Stromboli is one of the world’s most active volcanoes with near-continuous eruptive behaviour. The challenge was to identify patterns in monitoring data that could indicate changes in Predictive Volcanic Activity Modelling before they occur.

Approach
Using time-series analysis and machine learning, a predictive model was developed based on geophysical datasets. Close collaboration with domain experts ensured the outputs were scientifically sound and practically applicable for monitoring teams.

Result
A predictive model capable of supporting monitoring teams in understanding and anticipating changes in volcanic behaviour, grounded in historical measurement data.