Rock slope failure preparation paced by total crack boundary length

Abstract Gravitational mass wasting prediction requires understanding of the factors controlling failure. Prior to slope failure, cracks in weakened rock are thought grow and coalesce, eventually forming a continuous failure plane. Here, we apply hidden Markov machine learning model seismic data, revealing temporal evolution prior major rockslide event Swiss Alps. After prolonged linear increase crack cumulative number, an S-shaped rate pattern occurred day before rockslide. A simple mechanistic can explain this behaviour, showing that total boundary length is key factor plane immediately movement. Our findings imply should be treated as 2-D, rather than 1-D objects, driven predominantly by internal external processes. offers novel, physically based approach for early warning failures.