EXPLAINING A DEEP SPATIOTEMPORAL LAND COVER CLASSIFIER WITH ATTENTION AND REDESCRIPTION MINING

This study investigates how weather conditions influence rockfall events at a limestone cliff in the French Alps by developing a statistical approach linking events to weather patterns over time. Results show that the most significant trigger for large rockfalls is short-duration intense rainfall lasting one to three days. We also found that freezing depth increases the likelihood of rockfall occurrence over longer forecast horizons. These findings can be used to develop early warning systems.