Advances in the Quantitative Characterization of the Shape of Ash-Sized Pyroclast Populations: Fractal Analyses Coupled to Micro- and Nano-Computed Tomography Techniques

Abstract:

The quantification of shape parameters in pyroclasts is fundamental to infer the dominant type of magma fragmentation (magmatic vs. phreatomagmatic), as well as the behavior of volcanic plumes and clouds in the atmosphere.

In a case study aiming at reconstructing the fragmentation mechanisms triggering maar eruptions in two geologically and compositionally distinctive volcanic fields (West and East Eifel, Germany), the shapes of a large number of ash particle contours obtained from SEM images were analyzed by a dilation-based fractal method. Volcanic particle contours are pseudo-fractals showing mostly two distinct slopes in Richardson plots related to the fractal dimensions D1 (small-scale “textural” dimension) and D2 (large-scale “morphological” dimension). The validity of the data obtained from 2D sections was tested by analysing SEM micro-CT slices of one particle cut in different orientations and positions. Results for West Eifel maar particles yield large D1 values (> 1.023), resembling typical values of magmatic particles, which are characterized by a complex shape, especially at small scales. In contrast, the D1 values of ash particles from one East Eifel maar deposit are much smaller, coinciding with the fractal dimensions obtained from phreatomagmatic end-member particles. These quantitative morphological analyses suggest that the studied maar eruptions were triggered by two different fragmentation processes: phreatomagmatic in the East Eifel and magmatic in the West Eifel. The application of fractal analysis to quantitatively characterize the shape of pyroclasts and the linking of fractal dimensions to specific fragmentation processes has turned out to be a very promising tool for studying the fragmentation history of any volcanic eruption.

The next step is to extend morphological analysis of volcanic particles to 3 dimensions. SEM micro-CT, already applied in this study, offers the required resolution, but is not suitable for the analysis of a large number of particles. Newly released nano CT-scanners, however, allows the simultaneous analysis of a statistically relevant number of particles (in the hundreds range). Preliminary results of a first trial will be presented.