Abstract (eng)
This cumulative thesis consists of three papers in which I analyse the seismicity of Austria and Vienna Basin, focusing on the Vienna Basin Transfer Fault system, in terms of the completeness of earthquake records and aftershock sequences. The article by Nasir et al. (2013; Appendix A) assessed catalogue completeness for a composite catalogue for Austria and the surrounding regions using TCEF and statistical analyses according to Stepp (1972). Most of the earthquake data is pre-instrumental in nature. Therefore, to avoid the uncertainties resulting from intensity-magnitude conversions the computation and analysis is accomplished in intensity. When comparing both completeness corrections, it should be stressed that the TCEF includes all earthquakes of the higher intensity classes based on the assumption that all of the strongest earthquakes have been observed during the whole catalogue length. Due to this reason the highest intensity class is also included for Gutenberg-Richter analysis. The Stepp method (Stepp, 1972), on the other hand, only considers intensity classes for which stable recurrence periods can be calculated from the earthquake record. As this is not the case for earthquakes of higher intensity, the Stepp, 1972, test does not include the highest intensity class for Gutenberg-Richter analysis due to the fact that the historical observation period is too short to calculate a statistically meaningful recurrence interval for these events. Both completeness corrections therefore reveal significant differences in the resulting Gutenberg-Richter relations and different recurrence estimates for strong earthquakes. The GR results shows that there is only a minor difference in a- and b- values for the clustered and declustered catalogue. The GR a- and b-values after TCEF correction underestimates the lower intensities and overestimates the higher intensities. The Stepp test is regarded more reliable due to the fact that it is only considering intensities, for which stable recurrence periods can be computed from the time coverage of the underlying earthquake catalogue. Results suggest that data for earthquakes of I0=VII and VIII are only complete for the period after 1750, lower intensities are only complete from 1900 on. Comparison of the completeness of the whole catalogue with the Vienna Basin sub-region suggests that the completeness for the entire catalogue is overestimated by both applied methods. In the Vienna Basin, I0=VII and VIII is only complete since 1900; I0<VII earthquakes are even incomplete for the 20th century. Observation time is too short to determine recurrence intervals for I0=X (whole region) and I0=IX (Vienna Basin). It is concluded that historical observations are not significant for determining stable recurrence rates for large earthquakes in areas with slow and very slow faults. The study by Nasir et al. (2020; Appendix B) analysed the 1906 Dobrá Voda (M=5.7) earthquake which is the dominant earthquake at the northern part of the Vienna Basin and has influenced the local seismic activity in the 20th century. Environmental effects caused by the 1906 Dobrá Voda mainshock and its strongest aftershock in 1906 described by contemporary authors include hydrological effects and surface cracks. Application of the ESI 2007 intensity scale to these effects help better constrain the epicentral intensity for the Dobrá Voda mainshock and its strongest aftershock. Analyses reveal epicentral intensities of I0=IX and I0=VIII for the main shock and the strongest aftershock, respectively. The new intensity estimates are regarded more appropriate to describe the maximum intensity of the events than the value listed in the ACORN and CNEC catalogues (ACORN, 2004; Grünthal et al., 2009). Earthquakes prior to the main shock are sparse and mostly scattered around the entire region. After the mainshock, earthquakes are mainly concentrated within 13 km of epicentral distance for ca. 70 years. Comparison of aftershock determination methods suggests that the aftershock sequence might have lasted longer than predicted by the Omori law, as suggested by Stein & Liu (2009) for slow-moving faults in intraplate regions. Nasir et al. (submitted to Austrian Journal of Earth Sciences, JAES22_03; Appendix C) divided the VBTF regionally into eight arbitrary segments of approximately 50 km length each in order to identify potential long aftershock sequences comparable to the Dobrá Voda 1906 earthquake. Such aftershock sequences might result from unrecorded strong earthquakes. The segments overlap each other for not missing information regarding aftershock patterns that may result from earthquakes that occurred next to segment boundaries. The working hypothesis is that the cause of the documented heterogeneous earthquake activity with seismicity particularly concentrating along the southern part of the VBTF might reflect long aftershock sequences subsequent to one or several strong earthquakes that occurred prior to the time covered by the earthquake catalogue. Aftershock activity is generally triggered by a large earthquake. In the aftermath of a large earthquake, this aftershock activity leads to a local increase of seismic activity which later on decays back to a lower level labelled as ‘normal’ background seismicity (Stein & Liu, 2009). The length of an aftershock sequence can vary from a few months at plate margins to several years and even decades and centuries in intraplate regions (Stein & Lui, 2009), depending not only on the magnitude of the causing earthquake, but apparently also on the regional level of background seismicity. At the northern end of the VBTF, the 1906 Dobrá Voda mainshock (M=5.7) has caused elevated seismicity in its near vicinity which has been still recorded till today (Nasir et al., 2020). The analysed segments of the Vienna Basin Transfer Fault (apart from Dobrá Voda) and the Mur-Mürz Fault do not show seismicity patterns which could be interpreted as long aftershock sequences subsequent to a so-far unknown strong earthquake. The 1906 Dobrá Voda (M=5.7) earthquake lies in segment 2 causing the long aftershock pattern observed in segments 2 and 3. Segment 3 also revealed a temporal seismicity pattern indicating an about 10-20 years long aftershock sequence after a mainshock in 1890 (M=4.5). In segment 8 several moderate earthquakes occurred, e.g., in 1899 (M=4.1), 1995 (M=4.2) and 2004 (M=3.8), each characterized by a short aftershock sequence. Apparently, these earthquakes are not strong enough to cause long aftershock sequences.