Combining machine learning techniques, microanalyses and large geochemical datasets for tephrochronological studies in complex volcanic areas: New age constraints for the Pleistocene magmatism of central Italy

Abstract

Characterization, correlation and provenance determination of tephra samples in sedimentary sections (tephrochronological studies) are powerful tools for establishing ages of depositional events, volcanic eruptions, and tephra dispersion. Despite the large literature and the advancements in this research field, the univocal attribution of tephra deposits to specific volcanic sources remains too often elusive. In this contribution, we test the application of a machine learning technique named Support Vector Machine to attempt shedding new light upon tephra deposits related to one of the most complex and debated volcanic regions on Earth: the Pliocene-Pleistocene magmatism in Italy. The machine learning algorithm was trained using one of the most comprehensive global petrological databases (GEOROC); 17 chemical elements including major (SiO O 5 ) and selected trace (Sr, Ba, Rb, Zr, Nb, La, Ce) elements were chosen as input parameters. We first show the ability of support vector machines in discriminating among different Pliocene-Pleistocene volcanic provinces in Italy and then apply the same methodology to determine the volcanic source of tephra samples occurring in the Caio outcrop, an Early Pleistocene sedimentary section located in Central Italy. Our results show that: 1) support vector machines can successfully resolve high-dimensional tephrochronological problems overcoming the intrinsic limitation of two-and three-dimensional discrimination diagrams; 2) support vector machines can discriminate among different volcanic provinces in complex magmatic regions; 3) in the specific case study, support vector machines indicate that the most probable source for the investigated tephra samples is the so-called Roman Magmatic Province. These results have strong geochronological and geodynamical implications suggesting new age constraints (1.4 Ma instead of 0.8 Ma) for the starting of the volcanic activity in the Roman Magmatic Province.

Machine learning classifiers for attributing tephra to source volcanoes: an evaluation of methods for Alaska tephras

“…The relatively poor performance of SVM prob. on this dataset was unexpected given the high accuracy of a raw (non‐probabilistic) model of similar design (>0.90) that discerned volcanic sources in Italy (Petrelli et al ., ). Even averaging the predicted probabilities per sample, the predictive accuracy of SVM prob.…”

“…was hardly better than considering each shard independently (0.630 vs 0.628). However, Petrelli et al (2017) used P 2 O 5 and trace elements as predictors in addition to major elements, which undoubtedly contributed to their SVM's accuracy. Interestingly, the comparatively poor SVM prob.…”

“…Supervised classifiers for tephra correlation are rare. However, several examples have shown that the concept is viable for tephra classification, including applications of linear discriminant analysis (LDA) (Beaudoin and King, 1986;Stokes et al, 1992;Shane and Froggatt, 1994;Bourne et al, 2010), and support vector machines (SVM) (Petrelli et al, 2017).…”

Advancing tephrochronology as a global dating tool: Applications in volcanology, archaeology, and palaeoclimatic research

“…Others have developed further statistical approaches, such as simple correlation coefficients or multivariate methods using discriminant function analysis, principal component analyses, and kernel density estimates (Begét et al, 1991;Pollard et al, 2006;Pouget et al, 2014;Ramsey et al, 2015). Petrelli et al (2017) explore a novel machine-learning approach for statistical discrimination between Pleistocene Italian tephra. Machine learning is a term derived from the computing literature that refers to statistical methods such as cluster analysis or discriminant analysis.…”

“…Tomlinson et al, 2015), statistical methods for distinguishing multivariate datasets (e.g. Pouget et al, 2014;Blegen et al, 2015;Bronk Ramsey et al, 2015;Petrelli et al, 2017), new methods for visual assessment of tephra depositional processes and taphonomy (e.g. Griggs et al, 2014Griggs et al, , 2015Hopkins et al, 2015;Zawalna-Geer et al, 2016), and means to extract maximum information from even the finest cryptotephra deposits (e.g.…”