Characterization of Ground Response and Liquefaction for Kathmandu City Based on 2015 Earthquake Using Total Stress and Effective Stress Approach

1. Introduction

Earthquakes are very random in nature and can destroy a whole city in a matter of seconds, causing severe human and infrastructure losses. The severity of the earthquake occurrences is significantly dependent on the interaction of the strata with the ground motion. For example, safety of the structures is greatly governed by the response of foundation strata during an earthquake. An earthquake, occurs due to the rupture or convergence of the fault planes, generates primary and secondary waves. These waves propagate through the underlying bed rock and reaches to the ground surface. Various factors like soil stratification, index properties as well as dynamic characteristics of the soil strata, ground motion parameters and other geological parameters are responsible for the amplification or attenuation of the seismic waves. In order to evaluate the amplified or attenuated behavior of waves at ground surface, ground response analysis is a viable option. Ground response analysis is a powerful tool, which helps in the development of design response spectra and also helps in the evaluation of dynamic stresses and strains for liquefaction hazard analysis. Despite the existence of multi-dimensional (two-2D and three-3D) ground response analysis, one dimensional analysis is widely used in engineering practice due to its simplified nature and also efficient simulation of realistic soil response (Kramer, 1996; Garala and Madabhushi, 2018).

South Asian region is one of the most seismically active regions in the world. Several ground response studies have been reported in literature for this region. Kumar and Krishna (2013) and Dammala et al. (2017) have carried out a seismic response analysis of Guwahati city using equivalent linear method and reported that the Fourier amplifications of ground motion at surface are in the range of 4.14-8.99 for a frequency band of 1.75 Hz - 3.13 Hz and the maximum spectral acceleration varied in range of 1.0g to 4.71g at different locations. Basu et al. (2019) have also performed ground response analysis for Guwahati city and mentioned the importance of nonlinear properties of soil in ground response based on the non-Masing load/unload/reload characteristics. Ranjan (2005) carried out seismic response analysis of Dehradhun city and reported that the range of spectral acceleration was 0.06-0.37g at frequency range 1 Hz - 10 Hz. Similarly, Chaterjee et al. (2015), Phanikanth et al. (2011) and Mahmood et al. (2015) conducted the ground response analysis of Kolkata, Mumbai and Dhaka using one dimensional equivalent linear method respectively. Though, the estimation of ground response due to the local soil sites is a complex problem, it is of utmost significance for unplanned and haphazardly growing city like Kathmandu, capital of Nepal. Kathmandu city being surrounded by several thrust fault such as Main Boundary Thrust and Main Central Thrust (see Figure 1), is one of the highly active seismic regions of Nepal. The region had experienced many major earthquakes in the past such as 1934 Nepal-Bihar Earthquake and 2015 Nepal Earthquake. Few ground response studies related to the recent 2015 Nepal Earthquake were reported in the literature. Gautam et al. (2016) carried out the equivalent linear ground response analysis of soft sediments of Kathmandu City using EERA algorithm and reported that maximum spectral amplification was found to be occurring in the frequency range of 0.2 Hz – 0.6 Hz. Chamlagain et al. (2015) simulated M_w 7.8 Nepal earthquake at Kathmandu and reported that soil fundamental period ranges from 0.4s to 0.7s along with PGA variations of 0.11g to 0.46g. Paudyal et al. (2015) evaluated the predominant frequency of ground motion using the Horizontal-to-Vertical (H/V) spectral ratio technique. The predominant frequencies varied from 0.5 Hz to 8.9 Hz in the study area, whereas the second resonance frequency varied from 4 Hz to 6 Hz in the center and northern part of the valley. The result indicated that the center and northern part of the valley had a wide range of resonance frequency due to two levels of impedance contrast. This leads to the suggestion of considering the effects of surface and the lower layers during the planning and designing of infrastructures in the Kathmandu Valley. Furthermore, Okamura et al. (2015) reported that soil liquefaction occurred in few areas of Kathmandu like Manamaiju and Imadol during Nepal 2015 earthquake.