Image processing analysis in reconstructing fragmentation and transportation mechanisms of pyroclastic deposits. The case of Monte Pilato-Rocche Rosse eruptions, Lipari (Aeolian islands, Italy)

Abstract A quantitative study of particle morphology of samples from Monte Pilato-Rocche Rosse deposits, Lipari (Aeolian islands, Italy) shows a significant variability in particle structure and texture. The study consisted of image processing techniques and multivariate statistical methods in analysis of the quantitative particle morphology data, with the goal of identifying the genetic processes responsible for the formation of the deposits. From these investigations we numerically define the form of the clasts, thus avoiding subjectivity in the classification of clast morphology. Starting from this clast definition, we then construct diagrams in which clasts due to different genetic processes (i.e., magmatic vs. hydromagmatic fragmentation), are discriminated based on simple nondimensional parameters. The application of the methods used seems to be of general interest in discriminating the fragmentation and transportation dynamics responsible for the formation of pyroclastic deposits.