3:30 PM - 3:45 PM
[HDS07-19] S-wave velocity profiles obtained by active and passive surface wave methods at the Portuguese Bend Landslide
Keywords:Landslide, Surface wave method, Microtremors array measurements, S-wave velocity, Horizontal to vertical spectral ratio, California
We carried out active and passive surface wave measurements and estimated 1D S-wave velocity (Vs) profiles at the Portuguese Bend Landslide, at a coastal region of the Los Angeles metropolitan area, California, U.S. The landslide is placed at the dip slope of Miocene Monterey Formation mainly consisting of clay and tuff layers, uplifted by two parallel faults, the Palos Verdes Fault on the northeast and the San Pedro Fault offshore on the southwest. The landslide has a long history of movements. The area was identified as a potentially hazardous landslide area as early as the 1920s. The movement accelerated with real estate construction and development at the region in the 1940s and 1950s. Restriction of development and landslide prevention such as removing ground water from the landslide mass stabilized the movement in 1980s. In 2024, the landslide became active again and approximately 160,000 square meters of area moved toward to the Pacific Ocean. Displacement rate reached up to 30 cm in a week from spring to summer 2024 and many homes and roads were severely damaged.
Active (multichannel analysis of surface waves : MASW) and passive (microtremor array measurements : MAM) surface waves were measured at three sites, two sites (Cherry Hill Lane and South Bay Archery) on sliding body and one site (Abalone Cove Park) outside of the landslide. Active method (MASW) used a linear array of 90 m consisting of 12 geophones with the minimum and maximum receiver spacing of 1.5 m and 91.4 m respectively. A 10 kg sledge hammer generated P-SV wave motion. Passive method (MAM) used a 90 m linear array or a 45 m L-shaped array consisting of 12 geophones. We recorded approximately 30 min. of seismic ambient noise. Vertical component geophones with natural frequency of 2 Hz were used for both active and passive data acquisition. A phase shift and stack method and a spatial autocorrelation method calculates dispersion curves from active and passive surface wave data respectively. The active and passive dispersion curves were combined and one-dimensional (1D) Vs profiles were estimated by non-linear inversion. Clear dispersion curves were obtained between 2 Hz and 50 Hz, and Vs profiles down to 100 m deep were estimated at all three sites. We carried out single station three-component microtremor measurements at 10 sites inside and outside of the landslide including three sites where MASW and MAM were measured.
Figure 1 shows investigation site and horizontal to vertical spectral ratio (H/V) calculated from the single station three-component microtremor measurements. There are clear H/V peak between 1 and 10 Hz inside of the landslide whereas there is no clear peak outside of the landslide. Figure 2 compares dispersion curves at the three sites. Phase velocities obtained inside of the landslide (Cherry Hill Lane and South Bay Archery) were clearly lower than those outside of the landslide (Abalone Cove Park).
Figure 3 compares Vs profiles obtained by the inversion. There was clear difference between inside and outside of the landslide. At the outside of the landslide (Abalone Cove Park), Vs rapidly increased and reached 700 m/s, that can be considered as flesh Miocene bedrock, at approximately 20 m deep. In contrast, there were thick low velocity layers with Vs of 200~650 m/s inside of landslide (Cherry Hill Lane and South Bay Archery). Depth to the fresh Miocene bedrock appeared to be approximately 80 m deep at those sites. The thick low velocity layers inside the landslides implies that the landslide body was shattered by landslide movement and Vs came down significantly. The results demonstrated that the active and passive surface wave methods can be an effective tool to figure out landslide structure.
Active (multichannel analysis of surface waves : MASW) and passive (microtremor array measurements : MAM) surface waves were measured at three sites, two sites (Cherry Hill Lane and South Bay Archery) on sliding body and one site (Abalone Cove Park) outside of the landslide. Active method (MASW) used a linear array of 90 m consisting of 12 geophones with the minimum and maximum receiver spacing of 1.5 m and 91.4 m respectively. A 10 kg sledge hammer generated P-SV wave motion. Passive method (MAM) used a 90 m linear array or a 45 m L-shaped array consisting of 12 geophones. We recorded approximately 30 min. of seismic ambient noise. Vertical component geophones with natural frequency of 2 Hz were used for both active and passive data acquisition. A phase shift and stack method and a spatial autocorrelation method calculates dispersion curves from active and passive surface wave data respectively. The active and passive dispersion curves were combined and one-dimensional (1D) Vs profiles were estimated by non-linear inversion. Clear dispersion curves were obtained between 2 Hz and 50 Hz, and Vs profiles down to 100 m deep were estimated at all three sites. We carried out single station three-component microtremor measurements at 10 sites inside and outside of the landslide including three sites where MASW and MAM were measured.
Figure 1 shows investigation site and horizontal to vertical spectral ratio (H/V) calculated from the single station three-component microtremor measurements. There are clear H/V peak between 1 and 10 Hz inside of the landslide whereas there is no clear peak outside of the landslide. Figure 2 compares dispersion curves at the three sites. Phase velocities obtained inside of the landslide (Cherry Hill Lane and South Bay Archery) were clearly lower than those outside of the landslide (Abalone Cove Park).
Figure 3 compares Vs profiles obtained by the inversion. There was clear difference between inside and outside of the landslide. At the outside of the landslide (Abalone Cove Park), Vs rapidly increased and reached 700 m/s, that can be considered as flesh Miocene bedrock, at approximately 20 m deep. In contrast, there were thick low velocity layers with Vs of 200~650 m/s inside of landslide (Cherry Hill Lane and South Bay Archery). Depth to the fresh Miocene bedrock appeared to be approximately 80 m deep at those sites. The thick low velocity layers inside the landslides implies that the landslide body was shattered by landslide movement and Vs came down significantly. The results demonstrated that the active and passive surface wave methods can be an effective tool to figure out landslide structure.