OR/17/062 Aftershocks

Baptie, B, Ford, G, and Galloway, D. 2017. The Moidart earthquakes of 4 August 2017. British Geological Survey Internal Report, OR/17/062.

The earthquake was followed by at least four aftershocks, the largest of which had a magnitude of 3.4 ML and which occurred two minutes after the mainshock. The two largest aftershocks were also felt by people across the region. Measured P- and S-wave arrival times were used to determine hypocentres for each of four aftershocks using the HYPOCENTER location algorithm (Lienert et al., 1986) in the same way as for the mainshock. Similarly, local magnitudes were also calculated using measured amplitudes. Source parameters for all five earthquakes are shown in Table 3. There is some scatter in the calculated hypocentres, but in general the epicentres are within a few kilometres. The depths are rather more scattered. Estimated horizontal and vertical errors for the smaller events are significantly larger than for the bigger ones, primarily as a result of fewer measured phases.

Table 3    Source parameters for the mainshock and four aftershocks. Npha is the number of phases used for the location, GAP is the azimuthal gap, RMS is the RMS error in the earthquake location, and ERX, ERY and ERZ are the errors in the X, Y and Z directions in kilometres.

Date	Time (UTC)	Latitude	Longitude	Depth	ML	Npha	GAP	RMS	ERX	ERY	ERZ
04/08/2017	14:43:38.70	56.805	-5.888	12.2	4.0	33	153	0.41	7.0	1.7	4.9
04/08/2017	14:45:34.10	56.798	-5.869	10.6	3.4	14	178	0.28	6.7	1.7	4.7
04/08/2017	15:20:24.00	56.804	-5.871	8.8	1.1	7	181	0.31	14.4	2.4	13.6
04/08/2017	16:07:26.50	56.794	-5.893	8.2	1.2	6	221	0.23	16.3	1.8	8.5
04/08/2017	17:35:06.30	56.807	-5.874	10.1	2.2	14	178	0.35	8.3	1.8	5.9

Hypocentres and the 95% confidence ellipsoids for all five earthquakes are shown in Figure 9. Again, the horizontal error is much larger in the EW direction for all five events as a result of the distribution of stations and the large azimuthal gap.

Figure 9    Projections of 95% confidence ellipsoid for each of the hypocentres in: (a) the horizontal (XY) plane; (b) the YZ plane; and (c) the XZ plane. Each plane shows an area of 20 km by 20 km. (d) Shows the location of (a).

To further compare the events, we examined waveforms from two of the closest stations KPL (61 km) and INVG (121 km). Figure 10 and Figure 11 show the ground motions recorded at KPL and INVG, respectively. Each trace shows one second before and three seconds after the P-wave arrival. All five traces on both stations are very similar, which suggests that all of the earthquakes are closely located and have similar fault mechanisms. For example, the first motion of the initial P-wave arrival at KPL is down on all traces and the first motion of the initial P-wave arrival at INVG is up. There are also similarities in the P-wave codas of each event, although some differences are apparent. For example, the recording of the magnitude 1.2 ML event at 16:07 UTC at INVG has a high amplitude arrival approximately two seconds after the initial P-wave arrival that is not apparent on the other events. These differences may suggest some small variations in the event hypocentre and fault mechanism. In addition, the largest event shows a noticeably lower frequency content as might be expected for a higher magnitude.

Figure 10    Ground motions (nm/s) for the five earthquakes recorded at KPL at a distance of 61 km from the epicentre. Each trace shows one second before and three seconds after the P-wave arrival.

Figure 11    Ground motions (nm/s) for the five earthquakes recorded at INVG at a distance of 121 km from the epicentre. Each trace shows one second before and three seconds after the P-wave arrival.

Given the clustering of the hypocentres and the similarity of the observed waveforms, we used the double difference method (Waldhauser and Ellsworth, 2000) to determine relative locations for the five events in the sequence. The double difference method was used with both the measured phase-picks and precise relative travel-time differences obtained by cross-correlating the recorded waveforms at each station (e.g. Schaff and Richards, 2004). The double-difference residuals for pairs of earthquakes at each station were minimised by weighted least squares using singular value decomposition.

Figure 12 shows a comparison of the double difference locations with single event locations. The hypocentres are plotted in relation to the cluster centroid. The double differences locations are significantly more clustered than the single event locations, with some evidence for alignment of the hypocentres along an east-west trend at a depth of around 11 km. The relocated hypocentres all lie within a source with dimensions of 300 m, 100 m and 500 m in the X, Y and Z directions respectively.

Figure 12    A comparison of the single event locations (red squares) with the locations from the double difference method (black squares with error bars) in the (a) XY, (b) XZ and (c) YZ planes. Hypocentres are plotted relative to the cluster centroid.