The AlpArray Seismicity Catalog

Exploiting the new large seismic data set provided by the AlpArray Seismic Network (AASN) as part of the AlpArray research initiative (www.alparray.ethz.ch), we provide a highly consistent seismicity catalog with precise hypocenter locations and uniform magnitude calculations across the greater Alpine region (GAR) covering the period from 1st January 2016 to 31st December 2019. With a backbone of 715 broadband seismic stations (415 permanent, 300 temporary) and a uniform interstation distance of ~50 km, the AASN provides a unique opportunity to assess the laterally heterogeneous GAR seismicity distribution. Regularly, the GAR seismicity is monitored and reported by a dozen national and international observatories, requiring a challenging effort to create a uniform and reliable catalog to document and investigate the complex seismicity and tectonics of the GAR. To establish the highly consistent AlpArray Seismicity Catalog (AASC), we developed a new multi-step, semi-automated method. We applied the SeisComP3 (SC3) seismic-monitoring software and run it in playback mode to analyze the ~50 Tb of continuous data collected in 4 years for initial events detection and to calculate their hypocenter locations. We cleaned this preliminary, automatic seismic catalog from fake events and from events with an initial magnitude less than 2.0 MLv. We then made use of two additional software packages to refine phase picks and locations: the new ADAptive Picking Toolbox (ADAPT) Python library and the VELEST algorithm. The former was used to develop a new multi-picking algorithm for phase identification and precise arrival time determination. The latter was used to obtain the most reliable earthquakes locations, their quantitative error estimation and to reliably predict phase arrivals by solving the coupled hypocenter-velocity problem using the powerful joint-hypocenter determination technique (JHD). The JHD approach was also implemented as a filtering tool for outlier observations and to detect problematic events. We eventually recalculate the local magnitude (MLv) in a consistent and uniform way, obtaining a statistical magnitude of completeness of 2.4 MLv with different catalog-based techniques. The AASC is also regionally consistent up to 3.0 M+ with seismic bulletins provided by national and international agencies. Our final 4-year catalog contains 3293 precisely located earthquakes with magnitudes ranging between 0.4 - 4.9 MLv and it clearly delineates the major seismically active fault systems within the GAR. We additionally provide a new minimum 1D P-velocity model for the GAR and appropriate station delays, for both temporary and all permanent stations. These station delays for the permanent seismic station arrays, together with the velocity model, are key to consistently link the GAR past and future seismicity with our current catalog. This would allow the compilation of a broader consistent seismic catalog suitable for other seismological studies including, but not limited to, seismic hazard and a regional 3D local earthquake tomography.