A new seismological network for Bavaria and its application to the study of meteorologically triggered earthquake swarms

This thesis addresses to self-contained topics and is therefore
structured in tow parts. The first part describes the installation
of a new seismological network for Bavaria, whereas the second part
focuses on the investigation of rainfall induced seismicity near
Bad Reichenhall in southeastern Bavaria. Part 1 Because of the
focus of the existing seismological stations on teleseismic and
regional events, the capability of locating local earthquakes in
various regions of Bavaria was quite limited. To overcome this
situation 15 new stations were installed in the country.
Additionally, six existing stations were updated to
state-of-the-art technology in a second stage. The network geometry
closely oriented on the seismicity of Bavaria, with densly spaced
stations in areas were activity is high and promising scientific
problems could be addressed. The software concept was, as far as
possible, adopted from the GEOFON Project of th GeoFoschungsZentrum
Potsdam (GFZ), which is already used for the German Regional
Seismic Network (GRSN) and other European networks. This assures an
easy data transfer with other services and allows the future
incorporation of real-time data of the Bavarian network in an
European and global seismological networks. Part 2 In the second
part of this thesis, the focus will be laid on one of the dense
sub-networks in southeastern Bavaria. Here, near the town of Bad
Reichenhall, a connection between rainfall and seismicity is
suggested since the early 1970ties. However, because of the lack of
continuous high-quality data such a correlation could never be
tested by means of a physical model. In three papers the
observation of above average rainfall in 2002 and associated
earthquake swarms are studied in detail. Starting with an overview
paper with first location results and interpretation in Chapter 3,
the discussion of cluster analysis, high-precision relocation and
focal-mechanism analysis in Chapter 4, and finally by statistical
modeling of the relocated seismicity by means of point process
modeling and pore-pressure diffusion assuming a rate-state
earthquake friction model in Chapter 5. Although, seasonal
variability of seismicity related to ground water recharge and
precipitation has been previously observed on regional scales, a
statistically significant causal relationship between rainfall and
earthquake activity for an isolated region can be shown here for
the first time. The analysis of the high quality meteorological and
seismic data in the Mt. Hochstaufen region yields clear evidence
that tiny pore pressure changes induced by rainfall are able to
trigger earthquake activity even at 4 km depth via the mechanism of
fluid diffusion. Stress changes of the order of 5-13 mbar are found
to trigger earthquakes. This is much less than usually produced in
fluid injection experiments (several 100 bars and more), indicating
an extreme sensitivity of the crust with regard to tiny changes.
This might be an universal feature which can, however, only be seen
in the rare occasion of an isolated but critical system, like the
study area. Although, the derived focal mechanisms indicate an
influence of the Saalach Fault Zone on the stress regime of the
study area, the reason for the criticality of the seismogenic
volume in the Mt. Hochstaufen region is not yet finally resolved.
However, the high correlation between rainfall-induced pressure
changes at depth and seismicity opens the possibility of
forecasting future earthquake rates on the basis of rainfall data
in this region. Regarding the small volume in which the earthquakes
take place, the almost yearly occurring earthquake swarms and the
permanent, seismo-meteorological monitoring network, Mt.
Hochstaufen provides nearly controlled experimental conditions to
study fluid-induced seismicity and the physics of earthquake
swarms.