Equivalent linear and nonlinear site response analysis, study case for banda aceh, indonesia

Abstract

Earthquake is one of the most devastating natural hazards faced by various countries around the world. Indonesia has a reputation for being one of the world's most seismically active countries. Indonesia is the world's fourth most populous country, with a population of around 270 million people. Indonesia also has more earthquakes than the majority of other nations on earth. Indonesia is surrounded by over 18,000 km long of major tectonic plate boundaries. This is attributed to its geographical location, which is surrounded by three major active tectonic plates: Eurasian, Indo-Australian, and Philippine plates. Banda Aceh, a city located in the west part of Indonesia, that have experienced the most devastating earthquake in Indonesia. Banda Aceh geographically is located on 05º30′ – 05º35′ N and 95º30′ – 99º16′ E. At least there are 1,271 earthquakes occurred from 1907-2020 with a magnitude > 5 Mw happened around Banda Aceh area. The earthquakes of December 26, 2004 (Mw 9.1) and March 28, 2005 (Mw 8.7) are the two largest subduction earthquakes that have ever been recorded happened in the country. For such an area, better knowledge of seismic ground response is needed to design earthquake-resistant, sustainable, and cost-effective infrastructures. Both deterministic and probabilistic seismic hazard analyses can be used to objectively assess the seismicity of a particular location. These days, probabilistic seismic hazard analysis is commonly used since it carefully considers uncertainties such as the earthquake's location, time of occurrence, and magnitude. Probabilistic seismic hazard analysis has been used to assess the seismicity of Banda Aceh city. For this purpose, the earthquake catalog of the region was prepared by using the instrumental period data of Badan Meteorologi Klimatologi dan Geofisika (BMKG), Engdahl van der hilst and Bulland (EHB) Bulletin, and United States Geological Survey (USGS). All earthquake magnitudes in the earthquake catalog were converted to the same type of earthquake magnitudes using empirical formulas and the entire earthquake catalog was made homogeneous. Foreshock and aftershock are removed using a Z-Map with the criteria Gardner and Knopoff (1974), the software is also used to calculate b-value and rate recurrence mainshock. The subduction zone and shallow crustal are the seismic source models used. The seismotectonic conditions, local geology, and spatial data are used for basis of the characterization model. Seismic source and shallow earthquake data were modeled after ISC focal mechanism. Meanwhile, for the area that is not incorporated with the focal mechanism, the normal-slip mechanism is assumed to weight at 0.2, the reverse-slip mechanism at 0.4, and the strike-slip mechanism at 0.4. Because there are no specific attenuation functions for Indonesia region, attenuation function derived for other regions and worldwide was used. A computer software called EZ-Frisk used to perform the seismic hazard on Banda Aceh. PGA and SA in the period spectra 0.2 and 1 second was obtained. PSHA was carried out on the layer of bedrock with Vs30 at 750 m/s. Grid for a 50 km of the fault zone was set to be 0.1o x 0.1o and for 0.5o x 0.5o for the zone outside from the fault zone. The seismic acceleration from each grid was mapped using ArcGIS 9.3 software. The result obtained from the study was a seismic hazard map for Banda Aceh with the 2475 year return period. Based on the results in the form of hazard maps by the tremor in the bedrocks, the value of PGA in Banda Aceh and northern Sumatra ranged from 0.15 to 1.3g, SA 0.2 seconds value ranges from 0.35 to 2.8 g and SA 1 second ranging from 0.3 to 1.3g with the dominant source of the earthquake in the region is the megathrust fault and Sumatra-Andaman fault. This thesis focuses on understanding a phenomenon better known as site or ground response in which soil layers alter the amplitude, frequency content, and duration of earthquake-induced ground motion. Site Response Analysis is one of the most critical steps in geotechnical earthquake engineering. The amplification of seismic waves due to the geological structure of a particular site has been found to be an important topic for many researchers and civil engineers. The degree of the amplification caused by site conditions depends on the dynamic characteristics of the soil, the characteristics of the base rock motions, the impedance contrast between the soil profile and the base rock and the depth of semi-infinite half space. The objective of this research is to compare one-dimensional Equivalent Linear (EQL) and Nonlinear (NL) ground response analysis on some selected sites in the Banda Aceh region using Deepsoil software. Soil profile as several representatives from the region was selected and idealized. Due to the lack of shear wave velocity data, Vs profile was obtained from correlation. The earthquake time history used for this study was 2004 Sumatra-Andaman earthquake. Because the limited digitalized and recorded earthquake acceleration data, earthquake input motion used in this study was taken from simulated one produced by Sørensen et al in 2007. For equivalent linear analysis, the solution type selected is frequency domain. General Quadratic/Hyperbolic Model (GQ/H) model was used. Non-masing hysteretic loading was selected. For nonlinear analysis, the solution type selected is time domain. General Quadratic/Hyperbolic Model (GQ/H) model was used. Masing hysteretic loading was selected. Maximum strain values from Equivalent Linear (EL) and Nonlinear (NL) analyses showed only a few insignificant diffrences. However, the response spectra from the EL analyses are higher than those of the NL analyses, especially in soft soil. The site response analysis results reveal that the Banda Aceh soil profiles have a tendency to amplify the earthquake motions. In site response analysis studies, another issue that is as important as the effects of soils on the propagation of earthquake waves is the evaluation of the liquefaction potential of the soil under dynamic loadings. It is stated that the soil deposits with clay content may be evaluated as non-liquefiable. For this study, liquefaction analysis was conducted solely to saturated sand layers. For this study, the simplified method that compares CRR (Cyclic Resistance Ratio) which is the resistance of a soil layer against liquefaction to CSR (Cyclic Stress Ratio) was used to determine the factor of safety (FS) of the given soil. The formulas and table used were based on the method proposed by Idriss and Boulanger (2014). From the liquefaction analysis, the soil profile SPT-1 and SPT-2 are susceptible to liquefaction and SPT-3 sand layer was not liquefied.