Academia.eduAcademia.edu

Figure 10 – uploaded by Susan Beck

See full PDFdownloadDownload figure
Fig. 9. Plot of the error for the multistation iterative inversior versus assumed depth. The very lowest error corresponds to a deptl of 30 km with acceptable errors between 10 and 45 km.

Figure 9 Plot of the error for the multistation iterative inversior versus assumed depth. The very lowest error corresponds to a deptl of 30 km with acceptable errors between 10 and 45 km.

Related Figures (18)

Fig. 1. Map of the western Aleutian arc showing the epicenter (large solid diamond) of the February 4, 1965, Rat Islands main- shock. The small stars are the aftershocks with M > 5 relocated by Spence [1977], that occurred between February 4 and March 30, 1965. The focal mechanism for the mainshock is from the surface wave study of Wu and Kanamori [1973].
Fig. 1. Map of the western Aleutian arc showing the epicenter (large solid diamond) of the February 4, 1965, Rat Islands main- shock. The small stars are the aftershocks with M > 5 relocated by Spence [1977], that occurred between February 4 and March 30, 1965. The focal mechanism for the mainshock is from the surface wave study of Wu and Kanamori [1973].
Fig. 2a. Deconvolved single-station source time functions for the February 4, 1965, Rat Islands earthquake. The solid trace is the observed seismogram, and the dashed trace is the synthetic seismogram for the source time function shown for each station. The station azimuth (AZ), epicentral distance (DIST), horizontal component (N, E) if used, and seismic moment (Mo) are shown for each station. The source time functions show three main pulses of moment release (A, B, C). Within each pulse of moment release the consistent features that can be observed at many stations are shown by the arrows (Al, B1, C1). The first source time function has additional arrows showing the secondary features (A2, B2, C2) that are less consistent from station to station.
Fig. 2a. Deconvolved single-station source time functions for the February 4, 1965, Rat Islands earthquake. The solid trace is the observed seismogram, and the dashed trace is the synthetic seismogram for the source time function shown for each station. The station azimuth (AZ), epicentral distance (DIST), horizontal component (N, E) if used, and seismic moment (Mo) are shown for each station. The source time functions show three main pulses of moment release (A, B, C). Within each pulse of moment release the consistent features that can be observed at many stations are shown by the arrows (Al, B1, C1). The first source time function has additional arrows showing the secondary features (A2, B2, C2) that are less consistent from station to station.
Fig. 2b. The same as Figure 2a except that the P waves for these stations have a duration of 3 min or less.
Fig. 2b. The same as Figure 2a except that the P waves for these stations have a duration of 3 min or less.
Fig. 3. Best fit (least squares) straight lines for the delay time T, versus directivity parameter (p cos ¢); for each station for the three main features identified on the source time functions (Al, B1, C1). The slope of the line (X) gives the horizonta! distance to the feature and the zero intercept (tf) gives the actual delay time. The best rupture direction is 300° for each feature. The plot in the lower right shows the correlation coefficients of the best fit straight lines as a function of the assumed rupture azimuth for each feature.
Fig. 3. Best fit (least squares) straight lines for the delay time T, versus directivity parameter (p cos ¢); for each station for the three main features identified on the source time functions (Al, B1, C1). The slope of the line (X) gives the horizonta! distance to the feature and the zero intercept (tf) gives the actual delay time. The best rupture direction is 300° for each feature. The plot in the lower right shows the correlation coefficients of the best fit straight lines as a function of the assumed rupture azimuth for each feature.
Fig. 4. The top row is the single station inversion of the P wave for station COP which yields a seismic moment of 42 x 1027 dyn cm. In the second row, the surface wave seismic moment of 140 x 107 dyn cm is constrained to the three pulses of moment release in the source time function. The synthetic seismogram does not fit the observed data indicating the missing moment cannot be confined to the asperities. The bottom row shows the total seismic moment added in as a half sine wave with a duration of 240 sec to correspond to a fault length of 600 km. In this case the synthetic seismogram fits the data very well. Although this result is not unique, it indicates that the additional moment is released in a smooth fashion distrib- uted over a large area.
Fig. 4. The top row is the single station inversion of the P wave for station COP which yields a seismic moment of 42 x 1027 dyn cm. In the second row, the surface wave seismic moment of 140 x 107 dyn cm is constrained to the three pulses of moment release in the source time function. The synthetic seismogram does not fit the observed data indicating the missing moment cannot be confined to the asperities. The bottom row shows the total seismic moment added in as a half sine wave with a duration of 240 sec to correspond to a fault length of 600 km. In this case the synthetic seismogram fits the data very well. Although this result is not unique, it indicates that the additional moment is released in a smooth fashion distrib- uted over a large area.
Fig. 5. Plots showing the tomographic imaging error versus rupture azimuth and rupture velocity. For all rupture velocities the lowest error corresponds to a rupture azimuth of 300°. The rupture velocity is not well resolved.
Fig. 5. Plots showing the tomographic imaging error versus rupture azimuth and rupture velocity. For all rupture velocities the lowest error corresponds to a rupture azimuth of 300°. The rupture velocity is not well resolved.
Fig. 6. Final space-time image of the rupture process of the February 4, 1965, Rat Islands earthquake from the tomographic imaging technique. The data on the right are the previously deter- mined source time functions shown as the solid traces. The dashed traces are the synthetic source time functions for the rupture model shown on the left. The rupture image has spatial extent from 0 (corresponding to the epicenter) to 500 km in the direction of 300° and an a priori rupture velocity of 2.5 km/s. The rupture image shows three pulses of moment release.
Fig. 6. Final space-time image of the rupture process of the February 4, 1965, Rat Islands earthquake from the tomographic imaging technique. The data on the right are the previously deter- mined source time functions shown as the solid traces. The dashed traces are the synthetic source time functions for the rupture model shown on the left. The rupture image has spatial extent from 0 (corresponding to the epicenter) to 500 km in the direction of 300° and an a priori rupture velocity of 2.5 km/s. The rupture image shows three pulses of moment release.
Fig. 8. Moment rate density function from the tomographic imaging constrained to have a total seismic moment of 140 x 1027 dyn cm. The additional moment release is added as a long-period half sine wave. The moment rate density is in units of 107! dyn cm/km.
Fig. 8. Moment rate density function from the tomographic imaging constrained to have a total seismic moment of 140 x 1027 dyn cm. The additional moment release is added as a long-period half sine wave. The moment rate density is in units of 107! dyn cm/km.
Fig. 10. Results of the multistation iterative inversion. The resultant source time function has three pulses of moment release similar to the single station source time functions. The lower diagram is a space-time plot of the moment release. The epicenter is at 0 km and the spatial dimension extends from — 120 km (ESE of the epicenter) to +600 km (WNW of the epicenter). The temporal extent is from 0 to 240 s. The triangles represent the best time and spatial location of the subevents for each iteration. The size of the triangles is proportional to the seismic moment release in each iteration. The largest subevents locate on a line indicating a fairly constant rupture velocity.
Fig. 10. Results of the multistation iterative inversion. The resultant source time function has three pulses of moment release similar to the single station source time functions. The lower diagram is a space-time plot of the moment release. The epicenter is at 0 km and the spatial dimension extends from — 120 km (ESE of the epicenter) to +600 km (WNW of the epicenter). The temporal extent is from 0 to 240 s. The triangles represent the best time and spatial location of the subevents for each iteration. The size of the triangles is proportional to the seismic moment release in each iteration. The largest subevents locate on a line indicating a fairly constant rupture velocity.
FEB. 4, 1965, RAT ISLANDS EARTHQUAKE Fig. 11. The observed (top) seismograms and synthetic (bottom) seismograms for the model shown in Figure 10.
FEB. 4, 1965, RAT ISLANDS EARTHQUAKE Fig. 11. The observed (top) seismograms and synthetic (bottom) seismograms for the model shown in Figure 10.
The focal mechanism determined by Stauder [1968a] is shown in Figure 13 along with eight azimuthally distributed source time function-seismogram pairs. The- source time functions shown on the left were determined using the single-station deconvolution technique described by Ruff and Kanamori [1983] and assuming a point source depth of 15 km. The waveforms and source time functions appear to be very simple at most azimuths, consisting of a simple well-defined pulse of moment release. Stations to the south and west (POO, ADE, and HNR) appear to have more complicated waveforms, although a simple moment pulse still explains the data fairly well. The later arriving energy may be due to ocean reverberations which have been ampli- fied by a dipping ocean floor interface [sée Wiens, 1989]. The total duration of the single moment pulse is 4-8 s with no reliable directivity. -:— we = a «
The focal mechanism determined by Stauder [1968a] is shown in Figure 13 along with eight azimuthally distributed source time function-seismogram pairs. The- source time functions shown on the left were determined using the single-station deconvolution technique described by Ruff and Kanamori [1983] and assuming a point source depth of 15 km. The waveforms and source time functions appear to be very simple at most azimuths, consisting of a simple well-defined pulse of moment release. Stations to the south and west (POO, ADE, and HNR) appear to have more complicated waveforms, although a simple moment pulse still explains the data fairly well. The later arriving energy may be due to ocean reverberations which have been ampli- fied by a dipping ocean floor interface [sée Wiens, 1989]. The total duration of the single moment pulse is 4-8 s with no reliable directivity. -:— we = a «
Fig. 13. Focal mechanism for the large March 30, 1965, normal! fault earthquake. Representative single-station source time functions for azimuthally distributed stations are shown. The observed seismograms are the solid traces, and the synthetic seismograms for the source time function shown are the dashed traces.
Fig. 13. Focal mechanism for the large March 30, 1965, normal! fault earthquake. Representative single-station source time functions for azimuthally distributed stations are shown. The observed seismograms are the solid traces, and the synthetic seismograms for the source time function shown are the dashed traces.
Fig. 14. Deconvolved source time functions for six stations at assumed depths for the March 30, 1965, earthquake. Source time functions for depths down to 30 km are simple, but for depths deeper than 30 km the source functions have periodic ringing, indicating an overestimation of the depth. value at 18 km and a seismic moment of 2.1 x 102? dyn cm (Figure 15). Christensen and Ruff [1985] discuss the use of simplicity parameters applied to source time functions to help constrain depth. One such simplicity parameter, the nism shown in Figure 13. In Figure 15 the normalized errors between the observed and synthetic P waves are plotted for assumed depths between 2 and 60 km. The seismograms are best fit at depths between 8 and 22 km with a minimum error
Fig. 14. Deconvolved source time functions for six stations at assumed depths for the March 30, 1965, earthquake. Source time functions for depths down to 30 km are simple, but for depths deeper than 30 km the source functions have periodic ringing, indicating an overestimation of the depth. value at 18 km and a seismic moment of 2.1 x 102? dyn cm (Figure 15). Christensen and Ruff [1985] discuss the use of simplicity parameters applied to source time functions to help constrain depth. One such simplicity parameter, the nism shown in Figure 13. In Figure 15 the normalized errors between the observed and synthetic P waves are plotted for assumed depths between 2 and 60 km. The seismograms are best fit at depths between 8 and 22 km with a minimum error
Fig. 16. The top plot is the normalized error versus depth for the omni-linear inversion for the March 30, 1965, earthquake. The inversion was preformed for depths between 2 and 60 km. The resultant source time function from the multistation omnilinear inversion for a depth of 24 km is shown. The observed data are the solid traces, and synthetic seismograms for the source time function shown are the dashed traces.
Fig. 16. The top plot is the normalized error versus depth for the omni-linear inversion for the March 30, 1965, earthquake. The inversion was preformed for depths between 2 and 60 km. The resultant source time function from the multistation omnilinear inversion for a depth of 24 km is shown. The observed data are the solid traces, and synthetic seismograms for the source time function shown are the dashed traces.
Fig. 17. Map of the western end of the Aleutian arc showing the tectonic blocks from Geist et al. [1988] and the locations of the asperities or regions of concentrated moment release identified in this study. The tectonic blocks are bounded by faults and extensional basins.
Fig. 17. Map of the western end of the Aleutian arc showing the tectonic blocks from Geist et al. [1988] and the locations of the asperities or regions of concentrated moment release identified in this study. The tectonic blocks are bounded by faults and extensional basins.
Fig. 18. Schematic space-time plot for the western Aleutian Arc. The solid bar represents the aftershock area and the hachured bars represent the dominant asperities for the February 4, 1965, earthquake. The dots represent the locations of a series of smaller earthquakes between 1898 and 1929. The events at the turn of the century tend to cluster at the edges of the 1965 Rat Island mainshock.
Fig. 18. Schematic space-time plot for the western Aleutian Arc. The solid bar represents the aftershock area and the hachured bars represent the dominant asperities for the February 4, 1965, earthquake. The dots represent the locations of a series of smaller earthquakes between 1898 and 1929. The events at the turn of the century tend to cluster at the edges of the 1965 Rat Island mainshock.
Related topics:
Geology
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved