SEISMIC VELOCITY STRUCTURE BENEATH BANDUNG BASIN FROM AMBIENT NOISE TOMOGRAPHY (ANT) AND HORIZONTAL TO VERTICAL SPECTRUM RATIO (HVSR) ANALYSIS

• Main Author: Pranata, Bayu
• Format: Dissertations
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/39812
• Institution: Institut Teknologi Bandung

Bandung is located on the basin filled by a thick sediment, which is composed of volcanic rocks and the deposition of the rest of the ancient lake. High population density, vital infrastructures and the potential sources of earthquakes in the surroundings make Bandung vulnerable to a high earthquake impact. The geological structure of the sub-surface is one of the factors influencing infrastructure damage due to earthquake shaking in addition to the magnitude of an earthquake and the distance to the epicenter. To determine the seismic characteristics and the geological structure of the Bandung Basin, a study was conducted using two methods by utilizing the ambient seismic noise data. The first method is the Horizontal-to-Vertical Spectrum Ratio (HVSR) to obtain the dominant frequency (F0), H/V amplification (A) and seismic vulnerability index (Kg). The second method is the ambient seismic Noise Tomography (ANT) to image the shear-velocity (Vs) structure beneath the Bandung Basin. The HVSR method used nocturnal data on the second day after the installation of seismographs at 58 measurement points from the Bandung Experiment network run from March to October 2014 in the Bandung city and its surrounding areas. The HVSR analysis results show the value of F0 in the study area: 0.195-7.016 Hz, the value of A: 1.6-11.3 and the Kg value: 0.6-245.6. The spatial distribution of Kg values shows that the highest seismic vulnerability is spread in the eastern part of the Bandung Basin including Bojongsoang, Rancaekek, Ciparay, Rancasari, and Majalaya. The ANT also used all data recorded by the Bandung Experiment network. In this study, cross-correlation of seismic noise signals was carried out to obtain Green's function. We used multiple filter technique (MFT) on the dispersive wave packet in the Green's function signal to obtain group velocity of the Rayleigh wave. Then travel time inversion was carried out to construct tomograms of group velocity extraction for different periods. Furthermore, depth inversion was conducted using the Neighborhood Algorithm applied to group velocity curves of 1 to 10 seconds for 270 grid points with a spacing of 2.75 km to obtain Vs model. From this process, we obtained Vs structure at depths of 1 km - 10 km. Based on the Vs tomographic maps, we observed pronounced low velocity anomalies that correlate with sediment deposits beneath the middle of the Bandung Basin and to the West of Tangkuban Parahu. On the other hand, high velocity anomalies were identified to be correlated with the distribution of the Beser Formation and the areas along the northern, eastern and southern mountains surrounding the Bandung Basin.