Seismic instruments placed outside of spatially extensive hazard zones can be used to rapidly sense a range of mass movements. However, it remains challenging to automatically detect specific events of interest. Benford's law, which states that first non-zero-digit of given datasets follow a specific probability distribution, can provide a computationally cheap approach to identifying anomalies in large datasets and potentially be used for event detection. Here, we select raw seismic signals to derive the first-digit distribution. The seismic signals generated by debris flows, landslides, lahars, and glacier-lake-outburst floods follow Benford's law, while those generated by ambient noise, rockfalls, and bedload transports do not. Focusing on debris flows, our Benford's-law-based detector is comparable to an existing random forest method for the Illgraben, Switzerland, but requires only single station data and three non-dimensional parameters. We suggest this computationally cheap, novel technique offers an alternative for event recognition and potentially for real-time warnings.

Machine learning can improve the accuracy of identifying mass movements in seismic signals and extend early warning times. However, we lack a profound understanding of the effective seismic features and the limitations of different machine learning models, especially for debris-flow warning. Here, we investigate the importance of seismic features for the binary debris flow classification tasks using two ensemble models: Random Forest (RF) and eXtreme Gradient Boosting (XGB) models. We find that an established approach to training machine learning models for debris flow classification task based on more than seventy seismic signal features may be affected by redundant input information. These seismic features are derived from physical and statistical knowledge of impact sources and are grouped into waveform, spectrum, spectrogram, and network sets. Our results show that only six selected seismic features can perform similarly for the binary debris flow classification task compared to published benchmark results trained with seventy features. Considering models that aim to capture patterns in sequential data rather than focusing on information only in one given window as ensemble models, using the Long Short-Term Memory (LSTM) algorithm does not improve the performance of binary debris flow classification tasks over RF and XGB. As a debris flow alarm task, the LSTM model predicts debris flow initiation more consistently and generates fewer false warnings. Our proposed framework simplifies seismic signal-driven early warning for debris flows and provides an appropriate workflow for identifying other mass movements.

The way Alpine rivers mobilize, convey and store coarse material during high-magnitude events is poorly understood, notably because it is difficult to obtain measurements of bedload transport at the watershed scale. Seismic sensor data, evaluated with appropriate seismic physical models, can provide that missing link by yielding absolute time-series of bedload transport. Low cost and ease of installation allows for networks of sensors to be deployed, providing continuous, watershed-scale insights into bedload transport dynamics. Here, we deploy a network of 24 seismic sensors to capture the motion of coarse material in a 13.4 km2 Alpine watershed during a high-magnitude bedload transport event. First, we benchmark the seismic inversion routine with an independent time-series obtained with a calibrated acoustic system. Then, we apply the procedure to the other seismic sensors across the watershed. Spatially-distributed time-series of bedload transport reveal a relative inefficiency of Alpine watersheds in evacuating coarse material, even during a relatively infrequent high-magnitude bedload transport event. Significant inputs measured for some tributaries were rapidly attenuated as the main river crossed less hydraulically-efficient reaches, and only a comparatively negligible proportion of the total amount of material mobilized in the watershed was exported at the outlet. Cross-correlation analysis of the time-series suggests that a faster moving water wave (re-)mobilizes local material and bedload is expected to move slower, and over shorter distances. Multiple periods of competent flows are likely to be necessary to evacuate the coarse material produced throughout the watershed during individual source-mobilizing bedload transport events.