A discrete Markov Model for earthquake occurrences in

Abstract: A large number of stochastic models are currently available for the earthquake occurrence. The Markov model is applied to data from the area of southern Alaska (peninsula of Alaska and Shumagin islands) and the Aleutian Islands in order to investigate for great earthquake occurrence in spa...

Full description

Bibliographic Details
Main Authors: Theodoros M. Tsapanos, Ra A. Papadopoulou
Other Authors: The Pennsylvania State University CiteSeerX Archives
Format: Text
Language:English
Published: 1999
Subjects:
Online Access:http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.494.4935
http://www.balkangeophysoc.gr/online-journal/1999_V2/aug1999/v2-3-2.pdf
Description
Summary:Abstract: A large number of stochastic models are currently available for the earthquake occurrence. The Markov model is applied to data from the area of southern Alaska (peninsula of Alaska and Shumagin islands) and the Aleutian Islands in order to investigate for great earthquake occurrence in space and time. The model defines a process in which successive state occupancies are governed by the transition probabilities of the Markov process. Each element pij, of the transition probability matrix represents the probability that the state is j at the time t+1, given that the state was i at the time t, and the probability can be written as: pij=Prob[Xt+1=j | Xt=i]. The area of southern Alaska and Aleutian Islands is divided in three seismic zones that are defined as states in the present study. Thus the earthquakes, which migrate from zone to zone, i.e. from state to state, carry with them the number of the zone in which they occurred. In this way we can examine the genesis of the earthquakes in the investigated area in a quantitative way, through the transition probabilities of the defined process. A pattern for an east-west migration, in space and in time, of large (M>7.0) earthquakes is found. A two-state Markov model is applied in the three zones, which suggests periods of activity and quiescence. The application of this model makes it possible to establish whether in a specific time period a state is in an active or an inactive period and this is useful for seismic hazard analysis.