부산대학교 도서관

The role of fault geometry, Coulomb stress and elapsed time is investigated herein to determine whether earthquakes are clustered or not in the central Apennines, Italy. Two earthquake sequences are analysed to determine the relative importance of Coulomb stress transfer and elapsed time. The importance of fault geometry when modelling the Coulomb stress transfer is demonstrated. The earthquake sequences of interest can be partially explained by a combination of Coulomb stress transfer and elapsed time, thus demonstrating that earthquakes in the central Apennines are non-stochastic in nature. Considering the variations in fault geometry over short (hundreds of metres) length scale has been used to demonstrate that the surface bedrock fault scarps are active. Over longer scales (kilometres) the geometry of the faults is shown to affect the pattern of Coulomb stress transferred during earthquakes. A novel methodology outlined in this thesis is used to model Coulomb stress changes throughout the historical record onto faults with strike-variable geometry. It is shown that the Coulomb stress transfer is likely to have played a role in two earthquake se- quences of interest, the 1703 - 1706 A.D. and 2016 - 2017 A.D. sequences. However Coulomb stress transfer cannot fully explain these sequences. The elapsed time on faults of interest to these sequences is considered, and it is shown that faults with longer elapsed time rupture preferentially over faults with shorter elapsed time in both se- quences. When considered together, fault geometry, Coulomb stress and elapsed time considered together can explain the progression of the 2016 - 2017 A.D. earthquake sequence. The 1703 - 1706 A.D. sequence can be explained in a similar manner, how- ever the results are less conclusive due to a lack of elapsed time data. Elapsed time cannot be considered alone, without information about the mean recurrence intervals on faults of interest. The results presented herein have implications for estimations of seismic hazard.