Estimation of Shear Wave Structure and Horizontal to Vertical Spectral Ratio at Different Sites in Kathmandu Valley

The present study was carried out to evaluate resonant frequency of the ground and to characterize subsurface ground based on shear wave velocity structure. For this, five sites were selected such as Pulchowk, Chhauni, Gaushala, Buddhanagar and Bhainsepati. About 20 data were recorded in each site and then shear wave velocity structure and graph of amplification ratio with their spatial distribution has been established with the help of software i.e. Seisimager/Seismodule Controller. The results of both analysis methods were then compared to the amplitude of the Gorkha Earthquake and borehole data. All these data and study indicates that the Kathmandu Valley sediments are dependent on the frequency of the seismic waves and the wave velocity is greater in the peripheral region than in the central part of the Valley. The result had also shown that the presence of silty-sand, clay and loose gravel soil with low bearing capacity and elastic modulus in most of the sites are responsible for devastation. It was also noted that apart from few limitations, a non-intrusive microtremor analysis can be adopted for earthquake site characterization in the Kathmandu Valley which can be readily applied and expanded upon in future seismic hazard and microzonation efforts for Kathmandu.


Introduction
The Kathmandu Valley, falls in one of the most active tectonics zones of the Himalayan belt and has experienced many destructive earthquakes in the past  (Pandey et al. 1995). Geological exploration has revealed that the Valley is an ancient lake deposit which is made up of thick layers of clay, silt, sand and gravel in irregular layer of deposition, which measures several hundred meters at the deepest point. Figure 1 represents the location map of the study area where borehole as well as microtremor survey has been carried out. Since the Kathmandu Valley is the parts of active collisional orogenic belt, combining rapid crustal shortening and thickening, that causes frequent strong earthquakes. Also, the level of seismic hazard in the Valley is high (Wesnousky et al. 1999

Objectives
The objectives of the study are to evaluate the resonant frequency of the ground, to characterize subsurface ground based on shear wave velocity structure and to compute the relationship between Shear wave velocity structure, Horizontal-to-Vertical Spectral Ratio (HVSR) and borehole data

Methodology
Microtremor Array Measurement (MAM) survey with Horizontal to Vertical Spectral Ratio (HVSR) was carried out in selected sites. Data were processed and inversion was done with sophisticated software Seisimager. Tabulated value of both compressional wave velocity (Vp) and shear wave velocity (Vs) were used to compute different geotechnical parameters. About 20 data for microtremor and 10 data for HVSR were created. Shear wave velocity profile ( Figure 2) and HVSR model ( Figure 11) has been established. The HVSR model at each borehole was compared with the M w 7.8 Gorkha Earthquake recorded at four different seismic stations whereas the shear wave and compressional wave velocity was compared with borehole data (Secondary data).
Site classification system proposed by Ambraseys et al. (1996)

Shear Wave and Compressional Wave Velocity Profiles with Borehole Log
The maximum average value of Vs30 obtained at Chhauniis 233.8 m/s (as shown in Figure 3) whereas minimum value obtained at Buddhanagar is 124.4 m/s (as shown in Figure 4). The average values of Vs 30 up to 20 m at other sites are 174.9 m/s obtained at Pulchowk (as shown in Figure 2), 174.0 m/s obtained at Gaushala (as shown in Figure 5) and 142.2 m/s obtained at Bhainsepati (as shown in Figure 6). This shows that the parametric values of the ground motion are greater in the peripheral region than in the central part of the Valley.

Determination of Average Horizontal to Vertical Spectral Ratio (HVSR)
The maximum and minimum values of frequency of the peak average value of HVSR obtained at Bhainsepati (as shown in Figure 7) is between 22 -26 Hz and minimum values of frequency of the peak average value of HVSR obtained at Buddhanagar (as shown in Figure 8) is between 1 -5 Hz respectively. The values of frequency of the peak average value of HVSR obtained at other sites are Chhauni (6 -10 Hz) (as shown in Figure 9), Gaushala (6 -10 Hz) (as shown in Figure 10) and Pulchowk (5 -9 Hz) (as shown in Figure 11).         S-wave velocity and P-wave velocity of the particular area are obtained and is correlated with borehole data. The N-value obtained from existing bore-hole data (SPT Test Number) can also be compared with the N-value obtained from seismic method (MAM). The HVSR for ten site has a relatively clear peak frequency and has been used to determine predominant period and amplification of a site and these results are also compared with the Gorkha Earthquake 2015.
These results can only be considered as general indications of what has occurred during the earth quake as analysis help us to explain the pattern of damage observed in the Kathmandu Valley.

Conclusion
Microtremor Array Measurement (MAM) survey with Horizontal to Vertical Spectral Ratio (HVSR) was carried out in selected sites. Data were processed and inversion was done with sophisticated software Seisimager. Also, the data were used for the evaluation of the resonant frequency of ground to study the topsoil response and ground characteristics of the selected sites in the Kathmandu Valleyhave indicated that the Kathmandu Valley sediments are dependent on the frequency of the seismic waves. The minimum value of Vs30 recorded in Buddhanagar was 124.4 m/s and the maximum value of Vs30 recorded in Chhauni was 233.8 m/s. The average value of Vs30 recorded for most of the sites are near to 142.2 m/s, indicating that near surface deposits are relatively soft. The inversion velocity models have shown the presence of clayey sandy, silty-sandy to sands, gravelly sand and their mixtures. The low velocity model for most of the sites is due to presence of soft soil in that area are responsible for the destruction in most of the sites. The results also have shown that the parametric values of the ground motion are greater in the peripheral region than in the central part of the Valley.