High frequency GPS as a potential contribution for monitoring a seismogenic structure

The Global Positioning System (GPS) has become an essential complement to seismic instruments for seismological studies because of its high precision, its easy of deployment and its ability to make measurements of displacement that are local to global in scale. Although several studies have demonstrated the fundamental rule of GPS in monitoring deformation associated with earthquakes [Larson et al.,2003; Bock et al.,2004; Larson, 2009; Yokota et al., 2009], its potential use for Seismology has not been fully investigated. This dissertation addresses the use of GPS for earthquake studies from two different directions - understanding and taking advantages of the increasing sample rate of GPS data - understanding and taking advantages of the availability of low-latency ancillary products (orbit and clock estimates) provided by different public organizations. The access to GPS data with increasing sample rate represents an important source because data at high frequencies allow to detect more details about the kinematic processes. In this area, we considered 10 Hz GPS data acquired from two sites located just on the activated fault ("near-source"), during the Mw 6.3 L'Aquila earthquake. The comparison between GPS and strong motion time series along with their frequency contents, con firms, on the one hand, the GPS capability to detect the rst seismic arrivals (P waves) and, on the other hand, the need of GPS sampling rates greater than 2.5 Hz (i.e. 5 or 10Hz) in the near eld of moderate-magnitude events to provide "alias-free" solutions of coseismic dynamic displacements. These displacements are very useful to constrain the kinematic rupture history of the main shock[Avallone et al., 2011]. Besides high-rate data, we used also lower frequency data of both L'Aquila and 2010 Chile event, to exploit the potential use of publicly available products for the estimation of mean coseismic displacements in the shortest time following an earthquake. By using JPL Ultra-Rapid orbits and clocks, a static coseismic displacement (5 minute solution) can be retrieved within about few hours after an earthquake. Instead, JPL Rapid products which also include high-rate 30-second clock estimates (starting from August 2011) make possible the detection of dynamic coseismic displacements (30 s and lower solution) within about 1 day after an earthquake.