmc)





p_pol(npsta)
p_qual(npsta)
npsta
nmc




dang
maxout

- azimuth to station from event (degrees E of N)
- takeoff angle (degrees from vertical: up=0; <90 upgoing; >90 down-going)
- first motion, 1=up, -1=down
- quality, 0=impulsive, 1=emergent
- number of first motion observations
- number of trials, ie number of possible azimuth/
takeoff angle pairs given for each station
- angle spacing for grid search (degrees)
- maximum number of fault planes to return: if more
are found, a random selection will be returned
- number of additional misfits allowed
- minimum number of total allowed misfits

 nextra
 ntotal
Outputs:
 nf
 strike(min(maxout,nf))
 dip(min(maxout,nf))
 rake(min(maxout,nf))
 faults(3,min(maxout,nf))
 slips(3,min(maxout,nf))

- number of focal mechanisms found
- strike
(the choice of the “fault” and
- dip
“auxiliary” plane is arbitrary.)
- rake
- fault normal vector
- slip vector

For P-wave polarity and S/P amplitude ratio data:
subroutine FOCALAMP_MC(p_azi_mc,p_the_mc,sp_amp,p_pol,npsta,nmc,
dang,maxout,nextra,ntotal,qextra,qtotal,nf,strike,dip,rake,
faults,slips)

Inputs:
 p_azi_mc(npsta,nmc)
 p_the_mc(npsta,nmc)





sp_amp(npsta)
p_pol(npsta)
npsta
nmc



dang

- azimuth to station from event (degrees E of N)
- takeoff angle (degrees from vertical: up=0; <90 upgoing; >90 down-going)
- amplitude ratios, log10(S/P)
- first motion, 1=up, -1=down
- number of observations
- number of trials, ie number of possible azimuth/
takeoff angle pairs given for each station
- angle spacing for grid search

5



maxout

- maximum number of fault planes to return: if more
are found, a random selection will be returned
- number of additional polarity misfits allowed
- minimum number of total allowed polarity misfits
- additional amplitude misfit allowed
- minimum allowed total amplitude misfit

 nextra
 ntotal
 qextra
 qtotal
Outputs:
 nf
 strike(min(maxout,nf))
 dip(min(maxout,nf))
 rake(min(maxout,nf))
 faults(3,min(maxout,nf))
 slips(3,min(maxout,nf))

- number of fault planes found
- strike
(the choice of the “fault” and
- dip
“auxiliary” plane is arbitrary.)
- rake
- fault normal vector
- slip vector/ normal to auxiliary plane

2. Computing the preferred, or most probable, mechanism.
After the set of acceptable mechanisms is computed, they are used to determine the
preferred mechanism and some estimates of the quality of the solution. The preferred
solution, or the most probable solution, is the average of the acceptable fault plane
solutions after outliers have been removed. Two types of uncertainty are computed, the
RMS difference of the acceptable nodal planes from the preferred planes, and the
probability that the real mechanism is “close” to the preferred mechanism, with a userspecified angle defining “close.” If there are clustered outliers, alternative solutions (or
“multiples”) are found based on those outliers. You can set the minimum probability for
the multiples (ie ignore multiples with a low probability.)
subroutine MECH_PROB(nf,norm1in,norm2in,cangle,str_avg,
dip_avg,rak_avg,prob,rms_diff)

Inputs:
 nf
- number of fault planes
 norm1(3,nf) - normal to fault plane
 norm2(3,nf) - slip vector/ normal to auxiliary plane
 cangle
- mechanisms are “close” if less than this angle apart (degrees)
 prob_max
- cut-off probability for multiples
Output:
 nstln
- number of solutions (multiples) returned
 str_avg(nstln)
- strike of each solution
 dip_avg(nstln)
- dip of each solution
 rak_avg(nstln)
- rake of each solution
 prob(nstln)
- percent of mechanisms “close” to average mechanism
 rms_diff(2,nstln) - RMS angular difference (degrees) of all planes to average
plane (1=fault plane, 2=auxiliary plane)
3. Computing the data misfit for the preferred mechanism.
The final step is to find the data misfit for the preferred mechanism. Separate but similar
subroutines are used depending on whether S/P ratios are used along with P-polarity data.
6

The input is the set of polarity (and amplitude ratio) observations for the stations, the
azimuth and takeoff angle to each station (only one set of angles this time, use the
preferred values), and the preferred mechanism. The outputs are mfrac, the weighted
fraction of misfit polarities (like the weighted misfit returned by FPFIT); and stdr, the
station distribution ratio (also like the station distribution ratio of FPFIT.) If S/P
amplitude ratios are used, the average log10(S/P) misfit, mavg, is also computed.
For P-wave polarity data only:
subroutine GET_MISF(npol,p_azi_mc,p_the_mc,p_pol,p_qual,
str_avg,dip_avg,rak_avg,mfrac,stdr)

Inputs:
 npol
 p_azi_mc(npol)
 p_the_mc(npol)
 p_pol(npol)
 p_qual(npol)
 str_avg,dip_avg,rak_avg
Outputs:
 mfrac
 stdr

- number of polarity observations
- azimuths
- takeoff angles
- polarity observations
- polarity quality
- preferred mechanism
- weighted fraction misfit polarities
- station distribution ratio

For P-wave polarity and S/P amplitude ratio data:
subroutine GET_MISF_AMP(npol,p_azi_mc,p_the_mc,sp_ratio,p_pol,
str_avg,dip_avg,rak_avg,mfrac,mavg,stdr)

Inputs:
 npol
 p_azi_mc(npol)
 p_the_mc(npol)
 sp_ratio(npol)
 p_pol(npol)
 str_avg,dip_avg,rak_avg
Outputs:
 mfrac
 mavg
 stdr

- number of observations
- azimuths
- takeoff angles
- S/P ratios
- polarity observations
- preferred mechanism
- weighted fraction misfit polarities
- average log10(S/P) misfit
- station distribution ratio

Input and Data Preparation:
As mentioned earlier, we recommend loading your data into the input arrays in the most
efficient manner possible, given your file formats, etc. We have included some example
driver programs illustrating how one might do this. These examples use data from one of
the Northridge, California, similar event clusters discussed by Hardebeck & Shearer
(2002), Shearer et al. (2003), and Hardebeck & Shearer (2003).
In example 1, we assume that we have already computed the azimuth and takeoff angle to
each station, and have estimated the uncertainty of both of these angles. The input format
(file: north1.phase) is a modified version of the standard FPFIT input format, expanded to

7

include the azimuth and takeoff angle uncertainties. In this case much of the data can be
read directly into the input arrays. The azimuth and takeoff angle for each trial are
computed by randomly selecting values from normal distributions with the given average
value and standard deviation.
Because there are known to be periods of time in which some stations have reversed
polarity (up appears as down, and vice-versa), the subroutine CHECK_POL is called to
check each station for a polarity reversal at the time of the earthquake. The polarity
reversal file format (file: scsn.reverse) is exactly the same as for FPFIT.
A few control parameters are read in. The parameters dang, nmc, maxout, and cangle
are passed directly to the focal mechanism subroutines, and are described above. The
parameter badfrac is the user’s estimate of the error rate for (impulsive) polarities in the
data set, and is used to compute nextra and ntotal, as described above.
The additional parameters are used to select what data is included in the input arrays.
The parameter delmax is the maximum source-station distance. The minimum required
number of polarity data is given by npolmin, and the maximum acceptable data gaps in
the azimuth and takeoff angle directions are given by max_agap and max_pgap. Focal
mechanisms are not found for events not meeting these last two criteria.
In example 2, we assume that we don’t have a good idea of the uncertainty in the azimuth
and takeoff angles. A 1D ray-tracing routine is used to compute the takeoff angles for
plausible velocity models (the azimuth to a station is the same for any 1D model.) Each
trial uses a different combination of (gradient) velocity model and perturbed source depth
to compute a set of takeoff angles.
The polarity input is given in another format similar to FPFIT (file: north2.phase.)
Because the azimuth and takeoff angle to each station is being computed, a file with
station names and locations is needed (file: scsn.stations.) The subroutine GETSTAT_TRI
is used to look up the station locations.
The velocity models are read in (files: vz.socal, etc), and a set of take-off angle tables are
generated, by a call to the subroutine MK_TABLE. The takeoff angles for a given sourcestation range and event depth are found by a call to the subroutine GET_TTS. The first
argument to this subroutine indicates which of the 1D velocity models to use.
In example 3, S/P amplitude ratios are also used. An additional input file with the P and
S amplitudes, as well as the noise levels prior to the arrival of the P and S waves, is
included (file: north3.amp.) We apply a station correction of a constant shift in
log10(S/P) (Hardebeck & Shearer, 2003), and therefore need to input a file with these
corrections (file: north3.statcor.)
Two additional control parameters are also needed. We limit the S/P observations to
those waveforms with signal to noise ratio of at least ratmin. In order to estimate the
allowed log10(S/P) misfit (qextra and qtotal, above), an estimate of the log10(S/P)

8

uncertainty is given by qbadfrac. For this example, it is very important that each event
have a unique ID number, as the ID is used to match the P-polarity and S/P amplitudes
from different files.
NEW: Example 4 is an update to the most recent SCEDC formats, which include 5character stations names.
NEW: Example 5 uses output from the program SIMULPS (Evans et al., 1994) to get the
azimuth and take-off angles. This is useful for computing focal mechanisms with raytracing in a 3D velocity model.
You can edit these examples to match your data format. You will probably need to
change the file format for the location/polarity input files (these formats are indicated
with comments.) You will probably also need to alter the station subroutines
GETSTAT_TRI and CHECK_POL (file: station_subs.f) to match the format of your station
lists. The idea of these subroutines is to have two look-up tables, one to get the location
of a given station, and the other to check for known polarity reversals. Storing the
stations in alphabetical order allows for quicker searching.
Include Files:
param.inc – sets maximum size of input/output arrays
 npick0
- maximum number of picks per event
 nmc0
- maximum number of trials of location/take-off angles
 nmax0
- maximum number of acceptable mechanisms output
rot.inc – sets the minimum grid angle size, and the corresponding maximum number of
grid points and possible stored mechanisms (need to be changed together)
 dang0
- minimum grid spacing (degrees)
 ncoor
- number of test mechanisms
vel.inc – parameters for the 1D ray tracer (if used)
 nx0
- maximum number of rows in table (range direction)
 nd0
- maximum number of columns in table (depth direction)
 nindex - maximum number of 1D models/ tables
 dep1
- minimum event depth (km)
 dep2
- maximum event depth (km)
 dep3
- depth step for table (km), ((dep2-dep1)/dep3)+1 ≤ nd0
 del1
- minimum source-station range (km)
 del2
- maximum source-station range (km)
 del3
- depth step for range (km), ((del2-del1)/del3)+1 ≤ nx0
 pmin
- minimum ray parameter
 nump
- number of rays traced
Output:

9

The output format is something that you can alter to fit your needs. Our examples
typically generate two output files, one with a single line per earthquake giving the
preferred mechanism, and a second file with a list of all of the acceptable mechanisms for
each event. We have developed the following mechanism quality criteria, you may wish
to develop your own. Initial tests (Kilb & Hardebeck, 2005) show that the best singleparameter indicator of mechanism quality is the average RMS fault plane uncertainty,
0.5*(rms_diff(1)+rms_diff(2)), with values ≤ 35° indicating the best mechanisms.
quality average misfit RMS fault plane
station distribution mechanism
(qual) (mfrac)
uncertainty
ratio (stdr)
probability (prob)
----------------------------------------------------------------------------------------------------------A
≤ 0.15
≤ 25°
≥ 0.5
≥ 0.8
B
≤ 0.20
≤ 35°
≥ 0.4
≥ 0.6
C
≤ 0.30
≤ 45°
≥ 0.3
≥ 0.7
D
maximum azimuthal gap ≤ 90°, maximum takeoff angle gap ≤ 60°
E
maximum azimuthal gap > 90°, maximum takeoff angle gap > 60°
F
fewer than 8 polarities
FILE FORMATS:
These are the input and output file formats as used in the examples. If your data is in a
different format, we suggest editing the driver code to be consistent with your format,
instead of writing format conversion routines.
Input Files:
P-polarity Files:
Example 1: north1.phase, Similar to FPFIT input file:
Event line:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-10
5i2
origin time, year, month, day, hour, minute
11-14
f4.2
origin time, seconds
15-16
i2
latitude, degrees
17
a1
latitude, 'S'=south
18-21
f4.2
latitude, minutes
22-24
i3
longitude, degrees
25
a1
longitude, 'E'=east
26-29
f4.2
longitude, minutes
30-34
f5.2
depth, km
35-36
f2.1
magnitude
81-88
2f4.2
horizontal and vertical uncertainty, km

10

123-138
a16
event ID
-----------------------------------------------------------------------------------Polarity lines:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-4
a4
station name
7
a1
polarity : U, u, +, D, d, or 8
i1
quality: 0=good quality, 1=lower quality, etc
59-62
f4.1
source-station distance (km)
66-68
i3
takeoff angle
79-81
i3
azimuth
83-85
i3
takeoff angle uncertainty ** NOT in standard FPFIT files
87-89
i3
azimuth uncertainty
** NOT in standard FPFIT files
-----------------------------------------------------------------------------------Example 2: north2.phase, Similar to old SCEDC phase format:
Event line:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-4
i4
origin time, year
5-12
4i2
origin time, month, day, hour, minute
13-17
f5.2
origin time, seconds
18-19
i2
latitude, degrees
20
a1
latitude, 'S'=south
21-25
f5.2
latitude, minutes
26-28
i3
longitude, degrees
29
a1
longitude, 'E'=east
30-34
f5.2
longitude, minutes
35-39
f5.2
depth, km
89-93
f5.2
horizontal location uncertainty, km
95-99
f5.2
vertical location uncertainty, km
140-143
f4.2
magnitude
150-165
a16
event ID
-----------------------------------------------------------------------------------Polarity lines:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-4
a4
station name
6-7
a2
station network
10-12
a3
station component
14
a1
P onset, I or E
16
a1
P polarity : U, u, +, D, d, or -

11

-----------------------------------------------------------------------------------Example 4: north4.phase, Similar to NEW SCEDC phase format:
Event line:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-4
i4
origin time, year
5-12
4i2
origin time, month, day, hour, minute
13-16
f4.2
origin time, seconds
17-18
i2
latitude, degrees
19
a1
latitude, 'S'=south
20-23
f4.2
latitude, minutes
24-26
i3
longitude, degrees
27
a1
longitude, 'E'=east
28-31
f4.2
longitude, minutes
32-36
f5.2
depth, km
131-146
a16
event ID
148-150
f3.2
magnitude
-----------------------------------------------------------------------------------Polarity lines:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-5
a5
station name
6-7
a2
station network
10-12
a3
station component
14
a1
P onset, i, I, e or E
16
a1
P polarity : U, u, +, D, d, or -----------------------------------------------------------------------------------S/P Amplitude Files:
Example 3: north3.amp:
Event line:
-----------------------------------------------------------------------------------free format: event_ID number_of_observations
-----------------------------------------------------------------------------------Amplitude lines:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-4
a4
station name
6-8
a3
station network

12

10-11
a2
station component
29-38
f10.3
noise level before P arrival
40-49
f10.3
P amplitude (arbitrary units)
51-60
f10.3
noise level before S arrival
62-71
f10.3
S amplitude (same units as P)
-----------------------------------------------------------------------------------SIMULPS 3D ray-tracing Files:
Example 5: north5.simulps:
Event line:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------2-7
3i2
origin time, year, month, day
9-12
2i2
origin time, hour, minute
14-18
f5.2
origin time, seconds
20-21
i2
latitude, degrees
22
a1
latitude, 'S'=south
23-27
f5.2
latitude, minutes
29-31
i3
longitude, degrees
32
a1
longitude, 'E'=east
33-37
f5.2
longitude, minutes
39-44
f6.2
depth, km
48-51
f4.2
magnitude
56-63
a8
event ID
-----------------------------------------------------------------------------------Station lines:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------2-5
a4
station name
7-11
f5.1
range
12-15
i4
azimuth
16-19
i4
take-off angle
66-69
f4.0
azimuth uncertainty ** NOT in standard SIMULPS files
70-73
f4.0
take-off uncertainty ** NOT in standard SIMULPS files
-----------------------------------------------------------------------------------Velocity Model Files:
Velocity files, free format:
depth(km) P_velocity(km/s)
Station Files:

13

Station location files – must be alphabetical by station name
Old SCEDC format (examples 1-3, 5; station_subs.f; scsn.stations)
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-4
a4
station name
6-8
a3
station component
42-50
f9.5
station latitude (degrees, signed)
52-61
f10.5
station longitude (degrees, signed)
63-67
i5
station elevation (meters)
91-92
a2
network code
-----------------------------------------------------------------------------------NEW SCEDC format (example 4; station_subs_5char.f; scsn.stations_5char)
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-2
a2
network code
5-9
a5
station name
11-13
a3
station component
61-69
f9.5
station latitude (degrees, signed)
71-80
f10.5
station longitude (degrees, signed)
82-86
i5
station elevation (meters)
-----------------------------------------------------------------------------------Polarity reversal file (same format as FPFIT)
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-4(5)
a4(5)
station name
6-9
i4
beginning of reversal: year
10-11
i2
month
12-13
i2
day
15-18
i4
end of reversal: year
19-20
i2
month
21-22
i2
day
-----------------------------------------------------------------------------------Output Files:
Output File 1: preferred mechanisms (1 line per event)
-----------------------------------------------------------------------------------columns
format
value

14

-----------------------------------------------------------------------------------1-16
a16
event name (e.g. cuspid for SCEC events)
18-21
i4
origin time, year
23-24
i2
origin time, month
26-27
i2
origin time, day
29-30
i2
origin time, hour
32-33
i2
origin time, minute
35-40
f6.3
origin time, second
42
a1
event type (L-local, R-regional, T-teleseism
Q-quarry, D-dubious or duplicate event)
44-48
f5.3
magnitude
50
a1
magnitude type:
e energy magnitude
w moment magnitude
b body-wave magnitude
s surface-wave magnitude
l local (Wood-Anderson)
c coda amplitude magnitude
h heliocorder magnitude
d coda duration magnitude
52-60
f9.5
hypocenter latitude (decimal degrees)
62-71
f10.5
hypocenter longitude (decimal degrees)
73-79
f7.3
hypocenter depth (km)
81
a1
location quality (SCEC qualities)
A rms< 0.15, erh< 1 km erz< 2 km
B rms< 0.30, erh< 2.5 km erz< 5 km
C rms< 0.50, erh< 5 km --D greater than above
E
erh> 90 km or < 3 stations
83-89
f7.3
RMS from SCEC or L-2 norm type locations
91-97
f7.3
horizontal error (km)
99-105
f7.3
depth error (km)
107-113
f7.3
origin time error (sec)
117-120
i4
number of picks used in the location
122-125
i4
number of P picks
127-130
i4
number of S picks
132-135
i4
focal mechanism strike (degrees)
137-139
i3
focal mechanism dip (degrees)
141-144
i4
focal mechanism rake (degrees)
148-149
i2
fault plane uncertainty (degrees)
151-152
i2
auxiliary plane uncertainty (degrees)
154-156
i3
number of P first motion polarities
158-159
i2
weighted percent misfit of first motions
161
a1
focal mechanism quality: A-F
163-165
i3
probability mechanism close to solution
167-168
i2
100*(station distribution ratio)

15

170-172
i3
number of S/P ratios
174-176
i3
100*(average log10(S/P) misfit)
178
a1
multiple flag (* if there are multiples for this event)
-----------------------------------------------------------------------------------Output File 2: set of acceptable mechanisms (multiple lines per event)
Event line:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------1-4
i4
origin time, year
6-7
i2
origin time, month
9-10
i2
origin time, day
12-13
i2
origin time, hour
15-16
i2
origin time, minute
18-23
f6.3
origin time, second
26-28
f3.1
magnitude
30-38
f9.4
hypocenter latitude (decimal degrees)
40-49
f10.4
hypocenter longitude (decimal degrees)
51-56
f6.2
hypocenter depth (km)
58-65
f8.4
horizontal error (km)
67-74
f8.4
depth error (km)
76-80
i5
number of P first motion polarities
82-86
i5
number of acceptable mechanisms output
88-103
a16
event name
105-111
f7.1
focal mechanism strike (degrees)
113-118
f6.1
focal mechanism dip (degrees)
120-126
f7.1
focal mechanism rake (degrees)
128-133
f6.1
fault plane uncertainty (degrees)
135-140
f6.1
auxiliary plane uncertainty (degrees)
142-148
f7.3
weighted fraction misfit of first motions
151
a1
focal mechanism quality: A-F
153-159
f7.3
probability mechanism close to solution
161-164
f4.2
station distribution ratio
-----------------------------------------------------------------------------------Mechanism lines:
-----------------------------------------------------------------------------------columns
format
value
-----------------------------------------------------------------------------------6-14
f9.2
focal mechanism strike (degrees)
15-23
f9.2
focal mechanism dip (degrees)
24-32
f9.2
focal mechanism rake (degrees)
33-41
f9.4
normal vector to fault plane, x (N) component
42-50
f9.4
normal vector to fault plane, y (E) component

16

51-59
f9.4
normal vector to fault plane, z (D) component
60-68
f9.4
normal vector to auxiliary plane, x (N) component
69-77
f9.4
normal vector to auxiliary plane, y (E) component
78-86
f9.4
normal vector to auxiliary plane, z (D) component
-----------------------------------------------------------------------------------REFERENCES:
Evans, J. R., D. Eberhart-Phillips, C. H. Thurber, User's manual for SIMULPS12 for
imaging Vp and Vp/Vs; a derivative of the "Thurber" tomographic inversion
SIMUL3 for local earthquakes and explosions, U. S. Geological Survey Open-File
Report: OF 94-0431, 101 pp, 1994.
Hardebeck, J. L. and P. M. Shearer, A new method for determining first-motion focal
mechanisms, Bull. Seism. Soc. Am., 92, 2264-2276, 2002.
Hardebeck, J. L. and P. M. Shearer, Using S/P Amplitude Ratios to Constrain the Focal
Mechanisms of Small Earthquakes, Bull. Seism. Soc. Am., 93, 2434-2444, 2003.
Kilb, D. and J. L. Hardebeck, Fault Parameter Constraints Using Relocated Earthquakes:
Validation of First Motion Focal Mechanism Data, Bull. Seism. Soc. Am., 96, 11401158, 2006.
Reasenberg, P., and D. Oppenheimer, FPFIT, FPPLOT, and FPPAGE: FORTRAN
computer programs for calculating and displaying earthquake fault-plane solutions,
U.S. Geol. Surv. Open File Rep. 85-739, 109 pp, 1985.
Shearer, P. M., J. L. Hardebeck, L. Astiz, and K. B. Richards-Dinger, Analysis of similar
event clusters in aftershocks of the 1994 Northridge, California, earthquake, J.
Geophys. Res., 108, B1, DOI:10.1029/2001JB000685, 2003.

17



---

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.3
Linearized                      : No
Page Count                      : 17
Title                           : Microsoft Word - HASH_manual_v1.2.doc
Author                          : Jeanne Hardebeck
Producer                        : Mac OS X 10.10.5 Quartz PDFContext
Creator                         : Word
Create Date                     : 2018:12:04 23:54:43Z
Modify Date                     : 2018:12:04 23:54:43Z

EXIF Metadata provided by EXIF.tools