Lessons Learned from Regional MT Inversion of Small to Moderate Earthquakes Using the Dense Alparray Seismic Network (AASN)

Session: Earthquake Source Parameters: Theory, Observations and Interpretations [Poster]
Type: Poster
Date: 4/30/2020
Time: 08:00 AM
Room: Ballroom
Description: 

The AASN was operated from 2016 to 2019 by 36 European research institutes aiming for insights into the orogenesis of the Alps as well as into past and recent geodynamic processes. The network included above 620 temporary and permanent broadband stations with a spacing of 50 - 60 km and was accompanied by additional, denser networks. While the excellent station coverage provides great opportunities for seismicity studies, the large number of stations (up to 100) poses new challenges to MT inversions. Quality control and the choice of appropriate configurations becomes crucial. Weak magnitude events and the complex heterogeneous tectonic setting force us to push the limits of full waveform MT inversions. We develop semi-automatic, adaptive approaches for a standardized quality assessment of large seismic networks and for the selection of appropriate waveform fitting targets and frequency ranges. The source optimization framework ‘Grond’ uses a Bayesian bootstrap-based probabilistic inversion scheme with flexible integration of different waveform attributes to provide full or deviatoric MT solutions including uncertainties. We compare the inversion results of various methods and configurations in time domain (td) and frequency domain (fd) using different frequency ranges and epicentral distances. We inverted approx. 100 MT solutions for events down to Mw 3.1 occurring within the operating time of the AASN. For this magnitude range a combination of fd spectra and td waveform fitting of surface waves (Z, R , T comp., 0.02-0.07 Hz) provides most stable results. In case of distorted absolute amplitudes a combination with max. cross-correlation fitting proved to be useful. We find that for smaller events surface waves are not observed and higher frequency body waves are strongly influenced by complex heterogeneities along the travel path. To extend the source analysis to even weaker events the MT inversion is combined with network similarity cluster analyses, enabling the association of weaker events to larger ones and therefore the reconstruction of the geometry of active faults.

Presenting Author: Gesa M. Petersen

Authors