next up previous contents index
Next: 23.2 FPFIT Up: 23.1 FOCMEC Previous: 23.1.5 Program operation   Contents   Index

23.1.6 Making synthetic amplitudes and polarities to test FOCMEC and HASH

A synthetic input can be generated for FOCMEC. The purpose is, for a given fault plane solution with a given set of stations, to test how well 'correct amplitudes and polarities' can be inverted to the known fault plane solution. The procedure is: 1: Select an event with a set of observations and a fault plane solution.

2: Write SYNTET in the F-line in S-file where method is given, using e.g. FOCMEC, this is FOCMEC, replace with SYNTET. 3: Run FOCMEC from EEV. The synthetic polarities and amplitudes will then be generated for the same observations as given in the S-file. These observations are normally written out to the file focmec.dat as polarities and corrected amplitudes. The focmec.dat is then input to the original FOCMEC. The synthetic observations are now made by correcting the focmec.dat file to the theoretical polarities and amplitude ratios corresponding to the SYNTET fault plane solutions. The theoretical values are therefore at the source and not affected by Q and free surface. The original observations in the S-file are not changed. 4: Do the inversion in the normal way.

Since the theoretical values are generated by FOCMEC routines, FOCMEC should exactly return the SYNTET solution. However depending on the number and types of observations, this might not be the case. Before making the solution with FOCMEC, it is possible to edit the focmec.dat file to introduce errors or remove observations in order to test the sensitivity of the solution to errors in input or insufficient input. Vp/Vs is hardwired to 1.74 for calculating synthetics.

The same input is also used with HASH so HASH can also be tested in the same way. It is assumed that HASH calculates amplitudes the same way as FOCMEC but this has not been tested. HASH does not use S on Z-component and the P is assumed to be read on the R-component while SEISAN uses the Z-component. So some differences can be expected.

Computer limitations: Total number of polarities must be less than the dimension of array DATA (parameter max_data) for Nordic data (see file in INC directory).

Figure 23.1 shows an example of a fault plane solution calculated with FOCMEC.

Figure 23.1: Top: An example of a fault plane solution plot. Symbols are explained in the text. Bottom: A fault plane solution also showing the stations with corresponding polarities.

next up previous contents index
Next: 23.2 FPFIT Up: 23.1 FOCMEC Previous: 23.1.5 Program operation   Contents   Index
Peter Voss : Wed Aug 9 08:16:56 UTC 2017