Doctor of Philosophy
Abstract:This dissertation presents the results of applying three independent statistical techniques on the seismic catalog of Alaska and Aleutian subduction zone. I perform Visibility Graph Analysis and Multifractal Detrended Fluctuation Analysis respectively on the seismic catalogs of several defined seismogenic zones in surface and depth. Forecasting earthquake hazard is based on the assumption that the Gutenberg-Richter relation represents the size distribution of future earthquakes and we show that the series produced by these methods have properties with close correlation with the b-value of the Gutenberg-Richter law. Visibility graph analysis basically maps a time series into the networks of nodes and connection and we want to show that produced network keeps a relationship with seismic characteristics of the region. Same goes for the multifractal detrended fluctuation analysis which studies the multifractality of the seismic catalogs as magnitude time series. I am also trying to improve the spatial information of the catalog using the Condensation method based on the location error. It will produce a new catalog that differs with the original one by the new assigned weight to the events according to their accuracy relative to the neighboring events. Using this statistical method will contribute to the discovery of previously unknown active structures and a better understanding of seismic hazards in Alaska.
Dissertation or thesis originally submitted to ProQuest
Azizzadehroodpish, Shima, "Statistical Analysis of Seismicity Catalog of Alaska: The mysteries of the timeseries" (2018). Electronic Theses and Dissertations. 2881.