Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.770043
Title: Microseismic full waveform modeling and location
Author: Shi, Peidong
ISNI:       0000 0004 7660 6818
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Seismic waveforms generated by earthquakes contain valuable information about the Earth's interior. Effectively utilizing seismic waveforms is critical for understanding earthquake source mechanism, imaging subsurface structure and monitoring earthquake hazard. In contrast to large earthquakes, microearthquakes have much lower magnitudes and are difficult to detect. Recorded waveforms of microearthquakes have small amplitudes and can be easily contaminated by noise. In this thesis, I develop an automatic method to fully utilize seismic waveforms to locate earthquakes, especially microearthquakes. A seismic modeling tool which can simulate seismic wave propagation in complex media using various seismic sources is also developed to generate synthetic seismic waveforms for testing and analysis. As seismic anisotropy is common in shale and fractured rocks, I develop an efficient finite-difference full waveform modeling tool for simulating wave propagation in heterogeneous and anisotropic media. In order to model both double-couple and non-double-couple sources, an arbitrary moment tensor source is implemented in the forward modeling tool. The modeling tool can serve as an efficient Eikonal solver for the waveform migration used in source location or subsurface imaging. The modeling tool also provides an efficient way to obtain the Green's function in anisotropic media, which is the key of anisotropic moment tensor inversion and source mechanism characterization. The modeling tool can be used to generate seismic waveforms in complex, anisotropic models using various source-receiver geometries and source mechanisms. I generate and analyse synthetic datasets for vertical downhole arrays and surface arrays using this modeling tool. Due to the influence of seismic anisotropy, seismic location can have a deviation of a few hundred meters. Through analysing the synthetic seismic waveforms, I find that it is feasible to evaluate the seismic anisotropy of the subsurface and further estimate the orientation and density of potential cracks in the subsurface by examining the traveltimes and amplitudes of recorded seismic waveforms. I propose a novel waveform coherency-based method to locate earthquakes from continuous seismic data. The method can automatically detect (micro-)earthquakes and find the locations and origin times of seismic events directly from recorded seismic waveforms. By continuously calculating the coherency between waveforms from different receiver pairs or groups, this method greatly expands the available information which can be used for event location and has high imaging resolution. This method does not require phase identification and picking, which allows for a fully automated seismic location process. I have tested and compared this method to other migration-based methods (i.e. envelope, STA/LTA and kurtosis migration) in noise-free and noisy synthetic data. The tests and analysis show that the new developed method is very noise resistant and can obtain reliable location results even when the signal-to-noise ratio is below 0.1. By utilizing waveform coherency, the new method is able to suppress strong interference from other seismic sources occurring at a similar time and location and shows excellent performance in imaging weak seismic events. It can be used with arbitrary 3D velocity models and is able to obtain reasonable location results with smooth but inaccurate velocity models. Computational efficiency test shows the new method can achieve very high speedup ratio easily. This new method exhibits excellent location performance and can be easily parallelized giving it large potential to be developed as a real-time location method for very large datasets. I apply the new method to automatically locate the induced and volcano-tectonic seismicity using sparse and irregular monitoring arrays. Compared to other migration-based methods, in spite of the often sparse and irregular distribution of monitoring arrays, the new method shows better location performance and obtains more consistent location results with the catalogue obtained by manual picking. The new method successfully locates many volcano-tectonic earthquakes with local magnitude smaller that 0 beneath Uturuncu, where seismicity is triggered by the passage of surface waves caused by the M 8.8 2010 Maule earthquake. The case at Uturuncu demonstrates that this new method can be used to automatically detect and locate microseismic events in large or streaming seismic datasets, which are time consuming and difficult to manually pick. 98.25% of 114 triggered seismic events in the published catalogue have been successfully detected and located. In addition, the new location method also automatically detect and locate 322 verified additional seismic events whose magnitude is smaller than 0. Using this new location method, a more complete seismic catalogue with much lower magnitude threshold can be obtained, which can benefit further seismic analysis. The new location results at Uturuncu show that seismicity is likely deeper than previously thought, down to 7 km below the surface. The event location example at the Aquistore carbon capture and storage site shows that continuous and coherent drilling noise in industrial settings will pose great challenges for source imaging. However, automatic quality control techniques are used to automatically select high quality data, and can thus effectively reduce the effects of continuous drilling noise and improve source imaging quality. By utilizing this new location method in combination with the automatic quality control techniques, event location down to signal-to-noise ratios of 0.025 (1/40) is possible in tests. The location performance of the new method for synthetic and real datasets demonstrates that this new method can perform as a reliable and automatic seismic waveform analysis tool to locate microseismic events.
Supervisor: Nowacki, Andy ; Rost, Sebastian ; Angus, Doug Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.770043  DOI: Not available
Share: