Simulation of Strong Ground Motion and Site Characterization

Abstract

Strong motion studies play an important role in seismology and earthquake engineering. It is well-known fact that civil engineers face major problems in designing of buildings and other structures which can withstand safely without any damage during an earthquake. Strong motion prediction provides useful information or basic inputs which helps in designing of such buildings and structures in seismically active regions. This useful information required for such constructions can be obtained either from recorded or simulated strong motion time histories of earthquakes. Since strong motion recordings are very few or limited in any region therefore it is not always feasible to use observed records for designing earthquake resistant structures. In case of unavailability of observed records simulation of the strong motion is the only option left. In recent years researchers use many techniques to model earthquake ground motions. Each method has its advantages as well disadvantages. Among the methods Semi empirical method takes advantages of the stochastic and empirical method. The semi-empirical method has been evolved as important tool to simulate the high frequency ground motion. The method has been already used and tested for the simulation of ground motion by Midorikawa (1993), Joshi and Patel (1997), Joshi et al. (1999, 2001, 2012a, 2012b, 2014), Kumar et al. (1997), Joshi and Midorikawa (2004), Joshi and Mohan (2008). The modeling of the ground motion requires the source parameters of the earthquake, path effect and site effect. In the present thesis the source parameters of the Nepal main shock 25 April 2015 and its aftershock has been estimated using the grid search algorithm. These source parameters (moment magnitude, stress drop, source radius etc) are used to simulate the ground motion of the earthquake. In the earlier studies, simulation of the ground motion does not consider the site effect, but in the present thesis it is considered by calculating the shallow velocity profile. These Vs profile obtained i from the joint inversion of the horizontal to vertical spectral ratio (HVSR) and multichannel analysis of surface waves (MASW) is further used to simulate the ground motion of the Nepal Earthquake at the surface. The Vs profile is taken as input for the SHAKE 91 (Idriss & Sun 1993) and the hard rock simulation. The SHAKE 91 is used to simulate the ground motion of the Nepal earthquake at the surface. Strong ground motion of the Nepal earthquake recorded in the Kumaon Himalaya region is used to estimate the source parameters of the earthquake. The source parameters have been calculated from the source displacement spectra using Brune’s (1970) theory. The fitting of theoretical Brune spectra with the obtained source spectra using grid search algorithm gives direct estimate of corner frequency and long-term flat level. The value of long-term flat level and corner frequency from source displacement spectra calculated at each station is used to calculate stress drop, source radius and seismic moment of this earthquake. The present study indicate that the Nepal earthquake originated 12 km below the epicenter located at 27.93°N, 84.70°E. The source radius, stress drop and seismic moment of this earthquake estimated from source displacement spectra are (44.13 ± 3.85) km, (18.68 ± 5.93) bars and (3.53 ± 0.28) x1027 dyne cm, respectively. The strong ground motion of the Nepal earthquake has been simulated using the semi empirical method. First, the ground motion of the earthquake simulated at the hard rock. Second, we estimate the Velocity profile and it is used to simulate the ground motion at the surface. We had used strong motion data of the Nepal main shock (Mw=7.8) and aftershock (Mw=6 .6) to model the Nepal main shock which were recorded on far field station installed in Kumaon Himalaya region. Strong motion data of the mainshock and aftershock are used to determine the source parameters. The average value of the seismic moment and stress drop are 8.26x1025 dyn-cm and 10.48 bar respectively for the aftershock. First, near field strong motion data recorded at Kantipath station is modelled by changing the nucleation point, rupture ii velocity and the several possible locations. The modelling parameters are finalized based on the root mean square between the observed and simulated records and their response spectra. It is concluded that changing location, nucleation point and rupture velocity has more effect on the modeled time series. These final parameters are used to model the far field strong motion data recorded at stations in Kumaon Himalaya. The comparison of full waveform and its response spectra has been made to finalize the rupture parameters and its location. The comparison of observed and simulated records shows that this earthquake was triggered by a rupture propagating in NE-SW direction with a rupture velocity 3.0 km/sec from a distance of 80 km from Kathmandu at a depth of 12 Km. The 2015 Nepal earthquake has attracted the entire world’s attention to Himalaya due to heavy damage caused in the region. Damage caused due to site amplification of the ground shaking was clearly evident during the Nepal earthquake. Region itself has large possibilities of site amplification which can result high damage during future earthquakes. So, the site amplification factor and site predominant frequency has been estimated using the HVSR method for the various stations using the strong motion data. The spectral ratio method is among most simple and easiest method. The method has several advantageous over the Vs30 because it provides the complete description of the amplification at each frequency. It is observed from the result that some sites show the amplification at low frequency range which coincides with low natural frequency of high-rise buildings. The low predominant frequency is mainly controlled by the deeper low velocity layer. Whereas some sites show the amplification at high frequency range suggesting that local geology and shallow weathered layers of the rock dominate in the amplification. The entire Kumaon Himalaya region the has predominant frequency ranges from 1.4 Hz to 13.2 Hz. It is concluded that low frequency as well as high frequency can amplify in this area which can cause a heavy loss of lives and property.