[S16P] AM-P

[S16P-08] Estimation of V_S30 using microtremor array records in Thimphu and surrounding towns, the Kingdom of Bhutan

*Takumi Hayashida¹, Nityam Nepal², Phuntsho Pelgay², Shiro Ohmi³, Karma Namgay², Dowchu Drukpa², Hiroshi Inoue⁴ (1. IISEE, Building Research Institute, 2. DGM, Ministry of Economic Affairs, Bhutan, 3. DPRI, Kyoto Univ., 4. NIED)

Bhutan is a landlocked country located in the eastern Himalayas, one of the most seismically active regions in the world. Thimphu and surrounding towns (Isuna, Paro, and Ura) are located in the western central part of Bhutan at altitudes of >2000 m, and settlements are distributed along river valleys. The global V_S30 map calculated from the ground slope (Heath et al., 2020) indicates decreasing V_S30 values along the valleys. Recent seismic intensity observations also show large ground shakings in the valley areas compared to mountainous sites. To evaluate the effect of subsurface soil structure on the ground motion, we conducted microtremor measurements in the areas. We used four portable broadband seismometers (CMG-40T: Guralp Systems Ltd.) with 24-bit recorders (Centaur: Nanometrics Inc.) for the measurements. Microtremor arrays with radii of 5–30 m were deployed in the valleys and uphill areas, and the measurements were taken for >15 minutes for each array. We applied the spatial autocorrelation (SPAC) method (Aki, 1957; Okada, 2003) for velocity structure estimations and derived microtremor horizontal-to-vertical (H/V) spectral ratios to identify dominant frequencies at each site. We found the microtremor measurement below 10 Hz was difficult due to stiff subsurface soil as well as weak microtremor power in the target areas, but the SPAC results indicate dispersive characteristics of Rayleigh wave phase velocity above 10 Hz, and the estimated structure models show S-wave velocity profiles down to several tens of meters. We estimated V_S30 values using the estimated S-wave velocity models or by applying the directly V_S30 estimation method with SPAC coefficients (Hayashida et al., 2022; in review). Our results showed no apparent difference in deeper soil structures between the valley and uphill areas but a slight difference in the shallower part (several meters below the ground surface) due to fluvial sediments. The estimated V_S30 values range from 300–900 m/s, which are comparative with those calculated using topographic features, and our results contribute to the development of more detailed hazard maps.

Acknowledgments:
This study was supported by Japan Science and Technology Agency (JST) and Japan International Cooperation Agency (JICA) under the "Science and Technology Research Partnership for Sustainable Development (SATREPS): Project for Evaluation and Mitigation of Seismic Risk for Composite Masonry Buildings in Bhutan (FY2017-2022)".