The microtremor method is very useful in urban areas to estimate local site effects for microzonation purposes. As seismic waves propagate from bedrock to the ground, their characteristics are site specific due to the heterogeneity of substrata. Source, path and site effects are prime parameters that affect ground response. Many destructive earthquakes that occurred in the past (e.g. Bhuj 2001, Bingol 2003, Kashmir 2005, Haiti 2010, Tohoku 2011, Van 2011 and Yunnan 2012) clearly illustrate the effects of local soils on damage severity and pattern. The Indian sub-continent has more than 400 major faults that influence seismic activity and India has experienced several devastating earthquakes (Assam 1897 (M  =  8·7), Kangra 1905 (M  =  8·6), Bihar-Nepal 1934 (M  =  8·4), Assam-Tibet 1950 (M  =  8·7), Latur 1993 (M  =  6·4), Chamoli 1999 (M  =  6·8), Bhuj 2001 (M  =  7·6) and Sikkim 2011 (M  =  6·9)). The state of Andhra Pradesh is located in the central part of peninsular India (zones II and III according to Indian seismic code) and has a record of earthquakes along the coast. This study considered Vijayawada (zone III), the third largest city in the state. Microtremor surveys were carried out at 75 different locations in the Vijayawada urban area and analysis was carried out using the Nakamura technique. Dynamic characterisation was carried out by considering the shape of the response curve, horizontal to vertical (H/V) amplitude, predominant frequency and the characteristic soil profile at all the test sites. Based on this detailed analysis, a classification is proposed and a predominant frequency map of the study area was developed. It was found that areas in the north eastern and south eastern parts of the city, with silty and clayey sand formations, have comparatively high predominant frequencies (≥ 4 Hz). The northern and western parts of the city, with high silty clay and silty sand, are characterised by moderate frequency values (2–4 Hz). Low (< 2 Hz) values of predominant frequencies were observed at a few locations. The H/V amplitudes are high (2–3) in the eastern and western regions.

Shear modulus and damping ratio which are important parameters in any dynamic analysis of soil are related to the frictional behavior at inter particle contacts and rearrangement of grains. The strength loss of granular soil when subjected to cyclic loading is affected by particle size, shape and its distribution. Soil structure also termed as fabric which encompasses grains/particle distribution, particle orientation and arrangement, the voids and the fluid present in between the voids, continually changes during cyclic loading. Such changes are imperative of mechanisms that result in liquefaction of saturated sands and compaction of dry sands. Unlike, experimental and analytical testing methods, numerical methods are equipped to provide insight into these underlying mechanisms. To understand the interactions between particles that influence the dynamic behavior of soil, modeling of cyclic triaxial test was conducted using 3D DEM for different particle shapes and sizes. Before the testing, validation of the generated soil sample is conducted by checking the pressure dependent behavior and dilative response of the sample. Contact normals and contact forces were also used to validate the sample at different stages. From the cyclic testing, normalized shear modulus curves, contact normal and force distributions were calculated. It was observed that the rounded particles had more shear stiffness compared to other shapes. This behavior was not evident when the shear modulus was normalized and the spherical particles had the highest modulus. The effect of particle shape was verified with the change in the coefficient of uniformity (Cu) with shear modulus and it is observed that shear modulus decrease with an increase of Cu. Micromechanical expression for the same was also been discussed.

No abstract

No abstract