Kinematic Deformation Analysis Associated With Earthquake And Volcano Based On GPS And Seismic Observations

In the past thousands of years, humans' activities were strongly affected by various natural disasters and their living places were sometimes destroyed. In those natu-ral disasters, earthquake and volcano disaster, as two of the geological disasters, have the highest activity and the most significant cumulative impact in the history. Since 1950's, with the advanced development of multiple techniques of continuous measure-ments, the knowledge about mechanism and evolution of seismic and volcanic process is significantly improved. Continuous Global Positioning System (GPS) and modern seismometer show unique potentials in the multiple techniques because of their contin-uous operation in the time domain. They could provide comprehensive observations with a full band to studying any geophysical process. Different geophysical signals gen-erated by different seismic and volcanic processes would be extracted from a massive dataset with continuous multi-observation by different designed methods. Although the relevant methods have been well developed and numerous geophysical signals have been uncovered, it is still difficult to conclude if there are any unknown signals emitted from these two sources or if the well-known signals are observable by multiple measure-ment or if the precision of the relevant observation could be improved. To answer these questions, the revision or re-evaluation of multiple observations in different aspects are required. In this study, continuous observations from GPS and seismometer during a earthquake and in a volcanic region are solely or jointly discussed to extract and analyse the kinematic deformation signals associated with earthquake and volcano processes.The principles of GPS and seismometer are first introduced, which is followed by a comprehensive summary about their noise characteristics in the different frequency band when they are operating in the field. An idea to extract the geophysical signal from a noisy continuous dataset is introduced as well. Secondly, the potential of GPS to record seismic waves excited by the 2011 Tohoku-oki earthquake is systematically discussed. Based on the kinematic solution of GPS observation in Japan, the ability of recording P/S phases of the main-shock and after shocks, seismic phase from tsunami, direct surface wave, multiple surface waves and other phases are studied. The broadband seismograms and the theoretical geophysical model are applied to validate the results solely inferred from the GPS observations. GPS is concluded to be capable of recording a weak seismic phase, that is multiple surface waves, which could act as a complementary tool to study surface wave dispersion.Thirdly, with a analysis of a long-term continuous seismic observations, the signal characteristics of long-period tremor (LPT) emitted by the Naka-dake volcano, Aso Caldera from 2011 to 2015 are discussed, which evaluates the completeness of the LPT features in a volcanic system. Observations from two broadband and seven short-period seismometers are analysed. An automatic extracting algorithm based on the continuous wavelet transformation is applied here to pick up all possible LPT events in the five-year dataset. Monthly and weekly stacks of the picked LPT waveforms are systematically used to evaluate their signatures, which results in the uncovering of different kinds of LPT. In the presence of the topography and layered crustal medium, the source locations and mechanisms of different LPT signals are inverted by a grid-search method. The different LPTs are concluded to go against the theory that the signal characteristic in a single volcano system should be unique and fixed. Consequently, the temporal evolution of the number, energy and amplitude of different LPTs in the five years are discussed. Together with other geological observations, the original magma-transportation model in the hydrothermal system are modified. The relevant features of different kinds of LPTs are related to the transportations of the magma or heat from a deep-seated chamber to the reservoir or from the reservoir to the shallow conduit. It may act as a "magma meter" of a volcanic system.At last, a joint application of GPS and seismometer on monitoring broadband kinematic coseismic ground motion on-line is discussed, including the problems of the robustness and real-time performance of the solution. The weaknesses of low sensitiv-ity and aliasing problems of GPS and baseline error of seismometer from the rotation and tilt of the inertial sensor in the near field of a moderate or large earthquake are considered, especially the non-linear time-varying nature of baseline error. An adap-tive Kalman filter is introduced to loosely integrate the GPS displacement and the seismometer acceleration, which is implemented by the Helmert variance component method. The variant statistical characteristic of the system noise model introduced by the time-varying baseline error is well considered in this adaptive filter. It is also difficult to disperse when the number of epochs seriously increases, which is efficient to estimate the broadband kinematic coseismic signal in real time This method is applied to determine the real-time broadband coseismic signals excited by the 2011 Tohoku-oki earthquake in a simulated real-time mode.