Azimuth of arrival (back azimuth) from 3-component stations, h(Azim)
If a 3 component station is available, the azimuth of arrival can be determined using the method developed by Roberts et al. (1989). Display any of the 3 components and press h (Azim). Then select a zoom window around the P-arrival of a few secs duration for the analysis. The 3 components will now be displayed below in order Z, N and E and the calculated azimuth, apparent velocity and correlation will be displayed at the bottom line. In order to check the stability of the estimate, try different windows and filters. Often, a filter must be used to get reliable results. The displayed azimuth and apparent velocity is only saved in the S-file when an associated phase is picked. THAT PHASE MUST BE PICKED ON THE SINGLE UPPER TRACE SEEN ON THE SAME SCREEN. If there is none, use I or E. The velocity estimate is not very reliable and is dependent on the local velocities. In order to calculate the apparent velocity, the P-velocity of the top layer must be given. The default value is 5.0 km/sec, but another value can be set in the MULPLT.DEF file. To get a good estimate, the correlation coefficient should be as high as possible and positive. The quality of the obtained azimuth can be tested by locating the event with the calculated azimuth weighted out and observe the azimuth residual.
The output is different in new and old Nordic format, see exaple below:
old: STAT SP IPHASW D HRMM SECON CODA AMPLIT PERI AZIMU VELO AIN AR TRES W DIS CAZ7 ASK EZ EP 916 40.94 37.3 9.2 111 29.9 205 ASK EZ E 916 41.96 41.8 7.8 new: STAT COM NTLO IPHASE W HHMM SS.SSS PAR1 PAR2 AGA OPE AIN RES W DIS CAZ7 ASK EHZ EP 916 40.940 111.0-1.0510 29.9 205 ASK EHZ BAZ-P 916 40.942 37.3 9.3 BER jh ASK EHZ BAZ 916 41.962 41.8 7.8 BER jh
Two back azimuths (BAZ) have been picked, one associated with the P and one a bit after the P. In the old format, the one picked with the P is also stored with the P-phase line since there is space on every phase line for BAZ while the one picked later is put on a new phase line and marked with E.
In the new format, Nordic2, the 3 observations are on separate lines and the phase picked with the P is identified as BAZ-P and the BAZ pick with E is marked as BAZ and no phase. The advantage of 3 lines is that each observation can be changed independtenly of the others.
Figure 8.10 shows an example.
Azimuth and apparent velocity from array data, FK analysis f(FK)
Using this command, the traces seen on the screen will be put into the FK program and an FK plot will be displayed. The azimuth and apparent velocity with the highest correlation is selected by the program, however any other value can be manually selected. The values will ONLY enter the S-file if associated with a phase in the same way as amplitudes are picked. For more details, see section 40.
Option U(Rotat) will rotate the horizontal components for the next plot if the two horizontal components are available. The rotation will display the radial component instead of the N-component and the transverse component instead of the E-component. The back-azimuth used is displayed above the trace. All channels will be displayed rotated until u(Rotat) is pressed again This means that phases can be picked and spectra made with the rotated channel. When picking phases on rotated signals, these will appear in the S-file with components R or T instead of N and E respectively. This also means that only if the rotated signals are shown, will the phases read on rotated channels appear on the plot. The station back-azimuth is obtained in the following way: If a hypocenter is given in the header line, the angles are calculated using the current STATIONx.HYP file. If no hypocenter is available, the angle will be read from the S-file under column observed azimuth (47-51) (if not blank) and the azimuth residual will be added. This option permits the user to first determine the azimuth with the 3-component option and then rotate the signals with the determined azimuth. Finally, if no observed azimuth is available, the event to station azimuth + 180 deg. will be used if available (column 77-79). If no back-azimuth can be found, no rotation is done and an angle of 999 deg. is displayed. If in single trace mode and choosing the 3-component option AND the rotate option, the user will be prompted for a rotation angle and the rotated channels will be shown in the usual 3-component plot, however, the azimuth determined is done with the unrotated channels.
PROBLEM: In general, the R-channel will use the response of the N-channel and the T-channel will use the response of the E-channel so for instrument response removal to be correct, the 2 channels must have the same response curve.