We carry out a broad-band analysis of the seismic signals generated by a massive landslide that occurred near Icy Bay (Alaska) on 2015 October 17. The event generated seismic signals recorded globally. Using Instaseis, a recently developed tool for rapid computation of complete broad-band synthetic seismograms, we simulate the seismic wave propagation between the event and five seismic stations located around the landslide. By modeling the broad-band seismograms in the period band 5-200 s, we reconstruct by inversion a time-varying point force to characterize the landslide time history. We compute the broad-band spectrum of the landslide force history and find that it has a corner period of about 100 s, corresponding to the duration of sliding. In contrast with standard earthquakes, the landslide force spectrum below the corner frequency decays as ?, while the spectral amplitudes at higher frequencies is proportional to ω-2, similar to the rate of spectral decay seen in earthquakes. From the inverted force history and an estimate of the final run-out distance, we deduce the mass, the trajectory and characteristics of the landslide dynamics associated with the centre of mass, such as acceleration, velocity, displacement and friction. Inferring an effective run-out distance of ~900 m from a satellite image, we estimate a landslide mass of ~150 million metric tons.
Bibliographical noteKAUST Repository Item: Exported on 2022-06-09
Acknowledgements: Data were obtained from the Data Management Center of IRIS (Incorporated Research Institutions for Seismology), using the seismological community scientific library Obspy. Figures were made using Obspy [(e.g. Beyreuther et al. 2010; Krischer et al. 2015), www.obspy.org] and Matlab.We thank Tarje Nissen-Meyer, Martin van Driel and Lion Krischer for fruitful discussions about Instaseis during the First TIDESTraining School (Bertinoro, Italy), supported by COST (European Cooperation in Science and Technology). We acknowledge support from a Lamont-Doherty Earth Observatory Postdoctoral Fellowship and the Brinson Foundation. LG also acknowledges partial support from Princeton University and King Abdullah University of Science and Technology. This work was supported in part by the U.S. National Science Foundation Division of Earth Sciences and the Geopmorphology and Land-use Dynamics/ Geophysics programs under award 1227083
This publication acknowledges KAUST support, but has no KAUST affiliated authors.