Ground motion estimation and its application in the seismic hazard analyses of west Tennessee

There are two approaches to estimating ground motion characteristics from Fourier amplitude spectra: the random vibration theory (RVT) and the time domain simulation (TDS). The RVT method is computationally faster than TDS. However, the estimation of spectral values by the RVT method is not as good as the peak ground acceleration because the assumption on which it is based is not reasonable for the estimation of spectral values.; In the prediction of ground motion by RVT, the basic assumption is that the ground motion process is a band-limited stationary Gaussian white noise. For pseudo response spectral values (velocity and acceleration), the estimation is based on the response of a single-degree-of-freedom (SDOF) system due to the input of band-limited Gaussian white noise. The response of a SDOF system is a narrow-band random process. Therefore, the assumption of a band-limited random process (used to predict the peak ground acceleration and peak ground velocity) is no longer valid for the estimation of pseudo response spectral values. The property of a narrow-band process is significantly different from that of band-limited process. It is necessary to take properties of a narrow-band process into consideration in the estimation of ground motion.; In this study, a new theoretical method is proposed to estimate spectral values on the basis of a narrow-band stationary Gaussian process. A key feature of the method is the use of envelope crossings in lieu of process crossings of ground motion amplitude levels. Substitution of envelope crossings makes the estimation of the crossing number more reasonable for a narrow-band random process. This improves the estimation of spectral values for a wide range of frequencies. The strength and the limitation of the proposed approach are illustrated through several examples.; As an application of the estimation of ground motion, we estimated the hard rock seismic hazard in West Tennessee. In this region, the major seismic source zone is the New Madrid zone. However, there are many uncertainties in delineating geometric boundaries of seismic source zones and the determination of seismicity parameters because of the low seismicity and the complexity of the Mississippi Embayment. Three sets of seismic source zone boundaries and several alternatives for seismicity parameters are considered to account for uncertainties in the estimation of seismic hazard.; A comprehensive seismic hazard considering all alternatives can be obtained using the popular logic tree method. Results of this study are compared with the newest national seismic hazard zonation maps. Comparisons show that our results are generally close to results of national seismic hazard zonation for a return period of 2,500 years but are higher for a return period of 500 years. (Abstract shortened by UMI.)...