Advances in Earthquake Geotechnics Springer Tracts in Civil Engineering T G Sitharam Ravi S Jakka Sreevalsa Kolathayar Editors Advances in Earthquake Geotechnics Springer Tracts in Civil Engineering S.
Boominathan and R Vijaya
On January 26, 2001, a devastating earthquake struck the Kutch region of Gujarat, India, resulting in approximately 20,000 deaths and an economic loss of around $5 billion In response to this disaster, the Indian seismic code was revised, placing Kutch in seismic zone 5, indicating the highest risk Historical seismic data indicates that basin effects significantly influenced the damage distribution in the region Subsequent site response studies revealed considerable spatial variation in strong motion and high site amplification in certain areas However, existing research has largely overlooked the impact of subsurface geometry on site amplification While field studies provide evidence of basin effects, there is a lack of detailed numerical studies on Kutch Basin's site response This study aims to investigate the basin effects in Kutch through the development of numerical models, highlighting the complex wave propagation phenomena that alter ground motion characteristics across the basin surface.
1995), 1995 Hyogo-Ken Nanbu earthquake in Kobe Basin (Pitarka et al., 1997),
2001 Bhuj earthquake in Kutch Basin, etc., it is not transferred yet to the building design codes and provisions due to the lack of sufficient research and computational
Indian Institute of Technology Madras, Chennai, India e-mail: boomi@iitm.ac.in © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023
T G Sitharam et al (eds.), Advances in Earthquake Geotechnics, Springer Tracts in Civil Engineering, https://doi.org/10.1007/978-981-19-3330-1_2
The study focuses on the Kutch region by utilizing two distinct numerical tools to model its seismic behavior Initially, a three-dimensional simplified model is created to analyze the influence of the basin, followed by the development of a two-dimensional typical Kutch Basin model Ultimately, the most effective model for examining the seismic response of the Kutch Basin is identified A thorough parametric study is then conducted using this model to evaluate the seismic response of a generic basin.
Basin effect refers to the influence of two- or three-dimensional sedimentary basin structures on ground motions The basin effects were noticed initially by Hanks
The basin effect, initially noted during the 1971 San Fernando earthquake, gained prominence following the 1985 Michoacan earthquake, which significantly altered surface motion in Mexico City due to basin effects (Bard et al., 1988) Notably, the peak ground acceleration at the SCT station in Mexico City, located 400 km from the epicenter, exceeded that of the nearby Campos station Within the city, stark variations in ground motion intensity were evident, with SCT recording a peak ground acceleration five times higher than that of the UNAM station, and a spectral acceleration at 2 seconds that was approximately ten times greater (Semblat & Pecker, 2009) This pattern of damage was further highlighted by observations from the 1988 Armenia earthquake.
1994 Northridge Earthquake, 1995 Kobe Earthquake, 1995 Dinar earthquake and
1999 Kocaeli earthquake also reflected the multidimensional effects of the basin on ground motion
Research has demonstrated that the spatial variation of surface ground motion in sedimentary basins is primarily influenced by lateral heterogeneity, leading to significant seismic wave amplification (Kawase & Aki, 1989; Graves, 1995; Davis et al., 2000) Complex wave scattering phenomena—including wave trapping, basin edge effects, wave focusing, and double resonance—result from the multidimensional characteristics of these basins and significantly modify ground motion (Narayan & Kamal, 2015; Semblat et al., 2002) Numerical simulations of earthquakes in various valleys and basins worldwide reveal amplification factors ranging from 1 to 6 across basin surfaces (Frankel, 1993; Olsen, 2000; Smerzini et al., 2011) Consequently, understanding multidimensional basin effects is essential for accurate hazard assessment and effective seismic design of structures.
There are several basins in India, such as Kutch Basin, Indo-Gangetic Basin, Brahmaputra Basin and Talchir Basin which are located in regions of highest seismic
Numerical Modelling of Basin Effects on Earthquake Ground … 31 risk The basin amplification factors are developed for the Indo-Gangetic region by various numerical approaches (Bajaj & Anbazhagan, 2019; Bagchi & Raghukanth,
Limited studies have investigated the multidimensional effects of river basins on ground motion characteristics and structural responses in India As cities along these basins grow rapidly due to population migration, there is a critical need to analyze these effects through site response analysis and to incorporate basin factors into seismic codes Gujarat, one of India's wealthiest states, features three significant basins—Kutch, Cambay, and Narmada Notably, the Kutch Basin has experienced numerous moderate to major intra-plate earthquakes over the past 200 years, including the devastating 2001 Bhuj earthquake, which significantly impacted areas up to 350 km from the epicenter This event highlighted the crucial role of subsurface soil nature and geometry in modifying local seismic ground motions.
3 Kutch Basin: Seismotectonic Setting and Strong Motion Instrumentation
The rift basins of Gujarat, including the Kutch, Cambay, and Narmada Basins, were formed due to reactivated movements along major Precambrian trends The Kutch Basin, characterized by highlands and plains, exhibits the typical asymmetric rift basin geometry inclined towards the south Rifting began during the late Triassic period with the breakup of Gondwanaland and continued until the late Cretaceous pre-collision stage of the Indian plate Following the collision, the Kutch Basin evolved into a shear zone with strike-slip movements along sub-parallel rift faults, resulting in a series of half grabens from north to south The Kutch Mainland fault emerged as the principal fault, with significant earthquake epicenters, including the 2001 Bhuj earthquake, located in this area The basin's structural axis descends southwest, reflected in sediment thickness ranging from less than 500 m in the north to over 4000 m in the south, and from 200 m in the east to over 2500 m in the west The Kutch Basin is composed of Tertiary rocks along its periphery and Quaternary sediments within the basin.
Following the 2001 Bhuj earthquake, the Kutch region experienced intense aftershock activity, with occasional magnitudes of M 3 and M 4 To monitor these events, the National Geophysical Research Institute established a network of three-component broadband sensors and accelerographs Notably, five years post-mainshock, a moderate aftershock of M 5.6 occurred on April 6, 2006 Analysis of strong motion data allowed for the estimation of sediment thickness by examining the travel time differences between S and Sp waves, utilizing a velocity model derived from geophysical surveys.
4 3D Seismic Analysis of Kutch Basin by Spectral Element Method
The Kutch Basin in India, located between latitudes 23.00° N and 23.85° S and longitudes 69.55° E and 70.85° W, is the focus of this study (Mandal, 2006) Utilizing the spectral element code SPEED (Mazzieri et al., 2013), a 3D numerical model is developed to examine how the basin influences ground motion amplification and structural response The numerical model's dimensions are informed by existing data, with a maximum sediment thickness of 1.534 km leading to a consideration of a basin depth of 1.5 km.
As the highest frequency of aftershocks occurred at focal depth of 20–30 km and
The 2001 Bhuj earthquake was analyzed within a computational domain measuring 200 km × 140 km × 30 km, incorporating a simplified rectangular basin of dimensions 150 km × 90 km × 1.5 km The basin extends approximately 250 km in the east-west direction and 150 km in the north-south direction, with sediment thickness data considered for 150 km east-west and 90 km north-south, as reported by Mandal (2006) The depth of the half space is set at 30 km, aligning with findings by Mandal et al (2005) A geometric layout of the 3D model utilized in this study is illustrated in Fig 1.
Fig 1 3D simplified model of Kutch Basin along with the surface of monitored points (not to scale) (Vijaya et al., 2020)
Numerical Modelling of Basin Effects on Earthquake Ground … 33
The computational domain is discretized with hexahedral elements, ensuring accurate wave propagation by utilizing five grid points per minimum wavelength (Paolucci et al., 2016) The soil in the basin is modeled as a linear visco-elastic medium, defined by its average density (ρ) and shear wave velocity (Vs) In the Kutch region, the shear wave velocity, measured through Multichannel Analysis of Surface Waves (MASW) and PS-logging methods, ranges from 200 m/s at the surface.
At a depth of 60 meters, the shear wave velocity is measured at 800 m/s (Sairam et al., 2019) For this region, an average shear wave velocity (Vs) of 500 m/s is used for soil, while 1000 m/s is applied for rock The compressional wave velocity (Vp), damping (Q), and density (ρ) values are derived from the shear wave velocity Poisson’s ratio is set at 0.2 for fine dense sandy soil and 0.15 for rock, following existing literature (Sairam et al., 2019) Additionally, absorbing boundary conditions are established along the sides, utilizing the local P3 paraxial condition as proposed by Stacey.
In 1988, a second ricker wavelet with a maximum frequency of 1 Hz was utilized as the input motion at the base of the simulation The simulation was conducted at the Virgo Cluster at IIT Madras, leveraging 124 parallel processors The entire simulation for the basin required approximately 90 hours of computational time For more comprehensive insights into the 3D modeling of the Kutch Basin, refer to the works of Vijaya et al (2018, 2020).
The 3D ground response analysis of the Kutch Basin reveals acceleration time history data collected at 24 distinct points across the basin surface, as illustrated in Fig 1 These points are strategically positioned along four longitudinal cross sections (A–A, B–B, C–C, and D–D) marked in red, and three transverse cross sections (E–E, F–F, and G–G) marked in blue Due to the symmetrical nature of the basin, only one-fourth of the surface is monitored The horizontal acceleration time history data for sections A–A and D–D is depicted in Fig 2.
The analysis reveals that numerous peaks are present near the edge of the basin, with a noticeable decrease in the number of peaks as one approaches the center.
The analysis reveals that cross-sectional C–C and D–D exhibit a higher number of peaks compared to sections A–A and B–B, attributed to the basin edge effect caused by surface waves resulting from the interference of body waves and reflected waves at the edges In contrast, fewer peaks are observed towards the center due to the dampening of surface waves over distance Additionally, the duration of ground motion is notably longer near the edges than at the basin center, influenced by locally generated surface waves This finding aligns with observations made by Pitilakis et al (2004) during field studies at the EUROSEISTEST site in northern Greece.