Next: 34. Magnitude and spectral relations, Up: 33. Seismic risk related programs Previous: 33.3 SEISAN EXPLORER: Contents Index
The purpose of CLUSTER is to find and remove dependent events (foreshocks, aftershocks and swarm events) from a catalog. In contrast to the old CLUSER in SEISAN before version 10.0, this completely new version will search for dependent events using more parameters: Time difference, epicentral distance and depth distance. These parameters can be magnitude dependent and can be different before and after the main event. The program requires an input parameter file cluster.def which can be placed in working directory or DAT. An example is given in DAT and seen below.
For declaring an event as a dependent event, the dependent event must be occur at less than a given time difference from the main event, it must be smaller than a specified magnitude (relative to the main event), the epicentral or hypocentral distance between main and dependent event must be less than a given value and the difference in hypocentral depth between the two events must be less than a given value. Hypocentral distance is calculated as the direct distance between main shock and dependent event.. The time and distance limits are generally magnitude dependent so larger values are used for larger events. Different magnitude dependence can be used before and after the main event.
If the event has no location or magnitude, it will not be considered as a possible dependent event and will be included in the list of main shocks.
The magnitude used is selected among all prime magnitudes in the file (the 6 magnitudes on main header line, can be 2 lines if more than 3 magnitudes). The magnitudes can be given an order of priority in SEISAN.DEF:
KEYWORD............Comments.............Par 1.....Par 2 MAGNITUDE_ORDER LBER MAGNITUDE_ORDER LNAO
In this example, LBER (local magnitude calculated by BER) is first chosen, if not there, LNAO and if not there either, the first magnitude found in file. If no order is given in SEISAN.DEF, the magnitude used will be the first found for the event, irrespective of type or agency. It is possible to leave either magnitude or agency blank in which case the blank represents a wildcard.
Potentially all events can be a mainshock so all combinations, within the time limit, will be searched. A search is first made after the main event (for aftershocks) and then before the main event (for foreshocks).
Input:
A file in Nordic format. If only header lines, there must be a blank line between events.
Output:
All dependent events will be removed and put out in a file called cluster_reject.out while the cleaned catalog will be in file cluster_use.out. Optionally, a debug output can be made, either on screen or in a file (see cluster.def). An example is seen below
Main : 2012 210 1222 22.3 66.4 14.2 14.0 0.9 LTES Main : 2012 210 1316 8.1 66.3 14.9 0.0 1.1 LTES Main : 2012 210 1356 25.0 71.2 -8.1 13.0 3.5 LTES After: 2012 212 2332 7.8 71.2 -8.2 12.9 1.8 LTES M2.5 T 22 2 D 26 3 After: 2012 213 853 34.5 71.0 -8.0 19.7 1.3 LTES M2.5 T 22 2 D 26 19 After: 2012 216 1322 6.1 71.1 -7.7 0.1 1.5 LTES M2.5 T 22 5 D 26 18 Fore : 2012 124 027 10.9 71.1 -8.0 43.2 2.3 LTES M2.5 T 22 17 D 26 14 Fore : 2012 122 1248 16.6 71.1 -8.3 23.7 1.7 LTES M2.5 T 22 19 D 26 12 Fore : 2012 120 128 10.3 71.1 -7.5 13.2 1.8 LTES M2.5 T 22 21 D 26 22 Main : 2012 210 1543 54.2 73.3 7.8 15.0 2.8 LTES Main : 2012 210 1549 52.9 -27.6-176.6 33.0 5.0 BNAO Main : 2012 210 1654 31.8 2.2 122.2 33.0 5.4 BNAO
Each potential main (Main) event is listed. To be considered a potential main event, the event must have both location and magnitude. If a dependent event is found, its type is shown (After or Fore) with the parameters: M: The interpolated main shock magnitude used, T: The interpolated maximum time difference used and the actual time difference (days), D: The interpolated maximum distance (km) used and the actual distance. In the above example, the upper magnitude limit for dependent events is 2.5, main event is 3.5, the limit in distance is 26 km and in time 22 days. Directly following the Main event comes a listing of all dependent events. The next potential main event will then be the first unused event following the previous Main event and that Main event can then jump back in time since there might be several events following the previous Main event which were not declared as dependent events.
Example of a cluster.def file
This file is parameter file for cluster. The file can be in working
directory or DAT.
Only the lines with recognized keywords and a non blank field
under Par 1 will be read. The comments have no importance. Lines with
keywords MAGS AFTER DIST TIME and MAGS BEFORE DIST TIME must come grouped
together in increasing magnitude order. Parameters for these lines are:
Reference magnitude of main event, maximum magnitude of corresponding after/foreshock
(AFTER/FORE), distance(km) and time(days).
All number must be real numbers. Columns Par 1-Par 4 start in
columns 41,51, 61 and 71.
All keywords in capital letters.
Hypocentral distance is calculated as the direct distance between main shock and
dependent event.
Parameter values are interpolated for magnitudes between the ones given.
The values shown in debug output are the interpolated values.
If event is larger than largest event in table, the values for the largest event is
used. If event is smaller than smallest event in
table, no dependent events will be searched for.
The MAGS AFTER DIST TIME are magnitude of main shock, corresponding magnitude
of largest aftershock, distance and time. Similarly for the foreshocks.
The MAX DEPT DIFFERENCE will consider only events with a depth
difference less than MAX DEPT DIFFERENCE as a possible dependent event.
KEYWORD............Comments.............Par 1.....Par 2.....Par 3.....Par4.....
HYPOCENTRAL DIST 1.0 yes, 0.0 no 0.0
MAX DEPTH DIFF Difference in depth 50.0
DEBUG OUT 0.0 none, 1.0 screen 2.0 file 1.0
MAGS AFTER DIST TIME Mag. dep. par. 2.5 1.5 19.5 6.0
MAGS AFTER DIST TIME Mag. dep. par. 3.0 2.0 22.5 11.5
MAGS AFTER DIST TIME Mag. dep. par. 3.5 2.5 26.0 22.0
MAGS AFTER DIST TIME Mag. dep. par. 4.0 3.0 30.0 42.0
MAGS AFTER DIST TIME Mag. dep. par. 4.5 3.5 35.0 83.0
MAGS AFTER DIST TIME Mag. dep. par. 5.0 4.0 40.0 155.0
MAGS AFTER DIST TIME Mag. dep. par. 5.5 4.5 47.0 290.0
MAGS AFTER DIST TIME Mag. dep. par. 6.0 5.0 54.0 510.0
MAGS AFTER DIST TIME Mag. dep. par. 6.5 5.5 61.0 790.0
MAGS AFTER DIST TIME Mag. dep. par. 7.0 6.0 70.0 915.0
MAGS AFTER DIST TIME Mag. dep. par. 7.5 6.5 81.0 960.0
MAGS AFTER DIST TIME Mag. dep. par. 8.0 7.0 94.0 985.0
MAGS AFTER DIST TIME Mag. dep. par. 8.5 7.5 110.0 1000.0
MAGS BEFORE DIST TIME Mag. dep. par. 2.5 1.5 19.5 6.0
MAGS BEFORE DIST TIME Mag. dep. par. 3.0 2.0 22.5 11.5
MAGS BEFORE DIST TIME Mag. dep. par. 3.5 2.5 26.0 22.0
MAGS BEFORE DIST TIME Mag. dep. par. 4.0 3.0 30.0 42.0
MAGS BEFORE DIST TIME Mag. dep. par. 4.5 3.5 35.0 83.0
MAGS BEFORE DIST TIME Mag. dep. par. 5.0 4.0 40.0 155.0
MAGS BEFORE DIST TIME Mag. dep. par. 5.5 4.5 47.0 290.0
MAGS BEFORE DIST TIME Mag. dep. par. 6.0 5.0 54.0 510.0
MAGS BEFORE DIST TIME Mag. dep. par. 6.5 5.5 61.0 790.0
MAGS BEFORE DIST TIME Mag. dep. par. 7.0 6.0 70.0 915.0
MAGS BEFORE DIST TIME Mag. dep. par. 7.5 6.5 81.0 960.0
MAGS BEFORE DIST TIME Mag. dep. par. 8.0 7.0 94.0 985.0
MAGS BEFORE DIST TIME Mag. dep. par. 8.5 7.5 110.0 1000.0
There must be at least one "MAGS AFTER DIST TIME" line but
"MAGS BEFORE DIST TIME" are not needed.
In that case there will be no search for foreshocks.