Sensitivity of results inferred from microtremor observations

• Main Author: Daniel Lukas Mulyawan Jap
• Other Authors: Kusnowidjaja Megawati
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Civil engineering::Structures and design
• Online Access: http://hdl.handle.net/10356/39757
• Institution: Nanyang Technological University

Microtremor based observation to estimate subsurface soil structure and ground response, called Microtremor Survey Method, have become an interesting topic for research, mainly because of the advantages it offers such as low cost and non-destructive observation. The method requires an observation array and sensors to record microtremor waves amplitude which is then processed to estimate Rayleigh waves dispersion curve and subsurface soil layers. In this project, Spatial Autocorrelation (SPAC) and Capon’s Frequency-wavenumber (F-K) methods are used for the processing of microtremor recording. The field observations were conducted in Katong Park, Beatty School and Nanyang Technological University areas where soil profile data is available. The first investigation carried out was the applicability of f-k method for small array recording and it was found that f-k method is not suitable for small array. The accuracy and consistency of microtremor method was studied in this project using several observations in Katong Park and Beatty School area on 2008, 2009 and 2010. The results indicate that microtremor method provides good and consistent estimation. This study also investigated the impact of defective sensor on accuracy of phase velocity estimation. It was found that SPAC method remains accurate even when two sensors are defective while f-k method remains accurate up to one defective sensor located on the external corner of the array. This study also compares three data lengths (4096, 8192, and 16384) for both SPAC and f-k methods. The 8192 data length is chosen as the optimum as it could provide good subsurface estimation and practical field observation requirement. Various array configurations (circular, square, rectangle, and line) were constructed for the purpose of comparing accuracy obtained from SPAC method processing. It was found that circular array yields the most accurate phase velocity estimation while line array only yields satisfactory estimation. Furthermore, due to its practicality, 60 m line array was constructed and the phase velocity estimated implies that line array only capable in providing satisfactory estimation. The subsequent investigation on impact of near source noise towards SPAC method phase velocity estimation shows that when the source is located too close to the sensors that the noise is recorded concurrently on all sensors, the noise might cover the natural microtremor waves and SPAC method is unable to estimate the phase velocity. Finally, the applicability of microtremor observation for hard soil stratum was assessed in this project. The observation was conducted inside Nanyang Technological University which is on the Jurong formation (sedimentary rock). The result indicates that Microtremor Survey Method can be applied to this type of soil profile with excellent estimation. Hence, it can be concluded that Microtremor Survey Method is an excellent preliminary exploration method.