Evidence for faulting and fluid-driven earthquake processes from seismic attenuation variations beneath metropolitan Los Angeles

Seismicity in the Los Angeles metropolitan area has been primarily attributed to the regional stress loading. Below the urban areas, earthquake sequences have occurred over time showing migration off the faults and providing evidence that secondary processes may be involved in their evolution. Combining high-frequency seismic attenuation with other geophysical observations is a powerful tool for understanding which Earth properties distinguish regions with ongoing seismicity. We develop the first high-resolution 3D seismic attenuation models across the region east of downtown Los Angeles using 5,600 three-component seismograms from local earthquakes recorded by a dense seismic array. We present frequency-dependent peak delay and coda-attenuation tomography as proxies for seismic scattering and absorption, respectively. The scattering models show high sensitivity to the seismicity along some of the major faults, such as the Cucamonga fault and the San Jacinto fault zone, while a channel of low scattering in the basement extends from near the San Andreas fault westward. In the vicinity of the Fontana seismic sequence, high absorption, low scattering, and seismicity migration across a fault network suggest fluid-driven processes. Our attenuation and fault network imaging characterize near-fault zones and rock-fluid properties beneath the study area for future improvements in seismic hazard evaluation.


Figure 1 .
Figure 1.Map showing the locations of the seismic nodes (gray triangles) that were installed along 10 dense linear arrays labelled in the map during the BASIN experiment.The blue rectangle indicates the study area.Eight BASIN arrays (SG3, SG4, part of SB1, SB2, SB3, SB4, SB5, and SB6) were used in this study.The light blue triangles indicate the 15 broadband stations from the Southern California Seismic Network that were included in our dataset.The red line indicates the outline of the sedimentary basins.Faults are from the Southern California Earthquake Center Community Fault Model (CFM) 6.0 1 .

Figure 2 .
Figure 2. Histogram showing the number of three-component seismograms from each linear nodal array of the BASIN network and the regional network (REG).After the final selection during the peak delay processing, the BASIN network provides 3,315 three-component seismograms out of the total of 5,250.

Figure 3 .Figure 4 .
Figure 3. C-C' cross-section of the scattering model at 18 Hz.The location of the cross-section is shown in Fig. 3a.

Figure 5 .
Figure 5. Checkerboard tests (input in the left panels and output in the right panels) for Q −1 c anomalies at (a) 1.5 km and (b) 4.5 km depth at 18 Hz.

Figure 6 .
Figure 6.Spike test (input in the left panel and output in the right panel) for a high Q −1 c anomaly at 4.5 km depth at 18 Hz shown in Figure 4e.

Figure 7 .Figure 8 .
Figure 7. Hit count showing the well-resolved areas of the peak delay maps.Blocks crossed by less than 2 rays are shown in white.The horizontal and vertical slices are taken at 3.5 and 7 km depth, and -117.5°longitude,respectively.The black dashed line shows the outline of the basins.The black rectangles show the extent of the model in Fig. 3.The N-S profile corresponds to profile B-B' in Figure 3a.

Figure 9 .Figure 10 .
Figure 9. Upper panel: 10 km depth slice of the scattering model at 18 Hz.Bottom panel: hit count map at 10 km depth.Seismicity deeper than 10 km is shown in Fig. S13.The black line shows the basins' outline.

Figure 11 .
Figure 11.Distance-time evolution of earthquakes that occurred in June 2019 across the Fontana area (black rectangle Fig. 6a), from the SCSN catalog.Distance is relative to the grey square in Fig. 6a.The magenta line is the best fit line.Earthquakes are color-coded based on focal depth.

Figure 12 .Figure 13 .
Figure 12. 3D view (a) from the northwest of the modelled fault networks corresponding to the 2019 Fontana seismic sequence.We used the seismicity presented in Fig. 7.The circles indicate the preferred fault orientations and are colour-coded based on the cluster classification resulting from the analysis.3D view (b) from the northwest of the earthquakes, which are shown with different symbols based on the cluster classification resulting from the 3D fault plane imaging.The symbols are also colour-coded based on the time.

Figure 14 .
Figure 14.Example of low and high peak delay values for two vertical component velocity seismograms (a and d) recorded at two different seismic stations and characterized by approximately the same S-wave arrival time.(b) and (e): smoothing root-mean-square envelopes in the 12-24 Hz frequency band.The vertical blue and red dashed lines indicate the S-wave onset in the case of low and high peak delay values, respectively.Panel (c): Peak Delay values as a function of the S-wave arrival time.The red line is the result of the linear regression described in the Data and Methods section and the dashed red lines indicate the interval of 2 standard deviations used for selecting the peak delay values.The blue and red stars indicate the position of the low and high peak delay values with respect to the best fit line, respectively.The x-and y-axes labels are shown in a non-logarithmic format to aid the visualization.

Figure 15 .
Figure 15.Three-dimensional views of ray path coverage.(a) 3D view from the south-west.(b)-(c) views from the west and south, respectively.

Figure 16 .
Figure 16.Three component seismograms filtered in the 12-18 Hz frequency band (blue lines) and the energy envelopes (log(Et α )/(2π f 0 ), red lines).Best fit line (magenta lines) in the coda time window of length 4 s.

Figure 17 .
Figure 17.(a) Q −1 c as a function of the raylength at 18 Hz.The red dashed lines correspond to the mean values Q −1 c ± 2σ (σ is the standard deviation).(b) Moving average (blue line) and 2 x standard deviation (dashed blue lines) computed within a moving window of 500 Q −1 c values.