Limits of cross-correlation detection methods – an application to deep low-frequency earthquakes in the Eifel Region
Since 2013, when the deepest earthquakes (ca. 40 km) in Germany were measured in the Eifel region, several so called „deep low frequency“ (DLF) events occurred in the south of the Laacher See volcano (East Eifel Volcanic Field). These events are characterized by deep hypocenters (10 km to 40 km) and low frequencies (< 10 Hz). This type of earthquakes are attributed to mass transport in the volcanic plumbing system and are known from other volcanoes (eg. Kusatsu-Shirane, Galeras, Kilauea, Redoupt (Chouet, 2012)). During my master‘s thesis I analyzed these earthquakes – which were first described by Hensch et al. (2019) – and I tried to detect them with cross-correlation methods. I could categorize these DLF earthquakes into 3 classes, based on duration, frequency and visible Phases. Using 11 Masterevents I could detect approx. 50 – presumably – DLF earthquakes using data from one station (LAGB) from 2015-2019. These earthquakes are located in the WEVF and around the Laacher See volcano. Nevertheless the performance of this detector is bad, as it detects significantly more noise (false detections) than eartquakes (true detections). I will show that the cross-correlation detection method is not the best method for detecting DLF earthquakes because the amplitude is invariant when two signals with equal frequencies are cross-correlated.
Einladungen für das Zoom - Meeting werden per E-Mail an die gpistud-Liste und gpi-personal-Liste geschickt. Zusätzlich werden die Informationen in der Seminargruppe auf ILIAS gepostet.
Außenstehende kontaktieren bei Interesse bitte Mike Lindner.