Project

Back to overview

The architecture of subvolcanic reservoirs

English title The architecture of subvolcanic reservoirs
Applicant Caricchi Luca
Number 162503
Funding scheme Project funding
Research institution Département des sciences de la Terre Université de Genève
Institution of higher education University of Geneva - GE
Main discipline Mineralogy
Start/End 01.12.2015 - 31.03.2017
Approved amount 138'996.00
Show all

All Disciplines (2)

Discipline
Mineralogy
Mathematics

Keywords (4)

Petrology; Volcanology; Magmatic reservoirs; Correlation of chemical profiles

Lay Summary (Italian)

Lead
In sintesiL’attività vulcanica è alimentata da “camere magmatiche” locate a 10-15 km di profondità al di sotto di vulcani attivi. Dati esitenti suggeriscono che questi “contenitori” non sono delle cavità riempite di magma che può eruttare alla superficie (camere magmatiche), ma piuttosto delle regioni in cui volumi discreti di magma potenzialmente eruttabile sono distribuite all’interno di magma più cristallizzato e non eruttabile (reservoir magmatici). Abbiamo solamente un’idea qualitativa della struttura di queste regioni di accumulazione di magma e questo ci impedisce di fare inferenze sulla probabilità di un’eruzione o del suo potenziale volume.
Lay summary

In sintesi
L’attività vulcanica è alimentata da “camere magmatiche” locate a 10-15 km di profondità al di sotto di vulcani attivi. Dati esitenti suggeriscono che questi “contenitori” non sono delle cavità riempite di magma che può eruttare alla superficie (camere magmatiche), ma piuttosto delle regioni in cui volumi discreti di magma potenzialmente eruttabile sono distribuite all’interno di magma più cristallizzato e non eruttabile (reservoir magmatici). Abbiamo solamente un’idea qualitativa della struttura di queste regioni di accumulazione di magma e questo ci impedisce di fare inferenze sulla probabilità di un’eruzione o del suo potenziale volume. 

Soggetto e obiettivo

Il nostro obiettivo è di quantificare i volumi di magma con proprietà fisiche differenti all’interno di reservoir magmatici. Questo ci permetterà di stimare più precisamente il potenziale volume di un’eruzione dall’analisi dei dati di monitoraggio vulcanico.

Analizzeremo le variazioni di composizione chimica di minerali che hanno cristallizzato nel magma durante diverse migliaia di anni prima di una grande eruzione (Kilgore Tuff, USA; 2000 km3 di magma eruttato). La composizione chimica dei minerali cambia in funzione della pressione, temperature e composizione chimica dei magmi dai quali cristallizzano. Quindi i minerali registrano le variazioni delle condizioni fisiche all’interno di un reservoir magmatico nel tempo. Applicheremo un metodo matematico per quantificare in maniera statisticamente significativa le percentuali di minerali che hanno registrato le stesse variazioni fisico-chimiche. Questo ci permetterà di determinare l’architettura del reservoir magmatico che ha alimentato questa grande eruzione.

 

Implicazioni per la società

Una migliore comprensione della struttura delle zone di accumulazione di magma servirà per aumentare la nostra capacità di stimare l’impatto di una potenziale eruzione sulla nostra società e quindi di mitigarne gli effetti.

Direct link to Lay Summary Last update: 02.10.2015

Responsible applicant and co-applicants

Employees

Name Institute

Publications

Awards

Title Year
Prix d’Excellence de l’Association Genevoise des Femmes Diplômées des Universités 2016

Associated projects

Number Title Start Funding scheme
150204 The fate of magma in the Earth’s crust: Plutons and volcanic eruptions 01.04.2014 Project funding
172702 The build-up to volcanic eruptions 01.04.2017 Project funding

Abstract

Recent geophysical and geochemical evidence suggest that magmatic reservoirs are composed of multiple pockets of eruptible magma that were separated for most of their lifetime. These magmatic reservoirs are loci where magmas fractionate by the separation of residual melt from crystals, in addition to evolution by other processes such as assimilation and mixing. These magma reservoirs ultimately feed volcanic eruptions. However, so far, the accumulation of residual melt in these reservoirs has been modelled as a single accumulation in the upper portion of a single reservoir. The recent data call for a revision of our conceptual understanding of the thermal and chemical structure of magma storage regions in the crust. Understanding the magmatic interactions within and between these magma storage regions will inform models for eruption triggering, which are currently based on magma reservoirs with relatively simple geometry. The physical and chemical interactions in these magma reservoirs can be investigated through eruptive products that record time histories of their chemical and physical evolution. The time-evolution of magma volumes can be studied through both the stratigraphy of erupted lavas, and the stratigraphy-like information recorded in crystal zoning patterns. Core to rim variations in mineral chemistry reflect changes in pressure, and/or temperature, and/or magma chemistry occurring over time in a magmatic reservoir. Because these crystals record the evolution of the erupted magma, identification of populations of crystals with similar zoning patterns reflects relative volume of magma with similar chemical histories. Additionally, the variety of zoning profiles and crystal populations is linked to the physical dynamics of the reservoir. Therefore, a statistically analysis of zoned minerals collected at different locations along the stratigraphy of volcanic deposits can reveal the spatial distribution and relative volumes of magma at different thermal and physical conditions within the subvolcanic storage region, and evolution of that system through time. This was the original motivation for which we started developing an automated method for the cross-correlation of chemical profiles in minerals during the master project of Miss Line Probst in collaboration with Prof. Martin Gander (Section of Mathematics, UNIGE) and Dr Glen Wallace. However, a statistically significant number of analyses in minerals collected at different locations along the stratigraphy of volcanic deposits are required to determine the spatial distribution and relative volumes of magma at different thermal and physical conditions within the subvolcanic storage region. This was the original motivation for which we started developing an automated method for the analyses of chemical profiles in minerals during the master project of Miss Line Probst in collaboration with Prof. Martin Gander (Section of Mathematics, UNIGE) and Dr Glen Wallace. We selected different eruptions from two volcanic systems to carry out this project:1)The Kilgore Tuff eruption vented 1800 km3 of crystal poor magma (DRE=Dense Rock Equivalent) and generated a caldera of more than 50 km diameter. For this system zircon geochemistry and geochronology provide evidences for the presence of isolated magma pockets that ultimately merged and fed the eruption. However, both the relative proportions of such eruptible subvolumes and their spatial arrangement are unknown. 2)The 3 main eruptions of Cerro Machin volcano in Colombia. The first eruption of this volcano lead to the formation of a caldera of 3 km diameter and 2 main eruptions of 1 to 5 km3 DRE followed. Magmas are crystal rich (30-40 vol.% crystals) and, for this system no data exist to assess if magma feeding the eruption were extracted from a thermally chemically homogeneous reservoir or if the storage region was as complex as that from which the Kilgore tuff originated.These systems are chosen because they are both associated with caldera formation, but the size of the systems is different, they are located in different tectonic settings, and the erupted magmas have different chemistry and crystallinity; additional reasons and details are explained in detail in the following sections. We will collect samples along stratigraphic profiles at targeted locations around the volcanic systems. Mineral separates will be prepared from each stratigraphic level.The cross correlation of mineral profiles will utilize profiles from catholuminescence images, back scattered electron images, and elemental maps for plagioclase, feldspars, quartz, zircons and pyroxenes that are present in the erupted products. Once the populations of minerals are identified from the cross correlation of their zoning profiles, we will perform major, trace elements, and isotopic core-to-rim analyses of selected minerals from the identified populations to bolster the petrological interpretation, characterise the thermal, chemical, and physical evolution of these magmatic systems, and to identify the progression of events that finally triggered the eruptions.
-