Artículo Ea2

Indian Geotech J (April 2024) 54(2):358–393
https://doi.org/10.1007/s40098-023-00775-8
ORIGINAL PAPER
Nonlinear Seismic Ground Response Analysis for Site Classes D

and E of Bihar Region, India
Khushboo Kumari1 · Prabhakar Kumar2 ·
Shiv Shankar Kumar2
Received: 15 March 2023 / Accepted: 22 August 2023 / Published online: 11 September 2023
© The Author(s), under exclusive licence to Indian Geotechnical Society 2023
Abstract According to the NEHRP (2003), the soil site input acceleration) of seismic wave was found to be in the
classes C, D and E are more responsible for the seismic wave range 0.35–3.0. Further, the results obtained from nonlin-
amplification; however, among these three site classes, the ear GRA also indicated that the seismic wave is amplified
site class E is more responsible for soil liquefaction. Based by 180%, but the deamplification by 50%. The increase of
on the borehole profiles and SPT-N values, collected from all pore water pressure ratio up to 0.93, with increasing seis-
48 sites within Bihar region, it was observed that the entire mic energy or higher PGA input motion, indicates the initia-
Bihar region comes under the range of site classes D and tion of soil liquefaction at most of the soil site. The results
E with the variations of Vs30 ranging from 180 to 360 m/s. indicated that the seismic GRA is significantly affected by
Therefore, an attempt has been made to perform nonlinear input motion and the soil variability. It can be stated that
seismic ground response analysis (GRA) of site classes D this study can be useful for the geotechnical engineers to
and E of Bihar region using DEEPSOIL software, since the design the earthquake-resistant structures; however, more
entire Bihar region comes under the seismic zone of III, IV experimental investigations are required to understand the
and V. Three acceleration time histories of different peak mechanism of soil liquefaction in the entire Bihar region.
ground acceleration (PGA) of 0.1 g, 0.26 g and 0.45 g indi- Further, this study can also be utilized for the development
cating low, moderately high and very high seismic hazard of surface level ground motion attenuation relationship for
scenarios, respectively, have been chosen for this analysis. the Bihar region. Overall, this study is one of the preliminary
The results obtained from seismic GRA have been presented investigations toward seismic microzonation study of Bihar
in terms of the variations of acceleration, amplification/ region, which can be useful for the development of disaster
deamplification of seismic waves, shear strain, shear stress management plan by providing seismic microzonation map.
ratio and pore water pressure along with the depth. The high
shear strains (greater than 0.5%) observed within the soil Keywords Ground response analysis · Site classes D and
deposit may cause permanent deformation in the ground E · Input motion · Shear wave velocity · Soil variability ·
resulting catastrophic damage to the existing structures. The Uncertainty
amplification factor (i.e., ratio of output acceleration to the
* Shiv Shankar Kumar Introduction

[email protected]
Khushboo Kumari The recent deadliest 2023 Turkey ear thquakes
[email protected] (5.2 ≤ M w ≤ 7.7) once again indicated that the ground
Prabhakar Kumar shaking, during earthquakes, leads instability to the soil-
[email protected] supported structures which can cause severe damage to
1
National Institute of Technology (NIT) Patna, Bihar 800005, structure as well as loss of property. The damage to the
India structures depends on several factors such as soil condi-
2
Department of Civil Engineering, National Institute tions, quality of the constructional materials, frequency
of Technology (NIT) Patna, Bihar 800005, India of the earthquake motion as well as the natural frequency
13
Vol:.(1234567890)
Indian Geotech J (April 2024) 54(2):358–393 359
of structures, ground shaking and seismic magnitude. Fig- the effect of seismic motion prior to the occurrence of
ure 1a–d indicates the results of ground shaking due to the seismic events. Seismic GRA is used to evaluate the free-
earthquakes. Therefore, it can be stated that the sustainable field response of seismic waves on the surface level based
infrastructure always requires the mitigation techniques or on the characteristics of soil deposits and the bed rock
tools to minimize these damages, since the earthquakes motions. This is one of the significantly important ways to
are a natural activity and cannot be stopped. Thus, to mit- see the potential amplifications of bedrock motion at the
igate or minimize the damages, geotechnical engineers surface level, which requires for the design of any struc-
have developed different methods to foresee the potential tures prior to the ground shaking. This GRA can be done
consequences of any seismic activity at any site of inter- by either by one-dimensional or two-dimensional analysis
est prior to the construction. Seismic ground response using equivalent linear (frequency-domain analysis) and
analysis (GRA) is one of the effective tools to analyze nonlinear (time-domain analysis) approaches [1]. Several
Fig. 1 Effect of ground shaking a during and b after 2015 Nepal earthquake, c structural response during earthquake and d structural response
after the earthquake [93]
13
360 Indian Geotech J (April 2024) 54(2):358–393
computer programs such as DEEPSOIL, SHAKE 2000 and Previous Studies

EERA are available for one-dimensional GRA, which is
based on the consideration of infinitely extended soil and Several studies are available on the ground response analy-
bedrock layer in horizontal direction. The equivalent lin- sis, which has been carried out at several places in India, and
ear method is an approximate method in which nonlinear revealed that the ground response analyses are needed to be
behavior of the soil (i.e., strain-dependent shear modulus carried out for safe and sustainable design of the structures
and damping ratio) is modeled in terms of equivalent lin- due to ground shaking. Researchers have performed 1-D
ear properties by incorporating effective shear strain and equivalent linear and nonlinear seismic GRA to predict seis-
iterative procedure [1, 2]. mic design parameters and reported that the GRA is strongly
However, nonlinear time-domain analysis, based on the influenced by the local site geology as well as the strong
consideration of effective stress, incorporates the realis- motion characteristics [4–16]. The local soil geology is
tic nonlinear soil behavior where the pore water pressure mainly responsible for the amplification or deamplification
generation and dissipation can also be modeled. Moreo- of seismic motion from bedrock to surface level. It was also
ver, GRA is used to predict the ground surface motion reported that the stiffer soils produce higher spectral accel-
for the development of design response spectra, to evalu- erations as compared to softer soils as per Basu et al. [17],
ate the dynamic stresses as well as strains for the evalu- which shows a contradiction with the statement reported by
ation of liquefaction hazards and also to determine the Kramer [1]. Kumar et al. [4] have performed seismic GRA
earthquake-induced forces that can lead to instability and liquefaction study for Kalyani region, Kolkata, using
of earth and earth-retaining structures [1]. Since these equivalent linear and nonlinear analysis using PLAXIS-2D.
parameters depend on the variations of soil types as well It was reported that the equivalent linear analysis gives con-
as the soil conditions existing below the ground level, it is servative results as compared to the nonlinear analysis. Fur-
very much important to classify each soil site. According ther, Kumar et al. [4] have also reported that the simplified
to the National Earthquake Hazards Reduction Program method fails to predict the soil liquefaction for few regions,
(NEHRP), the site has been classified as site classes A, B, which was found to be liquefaction prone regions based on
C, D and E depending on the shear wave velocity within equivalent linear and nonlinear GRA. Pawirodikromo [18]
30 m (i.e., Vs30) depth from the existing ground level [3]. has performed nonlinear GRA for Yogyakarta, Indonesia,
The site class A reflects hard rock with V s30 > 1500 m/s, using computer program nonlinear earthquake site response
whereas the site classes B, C, D and E indicate the rock analyses (NERA) subjected to bedrock motion of PGA
strata, very dense soil or soft rock, relatively stiff soil, ranging from 0.254 to 0.289 g. It was reported that PGA
soft soil with the shear wave velocity 760 < Vs30 ≤ 1500, at ground surface varied from 0.398 to 0.412 g. The site
360 < Vs30 ≤ 760, 180 < Vs30 ≤ 360, < 180 m/s, respectively. amplification of seismic wave and the spectral site amplifi-
Among these soil site classifications, the site classes C, D cation was in the range of 1.40–1.43 and 1.22–1.39, respec-
and E are more responsible for the seismic wave amplifica- tively. Further, it was reported that the damage of building is
tion; however, the site class E is more susceptible to failure mostly caused by the high levels of ground shaking as well
or collapse or liquefaction due to seismic excitation. Based as low-quality material, since the site amplifications were
on the borehole profiles and SPT-N values, collected from in the low range of damage intensity. The summary of the
all 48 sites within Bihar region, it was noticed that the studies [19–37] is presented in Table 1. Moreover, it was
entire Bihar region comes under the range of site classes observed that the GRA has been performed for the most of
D and E with the variations of Vs30 ranging from 180 to the Indian cities, however; very few studies are available for
360 m/s. Therefore, this study is intended to perform the the Bihar region. As per IS 1893 [38], almost all districts
nonlinear seismic GRA, since almost all soil site of Bihar of Bihar region come under the seismic zones IV and V.
region comes under the range of site classes D and E. Therefore, it is utmost important to perform GRA for the
This study can be useful for the geotechnical engineers potential amplification of bedrock motion at ground surface
to design the earthquake-resistant structures; however, to design earthquake-resistant structures considering futur-
more experimental investigations are required to under- istic earthquake events.
stand the mechanism of soil liquefaction in the entire
Bihar region. Overall, this study is one of the prelimi-
nary investigations toward seismic microzonation study Study Area
of Bihar region, which can be utilized for the development
of disaster management plans for Bihar state by provid- Bihar is the eastern state of India falling between latitudes
ing seismic microzonation map with the consideration of 24° 20′ 10″ N and 27° 31′ 15″ N and longitudes 83° 19′
futuristic earthquakes. 50″ E and 88° 17′ 40″ E (shown in Fig. 2). As per IS: 1893
[38], eight districts of Bihar fall into seismic zone V, 24 fall
13
Table 1 Summary of the literature review
Reference GRA methodology and Software Region Remarks
Kumar et al. [4] Equivalent linear and nonlinear analysis, PLAXIS Kolkata, India Equivalent linear (EL) analysis gives conservative results as
compared to the nonlinear (NL) analysis
Simplified method fails to predict the liquefaction susceptibil-
ity at certain regions which are prone to liquefaction based
on EL and NL analyses
Ansari et al. [5] Equivalent linear using DEEPSOIL Jammu, India Emphasized on the significance of seismic response analyses
of foundation soil and superstructures for developing any
prospective earthquake-induced liquefaction mitigation
strategies to design and construction of any project in
Jammu and Kashmir
Silahtar [6] Nonlinear GRA, DEEPSOIL Arifiye (Sakarya District), Turkey Parameters required to design the earthquake-resistant
structures depends upon the ground conditions and selected

strong ground motions
Kawan et al. [7] Equivalent linear and nonlinear analysis DEEPSOIL Bhaktapur City, Nepal Amplification factor obtained from equivalent linear analysis
were in the range of 0.9 to 1.5, whereas, from nonlinear
analysis, it ranged from 0.80 to 1.30
Kumar et al. [8] Equivalent linear and nonlinear approach using DEEPSOIL Guwahati (India) Stiffer soil layers result in similar PGA from both the analy-
ses, and the response is strongly influenced not only by the
local site geology, but by the strong motion characteristics
itself
Kumar et al. [9] Study on the effect of ground motion parameters with Sikkim (India) Response spectrum, which is one of the most important
distance design parameters for the construction at any site, decreases
as the distance increase
Sisodiya et al. [10] 1-D nonlinear, DEEPSOIL Bettiah, Bihar (India) Development of high shear strains (up to 5%) may cause
permanent deformation in the ground, which might lead to
the significant destruction to the structures
Satyam and Towhata [11] Equivalent linear analysis using code ONDA Vijayawada city (India) Ground response analysis, peak acceleration, frequency and
amplification parameters were assessed and the city was
divided into different zones corresponding to each value
Liquefaction hazard of the city was quantified using SPT-
based method in combination with the dynamic analysis
Reddy et al. [14] Equivalent linear, DEEPSOIL Andhra Pradesh (India) Estimated PGA was varying from 0.19 to 0.26 g, which
indicates the amplification of input motion and the ranging
was 2.37 to 3.25
Effective amplification (i.e., Fourier amplification ratio)
occurs by a factor greater than 5.09–5.31 in the frequency
range of 2.15–3.71 Hz
The larger amplification factors were observed at the loca-
tions where shear wave velocity (vs) was low, especially in
filled-up soils; the higher amplification indicates that the
seismic potential is higher
13
361
Table 1 (continued)
362
13
Chandran and Anbazhagan [15] 2D nonlinear seismic site response using FLAC2D Peninsular India It was reported that the short-period amplification factor (Fc)
gives higher values than long period amplification factor
(Fs) in the case of stiff soil, whereas it shows reverse trend
in case of soft soil
The trend of Fc became more complex than that of the Fs,
especially in the case of stiff soil sites
Nguyen et al. [16] Nonlinear (NL) and equivalent linear (EQL) Normalized response spectra of EQL approach results are
higher than those of NL approach
Peak ground accelerations at the surface from EQL analyses
are greater than those of the NL method because the latter
generates a higher amount of nonlinearity
Shear strains obtained from NL method are closely matched
with those from the EQL method, the difference between
them increasing with a decrease in soil stiffness
Basu et al. [17] Equivalent linear and nonlinear analysis using DEEPSOIL Guwahati (India) Attenuation of ground acceleration is observed in most part
of the site for input motion peak bedrock acceleration
greater than 0.18 g. Higher value of spectral accelerations
was reported for stiffer soil site as comparison to the softer
soil
Pawirodikromo [18] Nonlinear GRA using computer Yogyakarta, Indonesia PGA at ground surface was in the range of 0.398–0.412 g
program SHModel Site amplification of seismic wave and spectral amplification
was in the range of 1.40–1.43 and 1.22–1.39, respectively
It was reported that the damage of building is mostly caused
by the high levels of ground shaking; however, low-quality
materials were also responsible
Dammala and Krishna [19] Nonlinear analysis using DEEPSOIL Northeastern Loose or soft soil deposits amplify the incoming seismic
India waves from bed rock
Loosely compacted sand layers of northeastern region may
exhibits liquefaction up to a depth of 15 m from the surface
for the input motion PGA > 0.10 g
Bhusal et al. [20] Equivalent linear (EQL) and nonlinear (NL) GRA with Dharahara, Kathmandu, Nepal Design spectrum calculated lies in between NBC 105: 2020
DEEPSOIL and IS 1893:2016 (soil type D); therefore, the site-specific
seismic GRA has been recommended
Yildiz [21] Equivalent linear and nonlinear approach using DEEPSOIL Malatya, Turkey region Liquefaction potential as well as site amplification map near
the surface layer can be utilized by the engineers for the
planning and urban development in Malatya, Turkey region
Mase et al. [22] Equivalent linear seismic GRA; DEEPSOIL Bengkulu City, Indonesia PGA at ground surface is relatively higher in comparison to
the input motion
Peak ground acceleration in the study area ranges from 0.2 to
0.8 g, while the spectral acceleration varies between 0.5 and
1.5 and 0.4 and 0.8 g for periods of 0.2 and 1 s, respectively
Amplification factor of the site was observed to vary from 0.5
to 1.6
Table 1 (continued)
Yildiz [23] Equivalent linear and Nonlinear analysis, DEEPSOIL Istanbul Nonlinear analysis leads to the lower values of PGA and
spectral acceleration compared to the equivalent linear
analysis due to the over-predicted stiffness and under pre-
dicted damping for shear strains greater than the effective
shear strain
Amplification and deamplification were reported at the site
where soil deposits are relatively loose and dense, respec-
tively
Puri and Jain [24] Nonlinear GRA using DEEPSOIL Haryana (India) Amplification factor for PGA was in the order of 0.531—
1.881
It was recommended that the site-specific design parameters
should be adopted since the soils in Haryana are capable of

amplifying the earthquake ground motions
Dammala et al. [25] Nonlinear effective stress analysis, DEEPSOIL Northeast India High amplification for low peak bed rock acceleration
(PBRA) ground motions (< 0.1 g) was observed and the
attenuation of seismic waves was observed beyond a PBRA
of 0.10 g near the surface stratum due to the high strains
induced within the soil resulting in high hysteretic damping
Also, increased excess pore pressure generation with
increased PBRA of the input motion was observed and the
considered soil stratum is expected to liquefy beyond a
PBRA of 0.1 g
Putti et al. [26] Equivalent linear using DEEPSOIL Vishakhapatnam (India) It was reported that the peak ground acceleration at surface
and bedrock, predominant frequency and H/V amplitude
significantly affected by local site condition
It was also reported that the estimated parameters considering
local soil conditions will be of great help in seismic design
of structures as well as retrofitting of the existing structures
to withstand against seismic hazards
Sil and Haloi [27] Equivalent linear using EERA Silchar (India) Local soil conditions are responsible for changing the ground
motion parameters
Bed rock motion significantly amplified, within the range of
frequency from 2 to 8 Hz, due to the presence of soil layers
and, the maximum amplification factor was found to be 4.6
Ahmed and Bhattacharjee [28] Nonlinear using DEEPSOIL Jorhat, Assam (India) PGA obtained at the surface level was varied in the range
0.13–0.19 g and, the amplification ratio varied in the range
of 1.04–1.37
13
363
Table 1 (continued)
364
13
Pandey et al. [29] Equivalent linear GRA using SHAKE2000 Uttarakhand (India) It was observed that the site amplification ratios were vary-
ing from 2.5 to 4.9 for different sites and, the normalized
response spectrum obtained from GRA was significantly
different than the response spectrum provided by IS
1893:2002 (Part-1)
It was also recommended that the study of site-effect need to
be carried out for all the strong motion stations since, local
soil conditions as well as local soil geology strongly influ-
ence the strong motion characteristics
Singhai et al. [30] Equivalent linear and nonlinear analysis using DEEPSOIL Guwahati (India) Generation of excess PWP indicates that there is a reduction
in stiffness and shear strength of soil resulting changes in
the soil behavior of soil. The PGA obtained during nonlin-
ear analysis with excess PWP dissipation is approx. 35–95%
lesser than equivalent linear analysis
Desai and Choudhary [31] Equivalent linear and nonlinear analysis using DEEPSOIL Mumbai (India) Seismic amplification varies from 2.53 to 4.14 for frequency
band of 1.75– 3.5 Hz
Akhila et al. [32] Equivalent linear, DEEPSOIL Kolkata (India) PGA of the city range was reported in the range of 0.1–0.34 g
Naik and Choudhury [33] Equivalent linear using DEEPSOIL Goa (India) Peak ground acceleration amplification factors for the same
earthquake motion differ from site to site
Amplification factors are found to be varying from 1.56 to
2.36
Shiuly et al. [34] Equivalent linear SHAKE 2000 Kolkata (India) PGA of Kolkata metropolitan district varies from 0.169 g to
0.414 g and the maximum amplification factor ranges from
2.2 to 3
Pallav et al. [35] Equivalent linear, SHAKE 99 Imphal (India) Mean and standard deviation of surface level spectral ground
acceleration at PGA and natural period of 0.3 and 1 s were
reported as contour maps for all the synthetic sample earth-
quake events
Thaker et al. [36] Equivalent linear using DEEPSOIL and SHAKE 2000 Kutch, Gujarat (India) Peak ground acceleration obtained at the surface level was
0.216 g for a given bed rock motion of PGA (0.088 g),
indicating that site is amplifying in nature
Ranjan [37] Equivalent linear using SHAKE2000 Dehradun (India) Spectral acceleration was in the range of 0.06–0.37 g at
frequency range 1–10 Hz. The spectral acceleration map of
the city has been compiled using interpolation technique
in GIS environment for 3 Hz, 5 Hz and 10 Hz frequency,
which further indicates the lesser or severe vulnerability of
structures during the earthquakes
Fig. 2 Classification of seismic zones of Bihar [45]
in zone IV and 6 fall in zone III, which reflects moderate fore-deep. IGP was formed after the upliftment of the Him-
to very severe earthquake zone (Fig. 2); hence, it can be alayas due to the collision of the Indian and the Eurasian
stated that the entire Bihar comes under the seismic zones plates [39]. The river Ganga which flows across the state
III, IV and V. Bihar lies in between the Indian shield and separates North Bihar from South Bihar and is also respon-
the Himalayas and constitutes a substantial portion of the sible for the deposition of alluvium soil. It also straddles the
Indo-Gangetic plain (IGP) also known as the Himalayan river Ghaghara, Burhi Gandak, Mahananda, Kosi, Gandak
13
and Son River. Most of its area consists of thick alluvium Methodology
deposited by Ganga and its tributaries, whereas the south-
ern part of the region has crystalline and metamorphic rock One-dimensional equivalent and nonlinear GRA is based
deposits. on the assumption of vertical propagation of shear waves
through a linear visco-elastic material which approxi-
mate the soil behavior as a Kelvin–Voigt solid [1]. The
stress–strain behavior of Kelvin–Voigt solid during shear-
Seismotectonic Feature and Seismic Zone of Bihar ing can be presented by Eq. (1).
The study area consists of many tectonically active fea- 𝜕𝛾

𝜏 = G𝛾+𝜂 (1)
tures such as Main frontal thrust (MFT), Main Boundary 𝜕t
thrust (MBT), Main Central thrust (MCT), East Patna where τ, G, γ (= 𝜕u∕ 𝜕z ) and η are the shear stress, shear
fault (EPF), West Patna fault (WPF), Sitamarhi fault (SIF), modulus, shear strain and the coefficient of viscous damping
Munger–Saharsa Ridge fault (MSRF), Munger–Saharsa parameter, respectively. Moreover, the equation of motion,
Ridge Marginal fault (MSRMF), Malda Kishanganj fault for one-dimensional vertically propagated (in z direction)
(MKF), Katihar-Nailphamari fault (KNF) and Purnia-Ever- shear wave, can be represented by Eq. (2).
est lineament (PEL) [40–43]. Along with these faults, there
are other two faults perpendicular to WPF, aligned in the 𝜌
𝜕 2 u 𝜕𝜏
= (2)
northwest (NW) direction, falling aside to the Sitamarhi and 𝜕t2 𝜕z
Siwan [41]. The EPF is oriented in a NE–SW direction from
Patna in the south to the Nepal border to the east of Madhu- 𝜕2u 𝜕2 u 𝜕3u
bani, shown in Fig. 2. WPF and MSRF run almost parallel
𝜌 2
=G 2 +𝜂 2 (3)
𝜕t 𝜕z 𝜕z 𝜕t
to it on the left from Arrah to the Nepal border and on the
right from Bihar Sharif to Morang in eastern Nepal [44]. Finally, on substituting Eq. (1) into Eq. (2), the wave
Among these, EPF is one of the most active faults since, equation yields for one-dimensional ground response
the several major earthquakes such as 1934 and 1988 earth- are expressed as Eq. (3), wherein 𝜌 is the mass density of
quakes (see Table 2) are associated with this fault and it can medium; G is the shear modulus; η = (2G/ω)*D is the vis-
lead to a number of EQs primarily due to interaction with cosity, where ω and D is wave frequency and damping ratio,
MFT [45–47]. Munger–Saharsa ridge denotes the prolonga- respectively; and u is the displacement of the medium along
tion of the Bundelkhand and Satpura massifs. Further, IGP lateral direction. Further considering the complex shear
represents the depressed part of the Peninsular India and modulus (G*), the equation of motion can be expressed like
contains several hidden east–west directed tectonic features Eq. (4) [1].
due to narrow MFT [48, 49]. The crustal thickness of this ∗ z) ∗ z)
u (z, t) = Aei(𝜔t+k + Bei(𝜔t−k (4)
region varies from 30 to 45 km and increases in the north
direction [50]. The sediment accumulation rate in north where u (z, t) is horizontal displacement at a depth z below
Bihar is very high, ranging from 0.7 to 1.5 mm/year, as com- ground level and at time t; ρ is the density; and. ‘A’ and ‘B’
pared to the average sediment deposition rate of 0.2 mm/year are the amplitudes of waves traveling in the –z (upward)
due to the web of rivers throughout the region [51–53]. The and + z (downward) directions, respectively. k* = 𝜔 𝜌∕ G∗
√
alluvial deposits have a thickness of 1.5–5.0 km, concealing is the complex wave number, G* = (G + iωη) is the complex
the solid geology of the basement [54]. Such continuous shear modulus. To eliminate the frequency dependency,
deposition of sediments in a layer, of thickness up to several the complex shear modulus can be expressed as: G* = G
kilometers, leads to the higher path attenuation [55, 56]. (1 + 2iD), where D is the damping ratio.
Table 2 Major earthquake in Bihar and adjoining area [45]

S. no. Earthquakes Date Time (UTC) Latitude Longitude Magnitude Depth (km) Fatalities
(Mw)
1. 1833 Bihar 26 Aug 18:05:00 NA NA 7.5 NA 500

2. 1934 Bihar–Nepal 15 Jan 08:43:25 26.885° N 86.589° E 8.0 15 10,600
3. 1988 Bihar–Nepal 20 Aug 23:09:09 26.71° N 86.62° E 6.9 62 1500
4. 2015 Nepal (Gorkha EQ) 25 April 06:11:25 28.230° N 84.731° E 7.8 8.2 9000
5. 2015 Nepal (aftershock) 12 May 07:05:19 27.809° N 86.066° E 7.3 15 100
13
Equivalent Linear Analysis and the input motion are required. All these parameters have
been accessed from the bore-log and laboratory test to per-
Equivalent linear analysis is one of the methodologies to form nonlinear seismic GRA using commercial software
conduct GRA, in frequency domain, by approximating the DEEPSOIL 7.0.
accurate nonlinear strain-dependent dynamic soil properties,
i.e., the variations of shear modulus and damping factors Location of Boreholes and Soil Type of Study Area
with shear strain. During this analysis, the obtained shear
strain histories, for each layer, are used to compute the effec- The characteristics of soil strata up to depth of 30 m are
tive shear strain, which is further used to estimate the com- significantly important in earthquake engineering, since
patible nonlinear strain-dependent dynamic soil properties amplification or deamplification of ground motion is more
using an iterative technique. The effective shear strain (γeff) prominent in this range [57]. For GRA of Bihar state, 48
in each layer is determined from the maximum shear strain representative boreholes have been collected from the dif-
(γmax), i.e., γeff = Rγ γmax , where Rγ is the strain reduction ferent construction sites which indicate the typical variation
factor and it is the ratio of effective shear strain (γeff) to the of soil profile in entire Bihar region. The locations of the
maximum shear strain (γmax). For the analysis, Rγ is com- borehole are presented in Fig. 3. From all borehole profiles,
monly calculated with the help of earthquake magnitude it was found that the soils present within the Bihar region are
(M), i.e.,R𝛾 = (M − 1)∕ 10. mostly sand and clay with some amount of silt. The water
table location was observed to be varying from 0 to 5 m
from the surface level. The variation of SPT-N value, rang-
Nonlinear Analysis ing from 2 to 50, along with the depth indicates the condi-
tions of soil such as loose and stiff soil. Few typical bore-
The estimation of accurate design parameters for earth- hole profiles, presented in Fig. 4, indicate the availability of
quake-resistant structures during ground shaking is one soil types in Bihar along with the soil’s density, shear wave
of the most common geotechnical earthquake engineering velocity and thickness of different soil layers.
problems, since the local soil conditions significantly change Keeping eye on the different seismotectonic faults
the characteristics of earthquake motion from bedrock to the (such as Main frontal thrust, East Patna fault, West
surface level. Therefore, geotechnical engineer has a major Patna fault, Sitamarhi fault, Munger Saharsa Ridge fault,
concern to develop an efficient technique to model the soil Munger–Saharsa Ridge Marginal fault, Malda Kishanganj
properties accurately, to measure the earthquake response fault, etc.) in and around Bihar and high sediment deposition
at surface level. In nonlinear methodology, the equations rate in Bihar region (might be responsible for the amplifica-
of motions for dynamic equilibrium, shown in Eq. (5), are tion of seismic waves), the selection of borehole was done
solved at each time step using time integration method [2] in such a way that it covers the seismic zones III, IV and V.
to find out the seismic response of soil in terms of the varia-
tions of acceleration, amplification/deamplification of seis- Dynamic Soil Properties: Modulus Reduction
mic waves, shear strain, shear stress ratio and pore water and Damping Ratio Curves
pressure along with the depth.
Past earthquake events have indicated that soils may expe-
[M]{̈u} + [C]{u}
̇ + [K]{u} = −[M]{I}ü g (5) rience low to high shear strain levels [58]. The responses
where [M], [C], [K], {ü}, {ů}, {u}, {I} and üg indicate of soil at high shear strains (γ > 0.01%) are substan-
the mass matrix, damping matrix, stiffness matrix, rela- tially destructive than the response at low shear strains
tive acceleration, relative velocity, relative displacement, (γ < 0.01%), mainly due to the nonlinear behavior of soil at
unit vector and acceleration of the base of the soil column, high shear strains [59–61]. Therefore, it is essential to evalu-
respectively. ate the dynamic properties of regional soils for site-specific
To perform nonlinear seismic GRA, several important GRA for wide range of shear strain.
parameters such as borehole profile along with SPT-N value, For GRA, the dynamic soil properties, generally repre-
strain-dependent dynamic soil properties (in terms of shear sented by the variation of shear modulus and damping ratio
modulus and damping ratio variation with shear strain), plas- with shear strains, needed to be evaluated for the specific site
ticity index of soil (which was in the range of 5–30% for prior to the performance of GRA. Due to the unavailability
clayey soil), shear wave velocity (which can be estimated of site-specific dynamic soil properties in the Bihar region,
based on empirical correlation considering SPT-N value), the dynamic soil properties such as modulus reduction and
pore water pressure (PWP) model (which reflects the genera- damping ratio curves existing in the DEEPSOIL have been
tion and dissipation of excess PWP during ground shaking) utilized in the analyses. In the present study, to characterize
the dynamic properties of sand soil and clay soil, existing
13
Fig. 3 Locations of the

boreholes used for GRA in this
study
Fig. 4 Typical borehole profiles at a Arariya, b Sitamarhi, c Patna, d Madhepura site
within the boreholes, the modulus reduction and damping and SPT-N values, presented in Table 3, were used to evalu-
ratio curve proposed by Seed and Idriss [62] and Vucetic and ate the shear wave velocity (Vs) profiles. It can be noticed
Dobry [63], respectively, have been considered, as shown that the significant amount of uncertainty can be incor-
in Fig. 5a,b [2]. To define the dynamic soil properties of porated while using any one of the empirical correlations
clay soil, the plasticity index was considered in the range of [64–85], from Table 3, therefore an averaging technique has
5–30% as per the borehole data. been adopted to estimate an average empirical correlation
of shear wave velocity for GRA, since the random selection
Shear Wave Velocity of empirical correlation between VS and SPT-N, due to the
unavailability of shear wave velocity profile at any particular
Shear wave velocity (Vs) is also one of the significantly site, might affect significantly the outcome of site-specific
important input parameters for GRA studies, which indicate seismic GRA.
the variations of soil stiffness along with the depth. In the Figure 6 presents the variations of shear wave veloc-
present study, the correlations of shear wave velocity (Vs) ity along with depth considering 23 different empirical
13
Fig. 5 a Modulus reduction curve and b damping ration curve used in GRA (redrawn after Vucetic and Dobry [62] and Seed and Idriss [63])
Table 3 Empirical correlation for the estimation of shear wave

velocity based on the observed SPT-N value
S. no. References Correlations
1. Imai and Yoshimura [64] Vs = 76*N0.33

3. Ohba and Toriumi [65] Vs = 84*N0.31
4. Ohsaki and Iwasaki [66] Vs = 82*N0.39
5. Fujiwara [67] Vs = 92.1N0.337
7. Ohta and Goto [69] Vs = 85.35*N0.348
8. Seed and Idriss [70] Vs = 61*N0.5
9. Imai [71] Vs = 97*N0.31
10. Sykora and Stokoe [72] Vs = 100.5*N0.329
11. Athanasopoulos [73] Vs = 107.6*N0.36
12. Zheng [74] Vs = 116.1*(N + 0.3185)0.202 Fig. 6 Variation of average shear wave velocity along with depth
13. Lee [75] Vs = 57.4*N0.49 considering different empirical correlations
14. Iyisan [76] Vs = 51.5*N0.516
15. Kiku [77] Vs = 68.3*N0.292
velocity, obtained from all 23 different empirical correlations,
16. Yokota et al. [78] Vs = 121*N0.27
has been developed based on the averaging technique. The R2
17. Mhaske and Choudhury [79] Vs = 72*N0.4
value for the average shear wave velocity variation was found
18. Hanumantharao and Ramana Vs = 82.6*N0.43
to be 0.9938.
[80]
19. Kalteziotis et al. [81] Vs = 76.2*N0.24
Stress–strain response model
20. Jafari et al. [82] Vs = 22*N0.85
21. Dikmen [83] Vs = 58*N0.39
To define the stress–strain response during cyclic load-
22. Hasancebi and Ulusay [84] Vs = 90*N0.309
ing–unloading phenomenon of soil mass, a pressure-depend-
23. Maheswari et al. [85] Vs = 95.64*N0.301
ent hyperbolic model has been used. The hyperbolic model,
initially proposed by Kondner and Zelasko [86], modified by
Matasovic [87] has been utilized in this study. This modified
correlations. From Fig. 6, it can be stated that due to unavail- hyperbolic model is also known as modified MKZ (Mataso-
ability of shear wave velocity profile at any particular site, vic–Kondner–Zelasko) expressed in Eq. (6).
the random selection of empirical correlation between VS and
G0 γ
SPT-N can affect significantly the outcome of site-specific 𝜏=
(6)
( )s
seismic ground response analysis. Therefore, to choose the 1+𝛽 γ
γ
suitable correlations for the estimation of shear wave velocity r
in this study, an average empirical correlation of shear wave
13
In Eq. (6), G0, τ, γ and γr indicate the initial shear modu- absence of undrained cyclic test results data, as per Hashash
lus, shear strength, shear strain and reference shear strain, et al. [2].
respectively, whereas β and s are the model parameters used
to adjust the shape of the backbone curve. Further to make Pore Pressure Model
the model pressure-dependent coupling between shear stress
and confining pressure has been allowed in DEEPSOIL [2] For the present study, the excess pore water pressure genera-
by incorporating Eq. (7). tion models such as Matasovic and Vecetic [87] for sand and
( )b Matasovic and Vucetic [88] for clay soil have been adopted.
𝜎v� The variations of excess pore water pressure in sandy soil
γr = Reference shear strain = a × � (7) and in clayey soil are generated using Eqns. (10) and (11),
𝜎ref
respectively.
where a and b are the curve fitting parameters. σ′v is the
p × f × Nc × F × (γc − γtvp )s
effective vertical stress at the middle of the soil layer, and uN = (10)
reference stress (σ′ref) is the confining stress of 0.18 MPa. 1 + f × Nc × F × (γc − γtvp )s
This model is also called as pressure-dependent hyperbolic
model because it is almost linear at small strains which result uN = ANc
−3s (γc −γtvp )r
+ BNc
−2s (γc −γtvp )r −s (γc −γtvp )r
+ CNc +D
in zero hysteretic damping at small strains. Small strain (11)
damping is needed to be computed separately to simulate the
where uN = normalized excess pore pressure (ru), Nc = equiv-
actual behavior of soil. The expression followed by Hashash
alent number of cycles, γ c = cyclic shear strain and
et al. [2] is mentioned in Eq. (8).
γtvp = threshold shear strain in both Eqns. 10 and 11. Further,
(
1
)d p, s, F curve fitting parameters for the sand soil proposed by
𝜉 = Small strain damping ratio = c × (8) Matasovic and Vucetic [87] and f, r, s, A, B, C, D are curve
𝜎v�
fitting parameters for clayey soil proposed by Matasovic and
where c and d are the curve fitting parameters and σ′v is the Vucetic [88]. To estimate these curve fitting parameters,
effective vertical stress at the middle of the soil layer. pore water pressure model needed to be developed based
on the undrained cyclic triaxial test data. All these curve
Pore Water Pressure Generation and Dissipation fitting parameters have been adopted from Hashash et al.
[2], due to the unavailability of laboratory undrained cyclic
The nonlinear methodology incorporates the generation triaxial test data. Further, to predict the nonlinear dynamic
as well as dissipation of excess pore pressure during GRA responses of different soil types, a reduction factor suggested
since it is based on the effective stress conditions, whereas by Darendeli [89] has been considered in this study, which
equivalent linear method is based on the total stress concept is mentioned in Eq. (12). The modulus reduction curves
[1]. Due to strong ground shaking, the rise of excess pore and damping ratio curves can be adjusted simultaneously
pressure in the soil causes loss of strength or excessive set- by considering both modulus reduction and damping factor
tlements or and other damages due to a decrease in effective (MRDF) pressure-dependent hyperbolic model [2]:
stress and shear strength in the soil. However, the genera- )P
(12)
(
tion and dissipation of excess pore pressure, during ground F(γm ) = P1 G(γm )∕G0 2
shaking, depend on the permeability of soil and drainage
where γm represents the maximum shear strain accomplished
conditions. The dissipation of excess pore water pressure is
at any given time, G(γm) represents the shear modulus and
assumed to be in the vertical direction only, since it is based
P1 and P2 represent the curve fitting parameters.
on the Terzaghi 1-D consolidation theory, represented in
Eq. (9).
Selected Strong Ground Motion
( 2 )
𝜕u 𝜕 u
= Cv (9) Three synthetically generated acceleration time histories of
𝜕t 𝜕z2
PGA = 0.1 g, 0.26 g and 0.45 g, corresponding to Mw6.7,
where Cv indicates the vertical coefficient of consolidation. Mw7.8 and Mw = 8.2, have been used for GRA, since Khatri
In the present study, the pore water pressure model (the [90] has reported that the high probability of the occurrence
variations in pore water pressure during dynamic loading) of an EQ of Mw > 8.0, in near future, in the central seismic
proposed by Matasovic and Vecetic [87] and Matasovic and gap. The details about the generation of synthetic accelera-
Vecetic [88] for sandy soil and the clayey soil, respectively, tion time histories are discussed in Kumar et al. [45]. Since
was adopted to determine the curve fitting parameters in the the central seismic gap comes under seismic zone V [38],
13
Nath et al. [91] have also reported that the average value of Figure 7 presents the acceleration histories and their Fou-
PGAs, for Mw > 8, can be 0.18 g, 0.23 g, 0.33 g, 0.50 g and rier amplitude spectrum of input strong motions. Frequency-
0.82 g for low, moderate, moderately high, high and very domain representation indicates the variation of energy con-
high seismic hazards zones, respectively. As per IS: 1893 tent over a frequency band. It is observed that the maximum
[38], the seismic zone V has been defined based on the con- energy contents of strong motion are congregated over a
sideration of effective peak ground acceleration, from the fundamental frequency band of 1.0–5.0 Hz. The ground
most severe earthquake (i.e., maximum credible earthquake, motion characteristics such as predominant period, mean
MCE), and the service life of the structure in that region. period, bracketed duration and significant duration derived
IS: 1893 [38] has also suggested that the design value of from SEISMOSIGNAL program [92] are shown in Table 4.
PGA = 0.36 g and 0.18 g considering maximum considered The mean period of strong ground motions is varying from
earthquake (MCE) and design basis earthquake (DBE, i.e., 0.27 to 0.31 s, which represents the frequency content char-
earthquake which is expected to occur at least once dur- acterization parameter of the ground motions.
ing the design life of the structure) scenarios, respectively,
should be adopted for the structural design. Thus, to observe
the response of soil as well the effect of PGA during GRA, Results and Discussion
three different acceleration time histories of PGA ranging
from PGA = 0.1–0.45 g, of the futuristic EQs in central seis- Nonlinear seismic GRA, a time-domain analysis, incorpo-
mic gap, has been chosen considering the low, moderate and rates the nonlinear strain-dependent dynamic soil properties
high seismic hazards scenario of the study region. such as shear modulus and damping ratio changes during
Fig. 7 Acceleration time history and their Fourier amplitude (FA) of input motion having PGA: a 0.1 g, b 0.26 g and c 0.45 g
13
Table 4 Strong motion parameters for simulated ground motion used in GRA [45]
Parameters Low Moderate High
Magnitude (Mw) 6.7 7.8 8.2

Station Apparent station
Site class A/B A/B A/B
Distance from source 65 60 65
Max. PGA (g) 0.1 g 0.26 g 0.45 g
Predominant period (s) 0.12 0.08 0.14
Mean period (s) 0.278 0.312 0.279
Frequency (Hz) 3.6 3.2 3.6
Maximum velocity 21.62 cm/s at t = 39.5 s 32.45 cm/s at t = 15.4 s 46.32 cm/s at t = 28.36 s
Maximum displacement 60.27 cm at t = 44.3 s 46.88 cm at t = 59.9 s 69.29 cm at t = 63.932 s
Arias intensity (m/s) 1.31 1.94 4.96
Specific energy density ( cm2/s) 3702.58 4288.40 5152.13
Cumulative absolute velocity (cm/s) 1667.4 1719.87 2662.77
vmax/amax (s) 0.124 0.128 0.109
ground shaking. Since the nonlinear analysis incorporates at surface level) was found to be higher than the input bed-
the effective stress concept, the variations in the pore water rock PGA = 0.1 g; whereas, lesser PGA, at surface level, was
pressure within the soil profile can provide the idea about observed for the bedrock PGA of 0.26 g and 0.45 g. Higher
the soil liquefaction. value of acceleration at surface level indicates the amplifica-
tion of seismic waves, whereas lesser value of acceleration
Influence of Input Motion on Seismic GRA at surface level indicates deamplification of seismic waves
shown in Fig. 8b.
Responses of the soil, in terms of acceleration, have been Further, it was noticed that the seismic waves get ampli-
observed for BH-2, situated at Bhagalpur city, Bihar (India), fied within the depth of 5.5 m from the bedrock location;
using bedrock motion of PGA = 0.1 g, 0.26 g and 0.45 g. this may be due to the occurrence of soil softening due to
The surface level PGA was found to be 0.144 g, 0.187 g and increase in pore water pressure. Figures 9, 10 and 11 present
0.241 g corresponding to the input bedrock PGA = 0.1 g, the variation of shear strain along with depth using differ-
0.26 g and 0.45 g, respectively, as shown in Fig. 8a. The ent bedrock motions of PGA = 0.1 g, 0.26 g and 0.45 g at
maximum value of acceleration at surface level (i.e., PGA Bhagalpur BH-2. It has been observed that the maximum
Fig. 8 a Variation of acceleration and b their amplification/deamplification response along with depth using different bedrock motions of
PGA = 0.1 g, 0.26 g and 0.45 g at Bhagalpur BH-2 based on nonlinear analysis
13
Fig. 9 Strain time history at all layers of Bhagalpur site considering bedrock PGA = 0.1 g from nonlinear analysis
value of shear strain was found to be 0.16%, 4.49% and is present. These higher values of shear strain, within the
7.13%, at the depth of nearly 5.5 m (layer 3 in Figs. 9, 10 ground, reflect the existence of soft soil as well as soil sof-
and 11), for bedrock PGA = 0.1 g, 0.26 g and 0.45 g, respec- tening due to rise of pore water pressure.
tively. The high value of shear strain within the ground indi- The high value of shear strain (up to 5–7%, see Figs. 9,
cates that the high PGA values of input bed rock motion 10 and 11) at layer 3 (i.e., nearly at the depth of 5.5 m)
possess high value of shear strain into the soil deposit, due using bedrock PGA of 0.26 g and 0.45 g indicates the sig-
to high amount of soil nonlinearity. As per the soil profile nificant amount of permanent deformation resulting high
of Bhagalpur site, soils mainly consist of clay and sand soil. value of pore water pressure (excess PWP ratio > 0.9, see
The high value of shear strain was observed at the depth Fig. 12 a-f) in sandy soil strata may cause soil liquefaction.
of nearly 5.5 m (i.e., layer 3), where mostly sandy strata Thus, it can be stated that the development of high value of
13
shear strain, at any depth, depends on the soil conditions strain-dependent dynamic soil behavior of the construction
as well as the amplitude of seismic excitations. site prior to the construction.
The development of high shear strains (i.e., greater Figure 12a–f constitutes the variations in excess PWP
than 1%) may cause permanent deformation in the ground, ratio with time at all layers of Bhagalpur site consider-
which might lead to the destruction of the structure and ing Bedrock motion of PGA = 0.1 g, 0.26 g and 0.45 g.
also be responsible for the liquefaction of sandy soil. Simi- The excess pore water pressure ratio (ru) was found to be
lar observations have been reported by Sisodiya et al. [10]. increased with increasing input motion. It was also observed
Since deformation of ground might be responsible for that layer 1 and layer 2 consist of clayey strata, whereas
the catastrophic damage, it is suggested to evaluate the layers 3, 4, 5 and 6 consist of sandy strata. Layer 3 (i.e.,
13
sandy soil), at Bhagalpur site, reflects high value of excess 0.26 g and 0.45 g, respectively. It can also be noticed that
PWP ratio (i.e., > 0.9) with input motion PGA = 0.26 g and the excess PWP ratio, at a depth of 5.5 m, was greater than
0.45 g. The variations in pore water pressure with the soil 0.9, for both bedrock PGA of 0.26 g and 0.45 g in com-
deposits are not only due to the input PGA motion but also parison with the bedrock PGA of 0.1 g. The higher PGA
depend on the, soil type, stiffness of soil, local site geology, input motions (i.e., PGA > 0.26 g) pronounce the onset of
i.e., location of water table, etc. It has been noticed that the liquefaction. Thus, from the results, it can be stated that the
upper layers 1 and 2 (i.e., clayey strata) restrict the rate of generation of excess PWP within the soil mass depends on
generation of PWP; hence, the flow liquefaction is restricted. the soil type, soil conditions and loading conditions. Similar
The excess PWP ratio was found to be 0.608, 0.913 and observations have been reported by Singhai et al. [30]. How-
0.929, at a depth of nearly 5.5 m, for bedrock PGA = 0.1 g, ever, the impact of other strong motion parameters such as
13
Fig. 12 a–f Variations in excess PWP ratio with time at all layers of Bhagalpur site considering bedrock PGA = 0.1 g, 0.26 g and 0.45 g
amplitude parameters and frequency content parameters over deamplified depending on the existing ground conditions.
the generation of excess PWP needed to be studied, which is The computed response on/near to the ground surface dur-
beyond the scope of the present study. ing seismic activity will help the engineers to design the
safe and sustainable soil-supported earthquake-resistant
Influence of Soil Variability on Seismic GRA structures, since the estimated responses on/near to the
surface level will already be associated with the soil prop-
Site-specific seismic ground response analysis (GRA) is erties at that site. However, the uncertainty in the selection
one of the significantly important ways to assess the effect of bedrock input motion and the soil variability across the
of local soil site variability on the bedrock motion during site, as well as the soil properties, will always be there.
an expected earthquake since, the bedrock motion propa- Therefore, to account the uncertainty in the selection of
gated through any site-specific soil deposits gets amplified/ input ground motions, engineers always use a bunch of
13
ground motions (ranging from low PGA to high PGA) as PGA ranging from 0.02 to 0.33 g have been reported by
an input motion. This study has used three input motions Dammala et al. [25].
of PGA = 0.1 g, 0.26 g and 0.45 g to observe the effect of Further, using bedrock PGA = 0.26 g (Fig. 15b), it can be
soil variability during seismic GRA. Figure 13a–c pre- observed that, at surface level, the seismic wave amplifies
sents the variation of peak value of acceleration obtained by 85.65%, 78.40% and 30.79% for BH-1 of Munger, BH-1
at the each soil layers, within the soil deposits, using input of Nalanda and BH-2 of Jamui site, respectively, whereas,
motion of PGA = 0.1 g, 0.26 g and 0.45 g, for all 48 bore- the strong motion gets deamplified by 55.75% and 54.22%
holes situated in Bihar region. It has been observed that for BH-1of Nwada and BH-1 of Madhubani site, respec-
PGA at surface level was found to be 0.277 g for BH-1 at tively. It can be stated that at surface level AF varied from
Munger site, 0.26 g for BH-1 at Nalanda site, 0.089 g for 0.442 to 1.86 (see Table 4), which also indicates that the
BH-1 at Sitamarhi site and 0.0872 g for BH-2 at Vaishali soils (in terms of dynamic soil properties) are one of the
site, using input motion PGA = 0.1 g (see Fig. 13a). Fur- contributory factors for the ground motion amplifications.
ther, on using input bedrock PGA = 0.26 g (Fig. 13b), the The maximum amplification and deamplification of seismic
surface level PGA was found to be 0.48 g for BH-1 at waves, at ground level, can be up to 60% and 90%, respec-
Munger, 0.46 g for BH-1 at Nalanda, 0.34 g for BH-2 at tively. Furthermore, on using input bedrock PGA = 0.45 g
Jamui, 0.119 g for BH-1 at Madhubani and 0.115 for BH-1 (Fig. 15c), it can be observed that the seismic waves get
at Nawada site. However, on using input bedrock motion amplified by 50.76%, 42.62%, 3.65% and 0.762% at sur-
of PGA = 0.45 g (Fig. 13c), the surface level PGA was face level for BH-1 at Munger site, BH-1 at Nalanda site,
found to be 0.68 g at BH-1 of Munger site, 0.64 g at BH-1 BH-1 at Katihar site and BH-1 at Muzaffarpur site, respec-
of Nalanda site, 0.47 g at BH-1 of Katihar site, 0.45 g at tively. The maximum amplification and deamplification of
BH-1 of Muzaffarpur site, 0.16 g at BH-2 Vaishali site seismic waves, at ground level, can be up to 50% and 70%,
and 0.15 g at BH-1 of Madhubani site, which is higher in respectively. In comparison with the AF obtained from input
comparison with the response obtained using input bed- motion PGA of 0.1 g and 0.26 g, it can be stated that the
rock PGA of 0.1 g and 0.26 g. Overall, the range of PGA, higher PGA (0.45 g) bedrock motion reflects relatively lesser
at surface level, was found to be varied from 0.087 g to amplification at the ground surface, which is attributed to the
0.277 g, 0.115 g to 0.482 g and 0.15 g to 0.68 g subjected high damping, indicating high energy dissipation, due to the
to the input bedrock motion of PGA = 0.1 g, 0.26 g and development of high shear strains within the soil deposits.
0.45 g for entire Bihar region, shown in Fig. 13a–c. Fur- Summary of AF thorough out the Bihar region, using input
ther, the variations of surface PGA throughout the Bihar bedrock motion of PGA = 0.1 g, 0.26 g and 0.45 g, is pre-
region, using input bedrock motion of PGA = 0.1 g, 0.26 g sented by contour map in Fig. 16a–c.
and 0.45 g, are presented by contour map in Fig. 14a–c. Figure 17a–c presents the variations of shear strain with
Figure 15a–c presents the seismic wave amplification/ depth, at all 48 boreholes of Bihar region, using input bed-
deamplification at all 48 boreholes of Bihar region using rock PGA of 0.1 g, 0.26 g and 0.45 g. The variations of
bedrock PGA = 0.1 g, 0.26 g and 0.45 g which also reflects shear strain within the soil deposits might be responsible
the importance of seismic GRA. From Fig. 15a, it can be for small to large ground deformation depending on the
observed that, at lower input PGA, most of the site in Bihar soil types, soil conditions and input motion. It has been
region shows amplifying behavior whereas, very few site observed that the variations in the shear strain were found
shows deamplifying behavior. The amplification/deamplifi- to be 0.21% for BH-2 of East Champaran site, 0.15% for
cation of seismic wave is indicated by the amplification fac- BH-2 at Bhagalpur site, 0.15% for BH-1 at Banka site and
tor (AF), which is the ratio of the intensity of observed PGA 0.10% for BH-1 at Madhubani site corresponding to the
at the soil surface to the input bedrock PGA. The amplifica- depth of 7.25 m, 5.25 m, 7.5 m and 10.25 m using bedrock
tion factor (AF) was varying ranging from 0.872 to 2.77 for motion of PGA = 0.1 g (see Fig. 17a). Further, using bed-
PGA of 0.1 g, which shows that the soil for specific site of rock PGA = 0.26 g (Fig. 17b), the maximum shear strain
Bihar region is capable of amplifying earthquake ground was observed to be 4.49% for BH-2 Bhagalpur, 2.47% for
motion. It was observed that, for BH-1 at Munger site, the BH-2 East Champaran, 2.06% for BH-1 Nawada, 1.90% for
percentage amplification of seismic wave at the surface level Sitamarhi and 1.85% for BH-2 Arwal at a depth of 5.25 m,
was found to be 180% whereas, at BH-2 of Vaishali site, 7.25 m, 14.0 m, 24.0 m and 7.5 m, respectively. Further-
the percentage deamplification of seismic wave was found more, using input bedrock PGA = 0.45 g (Fig. 17c), the
to be 13%. The maximum amplification and deamplifica- maximum value of shear strain was observed to be 7.13%
tion of seismic waves, at ground level, may be up to 180% for BH-2 Bhagalpur, 6.87% for BH-1 Sitamarhi site, 6.56%
and 10%, respectively (see Fig. 15a–c). Similar results of for BH-1 of Arwal site, 4.45% for BH-1 of East Champaran
amplification (up to 100%) as well as deamplification (up to site and 4.14% for BH-1 of Banka site at a depth of 5.25 m,
54%) of seismic waves considering different input motions 18.5 m, 4.0 m, 7.0 m and 7.5 m, respectively. The shear
13
Fig. 13 a–c Variations in

acceleration with depth at 48
boreholes using different input
bedrock motions from Nonlin-
ear analysis
13
Fig. 14 Contour map of surface

PGA of Bihar region using
different input bedrock motions:
a PGA = 0.1 g b PGA = 0.26 g
and c PGA = 0.45 g from non-
linear analysis
13
Fig. 15 a–c Amplification and

deamplification of seismic wave
at 48 boreholes using different
input bedrock motions from
nonlinear analysis
13
Fig. 16 Contour map of

amplification factor at surface
level of Bihar region using dif-
ferent input bedrock motions:
a PGA = 0.1 g, b PGA = 0.26 g
linear analysis
13
Fig. 17 a–c Variation in shear

strain along with depth at 48
bedrock motions from nonlinear
analysis
13
strain greater than 0.5% might be responsible for the moder- Sitamarhi, 1.00 at BH-2 of Muzaffarpur, 0.628 at BH-1 of
ate to large deformation of the ground during ground shak- Arwal, 0.624 at BH-1 of Nalanda and 0.574 at BH-2 of
ing. Similar observations have been reported by Kumar et al. Sitamarhi sites, subjected to input motion of PGA = 0.26 g
[60]. (see Fig. 19b). Further, using input bedrock PGA = 0.45 g
Figure 18a–c depicts the variations of the excess pore (Fig. 19c), the stress ratio was found to be 1.20 at BH-2 of
water pressure (PWP) ratio (ru) along the depth, for all 48 Muzaffarpur site, 1.16 at BH-1 of Sitamarhi site, 0.91 at
boreholes of Bihar region, using input bedrock motion of BH-1 of Muzaffarpur site, 0.91 at BH-1 of Arwal site, 0.84
PGA = 0.1 g, 0.26 g and 0.45 g. The increase in the excess at BH-1 of Nalanda site and 0.675 at BH-1 of Munger site.
PWP ratio (ru) at any depth indicates the soil softening as The stress ratio is greater than 1, which indicates that there is
well as soil liquefaction depending on the variation of r u a possibility of liquefaction even though the sandy strata are
from 0 to 1. From Fig. 18a, it can be observed that at BH-2 at deeper depth. Moreover, it can be stated that the variations
of Bhagalpur site, at BH-1 of Banka site, at BH-1 of Arwal in the stress ratio within the soil deposits depend on the soil
site and at BH-1 of Siwan site, the maximum value ru was type, soil conditions and input bedrock PGA.
found to be 0.61, 0.55, 0.32 and 0.21 at a depth of 5.25 m,
7.5 m, 4.0 m and 5.5 m, respectively, using bedrock motion Influence of Soil Variability of Response Spectrum
of PGA = 0.1 g. Summary of variations in ru at all 48 bore-
holes using bedrock motion of PGA = 0.1 g is presented in The design acceleration response spectrum is very useful
Table 5. tools in earthquake engineering to analyze the performance
Further, using bedrock motion of PGA = 0.26 g of soil-supported structures, which is conceptually differ-
(Fig. 18b), ru was found to be 0.99, 0.91, 0.90, 0.89, 0.84 and ent than the acceleration spectrum. Response spectrum is
0.81 at the depth of 7.5 m, 5.25 m, 4.0 m, 7.5 m, 5.5 m and the representation of maximum responses of a spectrum
4.5 m, respectively, for the corresponding boreholes BH-2 of SDOF with varying natural periods for the given set of
Arwal, BH-2 of Bhagalpur, BH-1 of Arwal, BH-1 of Banka, ground motions at the same damping ratio [38], which is the
BH-1 of Siwan and BH-1 of Madhepura. The summary of basis of the development of design response spectrum. The
variations in ru at all 48 boreholes using bedrock motion of design acceleration spectrum refers to an average smooth-
PGA = 0.26 g, as an input motion, is presented in Table 5. ened graph of the maximum acceleration as a function of nat-
Furthermore, using input bedrock motion of PGA = 0.45 g ural frequency or natural period of oscillation for a specified
(Fig. 18c), the ru was observed to be 0.99, 0.93, 0.92, 0.91, damping ratio (i.e., 5% of critical damping) for the expected
0.90 and 0.830 at a depth 7.5 m, 5.25 m, 4.0 m, 7.5 m, 9.0 m earthquake excitations at the base of SDOF. Nowadays, the
and 4.5 m for the boreholes BH-2 of Arwal site, BH-2 of design acceleration response spectrum is prominently being
Bhagalpur site, BH-1 of Arwal site, BH-1 of Banka site, used by the engineers worldwide for the structural design in
BH-1 of Madhepura site and BH-1 of Siwan site, respec- the seismic prone regions. The maximum response of the
tively. The summary of r u at different depths for all 48 structures can be estimated by reading the value from the
boreholes using bedrock motion of PGA = 0.45 g is also ground response spectrum plot for an appropriate frequency.
presented in Table 5. Thus, it can be stated that the soil The structural response of low-rise buildings during earth-
liquefaction depends on the soil conditions as well as mag- quakes is characterized by the fundamental mode; however,
nitude of seismic excitations. The higher value of ru in sandy most of the building codes permit design forces to be calcu-
strata might be responsible for liquefaction, whereas that in lated from the design spectrum on the basis of building fre-
clayey strata might indicate the cyclic mobility. The r u > 0.9 quency. Figure 20a–c presents the spectral acceleration (for
indicates high probability of occurrence of soil liquefaction 5% damping ratio) at surface level, for 48 boreholes of Bihar
in sand soil deposits. region using input bedrock PGA = 0.1 g, 0.26 g and 0.45 g,
Figure 19a–c presents the distribution of shear stress ratio along with the design acceleration response spectrum pro-
along with depth, at all 48 borehole of Bihar region, using posed by IS: 1983 [38] for different soil types. The response
bedrock PGA = 0.1 g, 0.26 g and 0.45 g. The shear stress spectrum, for 5% damping, at the surface level using bed-
ratio (i.e., cyclic shear stress ratio) is a useful way to predict rock PGA = 0.1 g (Fig. 20a), indicates that the peak spectral
the liquefaction capability of stratified soil in the field. From acceleration (PSA) of BH-1 Munger at the surface is 1.34 g
Fig. 19a, the stress ratio was found to be 0.41 at BH-1 of corresponding to the period of 0.10 s. Similarly, PSA of
Arwal site, 0.33 at BH-1 of East Champaran site, 0.36 at BH-1 at Nalanda, BH-1 at Bhojpur, BH-2 at Jamui, BH-1 at
BH-2 of East Champaran site and 0.35 at BH-1 of Nalanda Muzaffarpur, BH-1 at Jahanabad, BH-1 at Arwal and BH-1
site, which is lesser than 1. Therefore, it can be stated that at Madhepura was found to be 1.33 g, 1.13 g, 1.06 g, 0.99 g,
there is no chance of liquefaction at any site of Bihar region, 0.96 g, 0.92 g and 0.91 g, respectively, for the corresponding
when subject to an earthquake of input motion PGA = 0.1 g. spectral period of 0.30 s, 0.28 s, 0.30 s, 0.14 s, 0.30 s, 0.36 s
However, the stress ratio was found to be 1.04 at BH-1 of and 0.14 s, respectively.
13
Fig. 18 a–c Variation in excess

PWP ratio along with depth at
48 boreholes using different
nonlinear analysis
13
Table 5 Summary of maximum pore water pressure at all 48 boreholes using bedrock motions of 0.1 g, 0.26 g and 0.45 g
Borehole Maximum pore water pressure corresponding to input bedrock motion Borehole Maximum pore water pressure corresponding to input bedrock
of motion of
0.1 g 0.26 g 0.45 g 0.1 g 0.26 g 0.45 g
ru Depth (m) ru Depth (m) ru Depth (m) ru Depth (m) ru Depth (m) ru Depth (m)
Bh-1 Arariya 0.068 15.25 0.55 15.25 0.68 15.25 Bh-1 Madhubani 0.085 12.75 0.74 28.25 0.76 28.25
Bh-2 Arariya 0.011 15.25 0.46 15.25 0.46 15.25 Bh-2 Madhubani 0.047 24 0.60 24 0.74 24
Bh-1 Arwal 0.33 4 0.90 4 0.92 4 Bh-1 Munger – – 0 0.002 17
Bh-2 Arwal 0.055 15.25 0.99 7.5 0.99 7.5 Bh-2 Munger 0.035 15.5 0.54 15.5 0.67 15.5
Bh-1 Aurangabad – – – – 0.01 15.25 Bh-1 Muzaffarpur 0.034 18.75 0.62 18.75 0.82 10.5
Bh-2 Aurangabad – – 0.004 16.6 0.05 16.6 Bh-2 Muzaffarpur 0.022 21 0.81 24.75 0.84 24.75
Bh-1 Banka 0.55 7.5 0.89 7.5 0.91 7.5 Bh-1 Nalanda 0.006 3.3 0.034 3.3 0.65 16
Bh-1 Bhagapur – – – – 0.012 15.25 Bh-2 Nalanda 0.056 16 0.53 16 0.06 3.3
Bh-2 Bhagapur 0.61 5.25 0.91 5.25 0.93 5.25 Bh-1 Nawada 0.17 32 0.71 32 0.74 32
Bh-1 Bhojpur – – 0 9.25 0.02 6 Bh-2 Nawada 0.013 16 0.52 16 0.48 16
Bh-2 Bhojpur – – 0.003 16 0.05 16 Bh-1 Patna – – 0.012 23.5 0.26 20.5
Bh-1 East Champaran 0.001 10.25 0.31 10.25 0.65 13 Bh-2 Patna – – 0.015 9.15 0.06 9.15
Bh-2 East Champaran 0.008 13.25 0.48 7.25 0.66 13.25 Bh-1 Purnia 0.042 5.15 0.68 5.15 0.76 5.15
Bh-1 Gaya 0.026 15.75 0.53 15.75 0.69 9.75 Bh-2 Purnia 0.031 6.9 0.70 3.7 0.81 3.7
Bh-2 Gaya 0.0005 8.5 0.35 8.5 0.44 8.5 Bh-1 Sitamarhi 0.11 24 0.73 24 0.79 12.5
Bh-1 Jamui – – – – – – Bh-2 Sitamarhi 0.076 6.5 0.74 27 0.84 27
Bh-2 Jamui – – – – 0.004 16.25 Bh-1 West Champaran 0.056 13.37 0.68 13.37 0.71 13.37
Bh-1 Jehanabad – – – – 0.008 16.5 Bh-2 West Champaran 0.16 14.25 0.79 14.25 0.81 14.25
Bh-2 Jehanabad – – 0.004 17 0.025 17 Bh-1 Siwan 0.21 5.5 0.84 5.5 0.83 9
Bh-1 Katihar 0.038 3.77 0.58 3.77 0.73 3.77 Bh-2 Siwan 0.076 6.5 0.56 6.5 0.69 17.75
Bh-1 Lakhisarai 0.032 14.75 0.49 5.75 0.64 5.75 Bh-1 Vaishali – – 0.006 15.25 0.04 15.25
Bh-2 Lakhisarai 0.014 15.75 0.49 15.75 0.56 15.75 Bh-2 Vaishali – – 0.016 15.25 0.11 15.25
Bh-1 Madhepura 0.10 4.5 0.80 4.5 0.90 4.5 Bh-1 Rohtas – – 0.005 15.25 0.04 15.25
Bh-2 Madhepura 0.060 9 0.58 9 0.78 9 Bh-2 Rohtas – – 0.007 15.25 0.04 15.25
13
385
Fig. 19 a–c Variation in stress

ratio along with depth at 48
bedrock motions from nonlinear
analysis
13
Fig. 20 a–c Spectral accelera-

tion, for 5% damping ratio, near
the surface level at 48 boreholes
of Bihar region using different
nonlinear analysis and design
response spectrum proposed by
IS: 1983 [38] for different soil
types
13
However, the response spectrum for 5% damping at sur- and SPT-N value. All these empirical correlations are
face level, using bedrock PGA = 0.26 g (Fig. 20b), shows the not established for Indian region. Thus, it can be stated
PSA = 2.23 g corresponding to the period of 0.14 s at BH-1 that a regional empirical correlation between shear wave
of Munger site. Similarly, the PSA at BH-1 of Nalanda, velocity and SPT-N value needed to be developed.
BH-2 of Jahanabad, BH-2 of Jamui, BH-1 of Jahanabad, • The effect of earthquake frequency, predominant period,
BH-1 of Bhojpur, BH-1 of Muzaffarpur and BH-2 of Purnia arias intensity, bracketed duration and other earthquake-
was found to be 2.12 g, 1.95 g, 1.6 3 g, 1.56 g, 1.48 g, 1.25 g associated parameters on the amplification or deampli-
and 1.20 g, corresponding to the period of 0.32 s, 0.44 s, fication of seismic wave is beyond the scope of present
0.36 s, 0.36 s, 0.34 s, 0.18 s, 0.36 s and 0.50 s, respectively. study.
Further, using bedrock PGA = 0.45 g (Fig. 20c), the PSA • Therefore, it is recommended to evaluate the site-specific
of BH-1 Munger at the surface was found to be 3.23 g cor- strain-dependent dynamic soil properties, such as vari-
responding to the period of 0.13 s. Similarly, the PSA of ations of shear modulus and damping ratio with shear
other locations of Bihar such as BH-1 of Nalanda site, BH-2 strain, at the preliminary stage of the construction.
of Jehanabad site, BH-2 of Purnia site, BH-2 of Jamui site,
BH-1 of Bhojpur site and BH-1 of Madhepura site was found
to be 2.21 g, 2.12 g, 2.09 g, 1.92 g, 1.72 g and 1.49 g cor- Conclusions
responding to the spectral period of 0.36 s, 0.56 s, 0.56 s,
0.44 s, 0.56 s and 0.13 s. Further, PSA at surface level for Based on the borehole profiles and SPT-N values, collected
all 48 boreholes, using bedrock motion of PGA = 0.1 g, from all 48 sites within Bihar region, it was observed that
0.26 g and 0.45 g. Overall, from Fig. 20, it can be observed the entire Bihar region comes under the range of site classes
that the free-field spectral response, using input bedrock D and E with the variations of Vs30 ranging from 180 to
PGA = 0.1 g, at all 48 boreholes of Bihar region is lesser 360 m/s. Further, nonlinear seismic ground responses have
than the design response spectrum proposed by IS: 1983 been performed by incorporating the pore water pressure
[38] for soft as well as hard soil types. During the analysis, generation and dissipation for the site classes D and E of
it was found that the free-field spectral acceleration at all 48 Bihar, India, using three acceleration time histories of dif-
boreholes is not same either by input bedrock PGA = 0.1 g ferent peak ground acceleration (PGA) of 0.1 g, 0.26 g and
or, 0.26 g or, 0.45 g, which also indicates the response of 0.45 g indicating low, moderately high and very high seismic
input motion significantly affected by the regional soil vari- hazard scenarios, respectively. Based on the results obtained
ability. From Fig. 20a–c, it can be noticed that the intensity from seismic GRA in terms of the variations of acceleration,
of free-field spectral acceleration is higher for periods rang- amplification/ deamplification of seismic waves, shear strain,
ing from 0.05 to 0.8 s. Further, it can be noticed that the shear stress ratio and pore water pressure along with the
free-field spectral response increases with increasing inten- depth, the following conclusions as well as recommenda-
sity of input bedrock PGA, i.e., from 0.1 to 0.45 g. This tions have been drawn:
is attributed to the fact that the higher intensity of ground
shaking induces high shear strain within the soil deposits. • The variations in PGA, at surface level, were found to
The variations of PSA thorough out the Bihar region, using be in the range of 0.087 g-0.277 g, 0.115 g-0.482 g and
input bedrock motion of PGA = 0.1 g, 0.26 g and 0.45 g, are 0.15 g-0.68 g subjected to the input bedrock motion of
presented by contour map in Fig. 21a–c. PGA = 0.1 g, 0.26 g and 0.45 g for entire Bihar region,
which is the indication of amplification or deamplifi-
cation of seismic wave. The amplification of seismic
Limitations wave (i.e., amplification factor) was found to be in the
range 0.35–3.0, which further indicates the seismic
The 1-D nonlinear GRA has been performed, for Bihar wave amplifies at surface level by 180%but deamplifies
region, using DEEPSOIL7.0 based on some limitations: by 90%. Thus, it can be stated that the seismic GRA is
significantly affected by input motion, local site geology
• Strain-dependent dynamic soil properties such as modu- and the soil variability.
lus reduction and damping ratio curve proposed by Seed • The maximum amplification and deamplification of
and Idriss [62] and Vucetic and Dobry [63] for sandy soil seismic waves, at ground level, was found in the order
and clayey soil, respectively, have been used due to the of 180% and 10%, respectively, using input motion of
absence of regional dynamic soil properties data. PGA = 0.1 g, whereas the same was found to be 60%
• Shear wave velocities corresponding to SPT-N value, and 90%, and 50% and 70% with the input motion of
used in this study, are estimated based on the average of PGA = 0.26 g and 0.45 g, respectively. Overall, it can
23 empirical correlations between shear wave velocity be stated that there is ample chance of amplification of
13
Fig. 21 Contour map of peak

spectral acceleration at surface
level of Bihar region using dif-
ferent input bedrock motions:
a PGA = 0.1 g, b PGA = 0.26 g
linear analysis
13
seismic waves at Bihar region. Hence, it is recommended Funding Not applicable.

that GRA needs to be performed, prior to the construc-
tion at any site, to observe the potential amplification of Availability of data and materials The data of this study are avail-
able from the corresponding author upon reasonable request.
seismic waves. Further, this study can be utilized for the
development of surface level ground motion attenuation Declarations
relationship for the Bihar region.
• The shear strains, at many sites, found to be greater than Conflict of interest The authors declare that they have no known
0.5%, with in the soil deposits, might be responsible for competing financial interests or personal relationships that could have
the moderate to large deformation of the ground during appeared to influence the work reported in this paper.
ground shaking, which is similar to the results proposed Ethical Approval Not applicable.
by Kumar et al. [60].
• During nonlinear seismic GRA, it was found that most
of the soil sites in Bihar region are prone to liquefaction
since excess pore water pressure ratio (ru) is greater than References
0.9; however, it was also observed that the soil liquefac-
1. Kramer SL (1996) Geotechnical earthquake engineering. Prentice
tion depends on the soil conditions as well as magnitude Hall, New York
of seismic excitations. The higher value of ru in sandy 2. Hashash YMA, Musgrove MI, Harmon JA, Ilhan O, Xing G,
strata might be responsible for liquefaction, whereas Numanoglu O, Groholski DR, Phillips CA, Park D (2020) DEEP-
that in clayey strata might indicate the cyclic mobility. SOIL 7.0, User Manual. Urbana, IL, Board of Trustees of Univer-
sity of Illinois at Urbana-Champaign
Though this study can be helpful for the engineers to 3. Building Seismic Safety Council (BSSC 2003) NEHRP recom-
identify the site of soils liquefaction, more experimental mended provisions for seismic regulations for new buildings and
studies are needed to be carried for detailed understand- other structures. Report FEMA-450 (Provisions), Federal Emer-
ing about the liquefaction characteristics’ of soils of gency Management Agency (FEMA), Washington
4. Kumar S, Muley P, Madani SN (2022) Ground response analysis
Bihar region. and liquefaction for Kalyani region, Kolkata. Environ Sci Pollut
• Stress ratio, one of the indices to identify the soil lique- Res, pp 1–20
faction, was found to be greater than 0.5 within the soil 5. Ansari A, Zahoor F, Rao KS, Jain AK (2023) Seismic response
deposits at many at locations, which further indicates and vulnerability evaluation of Jammu Region (Jammu and Kash-
mir). Indian Geotech J 53(3):509–522
that there is a possibility of liquefaction even though the 6. Silahtar A (2022) Evaluation of local soil conditions with 1D non-
sandy strata are at deeper depth. Further, based nonlin- linear site response analysis of Arifiye (Sakarya District), Turkey.
ear seismic GRA subjected to the input bedrock motion Nat Hazards, pp 1–25
of PGA = 0.1 g, 0.26 g and 0.45 g, it can also be stated 7. Kawan CK, Maskey PM, Motra G (2022) A Study of local soil
effect on the earthquake ground motion in Bhaktapur city, Nepal
the variations in the stress ratio within the soil depos- using equivalent linear and non-linear analysis. Iran J Scien Techn
its depends on the soil type, soil conditions and input Trans Civil Eng. https://doi.org/10.1007/s40996-022-00858-1
motion/ PGA. 8. Kumar SS, Dey A, Krishna AM (2014) Equivalent linear and
• During the analysis, it was found that the free-field spec- nonlinear ground response analysis of two typical sites at Guwa-
hati city. In: Proceedings of Indian geotechnical conference (IGC-
tral acceleration for 5% damping, at all 48 boreholes, 2013), Kakinada, India
is not same for input bedrock PGA = 0.1 g or 0.26 g or 9. Kumar SS, Krishna AM, Anbazhagan P (2018) Study on the vari-
0.45 g, which indicates that the response of input motion ations of ground motion parameters with distance for Mw 6.9 Sik-
significantly affected by the regional soil variability. It kim 2011 earthquake. ISET J Earthq Technol 55:33–46
10. Sisodiya SK, Kumar P, Kumar SS (2023) Site-specific GRA to
can be noticed that the intensity of free-field spectral quantify ground motion amplification for Bettiah site: a case study.
acceleration is higher for periods ranging from 0.05 to Lect Notes Civil Eng 300(2001):453–466. https://d oi.o rg/1 0.1 007/
0.8 s. Further, it can be noticed that the free-field spectral 978-981-19-6998-0_39
response increases with increasing intensity of input bed- 11. Satyam ND, Towhata I (2016) Site-specific ground response
analysis and liquefaction assessment of Vijayawada city (India).
rock PGA i.e., from 0.1 to 0.45 g. This is attributed to the Nat Hazards 81(2):705–724
fact that the higher intensity of ground shaking induces 12. Chatterjee K, Choudhury D (2018) Influences of local soil condi-
high shear strain within the soil deposits. tions for ground response in Kolkata city during earthquakes. In:
Proceedings of the national academy of sciences, India Section
A: physical sciences. 88(4):515–528
13. Rathje EM, Kottke AR, Trent WL (2010) Influence of input
Author Contributions KK was involved in conceptualization, formal motion and site property variabilities on seismic site response
analysis and writing—original draft preparation; PK was involved in analysis. J Geotech Geoenviron Eng 136(4):607–619
formal analysis and development of synthetic ground motion; and SSK 14. Reddy MM, Hanumantha Rao C, Reddy KR, Kumar GK (2021)
was involved in conceptualization, formal analysis, writing—original Site-specific ground response analysis of some typical sites in
draft preparation, writing—review and editing and supervision.
13
Amaravati region Andhra Pradesh, India. Indian Geotech J 34. Shiuly A, Sahu RB, Mandal S, Roy N (2018) Local site
52(1):39–54. https://doi.org/10.1007/s40098-021-00562-3 effect due to past earthquakes in Kolkata. J Geol Soc India
15. Chandran D, Anbazhagan P (2020) 2D nonlinear site response 91(4):400–410
analysis of typical stiff and soft soil sites at shallow bedrock 35. Pallav K, Raghukanth STG, Singh KD (2010) Surface level
region with low to medium seismicity. J Appl Geophy 179:104087 ground motion estimation for 1869 Cachar earthquake (Mw 7.5)
16. Nguyen VQ, Aaqib M, Nguyen DD, Luat NV, Park D (2020) A at Imphal city. J Geo Phys Eng 7(3):321–331
site-specific response analysis: a case study in Hanoi, Vietnam. 36. Thaker TP, Rao KS, Gupta, KK (2009) Ground response and
Appl Sci 10(11):3972. https://doi.org/10.3390/app10113972 site amplification studies for coastal soil, Kutch, Gujarat: a case
17. Basu D, Dey A, Kumar SS (2017) One-dimensional effective study. In: Proceeding of International Conference at BITS Pilani
stress non-masing nonlinear ground response analysis of IIT (ACSGE), India
Guwahati. Int J Geotech Earthq Eng 8(1):1–27. https://doi.org/ 37. Ranjan R (2005) Seismic response analysis of Dehradun City,
10.4018/IJGEE.2017010101 India. M.Sc Thesis, International Institute for Geo-Information
18. Pawirodikromo W (2022) Ground Motions, site amplification and Science and Earth Observations–Enschede. Netherlands, p 92
building damage at near source of the 2006 Yogyakarta, Indonesia 38. IS 1893 Part-1 (2016) Criteria for earthquake resistant design
earthquake. Geotech Geolog Eng 40:1–18 of structures-general provision and buildings (Sixth Revision).
19. Dammala PK, Krishna AM (2022) Nonlinear seismic ground Bureau of Indian Standards, New Delhi
response analysis in Northeastern India considering the compre- 39. Dewey JF, Bird JM (1970) Mountain belts and the new global
hensive dynamic soil behavior. Indian Geotech J 52:650–674. tectonics. J Geophys Res 75:2625–2647. https://doi.org/10.1029/
https://doi.org/10.1007/s40098-022-00598-z JB075i014p02625
20. Bhusal B, Aaqib M, Paudel S, Parajuli HR (2022) Site specific 40. Dasgupta S, Mukhopadhyay M, Nandy DR (1987) Active trans-
seismic hazard analysis of monumental site Dharahara, Kath- verse features in the central portion of the Himalaya. Tectonophys-
mandu, Nepal. Geomat Nat Hazards Risk 13(1):2674–2696 ics 136:255–264. https://doi.org/10.1016/0040-1951(87)90028-X
21. Yildiz Ö (2022) Seismic site characterization of Battalgazi in 41. Dasgupta S, Narula PL, Acharyya SK, Banerjee J (2000) Geologi-
Malatya, Turkey. Arab J Geosci 15(9):1–17 cal Survey of India. Seismotectonic atlas of India and its environs,
22. Mase LZ (2022) Local seismic hazard map based on the response New Delhi
spectra of stiff and very dense soils in Bengkulu city, Indonesia. 42. Burnwal ML, Burman A, Samui P, Maity D (2017) Deterministic
Geod Geodyn 13(6):573–584 strong ground motion study for the Sitamarhi area near Bihar–
23. Yıldız Ö (2021) Nonlinear and equivalent linear site response Nepal region. Nat Hazards 87:237–254. https://doi.org/10.1007/
analysis of Istanbul soils. NATURENGS MTU J Eng Nat Sci Ozal s11069-017-2761-2
Univ 1:88–101. https://doi.org/10.46572/naturengs.895283 43. Mridula S, Sinvhal A, Wason HR, Rajput SS (2016) Segmentation
24. Puri N, Jain A (2020) Estimation of Local site effect on earthquake of main boundary thrust and main central thrust in Western Hima-
ground motions for sites in the state of Haryana, India. In: Emerg- laya for assessment of seismic hazard. Nat hazards 84:383–403
ing trends in civil engineering. Springer, Singapore, pp 101–110 44. Ghosh T, Mukhopadhyay A (2012) Schematic natural hazard
25. Dammala PK, Kumar SS, Krishna AM, Bhattacharya S (2019) zonation of Biharusing geoinformatics. International Growth
Dynamic soil properties and liquefaction potential of northeast Centre, London, p 82
Indian soil for non-linear effective stress analysis. Bull Earthq 45. Kumar P, Kumar SS, Harinarayan, HN (2023) Development of
Eng 17:2899–2933. https://doi.org/10.1007/s10518-019-00592-6 synthetic ground motion-based attenuation relationship for bihar
26. Putti SP, Devarakonda NS, Towhata I (2019) Estimation of ground region for seismic ground response analysis considering central
response and local site effects for Vishakhapatnam, India. Nat seismic gap. Manuscript accepted for publications in Annals of
Hazards 97(2):555–578 Geophysics
27. Sil A, Haloi J (2018) Site-specific ground response analysis of 46. Banghar AR (1991) Mechanism solution of Nepal–Bihar eartb-
a proposed bridge site over Barak River along Silchar bypass quake of August 20, 1988. J Geol Soc India 1–78(37):25–30
road, India. Innov Infrastruct Solut 3:63. https://doi.org/10.1007/ 47. Verma AK, Pati P, Sharma V (2017) Soft sediment deformation
s41062-018-0167-y associated with the East Patna Fault south of the Ganga River,
28. Ahmad S, Bhattacharjee A (2017) Seismic ground response analy- northern India: Influence of the Himalayan tectonics on the south-
sis and pore pressure evaluation at selected locations of Jorhat ern Ganga plain. J Asian Earth Sci 143:109–121
city. In: The proceedings of Indian geotechnical conference, 48. Gansser A (1974) Himalaya geology. Soc Lond Spec Publ
GeoNEst, 14–16 December 2017, IIT Guwahati, India 4:267–278
29. Pandey B, Jakka RS, Kumar A (2016) Influence of local site con- 49. Valdiya KS (1976) Himalayan transverse faults and folds and their
ditions on strong ground motion characteristics at Tarai region of parallelism with subsurface structures of North Indian plains. Tec-
Uttarakhand, India. Nat Hazards 81(2):1073–1089 tonophysics 32:353–386. https://doi.org/10.1016/0040-1951(76)
30. Singhai A, Kumar SS, Dey A (2016) Site-specific 1-D nonlin- 90069-X
ear effective stress GRA with pore water pressure dissipation. 50. Choudhury SK (1975) Gravity and crustal thickness in the Indo-
In: Proceeding of 6th International conference on recent advance Gangetic plains and Himalayan region, India. Geophys J R Astron
in geotechnical earthquake engineering and soil dynamic, New Soc 40:441–452. https://d oi.o rg/1 0.1 111/j.1 365-2 46X.1 975.t b041
Delhi, pp 1–11 41.x
31. Desai SS, Choudhury D (2015) Site-specific seismic ground 51. Joshi DD, Bhartiya SP (1991) Geomorphic history and
response study for nuclear power plants and ports in Mumbai. lithostratigraphy of a part of Eastern Gangetic plain, Uttar
Nat Hazards Rev 16(4):1–13 Pradesh. J Geol Soc India 1–78(37):569–576
32. Akhila M, Ghosh C, Satyam DN (2012) Detailed ground response 52. Sinha R, Friend PF, Switsur VR (1996) Radiocarbon dating and
analysis at park hotel in Kolkata city India. 15WEEE, Lisboa sedimentation rates in the Holocene alluvial sediments of the
33. Naik N, Choudhury D (2013) Site specific ground response northern Bihar plains, India. Geol Mag 133:85–90. https://doi.
analysis for typical sites in Panjim city, Goa. In Proceedings of org/10.1017/S0016756800007263
Indian Geotech Conference, Roorkee, India
13
53. Chandra S (1993) Fluvial landforms and sediments in the north- 74. Zheng J (1987) Correlation between seismic wave velocity and
central Gangetic Plain, India. Darwin College, University of the number of blow of spt and depths. In: Proceedings of Chi-
Cambridge, Cambridge nese journal of geotechnical engineering—1985, China. ASCE,
54. Quittmeyer RC, Jacob KH (1979) Historical and modern seis- pp 92–100
micity of Pakistan, Afghanistan, northwestern India, and south- 75. Lee SHH (1990) Regression models of shear wave velocities in
eastern Iran. Bull Seismol Soc Am 69:773–823 Taipei basin. J Chin Inst Eng 13(5):19–532. https://doi.org/10.
55. Sinha R, Tandon SK, Bhattarcharjee PS, Dasgupta AS (2005) 1080/02533839.1990.9677284
Late Quaternary geology and alluvial stratigraphy of the Ganga 76. Iyisan R (1996) Correlations between shear wave velocity and
basin. Himal Geology 26:223–240 in-situ penetration test results. Teknik Dergi-tmmob Insaat
56. Harinarayan NH, Kumar A (2020) Ground motion prediction Muhendisleri Odasi 7:371–374
equation for north India, applicable for different site classes. 77. Kiku H (2001) In-situ penetration tests and soil profiling in
Soil Dyn Earthq Eng. https://doi.org/10.1016/j.soildyn.2020. Adapazari, Turkey. In: Proceedings of the 15th ICSMGE TC4
106425 satellite conference on lessons learned from recent strong earth-
57. Nandy DR (2007) Need for seismic microzonation of Kolk- quakes, August 25, 2001, Istanbul, Turkey, pp 259–265
ata megacity. In: Proceedings of workshop on microzonation. 78. Yokota K, Imai T, Konno M (1981) Dynamic deformation char-
Indian Institute of science, Bangalore, India, p 2627 acteristics of soils determined by laboratory tests. OYO Tec Rep
58. Kumar SS, Krishna AM (2013) Seismic ground response analy- 3:13–37
sis of some typical sites of Guwahati City. Int J Geotech Earthq 79. Mhaske SY, Choudhury D (2010) Gis-based soil liquefaction
Eng 4:83–101 susceptibility map of Mumbai city for earthquake events. J
59. Ishihara K (1996) Soil behaviour in earthquake geotechnics. Appl Geophys 70(3):216–225. https://doi.org/10.1016/j.jappg
Clarendon Press, London eo.2010.01.001
60. Kumar SS, Dey A, Krishna AM (2018) Response of saturated 80. Hanumantharao C, Ramana G (2008) Dynamic soil properties
cohesionless soil subjected to irregular seismic excitations. Nat for microzonation of Delhi, India. J Earth Syst Sci 117(2):719–
Hazards 93(1):509–529 730. https://doi.org/10.1007/s12040-008-0066-2
61. Kumar SS, Krishna AM, Dey A (2018) High strain dynamic 81. Kalteziotis N, Sabatakakis N, Vassiliou J (1992) Evaluation of
properties of perfectly dry and saturated cohesionless soil. dynamic characteristics of Greek soil formations. In: Second
Indian Geotech J 48:549–557. https:// d oi. o rg/ 1 0. 1 007/ Hellenic conference on geotechnical engineering 2:239–246
s40098-017-0255-5 82. Jafari MK, Shafiee A, Razmkhah A (2002) Dynamic properties
62. Seed HB and Idriss IM (1970) Soil moduli and damping factors of fine grained soils in south of Tehran. J Seismo Earthq Eng
for dynamic response analysis. Reoprt, pp EERC-70 4(1):25
63. Vucetic M, Dobry R (1991) Effect of soil plasticity on cyclic 83. Dikmen Ü (2009) Statistical correlations of shear wave veloc-
response. J Geotech Eng 117:89–107 ity and penetration resistance for soils. J Geophy Eng 6(1):61.
64. Imai T, Yoshimura Y (1970) Elastic wave velocity and soil prop- https://doi.org/10.1088/1742-2132/6/1/007
erties in soft soil. Tsuchito-Kiso 18(1):17–22 84. Hasancebi N, Ulusay R (2007) Empirical correlations between
65. Ohba S, Toriumi I (1970) Dynamic response characteristics of shear wave velocity and penetration resistance for ground shak-
Osaka plain. In: Proceedings of the annual meeting AIJ, Wash- ing assessments. Bull Eng Geol Environ 66(2):203–213
ington, D.C. (in Japanese) 85. Maheswari RU, Boominathan A, Dodagoudar G (2010) Use of
66. Ohsaki Y, Iwasaki R (1973) On dynamic shear moduli and pois- surface waves in statistical correlations of shear wave velocity
son’s ratios of soil deposits. Soils Found 13(4):61–73. https:// and penetration resistance of Chennai soils. Geotech Geolog
doi.org/10.3208/sandf1972.13.4_61 Eng 28(2):119–137. https://doi.org/10.1007/s10706-009-9285-9
67. Fujiwara T (1972) Estimation of ground movements in actual 86. Kondner RL, Zelasko JS (1963) A hyperbolic stress‐strain for-
destructive earthquakes. In: Proceedings of the fourth European mulation of sands. In: The Proceedings of the 2nd Pan-Amer-
symposium on earthquake engineering. London, pp 125–132 ican conference on soil mechanics and foundation engineer-
68. Imai T, Yoshimura M (1972) The relation of mechanical proper- ing, Brazilian Associate Soil Mechanics. São Paulo, Brazil, pp
ties of soils to p and s wave velocities for soil ground in Japan. 289–324
Urana Research Institute, OYO Corp 87. Matasović N, Vucetic M (1993) Cyclic characterization of liq-
69. Ohta Y, Goto N (1978) Empirical shear wave velocity equations uefiable sands. J Geotech Eng 119(11):1805–1822
in terms of characteristic soil indexes. Earthq Eng Struct Dyn 88. Matasović N, Vucetic M (1995) Generalized cyclic-degrada-
6(2):167–187. https://doi.org/10.1002/eqe.4290060205 tion-pore-pressure generation model for clays. J Geotech Eng
70. Seed HB, Idriss IM (1981) Evaluation of Liquefaction potential 121(1):33–42
of sand deposits based on observations of performance in previ- 89. Darendeli MB (2001) Development of a new family of nor-
ous earthquakes. Pre-print 81-544, Session on In-situ Testing to malized modulus reduction and material damping curves. The
Evaluate Liquefaction Susceptibility, ASCE National Conven- University of Texas at Austin
tion, St. Louis, Missouri 90. Khattri KN (1999) Probabilities of occurrence of great earth-
71. Imai T (1982) Correlation of n value with s wave velocity and quakes in the Himalaya. In: Proceeding of Indian academic and
shear modulus. In: Proceedings 2nd ESOPT, Amsterdam, pp science (Earth Planet Science) 108:87–92. https://doi.org/10.
57–72 1007/BF02840486.
72. Sykora D, Stokoe K (1983) Correlations of in situ measurements 91. Nath SK, Raj A, Sharma J et al (2008) Site amplification, Qs,
in sands of shear wave velocity, soil characteristics, and site and source parameterization in Guwahati Region from seismic
conditions. Report gr 83–33. Civil Engineering Department, and geotechnical analysis. Seismol Res Lett 79:526–539. https://
University of Texas at Austin doi.org/10.1785/gssrl.79.4.526
73. Athanasopoulos G (1970) Empirical correlations Vso-NSPT for 92. SEISMOSIGNAL program (https://s eismo soft.c om/p roduc t/
soils of Greece: a comparative study of reliability. WIT Trans seismosignal/). Accessed 03 Nov 2022
Built Environ. https://doi.org/10.2495/SD950031
13
93. Mian JF, Kontoe S, Free M (2013) Assessing and managing the Springer Nature or its licensor (e.g. a society or other partner) holds
risk of earthquake-induced liquefaction to civil infrastructure. exclusive rights to this article under a publishing agreement with the
In: Woodhead Publishing Series Civil Structural Engineering, A author(s) or other rightsholder(s); author self-archiving of the accepted
Handbook of Seis Risk Ana Manag Civil Infra Sys. Woodhead manuscript version of this article is solely governed by the terms of
Publishing, pp 113–138 such publishing agreement and applicable law.
Publisher’s Note Springer Nature remains neutral with regard to

jurisdictional claims in published maps and institutional affiliations.
13

Artículo Ea2

Uploaded by

Copyright:

Available Formats

Artículo Ea2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Artículo Ea2

Uploaded by

Copyright:

Available Formats

Indian Geotech J (April 2024) 54(2):358–393

Nonlinear Seismic Ground Response Analysis for Site Classes D

* Shiv Shankar Kumar Introduction

computer programs such as DEEPSOIL, SHAKE 2000 and Previous Studies

structures depends upon the ground conditions and selected

Reference GRA methodology and Software Region Remarks

should be adopted since the soils in Haryana are capable of

Reference GRA methodology and Software Region Remarks

Fig. 2 Classification of seismic zones of Bihar [45]

The study area consists of many tectonically active fea- 𝜕𝛾

Table 2 Major earthquake in Bihar and adjoining area [45]

1. 1833 Bihar 26 Aug 18:05:00 NA NA 7.5 NA 500

Fig. 3 Locations of the

Fig. 4 Typical borehole profiles at a Arariya, b Sitamarhi, c Patna, d Madhepura site

Table 3 Empirical correlation for the estimation of shear wave

1. Imai and Yoshimura [64] Vs = 76*N0.33

suitable correlations for the estimation of shear wave velocity r

in this study, an average empirical correlation of shear wave

Magnitude (Mw) 6.7 7.8 8.2

Fig. 13 a–c Variations in

Fig. 14 Contour map of surface

Fig. 15 a–c Amplification and

Fig. 16 Contour map of

Fig. 17 a–c Variation in shear

Fig. 18 a–c Variation in excess

Fig. 19 a–c Variation in stress

Fig. 20 a–c Spectral accelera-

Fig. 21 Contour map of peak

seismic waves at Bihar region. Hence, it is recommended Funding Not applicable.

Publisher’s Note Springer Nature remains neutral with regard to

You might also like