MLOC is a multiple-events hypercentroid decomposition relocation program developed and maintained by Eric Bergman since 1990 whose underlying idea is based on Jordan and Sverdrup (1981). This tutorial is a skeleton iteration based on the MLOC Training Workshop organized by Eric and Ezgi that was held in Albuquerque, NM from 10/15/18 to 10/19/18. Further update of this software will be reflected upon Eric's notification.
Eric has never personally distributed this software community-widely but he is fine with redistributing it by those who attended his MLOC workshop. Therefore, I keep the track of this software under my personal private repository and impose GPLv3 license here.
Before start, some prerequisites, including GMT5 and a Fortran compiler, must be met, for instance:
➜ MLOC git:(master) ✗ gmt --version
5.4.4
➜ MLOC git:(master) ✗ gfortran --version
GNU Fortran (Homebrew GCC 8.2.0) 8.2.0
Copyright (C) 2018 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.First clone this repository to your local:
git clone https://github.com/shipengcheng1230/MLOC.gitThen download the tables file here and put it under MLOC/mloc_working/tables. Your directory should look like this:
├── Docs
│ ├── Focal Depth
│ ├── HD calibration cartoons
│ ├── Hi-Rez topographic datasets.rtfd
│ └── Papers
├── MNF utilities
│ ├── mnf search
│ └── to mnf
├── clusters
│ └── Tunisia
├── mloc utilities
│ ├── lres
│ ├── null_test
│ ├── rstat
│ └── xdat
├── mloc_gfortran
├── mloc_intel
├── mloc_src
└── mloc_working
├── Utilities
├── jsa4
└── tablesA brief summary of this program is as below:
Docscontains all the related references and manuals.MNF utilitiescontains functionalities for transforming ISC bulletin to.mnfformat.mloc utilitiescontains functionalities for MLOC processing procedures.mloc_gfortranandmloc_intelcontainmakefilefor buildingmlocprogram .mloc_srccontains the source code.mloc_workingwill be your primary working directory, in whichtablehas all the geological models used by this program.
Now compile source code and any other utilities, for example using gfortran:
cd mloc_gfortran && make && cp ./mloc_g ../mloc_workingModify mloc.conf such that WORKING_DIR points to your mloc_working directory and AUTHOR to your name not exceeding 8 characters (for more configuration see docs):
WORKING_DIR: /Users/spc/Softwares/mloc/mloc_working
AUTHOR: SPC
Now it should not have problem running the program:
➜ mloc_working git:(master) ✗ ./mloc_g
mloc v10.4.5, release date 9/28/2018
Current program limits:
nevmax = 200
nqmax = 4000
ntmax1 = 35000
Enter a basename for this run: If you cannot come to this step, please pull up an issue.
We will use ISC bulletin data which contain all the possible phases identified by the agency. It is also desired to use own waveform data to pick phases manually to proliferate our data coverage but we will not cover this here for now.
The database contains two type: Reviewed ISC Bulletin and ISC Bulletin where the first one are those reviewed for correction. For output format we will choose ISF Bulletin. QuakeML cannot work with MLOC for now. Within the search region box, it falls on your own choice to draw the region and time period you prefer. At last, check Output phases which we will use later and there is no need to check Output web links. After clicking Search bulletin, you will be redirected towards the results page. Copy and paste all the contents and save to your local in which the file looks like:
International Seismological Centre
ISC: On-Line Bulletin
Any use of data from the ISC should be cited. The correct format for citations may be found on our citation page.
The ISC Bulletin has been rebuilt for the period 1964-1979. All ISC searches will now return the upgraded set of data for this period. The work on the Rebuild project continues for the period 1980-2010 and will result in further gradual bulletin updates in due time.
Once the search has completed, a compressed KML file will be available to view the results in Google Earth.
Make an event map
Search summary:
Database: ISC Bulletin
Search type: Circular search
Central latitude: 34.1
Central longtitude: 9.9
Radius: 200 km
Start date: 1960-01-01 00:00:00
End date: 2018-10-19 00:00:00
Events found: 215
DATA_TYPE BULLETIN IMS1.0:short
ISC Bulletin
Event 876000 Tunisia
Date Time Err RMS Latitude Longitude Smaj Smin Az Depth Err Ndef Nsta Gap mdist Mdist Qual Author OrigID
1961/01/21 03:45:25 35.2500 10.5000 uk BCIS 1901692
For simplicity, you will find a piece of sample data named test.dat in MLOC/clusters/Tunisia/Test/Data that centered somewhere in Tunisia with radius of 200 km dated from 1960 to 2018.
We will use the utility named isc_ims2mnf to obtain a .mnf style data. It is recommended to recompile them from MLOC/MNF utilities to suit your local library.
➜ Data git:(master) ✗ ./isc_ims2mnf
Release date April 6, 2018, writing MNF v1.3.3
Enter input filename:
test.dat
Bulletin output?:
y
Bulletin comment:
Tunisia test data
EOF reached after 215 eventsThis will generate a single .mnf bulletin if you choose y for Bulletin output?, which looks like:
B Tunisia test data
F MNF v1.3.3
E Tunisia
I ISC 876000
H = 1961/01/21 03 45 25.00 35.2500 10.5000 0.0 BCIS 1901692
STOP
E Tunisia
I ISC 853630
H = 1965/09/05 22 06 55.58 34.1967 8.6501 10.0 ISC 00876034
M 4.3 mb ISC 00876034
P ISO 10.05 353 Pn 1965 9 5 22 9 20.8 -1 0.8 Pn . .ISO . . ISC 28078860
P IFR 11.48 271 Pn 1965 9 5 22 9 40.6 -1 0.9 Pn . .IFR . . ISC 28078861
P WRM 15.81 352 P 1965 9 5 22 10 48.6 -1 6.6 P . .WRM . . ISC 28078862
P DOU 16.17 351 P 1965 9 5 22 10 48.3 -1 2.3 P . .DOU . . ISC 28078863
P EKA 22.67 342 P 1965 9 5 22 11 57.0 -1 0.1 P . .EKA . . ISC 28078864
P UPP 26.34 10 P 1965 9 5 22 12 30.8 -1 -0.6 P . .UPP . . ISC 28078865
P UME 30.50 10 P 1965 9 5 22 13 6.6 -1 -1.9 P . .UME . . ISC 28078866
P KIR 34.36 8 P 1965 9 5 22 13 35.3 -1 -7.0 P . .KIR . . ISC 28078867
P SHI 37.29 85 P 1965 9 5 22 14 9.0 -1 0.9 P . .SHI . . ISC 28078868
P SCH 54.77 317 P 1965 9 5 22 16 23.9 -1 -1.4 P . .SCH . . ISC 28078869
P CPO 74.07 302 P 1965 9 5 22 18 31.0 -1 -1.0 P . .CPO . . ISC 28078870
P WMO 83.41 308 P 1965 9 5 22 19 23.0 -1 -0.4 P . .WMO . . ISC 28078871
P UBO 86.51 318 P 1965 9 5 22 19 40.0 -1 1.0 P . .UBO . . ISC 28078872
P BMO 87.30 325 P 1965 9 5 22 19 41.0 -1 -1.7 P . .BMO . . ISC 28078873
P TFO 91.65 314 P 1965 9 5 22 20 5.0 -1 1.6 P . .TFO . . ISC 28078874
STOP
We then use mnf_search to refine our search, for instance removing those without much phases read:
➜ Data git:(master) ✗ ./mnf_search
Release date October 7, 2018
Enter input filename:
test.dat.mnf
Create a new bulletin (1) or individual event files (2)?
2
Create mloc command file?
y
Enter command file basename:
test1.1
Use lat-lon limits?
n
Use focal depth limit?
n
Use nearest station distance?
n
215 events read
205 events pass the search criteria
80 events that pass the search criteria and have 10 or more phase readings
51 events that pass the search criteria and have 20 or more phase readings
41 events that pass the search criteria and have 30 or more phase readings
32 events that pass the search criteria and have 40 or more phase readings
30 events that pass the search criteria and have 50 or more phase readings
24 events that pass the search criteria and have 60 or more phase readings
21 events that pass the search criteria and have 70 or more phase readings
18 events that pass the search criteria and have 80 or more phase readings
15 events that pass the search criteria and have 90 or more phase readings
13 events that pass the search criteria and have 100 or more phase readings
11 events that pass the search criteria and have 110 or more phase readings
11 events that pass the search criteria and have 120 or more phase readings
11 events that pass the search criteria and have 130 or more phase readings
11 events that pass the search criteria and have 140 or more phase readings
10 events that pass the search criteria and have 150 or more phase readings
8 events that pass the search criteria and have 160 or more phase readings
7 events that pass the search criteria and have 170 or more phase readings
6 events that pass the search criteria and have 180 or more phase readings
6 events that pass the search criteria and have 190 or more phase readings
6 events that pass the search criteria and have 200 or more phase readings
Minimum number of phase arrivals:
50
Event number selection: beginning and end numbers:
1 215
EOF reached after 215 events
30 events selectedNow we will have a list of .mnf file each of which contains one event with at lease 50 phases identified. Those will be the events we are going to relocate.
You should now see a command file (all command file are ended with .cfil) named test1.1.cfil as you previously specified in your mnf_search, which looks like a bunch of repetition of:
memb
even 19670126.1611.42
inpu 19670126.1611.42.mnf
memb
even 19680423.2230.24
inpu 19680423.2230.24.mnf
memb
Here, memb, even and input together specify each of the input events with their unique names and input .mnf files, which will be used by mloc for obtaining phase information as well as travel time.
First of all, let's make a new folder named test in our primary working directory, again which is MLOC/mloc_working, and copy paste all the .mnf file as well as the command file .cfil there from previous data folder. Let's take a look at cfil header.txt which is in the working folder:
pltt 1 2 4 5 6 7
dem1 globe
bdps tables/stn/bdps.dat
ppri pP
ppri sP
dcal on
phyp off
hlim 0. 1.0
clim 0. 180.
wind 3 4
frec 1 1 0 1
freh 1 1 0 1
depc 16
This file serves as a list of common commands that will be used almost everywhere in your analysis with MLOC. A brief summary is as below:
pltt: Control which kind of plot will be produceddem1: Determine which global velocity model will be used, here is AK135. For regions in oceans,etopo1is more preferential.bdps: Provide a list of stations that are known to produce bogus phase information.ppri: Prevent phase re-identification. MLOC has a number of procedures that will re-identify phases that are not reasonable w.r.t. velocity models we use.dcal: Direction calibration. Since we have local data with numerous Pn/Pg, it is possible to do direct calibration as oppose somewhere in the middle of ocean where only tele-seismic phase are available.phyp: P phases only for hypercentroid. For some places where S phases are prone to large reading errors, it is better to be turned on.hlim: Distance limit (in degrees) in data used for hypercentroidclim: Distance limit (in degrees) in data used for cluster vector (all data are used here)wind: Determine the weight coefficient w.r.t. sigma (confidential interval)frec: Free parameters for clusters: latitude longitude depth origin_time. It is often recommended not to do free depth relocation at first before removing outliers.freh: Free parameters for hypercentroid, the same as above.depc: Assign a uniform depth to all the events otherwise depths information in.mnffile will be used instead.
Relocating events using MLOC isn't of a single running. Instead, you need probably several or even dozens of running to obtain a much more reliable results. That is the reason we will use a command file to simplify the input of those same commands through each running. Now, copy paste those command in the beginning of test/test1.1.cfil under your working directory. Also copy paste utilities used for removing outliers. Again, recompile those utilities from source code first:
cat ../cfil\ header.txt | cat - ./test1.1.cfil > /tmp/out && mv /tmp/out ./test1.1.cfil
cp ../Utilities/* .
Now sitting at your working directory and run mloc program:
➜ mloc_working git:(master) ✗ ./mloc_g
mloc v10.4.5, release date 9/28/2018
Current program limits:
nevmax = 200
nqmax = 4000
ntmax1 = 35000
Enter a basename for this run: First of all, let's name our first attempt as test1.1
Enter a basename for this run: test1.1Then specify the data folder relative to your working directory, which is test:
Enter the name of the data directory: test
The commands are:
anno
bdps bias bloc bptc
cal ccat cfil clim comm corr cptf ctyp cvff cvou cvtt
datf dbug dcal dem1 dem2 dep diff
ellp epap eplt even
fdhp flag fmap fplt frec freh
help hlim
inpu
kill
lat lgtt lmod long lonr lres
mare mdou mech memb
nsmd
oldr
pert phid phyp plot pltt ppri pttt puke
radf rdpp rels revi rfil rhdf run
secv shcl skip splt sstn star stat step stop subc
taup tfil tikh time tomo tptt tt5e tt5s ttou
vect vlog vscr
weig wind
xsec
For more information, follow the "help" command with a command name
Enter commands:Next, we will fill in the commands we previously grouped that is for reading and processing data using cfil:
: cfil test1.1.cfil
End of command file reached, continue with interactive commands:Then, we add a few more commands, for instances, turn off the reading weights and define outliers that exceeds 3 sigma and finally enter run:
: weig off
: lres 3.0
: runYou will now see some information pops out in your terminal:
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
FYI from mloc_main: 30 events will be relocated
Station coordinates...
FYI from redsta: total number of stations 21118
Travel time tables...
Travel times spreads...
Reading errors...
Stations suspected of reporting bogus depth phases...
Phase data for each event...
FYI from mlocset: event 1 19670126.1611.42 test/19670126.1611.42.mnf
FYI from mlocset: event 2 19720519.0113.39 test/19720519.0113.39.mnf
FYI from mlocset: event 3 19740929.2334.56 test/19740929.2334.56.mnf
FYI from mlocset: event 4 19770904.0818.24 test/19770904.0818.24.mnf
FYI from mlocset: event 5 19780208.1614.38 test/19780208.1614.38.mnf
......
Within these contains information about how the results converged, possible one-minute error, possible outliers, etc. We will not discussed them in detail. You probably notice that a bunch of pdf shown in your data folder, for example, the base plot:
It is also worth mentioning the International Registry of Seismograph Stations (IR). Currently the update-to-date list is stored in MLOC/mloc_working/tables/stn/master_stn.dat which looks like:
# Deployment codes
# INSN: Iran National Seismograph Network
# ITSN: Iran Telemetered Seismograph Network
# Kharkheh: Kharkheh Dam Network, Iran
# TA: Transportable Array <http://ds.iris.edu/ds/nodes/dmc/earthscope/usarray/#transportable_array>
#
034A 27.0647 -98.6833 155 IR IRIS .TA . Hebbronville, Texas, U.S.A.
035A 26.9379 -98.1023 29 IR IRIS .TA . Encino, Texas, U.S.A.
035Z 26.46300 -98.06831 19 IR IRIS .TA . Hargill, Texas, U.S.A.
To supplement your own stations, put them in
Use the help command within mloc program to view more or refer manuals in the MLOC/Doc