S-wave Velocity Based On Microtremor Array
Download as PPTX, PDF1 like709 views
1) The study performed three-component microtremor measurements and microtremor array measurements at sites across Mexico City to delineate the S-wave velocity structure down to 200m depth. 2) Analysis of the H/V spectra and dispersion curves using a joint inversion technique produced S-wave velocity models for six sites showing significantly lower velocities in Mexico City compared to sites in the US, with velocities less than 150m/s down to 70m depth. 3) The peak frequencies of the H/V spectra across Mexico City from 0.25-1Hz correspond to a low velocity layer shallower than 100m depth, which helps explain the amplification and damage from earthquakes.
S-wave Velocity Based On Microtremor Array
- 1. S-wave Velocity Structure of Mexico City Obtained from Three-component Microtremor Measurements and Microtremor Array Measurements 1 Koichi Hayashi*, Geometrics Atsushi Nozu, Port and Airport Research Institute, Masanori Tanaka, Port and Airport Research Institute Haruhiko Suzuki, OYO Corporation Efraín Ovando Shelley, Universidad Nacional Autonoma de Mexico
- 2. Outline 2 • Introduction • Investigation site • Data acquisition Equipment H/V spectrum Dispersion curve • Analysis results • Comparison with United States • Conclusions
- 3. Introduction 3 • The earthquake that struck Mexico on 19 September 1985 caused severe damage in Mexico City although the city is located 400km away from the epicenter. • The main reason for this damage is that the city is located on a basin filled with very soft sediments. • Distribution of these soft sediments has been delineated by drillings and microtremor measurements. • A small number of attempts have been made to image the S-wave velocity structure of the basin using downhole seismic loggings. • In order to delineate S-wave velocity structure of the basin down to depth of approximately 200m, we have performed three-component micro-tremor measurements and microtremor array measurements.
- 4. 1985 Mexico Earthquake Mexico City 400km Mw=8.3 4
- 5. 1985 Mexico Earthquake 5 • Mw=8.3 • Most-often cited number of deaths is an estimated 10,000 people but experts agreed that it could be up to 40,000. • Damage area corresponds to the western part of the lake zone within 2 to 4 kilometers of the Alameda Central. • 6 to 15 story buildings are mainly damaged in the city due to a frequency range of 0.25 to 0.5Hz (period of 2 to 4seconds)
- 6. > 2sec Natural Period (H/V) of Mexico Basin Lermo and Chavez- Garcia (1994) Central Mexico City Airport Damage Area > 4sec 6
- 7. Investigation Site 7 • Investigation site is placed at the downtown of Mexico City. • 30km length survey line crosses the basin with a west-southwest to east-northeast direction. • 3 component microtremor measurements were performed at more than 10 sites on the line. • Microtremor array measurements were performed at 6 sites on the line. • Microtremor array measurements used 25 to 650m equilateral triangular arrays.
- 8. Chapultepec Aragon Texcoco No.7 Almeda Texcoco No.8 Texcoco TXC Investigation Site Central Mexico City Airport 8
- 9. Example of Array Configuration 19.422938 -99.182564 19.423838 99.184163 19.423818 99.180783 19.42138 99.182296 330m 19.462283 99.067562 19.463962 99.066875 19.4629 99.063903 19.464731 99.069911 19.461818 99.06931 19.460988 99.066317 19.459238 99.068806 650m Chapultepec 9 Aragon
- 10. Data Acquisition 10 • Data acquisition was carried out during the daytime in December 2008 and December 2009. • Microtremor measurement systems (JU210) made by Hakusan Corporation and data loggers (GPL-6A3P) made by Mitsutoyo Corporation were mainly used for data acquisition. • Both systems use accelerometers for the sensors. • In order to verify applicability of the accelerometers, servo- type velocity meters made by Katsujima Corporation (SD- 110) and Tokyo Sokushin Corporation (VSE11F, VSE12F) were also used in the 3 component microtremor measurements. • H/V spectra obtained through the accelerometers and the velocity meters were compared. • 30 min. to 1 hour of microtremors were recorded for each three component measurement or array measurement.
- 12. Comparison of H/V It is clear that peak frequency of H/V decrease from west to east (from edge to center of basin) 12
- 13. 600 500 ,-... (/') .'..E....., 4oo >-.+-' (.) 0 300(J) >I (J) 200 a.. 100 0 0.1 • Texcoco Main No.7(2008) • Texcoco North No.8(2008) • Aragon(2009) • Almeda(2009) <>Chapultepec(2009) c 1 Frequency(Hz) 10 100 Sites where the peak frequency of H/V spectra is higher, the phase velocity of the dispersion curve is also higher 13 , • 9 <> : - ·l.llt> .o.c.
- 14. Analysis (1) 14 • A joint inversion was applied to the observed H/V spectra and dispersion curves, and S-wave velocity models were analyzed for six sites. • In the inversion, phase velocities of the dispersion curves and peak frequencies of the H/V spectra were used as the observation data. • Unknown parameters were layer thickness and S-wave velocity. • A Genetic algorithm was used for optimization.
- 15. Analysis (2) 15 • Initial models were created by a simple wavelength transformation in which wavelength calculated from phase velocity and frequency is divided by three and plotted at depth. • Theoretical H/V spectra and phase velocities are generated by calculating the weighted average of the fundamental mode and higher modes (up to the 4th modes) based on medium response. • Rayleigh-Love ratio (R/L) is fixed as 0.7
- 16. S-wave Velocity Model（East） Texcoco No.7 Aragon 0 50 100 150 200 250 300 350 400 450 500 0 0.2 0.4 0.6 Vs(km/ s) Depth(m )
- 18. Chapultepec Aragon Texcoco No.7 Almeda Texcoco No.8 Texcoco TXC Peak Frequency of H/V and S-wave Velocity Model 0.7s 90m/s 1.5s 80m/s 3.9s 60m/s 3.3s 30m/s 1.1s 60m/s ＞400m/s450m <100m/s ≒200m/s 18
- 19. Comparison with United States H/V Spectra Dispersion Curve S-wave Velocity Model S-wave velocity of Mexico is extremely low compare with San Jose and Redwood City. Mexico City San Jose (William St. Park), CA Redwood City, CA 19 0 50 100 150 200 250 300 350 0.1 1 10 100 Phase-velocity(m/sec) Frequency(Hz) Mexico San Jose Redwood City 0.1 1 10 0.1 1 10 H/Vspectra Period(sec) Mexico San Jose Redwood City 0 20 40 60 80 100 120 140 160 0 S-wave velocity (m/sec) 100 200 300 400 500 Depth(m) Mexico San Jose Redwood City
- 20. Conclusions • We have performed the three-component microtremor measurements and microtremor array measurements in the Mexico basin and estimated the S-wave velocity models down to a depth of 200m. • S-wave velocity in the middle of the Mexico basin is lower than 150m/s to a depth of 70m and much lower than typical alluvial plains in Japan and United States. • Peak frequencies of the H/V spectra in Mexico City vary from 0.25 to 1Hz and it seems that these peak frequencies are mainly due to the low-velocity layer shallower than a depth of 100m. 20