Abstract
dc:description.abstractVillarrica volcano is one of Chile's most active volcanoes with over one hundred eruptions since 1558. Moreover, eruptions at Villarrica have a large range in eruption intensity and magnitude, from mafic-ignimbrite-forming eruptions to intense lava fountaining, but this diversity is not reflected in its dominantly basaltic to basaltic andesite erupted products. In this thesis, I use two different methodologies to investigate if this homogeneity is merely a reflection of ‘chemical averaging’ of whole-rock analysis, focusing on complete and transparent error propagation to support its findings. First, I consider the variable compositions of erupted crystal cargoes at Villarrica, utilising unsupervised machine learning techniques to group mineral compositions and relate their complexity to eruptive behaviour using thermodynamic modelling. Mixing between evolved and primitive melts is identified as the driving force for Villarrica's most explosive eruptions. Then I examine the assumption that melt inclusion saturation pressures have low uncertainties compared to traditional mineral-based barometers. Volatile-saturation pressures require multiple analytical techniques and complex processing to account for CO2 in both the melt inclusion glass and vapour bubble, making error propagation non-trivial. Errors in melt inclusion volumes estimated using 2D techniques dominate the total error, leading to comparable errors to classic mineral and mineral-melt barometers. With additional constraints on volume from micro-x-ray tomography or strategic polishing, the errors are several times lower. Finally, I investigate the major-, trace- and volatile-element contents of melt inclusions from four eruptions of Villarrica. Volatile saturation pressures demonstrate that Villarrica's magmatic system is vertically extensive. Trace elements show substantial variety despite homogeneous major elements, but show no trends with eruption. In combination, this strongly supports the application of the trans-crustal magmatic system model at Villarrica, whereby crystals from chemically distinct mushy reservoirs are assembled throughout the crust before their eruption.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Leeds
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Boschetty, Felix Oliver
- Advisors dc:contributor.advisor
-
- Ferguson, David
- Morgan, Daniel
- Ebmeier, Susanna
- ORCID person:identifier.orcid
Rights
- Language dc:language.dcterms_iso639_2
- eng
Identifiers
dc:identifier.*- Dc Identifier Grantnumber
- NE/S007458/1