Conceptualization of the Dike Distribution Analysis Aiming at Identification of Eruptive Centers

Introduction

Volcanic eruptions are accompanied by intrusion of dikes that often indicate the location of past eruptive centers. This has led Carniel et al. (2017) to develop an algorithm aiming to identify the location of buried or denudated eruptive centers based on the documented dikes where their density allows to do so. The algorithm was implemented as an R opensource script entitled FIERCE (FInding volcanic ERuptive CEnters) and tried at a number of volcanoes and volcanic areas, both modern and ancient, with encouraging results. Position of the vents computed by this algorithm coincided with the observed (for modern centers) and was in concordance with the distribution of volcanic products and other evidence of their location. Naturally, in every case the reliability of the computed position of eruptive center should be supported by the interpretation of volcanological and geological history of the volcano.

Every time this interpretation was made ad hoc, based on the geological data available, and it seems to be a good luck that so far we have been finding a self-consistent and palpable explanation for the results of computation. Still, this good luck is more or less explained by that (i) we have addressed a very small number of volcanoes and that (ii) almost all of these volcanoes have had a long record of geologic research. Query, what happens if we interpret dozens or hundreds of datasets? Wouldn’t there be some typical interpretations? Wouldn’t different scenarios of geological evolution be potentially responsible for the same dike pattern? Is the diversity of these patterns infinite or it can be somehow constrained?

To answer these questions, one should systematize the existing interpretations and put them in a coherent framework of volcano-tectonic knowledge. To build such framework is the purpose of this work.

Below in this chapter we will briefly overview the FIERCE methodology and the results achieved, suggest a knowledge framework to host these results and discuss its use in volcanology.