Towards Automated Real-Time Detection and Location of Large-Scale Landslides through Seismic Waveform Back Projection

Abstract

Rainfall-triggered landslides are one of the most deadly natural hazards in many regions. Seismic recordings have been used to examine source mechanisms and to develop monitoring systems of landslides. We present a semiautomatic algorithm for detecting and locating landslide events using both broadband and short-period recordings and have successfully applied our system to landslides in Taiwan. Compared to local earthquake recordings, the recordings of landslides usually show longer durations and lack distinctive P and S wave arrivals; therefore, the back projection method is adopted for the landslide detection and location. To identify the potential landslide events, the seismic recordings are band-passed from 1 to 3 Hz and the spectrum pattern in the time-frequency domain is used to distinguish landslides from other types of seismic sources based upon carefully selected empirical criteria. Satellite images before and after the detected and located landslide events are used for final confirmation. Our landslide detection and spatial-temporal location system could potentially benefit the establishment of rainfall-triggered landslide forecast models and provide more reliable constraints for physics-based landslide modeling. The accumulating seismic recordings of landslide events could be used as a training dataset for machine learning techniques, which will allow us to fully automate our system in the near future.

Laboratory Landquakes: Insights From Experiments Into the High‐Frequency Seismic Signal Generated by Geophysical Granular Flows

“…To extract more information and infer these properties, previous authors suggest using the high‐frequency component of landquakes, generated by the rapidly fluctuating forces exerted by the flow and associated with the accelerations of individual particles within it. The spectrogram of this high‐frequency component and its envelope have distinctive shapes (Suriñach et al., 2005) which can be used to detect landslides (e.g., Dammeier et al., 2016; Fuchs et al., 2018; Hibert et al., 2014; Lee et al., 2019). Furthermore, the properties of this envelope can be related to those of the landslide: the envelope's duration to the landslide's duration and hence its loss of potential energy (Deparis et al., 2008; Hibert et al., 2011; Levy et al., 2015); the envelope's amplitude to the seismic energy emitted by the landslide and hence its volume (Hibert et al., 2011; Levy et al., 2015; Norris, 1994), its work rate against friction (Levy et al., 2015; Schneider et al., 2010), and its momentum (Hibert et al., 2015, 2017); and envelope scale and shape parameters to the landslide's geometry via multilinear regression (Dammeier et al., 2011).…”

Locating Rockfalls Using Inter‐Station Ratios of Seismic Energy at Dolomieu Crater, Piton de la Fournaise Volcano

“…The rockfall seismic signals are analyzed in sliding time windows, making it possible to follow the rockfall trajectory over time. The method can therefore potentially be used for continuous monitoring in real time, in parallel with existing methods that detect and classify rockfall seismic signals (e.g., Dammeier et al, 2016;Dietze et al, 2017;Hibert, Provost, et al, 2017;E.-J. Lee et al, 2019;Maggi et al, 2017;Provost et al, 2017).…”

“…Growing networks of seismic stations offer the opportunity to continuously monitor large regions of interest. Landslide events can be detected, characterized, and located using the seismic signals they generate (e.g., Hibert et al, 2014;E.-J. Lee et al, 2019;Provost et al, 2017;Suriñach et al, 2005).…”

“…Landslide events can be detected, characterized, and located using the seismic signals they generate (e.g., Hibert et al., 2014; E.‐J. Lee et al., 2019; Provost et al., 2017; Suriñach et al., 2005). This helps in creating catalogs of landslides that allow statistical analysis of their spatial and temporal activity and estimation of their probability of occurrence.…”

Simultaneous state–parameter estimation of rainfall-induced landslide displacement using data assimilation

“…Li and Kong (2014) carried out a genetic algorithm and support vector machine (GA-SVM) method to establish a mathematical function prediction model. Although the above methods have certain practicability in the prediction of landslides, it is still problematic to carry out forecasts of rainfall-induced landslides in real time (Yin et al, 2010) -for the reason that surveillance photographs or optical remote-sensing satellites are not immediately available (Lee et al, 2019). It may take days, even months, to obtain field data and establish a process model of the study area.…”