Classification of Mt Etna (Italy) Volcanic Activity by Machine Learning Approaches

Alireza Hajian,Filippo Greco,Flavio Cannavò,Giuseppe Nunnari

doi:10.4401/ag-8049

Alireza Hajian, Filippo Greco + Show 2 more

Open Access

https://doi.org/10.4401/ag-8049

Copy DOI

Abstract

Assessment of the ongoing activity of volcanoes is one of the key factors to reduce volcanic risks. In this paper, two Machine Learning (ML) approaches are presented to classify volcanic activity using multivariate geophysical data, namely the Decision Tree (DT) and K-Nearest Neighbours (KNN). The models were implemented using a data set recorded at Mount Etna (Italy), in the period 01 January 2011 – 31 December 2015, encompassing lava fountain events and intense Strombolian activity. Here a data set consisting of five geophysical features, namely the root-mean-square of seismic tremor (RMS) and its source depth, counts of clustered infrasonic events, radar RMS backscattering power and tilt derivative, was considered. Model performances were assessed by using a set of statistical indices commonly considered for classification approaches. Results show that between the investigated approaches the DT model is the most appropriate for classification of volcano activity and is suitable for early warning systems applications. Furthermore, the comparison with a different classifier approach, reported in literature, based on Bayesian Network (BN), is performed.

Highlights

One of the key challenge to modern volcanology is to identify and characterize volcano activity, based on parameters recorded by the monitoring network that might be useful for hazard assessment and risk mitigation
A wide data set ranging from seismic, geodetic, gravimetry, geochemical, video etc., most of them collected in real time, were able to set strict constraints on the timing of the paroxysmal events which occurred in the last decade at Etna volcano [i.e. Aloisi et al, 2018; Bonaccorso et al, 2011; Greco et al, 2016]
The main purpose of this paper is to implement a classifier capable of recognizing different phases of the volcanic activity which occurred at Mount Etna between 01 January 2011 and 31 December 2015, through a set of geophysical parameters recorded in this area by the multidisciplinary monitoring network (Figure 1)

Summary

Introduction

One of the key challenge to modern volcanology is to identify and characterize volcano activity, based on parameters recorded by the monitoring network that might be useful for hazard assessment and risk mitigation. This topic has maximum priority especially for volcanoes located close to densely urbanized areas, such as Mount Etna volcano, Italy. The characterization of volcanic activity based only on one geophysical and/or geochemical parameter, may be lead to ambiguous forecasts One solution to this problem is through using a combination of different parameters to reduce the level of ambiguity and to enhance the quality of interpretations of the volcano activity

Objectives

Methods

Results