Apr 5, 2016 - Textural analysis have been carried out on pumice samples produced ...... PSD results show that all samples analyzed are mesoporous (4â50 nm). ...... Images were cleaned and made binary using Adobe Photoshop CS3 and.
Thesis
Morphology and microtextures of tephra particles
ALFANO, Fabrizio
Abstract The purpose of this PhD thesis is to give insights into the dynamics of highly explosive eruptions and on both tephra dispersal and sedimentation, based on the characterization of the erupted material. A systematic and detailed morphological characterization of tephra particles from different eruptions and the calculation of their terminal fall velocity is carried out in order to give new insights in understanding the influence of the morphology of volcanic particles on the settling process. The May 2008 Chaitén eruption has been taken as a case study. The eruption has been studied based on the data collected in the field on January 2009 in order to describe its stratigraphy and to determine the main eruptive parameters (i.e., volume, column height, mass eruption rate). On May 6th the eruption reached its climax producing a tephra layer composed mainly of lithic lapilli and blocks (> 2 mm) and a 19 km high subplinian column. Textural analysis have been carried out on pumice samples produced in this phase and a fractal analysis of the associated ash have also been conducted in order to study the influence of the [...]
Reference ALFANO, Fabrizio. Morphology and microtextures of tephra particles. Thèse de doctorat : Univ. Genève, 2011, no. Sc. 4374
Available at: http://archive-ouverte.unige.ch/unige:18489 Disclaimer: layout of this document may differ from the published version.
[ Downloaded 04/05/2016 at 03:18:29 ]
Université de Genève Dèpartment de Minéralogie
Faculté Des sciences Prof. Costanza Bonadonna
Morphology and Microtextures of Tephra Particles Thèse Présentée à la Faculté des Sciences de l’Université de Genève pour obtenir le grade de Docteur ès sciences, mention Sciences de la Terre
par
Fabrizio ALFANO de San Gavino Monreale (Italie)
Thèse n°4374
Genève Atelier de reprographie ReproMail 2012
Ai miei genitori
Acknowledgements
First of all, I would like to acknowledge the financial support I had from the Master and Back program (grant code MAB 2.2A2005-219) of the Regione Autonoma della Sardegna (Italy), by the Department of Mineralogy of the University of Geneva and by the Société Physique et d’Histoire Naturelle (SPHN) society of Geneva that founded the fieldwork on the Chaitén area. I am also grateful to the Fondation Schmidheiny for the financial support for my participation in many international congress and workshop. Very special thanks to prof. Raffaello Cioni, without whom I would have not been here in Geneva. Many thanks to Chuck and Laura Connor, Alain Volentik, David Pyle, Sebastian Watt, Licia Costantini, Pierre Delmelle and Lucia Gurioli for their support and collaboration and for giving me an important help to improve my knowledge and learn new techniques and approaches in studying volcanoes. Special thanks to prof. Bernard Grobety and Dr. Daniel Aritzegui for being in my PhD commission and for their suggestion and hints about my research work. I wish to thank Laura, Irene, Riccardo, Licia, Wim, Andrea, Sebastian, Cristina, Linda, Mohssen, Caroline, Flora, Kae and all the friends and colleagues of the Section de Sciences de la Terre et de l’Environnement for nice time, useful and interesting discussion and, moreover, for all the coffee breaks we could share together. Many thanks to all the person that have crossed my path. To my family, and in particular to my mother and my father that have supported me and made me who I am, to Andrea, for our long and brotherly friendship, and to Alişya, for her lovely support. Last but not least, I am deeply indebted with my advisor prof. Costanza Bonadonna for all the opportunities she provided during my PhD, for fieldworks and conferences, for helping me find the path for my future, for patience, disponibility, support and for having pushed me to improve myself and opening my mind to new perspectives.
Table of contents
Summary
v
Résumé
vii
Riassunto
ix
Chapter 1: Introduction
1
1.1
Explosive volcanism
1
1.2
Overview on the main topics discussed in this thesis
5
References
Chapter 2:
9
Insights on tephra settling morphological observations
velocity
from
11
2.1
Introduction
11
2.2
The shape of tephra particles and its influence on terminal fall velocity
12
2.3
Tephra samples
13
2.4
Morphological characterization of tephra particles
14
2.4.1
Density measurements
14
2.4.2
2D image analysis
15 i
2.5
2.6
2.7
2.4.3
Gas adsorption
15
2.4.4
3D scan
17
2.4.5
Particle morphological parameters
17
Results
20
2.5.1
Morphological characterization
20
2.5.2
Pore size distribution
22
2.5.3
Terminal fall velocity characterization
24
Discussion
31
2.6.1
Morphological characterization
31
2.6.2
Terminal fall velocity of the particles
33
Conclusions
35
Appendix
38
References
50
Chapter 3: The May, 2008, Chaitén eruption
55
3.1
Introduction
55
3.2
The tephra fall deposit
57
3.2.1
Stratigraphy of the southeast sector
60
3.2.2
Stratigraphy of the north sector
61
3.3
Chronology of the eruption and correlation with tephra dispersal
64
3.3.1
64
3.3.2 3.3.3
3.4
Chronology of May 2008 Events Correlation of tephra layers with the chronology of the eruption The total deposit
Results
66 67
67
ii
3.5
3.6
3.4.1
Tephra chemistry
67
3.4.2
Volume of the deposits
69
3.4.3
Eruptive parameters and classification
73
Discussion
76
3.5.1
Eruptive parameters
76
3.5.2
Eruption style
78
3.5.3
Hazard implications
79
Conclusions
80
Appendix
82
References
88
Chapter 4:
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
93
4.1
Introduction
93
4.2
The May 2008 Chaitén eruption
96
4.3
Samples and methods
98
4.3.1
Density and vesicularity of pumices
98
4.3.2
Textural analysis on pumices
98
4.3.3
Fractal analysis on ash particles
99
4.4
4.5
Results
100
4.4.1
Density measurements
100
4.4.2
Description of the thin sections
101
4.4.3
Vesicle size distribution
106
4.4.4
Fractal analisys on ash particles
113
4.4.5
Decompression rate
114
Discussion
114
4.5.1
114
Interpretation of the textures iii
4.6
4.5.2
Influence of vesicularity on particle morphology
116
4.5.3
Insights into eruption dynamics from textural observations
116
Conclusions
119
Appendix
121
References
156
Chapter 5: Concluding remarks
163
References
168
iv
Summary
Explosive volcanism is an important and complex natural phenomenon. Typically, explosive eruptions involve viscous magmas, rich in silica and volatiles, which violently fragment producing several complex phenomena (e.g., pyroclastic density currents, tephra fallout, lahars) that represent a serious source of risk and produce disruption to critical human activities. Understanding the dynamics of these phenomena represents a crucial issue in order to build useful tools to be used for hazard forecasting and risk analysis, management and mitigation. In particular, the study of tephra fallout, ranging from the analysis of the morphological and textural characteristics to the stratigraphic study of the deposits, can provide a large amount of information on the dynamics of the eruption and on the most relevant eruptive parameters. The purpose of this PhD thesis is to give insights into the dynamics of highly explosive eruptions and on both tephra dispersal and sedimentation, based on the characterization of the erupted material. A systematic and detailed morphological characterization of tephra particles from different eruptions and the calculation of their terminal fall velocity (TFV) is carried out. This study gives new insights in understanding the influence of the morphology of volcanic particles on the settling process. According to our results, 2D morphological characterization of volcanic particles is a fast and simple application for a wide range of particle size. 3D scanning also provides a promising tool for lapillisized tephra (> 2 mm). In contrast, gas-adsorption-derived surface area is not suitable for the calculation of TFV of volcanic particles mostly because it mainly describes the surface contribution of nanometric pores that are not expected to affect significantly TFV and because bulk-sample analysis is representative of neither individual particles nor of the whole particle population. As a result, the influence of particle shape on TFV increases with particle size. In particular, calculated TFV converges at small particle sizes (≥ 125 m), suggesting that the spherical assumption is appropriate for this size fraction.
v
Summary The May 2008 Chaitén eruption has been taken as a case study. The eruption has been studied based on the data collected in the field on January 2009 in order to describe its stratigraphy and to determine the main eruptive parameters (i.e., volume, column height, mass eruption rate). The activity was characterized by several explosive events each associated with plumes which reached up to about 19 km above sea level. The products were dispersed across a wide area, with the finest ash reaching the Atlantic coast of Argentina. Field observations in the proximal-medial area (3–25 km from the vent) indicate that the May 2008 tephra deposit consists of numerous layers variyng from extremely fine grained ash to layers of lapilli and blocks, composed of both juvenile and lithic material. The deposits are mostly associated with the three main explosive phases which occurred on 1st–2nd May, 3rd–5th May and 6th May, with an estimated bulk tephra volume of 0.5–1.0 km3. The 6th May event, represents the climactic phase of the Chaitén eruption, and produced a tephra layer composed mainly of lithic lapilli and blocks (> 2 mm). It is associated with a 19 km high subplinian column. Textural analysis have been carried out on pumice samples in order to study their vesicularity and the vesicle size distribution. A fractal analysis of the associated ash have also been conducted in order to study the influence of the vescicularity on particle morphology. Pumices show unimodal density distribution (400-1200 kg/m3; main mode at 600 kg/m3), high-density tail and vesicularity of 73%. They have glassy groundmass with no microcrystals and vesicles with dimension in the range 0.001-2.000 mm. As a result, vesicle number density of the products of this explosive event is estimated to be 1.3 ± 0.5 mm-3. Vesicularity have a different influence on particle morphology in relation to their size. The fractal analysis on ash particle show higher values of the fractal dimension for particle with diameters in the range 120-160 m. The unimodal size distribution of the vescicles and their relatively small dimensions suggest a rapid nucleation event produced during the late phases of magma rise. Vesciculation and fragmentation have been produced by a sudden decompression of the melt which is estimated to be about 10 MPa s-1, consequent to the opening of the volcanic conduit.
vi
Résumé
Le volcanisme explosif est un phénomène naturel complexe et pouvant comporter d’importantes conséquences. Ce type d’éruptions est généralement lié à la fragmentation violente de magmas visqueux riches en silicates et en éléments volatifs résiduels, pouvant aboutir à la formation d’aléas (coulées pyroclastiques, retombées de cendres, lahars) représentant un risque important pour les populations. Comprendre la dynamique de ces phénomènes est une tache importante afin d’établir des outils nécessaires pour l’évaluation et la réduction des aléas et du risque en résultant. En particulier, l’étude des tephra – allant de l’analyse des caractéristiques morphologiques et texturales à la description des dépôts sur le terrain – fournit quantités d’informations sur la dynamique d’éruption. Le but de ce doctorat est de donner un aperçu de la dynamique des éruptions explosives ainsi que sur la dispersion et la sédimentation des tephra, basé sur caractérisation des matériaux produits. Une étude approfondie et systématique de la morphologie de particules de tephra provenant de différentes éruptions ainsi que le calcul de leurs vitesses de chute limite (VCL) est conduite. Cette analyse fournit des nouveaux indices pour mieux comprendre l’influence de la morphologie des particules volcaniques sur leurs processus de déposition. Nos résultats suggèrent qu’une caractérisation morphologique des particules en 2D est une méthode simple et rapide pour une grande variété de particules. L’analyse en 3D représente un outil promettant pour les lapilli (taille > 2 mm). En revanche, la mesure de la superficie utilisant les méthodes de adsorption de gaz n’est pas adaptée au calcul de VCL pour deux raisons ; premièrement, cette méthode évalue la contribution de pores nanométriques, n’ayant pas ou peu effet sur la VCL; deuxièmement, l’analyse de l’ensemble d’un échantillon n’est représentatif ni de particules individuelles ni d’une population entière. Au final, l’influence de la morphologie des particules sur la VCL augmente avec la taille des particules, et les VCL calculées convergent pour les plus petites particules (≥ 125 m), ce qui suggère qu’une approximation à une sphère est appropriée pour cette fraction de taille.
vii
Résumé L’éruption de Chaitén en mai 2008 a été utilisée comme cas d’étude. Cette éruption a été étudiée sur la base de données collectées sur le terrain en Janvier 2009 afin de décrire la stratigraphie et de déterminer les paramètres éruptifs (volume, hauteur du panache, taux d’éruption). Cet épisode éruptif a été marqué par plusieurs explosions provoquant des panaches pouvant atteindre jusqu’à 19 km de hauteur. Les cendres se sont dispersées sur de larges zones, la partie la plus fine atteignant la côte Atlantique d’Argentine. Les observations sur le terrain dans la zone proximale (3-25 km du cratère) indiquent que de nombreuses couches de granulométrie variable (de cendres aux lapilli et bloques) résultent de l’éruption de mai 2008. Les dépôts sont pour la plupart le résultat de trois phases explosives dominantes, ayant eu lieu les 1-2 mai, 3-5 mai et le 6 mai, avec un total volume de tephra estimé à 0.5-1 km3. L’évènement du 6 mai représente le point culminant de l’éruption de Chaitén et a produit une couche de tephra composée principalement de lapili et de bloques lithiques (> 2 mm), associée à un panache plinien d’une hauteur de 19 km. Des analyses texturales ont été faites sur des échantillons de ponces afin d’évaluer leur vésicularité et la distribution de la taille des vésicules. Une analyse fractale des cendres a aussi été effectuée afin d’évaluer l’influence de vésicularité sur la morphologie des particules. Les ponces sont caractérisées par une distribution uni-modale de leur densité (400-1200 kg/m3 ; mode principal à 600 kg/m3) et une vésicularité de 73%. Elles ont matrice riche en verre dépourvue de microcristaux, et des vésicules de tailles variant entre 0.0012.000 mm. Au final, la densité de vésicules de cet évènement éruptif est estimée à 1.3 ± 0.5 mm-3. En revanche, la vesicularité a un different impact sur la morphologie des particules en fonction de la taille. L’analyse fractale sur les particules de cendres présente des valeurs de dimension fractale plus élevées pour les particules avec des diamètres de 120-160 m. La distribution uni-modale de la distribution de taille des vésicules et leurs tailles relativement petite suggère un processus de nucléation rapide et ayant eu lieu durant la phase finale de l’ascension du magma. La vésiculation et la fragmentation ont été déclenchées par une décompression soudaine du liquide magmatique, estimée à 10 MPa s-1, après l’ouverture du conduit volcanique.
viii
Riassunto
Il vulcanismo esplosivo è un fenomeno naturale molto importante e complesso. In genere è prodotto da magmi viscosi, ricchi in silice e specie volatili, che frammentano violentemente dando luogo a diversi fenomeni (es., flussi piroclastici, caduta di tefra, lahar) che possono rappresentare un serio rischio ed influenzare seriamente diverse importanti attività umane. La comprensione delle dinamiche che governano questi fenomeni è molto importante al fine di ottenere strumenti utili per la stima del hazard e per l’analisi, la gestione e la mitigazione del rischio ad essi connesso. In particolare, lo studio dei processi di dispersione e deposizione dei tefra, dalla caratterizzazione morfologica e tessiturale dei singoli piroclasti allo studio della stratigrafia dei depositi, può fornire una enorme quantità di informazioni sulle dinamiche di eruzione e sui più rilevanti parametri eruttivi che le caratterizzano. L’obiettivo di questa tesi di dottorato è di fornire nuove conoscenze sulle dinamiche che governano le eruzioni ad elevata esplosività e sui processi di dispersione e sedimentazione dei tefra sulla base della caratterizzazione dei materiali eruttati. È stato effettuato uno studio sistematico e dettagliato della morfologia dei frammenti di tefra prodotti da diverse eruzioni, e ne è stata calcolata la velocità terminale di caduta (VTC). Questo studio fornisce un incremento delle conoscenze sulla influenza della morfologia dei frammenti vulcanici sul loro processo di deposizione. Sulla base dei risultati, la caratterizzazione morfologica 2D dei frammenti vulcanici rappresenta un metodo semplice e rapido che può essere applicato a particelle in un ampio intervallo di dimensione. Anche la caratterizzazione morfologica 3D rappresenta un metodo promettente per l’analisi di frammenti di tefra di dimensione superiore ai lapilli (> 2 mm). Al contrario, l’applicazione del metodo della gas-adsorption nella misura della superficie delle particelle non è adatta al calcolo della VTC delle particelle vulcaniche, in quanto la misura include elementi della superficie e pori di dimensione nella scala dei nanometri, che non si considera abbiano una significativa influenza sulla VTC, ma anche perché fornisce analisi sul campione totale che non possono essere considerate rappresentative di particelle individuali o della intera popolazione di ix
Riassunto particelle che lo compongono. Si conclude che l’influenza della morfologia delle particelle sulla loro VTC aumenta all’aumentare della loro taglia. In particolare, la VTC calcolata converge per particelle di piccole dimensioni (≥ 125 m), suggerendo che per questo intervallo di taglia la morfologia può essere approssimata alla forma di una sfera. L’eruzione del vulcano Chaitén iniziata nel maggio 2008 è stata presa in considerazione. L’eruzione è stata studiata sulla base dei dati raccolti sul terreno nel gennaio 2009, con i quali è stata ricostruita la stratigrafia dei depositi e sono stati determinati i principali parametri eruttivi (es., volume di materiale eruttato, altezza della colonna, flusso di massa eruttata). L’attività eruttiva si è sviluppata attraverso diverse fasi esplosive associate a colonne eruttive che hanno raggiunto un’altezza massima di 19 km sopra il livello del mare. I prodotti sono stati dispersi su un’ampia area, e le ceneri fini hanno potuto raggiungere le coste atlantiche dell’Argentina. Le osservazioni di terreno nelle aree prossimali-mediali (3-25 km dal cratere) indicano che il deposito di tefra prodotto nel maggio 2008 consiste di numerosi livelli composti da frammenti di dimensione variabile da ceneri sottili, a lapilli e blocchi, di origine juvenile e litica. I depositi sono principalmente associati a tre fasi eruttive principali avvenute il 1-2 maggio, 3-5 maggio ed il 6 maggio, eruttando una quantità di tefra stimata tra 0.5 e 1.0 km3. L’esplosione del 6 maggio rappresenta la fase climatica dell’eruzione del Chaitén, e ha prodotto un livello di tefra composto principalmente da lapilli e blocchi litici (> 2 mm). Questa fase eruttiva è associata ad una colonna eruttiva subpliniana di 19 km. Le tessiture dei campioni di pomici prodotte durante questa fase eruttiva sono state analizzate al fine di studiare la vescicolarità di questi prodotti e la distributzione di taglia delle bolle. In aggiunta, le ceneri associate a questa fase eruttiva sono state sottoposte ad una analisi frattale al fine di studiare l’influenza della vescicolarità sulla loro morfologia. Le pomici sono caratterizzate da una distribuzione di densità unimodale (400-1200 kg/m3; moda a 600 kg/m3), con una coda di alta densità e una vescivolarità media del 73%. La pasta di fondo è vetrosa e priva di microcristalli e le bolle presentano dimensioni comprese tra 0.001 e 2.000 mm. Il numero di bolle per unità di volume è stimato a circa 1.3 ± 0.5 mm-3. La vescicolarità ha una influenza sulla morfologia che dipende dalla taglia del frammento considerato. L’analisi frattale sulle particelle di cenere mostra che le particelle di dimensione compresa tra 120-160 m sono caratterizzate da una dimensione frattale più elevata. La distribuzione di taglia x
Riassunto
unimodale delle bolle e le loro piccole dimensioni suggeriscono un evento di nucleazione rapido prodotto durante le fasi tardive della risalita del magma. La vescicolazione e la frammentazione si sono verificate in seguito ad un improvvisa decompressione del fuso magmatico stimata a circa 10 Mpa s-1, conseguente all’apertura del condotto vulcanico.
xi
Chapter 1
Introduction
1.1 – Explosive volcanism Explosive volcanism is a complex phenomenon which can have a significant impact on human society. It is characterized by the production and ejection of large amounts of gas and pyroclastic fragments originated by both the fragmentation of fresh magma (juvenile) and by the disruption of preexisting rocks (lithic) [Cas and Wright, 1987]. Depending on the degree of explosivity, volcanic eruptions can be classified in several typologies, ranging from Plinian eruptions, involving high viscosity magma characterized by a high degree of fragmentation and associated with eruptive column up to 40 km high, to Hawaiian eruptions, involving low viscosity magma and a low degree of fragmentation, and producing eruptive columns, called lava fountains, not higher than 300 m [Pyle, 1989; Walker, 1973]. Explosive eruptions are characterized by a specific process called “magmatic fragmentation”. This process involves the disruption of the magma rising up through the volcanic conduit, which transforms from a liquid system containing gas bubbles and crystals, to a gas with dispersed magma drops and solid particles [Cashman et al., 2000]. Fragmentation depends on several factors, such as the rheology and composition of the magma, the characteristics of the conduit, the ascent rate and the flow dynamics of the magma rising [Cashman et al., 2000; Cioni et al., 2000]. The results of this process is the production of volcanic particles which are solid fragments of the rising 1
Chapter 1 magma body, and that are characterized by dimensions ranging from less than a micron (ash, d < 2mm), to few millimiters (lapilli, d < 64 mm) up to a few meters (bombs, d > 64 mm) [Cas and Wright, 1987]. Explosive eruptions are catastrophic events and can be the source of several hazards.
Fig. 1.1. a) A pyroclastic density current on June 23, 1993, at Unzen volcano in southern Japan (photo by Setsuya Nakada, 1993; Kyushu University). b) Pyroclastic density current after the June 17, 1991 Pinatubo (Philippines) eruption (photo by Alberto Garcia Corbis). c) pyroclastic density current deposit, Montserrat, February 11, 2009 (photo by Montserrat Volcano Observatory). d) Effects of a pyroclastic density current on some buindings, Montserrat, February 11, 2009 (photo by Montserrat Volcano Observatory).
The most destructive phenomena associated with explosive eruptions are pyroclastic density currents (PDC; Fig. 1.1), which consist of a mixture of volcanic particles and gas that move rapidly on the ground under the influence of gravity. These currents have a runout that can vary from few kms to more than 100 kms. They are highly destructive being able to devastate the areas they cross and are responsible of the death of more than 10 thousand people since 1700 [Connor et al., 2009; Druitt, 1998]. They can origin from gravitational collapse of an eruptive column, by the fallback of pyroclastic fountains, by lateral blasts or by lava domes gravitational or explosive collapse [Branney and Kokelaar, 2002; Connor et al., 2009; Druitt, 1998; Sparks et al., 1978].
2
Introduction
Fig. 1.2. a) Plume produced by the 1991 Mount Pinatubo (Philippines) eruption (photo by USGS). b) Ashfall from the 1991 Mount Pinatubo (Philippines) eruption that caused collapse of the structure at Clark Air Base (photo by R. Batalon, US Air Force). c) Ballistic bomb from the 188889 eruption of Vulcano La Fossa (Italy) deforming pre-existing tephra layers. d) Sublpinian pumice fall deposit from one of the post AD 1000 events of Vulcano La Fossa (Italy). e) Ashfall deposit from the 2008 Chaitén (Chile) eruption.
A second process related to explosive eruption is tephra dispersal and sedimentation (Fig. 1.2). As an explosive eruption occurs, magma fragments and gas are injected into the atmosphere. These fragments can follow ballistic trajectories or they can be carried up producing a volcanic plume that can rise into higher levels of the atmosphere and spread laterally for hundreds or thousands of kilometers, and stay suspended for months or years before sedimenting. The term tephra indicates the totality of the volcanic fragments that fall through the atmosphere and sediment on the ground, regardless of their dimension or composition [Thorarinsson, 1944]. Tephra dispersal and sedimentation is a very complex phenomenon that can affect wide areas. Tephra dispersal and sedimentation depend on many factors, such as the characteristics of the eruption (i.e., height of the column, volume of magma erupted, mass eruption rate), the characteristics of the fragments (i.e., density, size, shape) and the characteristics of the atmosphere (i.e., vertical density profile, wind speed and direction). As the distance 3
Chapter 1 from the vent increases, the dimension of the sedimented particles decreases, and the finest ash particles are able to travel far from the vent, over distances larger than several hundred km [Bonadonna and Phillips, 2003; Bonadonna et al., 1998; Bursik et al., 1992; Sparks et al., 1992]. Although tephra fallout does not represent an immediate threat for human life, it can produce several problems to human activities. The possibility of tephra to stay suspended into the atmosphere and spread over vast region can affect seriously the aviation system, as happened during the last eruption of Eyjafjallajokull (Iceland) in 2010 and of Puyehue-Cordo Caulle (Chile) in 2011. Sedimentation of tephra particles can damage critical infrastructures and buildings due to the accumulation of the material over building roofs, as they can charge the structure and eventually cause it to collapse, or they can create the disruptuin of vegetation and crops and contaminate water supplies. But also they can create health problems to people, as the fine ash can affect the human respiratory system and to cattle, that can breathe volcanic ash and also eat ash deposited on the ground. Proximal areas are also affected by ballistic bombs that, due to their kinetics energy, can cause serious damage to bundlings and infrastructures and represent a serious threat for people’s safety [Blong, 1984; Cas and Wright, 1987; Connor et al., 2009].
Fig. 1.3. Example of the damages and the deposit produced by a lahar that invested Chaitén on May 2008 through the remobilization of the tephra deposit produced by the eruption.
Pyroclastic materials can represent a potential hazard also after their deposition (Fig. 1.3). In fact, pyroclastic deposits can be highly unstable and be subjected to remobilization producing debris flows and lahars. These phenomena consist of the 4
Introduction
production of a mixture of rocks, water, delaminated soil and vegetation that flows on the surface driven by the gravity force, reaching velocities up to 10-20 m/s and reaching high distances downstream up to more than 150 km [Connor et al., 2009; Iverson et al., 1998]. Understanding volcanic processes is a key challenge, necessary to obtain the basic knowledge to develop useful tools for forecasting the consequences of volcanic explosive eruptions and to achieve successful hazard assessments and mitigation of volcanic risk. Among all these volcanic processes tephra deposits retain a large amount of information that can help characterizing volcanic eruptions [Bonadonna and Houghton, 2005; Carey and Sparks, 1986; Pyle, 1989; Sparks et al., 1992; Wilson et al., 1978]. This thesis focuses on detailed studies of tephra deposits for a better understanding of both the interaction between fragmentation and ash generation and settling processes. This study has been carried out under the supervision of Prof. Costanza Bonadonna, and in collaboration with experts in the characterization of volcanic particles, and in particular on gas adsorption measurements (Dr. Pierre Delmelle and Dr. Licia Costantini), on the study of internal textures of pumices (Dr. Lucia Gurioli), and experts in the field characterization (Prof. Chuck Connor, Prof. David Pyle, Dr. Alain Volentik and Dr. Sebastian Watt).
1.2 – Overview on the main topics discussed in this thesis This thesis is divided in 4 main parts (chapters 2 to 5) preceded by a short introduction (chapter 1): Chapter 2 is a paper published in Journal of Volcanology and Geothermal research [Alfano F., Bonadonna C., Delmelle P., Costantini L., in press]. In this study we present a systematic and detailed morphological characterization of tephra particles from different eruptions (Fontana Lapilli, Masaya, Nicaragua; Keanakāko’i Formation, Kilauea, USA; recent dome explosions of Soufriere Hills volcano, Montserrat) and the calculation of their terminal fall velocity (TFV) as obtained based on different drag prediction models [i.e., Dellino et al., 2005; Ganser, 1993; Haider and Levenspiel, 1989; Wilson and Huang, 1979]. In particular, particle sphericity, and, therefore, 5
Chapter 1 particle surface area, is essential for the calculation of TFV of irregular-shape particles but is complex to determine. Various attempts have been proposed. According to our results, 2D morphological characterization of volcanic particles is a fast and simple application for a wide range of particle sizes and provides consistent sphericity and settling-velocity values. 3D scanning also provides a promising tool for lapilli-sized tephra (> 2 cm). In contrast, gas-adsorption-derived surface area is not suitable for the calculation of TFV of volcanic particles mostly because it mainly describes the surface contribution of nanometric pores that are not expected to affect significantly TFV and because bulk-sample analysis is representative of neither individual particles or of the whole particle population. Settling velocities calculated using values of surface area derived from gas adsorption analyses are up to two orders of magnitude lower than the values obtained through 2D analysis. Our results also show how the influence of particle shape on TFV increases with particle size. In particular, calculated TFV converges at small particle sizes (≥ 3 ) regardless of the model applied, suggesting that the spherical assumption is appropriate for this size fraction (discrepancies with the spherical model are within 10 %). Discrepancies with the spherical model increase with particle size up to about 50 % and depend on the choice of both the TFV model and the morphological parameterization used. In particular, the drag prediction model of Ganser (1993) is sensitive to the effect of particle morphology on TFV and is well suited for all sizes and Reynolds numbers of typical tephra particles. Finally, our results show how individual size categories (whole- and half-) are not associated with individual TFV values but with a range of values, which increases with class size. Nonetheless, the half system is associated with a smaller standard deviation than the whole- system, and is therefore more appropriate for the modeling of tephra dispersal. In any case, for dispersal modeling purposes, it is more appropriate to indicate a range of settling velocities for each size class rather than giving an average value. Chapter 3 is a paper published in Bullettin of volcanology which presents the results of the characterization of the May, 2008, Chaitén eruption based on the data collected in the field on January 2009 [Alfano F., Bonadonna C., Volentik. A.C.M., Connor C.B., Watt S.F.L., Pyle D.M., Connor L.J., 2011]. On May 1st 2008 Mount Chaitén (southern Chile) interrupted a long period of quiescence, generating a sequence of explosive eruptions and causing the evacuation of Chaitén town located a few kilometers south of the volcano. The activity was characterized by several explosive events each associated with plumes which reached up to about 19 km above sea level. 6
Introduction
The products were dispersed across a wide area, with the finest ash reaching the Atlantic coast of Argentina. Our field observations in the proximal-medial area (3–25 km from the vent) indicate that the May 2008 tephra deposit consists of numerous layers, most of which can be correlated with individual eruptive events. These layers vary from extremely fine grained ash to layers of lapilli and blocks, composed of both juvenile and lithic material. Here we describe the stratigraphy and physical characteristics of the May 2008 deposits, and propose a reconstruction of the timing of the May 2008 events. The deposits are mainly associated with the three main explosive phases which occurred on 1st–2nd May, 3rd–5th May and 6th May, with an estimated bulk tephra volume of 0.5–1.0 km3 (integration of both exponential and power-law fitting). For the 6th May event, represented by a layer composed mainly of lithic lapilli and blocks (> 2 mm), an isopleth map was compiled from which a 19 km plume height was determined, which is in good agreement with satellite observations. Chapter 4 is a article in preparation for Bulletin of Volcanology [Alfano F., Bonadonna C., Gurioli L., in preparation] and it presents a study of the textural and physical features of the juvenile clasts erupted during the climactic phase of the 2008 eruption of Chaitén (population of pumices with diameters in the range 2-6 cm deposited about 5 km far from the vent). Pumices show unimodal density distribution (400-1200 kg/m3; main mode at 600 kg/m3), high-density tail and vescicularity of 73%. They have glassy groundmass with no microcrystals and vescicles with dimension in the range 0.001-2.000 mm. Vescicles show textural variability with highly deformed morphologies, with convex and regular outlines in the low-density samples, and elongated shapes in the high-density samples, with occurrence of indented outlines due to collapse processes. Vescicle wall thickness increases with density. Modal thickness values are in the range 4.5-8.0m and 10.1-14.8 m respectively for low and high density pumices. All analyzed pumices show a unimodal vescicle size distribution, with most frequent vescicle size in the range 0.05-0.08 mm. Main differences among the samples are in the content of large vescicles (d > 1 mm) which are present only in the low density pumices. As a result, the estimated vescicle number density for the pumice products of this eruptive event is 1.3 ± 0.5 x 105 mm-3. The unimodal size distribution of the vescicles and their relatively small dimensions suggest a rapid nucleation event produced during the late phases of magma rise. This is confirmed by the absence of microcrystals that may have delayed vescicle formation which allowed the magma to maintain a low viscosity and a supersaturation in volatiles. Vesciculation and 7
Chapter 1 consequent fragmentation have been produced by a sudden decompression of the melt, related to the opening of the volcanic conduit, which is estimated to be ~ 10 MPa s-1. Chapter 5 presents a final discussion on the topics presented in this thesis giving a general overview of the work carried out, and drawing some general conclusions and future perspectives.
8
Introduction
References Blong, R. J. (1984), Volcanic Hazards. A sourcebook on the effects of eruptions, 424 pp., Academic Press, Sidney. Bonadonna, C., and J. C. Phillips (2003), Sedimentation from strong volcanic plumes, J Geophys Res, 108(B7), 2340-2368. Bonadonna, C., and B. F. Houghton (2005), Total grainsize distribution and volume of tephra-fall deposits, Bulletin of Volcanology, 67, 441-456. Bonadonna, C., G. G. J. Ernst, and R. S. J. Sparks (1998), Thickness variations and volume estimates of tephra fall deposits: the importance of particle Reynolds number, Journal of Volcanology and Geothermal Research, 81(3-4), 173-187. Branney, M. J., and B. P. Kokelaar (2002), Pyroclastic density currents and the sedimentation of ignimbrites, Geological Society, London. Bursik, M. I., R. S. J. Sparks, J. S. Gilbert, and S. N. Carey (1992), Sedimentation of Tephra by Volcanic Plumes .1. Theory and Its Comparison with a Study of the Fogo-a Plinian Deposit, Sao-Miguel (Azores), Bulletin of Volcanology, 54(4), 329-344. Carey, S. N., and R. S. J. Sparks (1986), Quantitative models of the fallout and dispersal of tephra from volcanic eruption columns, Bulletin of Volcanology, 48, 109-125. Cas, R., and J. Wright (1987), Volcanic successions: modern and ancient: a geological approach to processes, products and successions, 1 edition ed., 544 pp., Springer. Cashman, K. V., B. Sturtevant, P. Papale, and O. Navon (2000), Magmatic Fragmentation, in Encyclopedia of volcanoes, edited by H. Sigurdsson, Academic Press, New York. Cioni, R., P. Marianelli, and R. Santacroce (2000), Plinian and Subplinian Eruptions, in Encyclopedia of volcanoes, edited by H. Sigurdsson, Academic Press, New York. Connor, C. B., R. S. J. Sparks, M. Diez, A. C. M. Volentik, and S. C. P. Pearson (2009), The Nature of Volcanism, in Volcanic and Tectonic Hazard Assessment for Nuclear Facilities, edited by C. B. Connor, N. A. Chapman and L. J. Connor, pp. 74-115, Cambridge University Press, New York. Dellino, P., D. Mele, R. Bonasia, G. Braia, L. La Volpe, and R. Sulpizio (2005), The analysis of the influence of pumice shape on its terminal velocity, Geophys Res Lett, 32(21).
9
Chapter 1 Druitt, T. H. (1998), Pyroclastic density currents, in The Physics of Explosive Eruptions, edited by J. S. Gilbert and R. S. J. Sparks, pp. 145-182, Geological Society, London. Ganser, G. H. (1993), A rational approach to drag prediction of spherical and nonspherical particles, Powder Technol, 77(2), 143-152. Haider, A., and O. Levenspiel (1989), Drag Coefficient and Terminal Velocity of Spherical and Nonspherical Particles, Powder Technol, 58(1), 63-70. Iverson, R. M., S. P. Schilling, and J. W. Vallance (1998), Objective delineation of lahar-inundation hazard zones, Geol Soc Am Bull, 110(8), 972-984. Pyle, D. M. (1989), The thickness, volume and grainsize of tephra fall deposits, Bulletin of Volcanology, 51(1), 1-15. Sparks, R. S. J., L. Wilson, and G. Hulme (1978), Theoretical Modeling of Generation, Movement, and Emplacement of Pyroclastic Flows by Column Collapse, J Geophys Res, 83(Nb4), 1727-1739. Sparks, R. S. J., M. I. Bursik, G. J. Ablay, R. M. E. Thomas, and S. N. Carey (1992), Sedimentation of Tephra by Volcanic Plumes .2. Controls on Thickness and Grain-Size Variations of Tephra Fall Deposits, Bulletin of Volcanology, 54(8), 685-695. Thorarinsson, S. (1944), Petrokronologista Studier pa Island, Geographes Annuales Stockholm, 26, 1-217. Walker, G. P. L. (1973), Explosive volcanic eruptions - a new classification scheme, Geologische Rundschau, 62, 431-446. Wilson, L., and T. C. Huang (1979), The influence of shape on the atmospheric settling velocity of volcanic ash particles, Earth and Planetary Sciences Letters, 44, 311324. Wilson, L., R. S. J. Sparks, T. C. Huang, and N. D. Watkins (1978), The control of volcanic column height by eruption energetics and dynamics, J Geophys Res, 83, 1829-1836.
10
Chapter 2
Insights on tephra settling velocity from morphological observations1
2.1 – Introduction Tephra is a collective term used to describe all particles ejected during explosive eruptions and transported in the atmosphere irrespective of size, shape and composition [Thorarinsson, 1944]. Blocks and bombs (i.e., diameter > 64 mm) and lapilli-sized particles (i.e., diameters comprised between 64 and 2 mm) typically sediment close to the vent, whereas ash particles (i.e. diameters < 2 mm) can stay suspended for long times and can travel large distances [Durant et al., 2010] spreading over large areas up to several hundreds of squared kilometers [e.g., Mt St Helens 1980, Chaitén 2008; Alfano et al., 2011; Holasek and Self, 1995]. Settling of volcanic particles depends on their terminal fall velocity (TFV), which represents the dynamical balance between the gravity force, that accelerates the object downward, and the aerodynamic drag forces, that are opposed to the falling motion [e.g., Dellino et al., 2005; Ganser, 1993; Haider and Levenspiel, 1989; Kunii and Levenspiel, 1969; Wilson and Huang, 1979]. The precise determination of the 1
Published as: Alfano F., Bonadonna C., Delmelle P., Costantini L. (in press) Insights on settling velocity from morphological observations. Journal of Volcanology and Geothermal Research, DOI: 10.1016/j.jvolgeores.2011.09.013.
11
Chapter 2 aerodynamic drag forces requires a detailed parameterization of particle morphology [e.g., Ganser, 1993; Riley et al., 2003]. A widely used shape parameter is particle sphericity (), defined as the ratio between the surface area of a sphere with the same volume as the particle (equivalent sphere) and the surface area of the particle. Sphericity can be mainly derived based on approximations to simple equivalent geometric shapes [e.g., Aschenbrenner, 1956], from 2D image analysis of the projected surface of the particles [Riley et al., 2003] or 3D scan analysis, and from direct measurements of surface area through gas adsorption [Dartevelle et al., 2002; Riley et al., 2003]. The aim of this study is to compare and assess the use of various existing models for the calculation of TFV [i.e., Dellino et al., 2005; Ganser, 1993; Haider and Levenspiel, 1989; Wilson and Huang, 1979] and various approaches of morphological characterization (i.e., 2D and 3D). Tephra particles from three different eruptions are considered (Fontana Lapilli, Masaya, Nicaragua; Keanakāko’i Formation, Kilauea, USA; recent dome explosions of Soufriere Hills volcano, Montserrat).
2.2 – The shape of tephra particles and its influence on terminal fall velocity An object which falls through the atmosphere accelerates until it reaches a maximum constant velocity, TFV. Particles with low TFV values can stay suspended in the atmosphere for longer times, and thus can travel greater distances, than particles with high TFV. TFV is defined by the so-called Impact Law [Dellino et al., 2005; Wilson and Huang, 1979]:
TFV
4 gd ( s ) 3C D
[2.1]
where g is the gravitational acceleration, d is the diameter of the object, ρs and ρ are the density of the object and the density of the surrounding fluid, and CD is the drag coefficient (see Appendix A1 for all symbols, abbreviations and dimensions). The drag coefficient is a dimensionless number, which depends on particle shape and on the flow 12
Insights on tephra settling velocity from morphological observations regime of the fluid around the particle. The flow regime is related to the particle Reynolds number (Re), which is the ratio between inertial resistance of the particle and viscous resistance of the surrounding fluid. High particle Reynolds numbers (> 500) are associated with turbulent settling regime, whereas low particle Reynolds numbers (< 0.4) are associated with laminar settling regime [Kunii and Levenspiel, 1969]. For intermediate values of particle Reynolds number, the flow is transitional and changes progressively from laminar to turbulent (intermediate settling regime). Irregularly-shaped particles are characterized by higher aerodynamic drag forces than regularly-shaped particles of the same size and density. The assumption of spherical shape for irregular particles only holds in the case of a laminar sedimentation regime [e.g., Bonadonna and Costa, in press], but it is largely used in dispersal modeling because it provides analytical solutions for TFV, which are computationally more practical [e.g., Arastoopour et al., 1982; Kunii and Levenspiel, 1969]. Attempts to provide analytical solutions of settling velocity for irregular particles have relied on simple parameterizations of morphology, such as the ratio of the three orthogonal axes (F = I + S / 2 L, with L ≥ I S; Wilson and Huang 1979) and sphericity [Dellino et al., 2005; Ganser, 1993].
2.3 – Tephra samples Tephra particles produced by three different eruptions were used in this study in order to investigate a wide range of clast typology (i.e., lithics, juvenile) and texture (i.e., various degrees of vesicularity). All samples were dry sieved in order to separate particles in grain size classes according to a logarithmic distribution of particle diameters ( = - log2 d). The analyzed samples are: i) Fontana Lapilli deposit (Masaya, Nicaragua; FL1 and FL2); ii) Mystery Unit of the Keanakāko’i formation (Kilauea, USA; KMU); iii) 18th July 2005 explosion of Soufriere Hills volcano andesitic dome (Monteserrat, West Indies; SHV). The Fontana Lapilli deposit was erupted in the late Pleistocene (~ 60 ka) from a vent or multiple vents located near Masaya volcano, Nicaragua, and it represents the product of one of the few basaltic Plinian eruptions studied so far [Bice, 1985; Costantini et al., 2009; Wehrmann et al., 2006; Williams, 1983], and it is characterized by a predominant juvenile component (>96 wt.%; SiO2 ~53 wt.%). Bubble number 13
Chapter 2 density value is ~ 107 cm-3 for the whole eruption [Costantini et al., 2010]. In this study, we analyzed two samples from two different eruption phases: a moderate explosive phase at the beginning of the eruption (sample FL1), and the main eruption phase, which is of Plinian intensity (sample FL2). FL1 shows a bimodal grain size distribution and poor sorting (MdΦ = - 1.75; σΦ = 2.20), while FL2 shows a unimodal grain size distribution and moderate sorting (MdΦ = - 2.49; σΦ = 1.60) [parameters from Inman, 1952]. The Mystery Unit deposit is one of the youngest eruption units of the Keanakāko’i Ash Formation [McPhie et al., 1990] and was erupted in the late 18th century toward the end of 300 years of sporadic explosive activity of Kilauea (Swanson, personal communication). The KMU sample corresponds to one of the most widespread layers of the Mystery Unit deposit (probably associated with phreatic activity), which is characterized by a dominance of lithic clasts (mostly lava fragments) and a basaltic juvenile content - 1 (i.e., d < 4 mm).
2.4 – Morphological characterization of tephra particles 2.4.1 – Density measurements Bulk density was measured using a helium pycnometer (Micrometrics Acoupyc 1330) at the Powder Technology Laboratory of the EPFL (Lausanne). Measurements have been carried out on the fine ash fraction ( > 4) of the samples FL1, KMU and SHV, resulting in density values of 2953 ± 15 kg/m3, 3280 ± 21 kg/m3 and 2870 ± 3 kg/m3 respectively. Pumice density was measured through hydrostatic weight on particles size in the range – 5 ≤ ≤ - 3 for the samples FL1 and KMU resulting in an average density value of 990 ± 380 kg/m3 and 2640 ± 320 kg/m3 respectively. 14
Insights on tephra settling velocity from morphological observations The density distribution in relation with the grain size has been calculated following Bonadonna and Phillips [2003]. Juvenile density has been assigned to particles with ≤ - 1, assuming a negligible variation of density in this range of dimension. A linear increase of the density with decreasing particle size (up to the value of bulk density) was assumed for particles in the size range 4 ≥ ≥ - 1. Bulk density has been assigned to particles with ≥ 4 [Bonadonna and Phillips, 2003]. For the sample SHV (dome-collapse tephra) we assumed a negligible vesicularity and, therefore, a constant bulk density for all particles.
2.4.2 – 2D image analysis 2D images were taken using different devices depending on the dimension of the particles. For grain size classes > 3 (d < 125 m) images were taken using the integrated optical system of the particle analyzer CILAS 1180 (http://www.cilas.com/), which allows for an image resolution of 1428 pix/mm. For grain size classes 3 ≥ ≥ - 1, images were taken with an optical microscope which allows for image resolution of 112.9 pix/mm. For grain size classes > - 1 images were taken using a digital scanner which allows for a resolution of 63.0 pix/mm. Binary images were generated by automatic thresholding and manual corrected to reduce errors, then analyzed using the image analysis toolbox Jmicrovision 1.2.7 (http://www.jmicrovision.com/) developed at the University of Geneva. From the analysis of the images several parameters were determined for each particle: projected area (AP); projected perimeter (PP); 90 feret diameters [Riley et al., 2003]. Image analysis of tephra particle images was carried out on individual particles of the samples FL1, SHV and KMU.
2.4.3 – Gas adsorption Gas adsorption is widely used in material science and catalysis and it has only recently been applied to volcanic particles [Dartevelle et al., 2002; Delmelle et al., 2005; Mills et al., 2007; Riley et al., 2003; Witham et al., 2005]. The technique consists of measuring the amount of gas that can be adsorbed by a sample at different pressure conditions, which is related to the total exposed surface, expressed in terms of specific surface area (as, m2/g). For this study, nitrogen gas adsorption analyses at 77 K were 15
Chapter 2 performed using an Autosorb-1 V1.25 instrument (Quantachrome inc.) at the University of Geneva. Prior to each measurement, tephra samples were outgassed in situ overnight at 300 °C at a residual pressure < 0.1 Pa. The Brunauer, Emmett and Teller (BET) model [BET; Brunauer et al., 1938] is widely used to derive as from the gas adsorption data. The model is valid in the range 0.05 ≤ P / P0 ≤ 0.20 (where P and P0 are the equilibrium pressure and the saturation pressure of the gas, respectively). Using the as determinations on a known number of particles (i), the average surface area, Ā, of the particles can be derived if the total mass, m, of the particles analyzed is known (Ā = as m / i). If the analyses refer to individual particles (i = 1), the absolute surface area (A = as m) of that particle is derived. Pore size distribution analysis was also performed through complete cycles point-by-point measurements of volume of gas adsorbed (VGA) at P / P0 values in the range 0 > P / P0 > 1. The presence of micropores was investigated using the t-plot method [Deboer et al., 1966]. The analysis is conducted by plotting the volume of gas adsorbed at each step of P / P0 versus the statistical thickness (t) of the adsorbed layer, assuming that this layer behaves as a normal liquid N2 layer, with a density value set by temperature and a hexagonal dense packing. The resulting plot has a shape which depends on the pore size distribution of the sample. A solid which does not contain micropores produces a straight line in a t-plot diagram [Deboer et al., 1966]. The total volume of mesopores (Vmeso) can be estimated from the total amount of N2 adsorbed at P / P0 = 0.95, when all the pores are assumed to be filled with condensed N2. Under these conditions, the volume of N2 adsorbed relates to the volume of condensed N2, and knowing the molar volume of N2 at ambient pressure (34.7 cm3/mol), Vmeso can be calculated [Rouquerol et al., 1999]. Combining Vmeso and as and assuming a cylindrical geometry of pores, the average diameter for the mesopores (dav) can be calculated (dav = 2 Vmeso / as) [Rouquerol et al., 1999]. The pore size distribution of the mesopores in the tephra samples was calculated using the Density Functional Theory [DFT; Rouquerol et al., 1999]. This method consists in the determination of a mass density profile (r), number of pores with radius r per unit mass) for a given set of pores, and the calculation of the associated isotherm plot. The pore size distribution is determined based on the best fit between computed and measured adsorption isotherms [Rouquerol et al., 1999].
16
Insights on tephra settling velocity from morphological observations BET surface analyses have been carried out on bulk grain size fractions of samples FL1, FL2, KMU ( = - 1, 0, 1, 2, 3, 4, > 4) and SHV ( = 2, 3, 4, > 4), and on individual or known numbers of lithic and juvenile clasts of the KMU sample ( = - 1, 2, - 3, - 4). Pore size distribution analyses have been carried out on the bulk grain size fraction of the samples FL2 ( = - 1, > 3) and SHV ( = 2, 3, 4, > 4) and on individual lithic and juvenile clasts of sample KMU ( = - 1).
2.4.4 – 3D scan One sample (KMU - 4 lithic) was also analyzed with a 3D scanning system at the HEPIA (Haute École du Paysage d’Ingénierie et d’Architecture de Genêve), which provides a 3D digitalization of a solid object based on a simple triangulation system, and is able to give a measurement of its main morphological parameters, such as volume (V3D) and surface area (A3D). According to this method, objects with d > 2 cm are digitized with an ATOS II structured light scanning system. The digital model is obtained by recording the fringe pattern projected on the object surface with a dedicated camera [Barbero and Ureta, 2011; Brajlih et al., 2007; Sugar et al., 2008]. Two images are registered by the device in order to cover the whole surface of the object. The images are elaborated by a software (gom-atos) which determines the 3D coordinates of each pixel composing the image, producing a polygon mesh of the surface. The polygon mesh is again elaborated using a software (Imageware package) which allows the generation of a 3D digital model (contact surface computation) and scale external dimension and volume. Unfortunately, the 3D scan measurement could not be carried out on particles < 2 cm.
2.4.5 – Particle morphological parameters The calculation of particle TFV is based on the determination of several parameters and shape factors that characterize particle morphology. In detail, the diameter of the equivalent sphere (diameter of the sphere with the same volume of the particle), the shape factor (F) of Wilson and Huang [1979] and particle sphericity () based on various models, were calculated from the morphological analyses described above (i.e., 2D image analysis, gad adsorption, 3D scan). 17
Chapter 2
Figure 2.1. Semi-log plot of the average values of the shape factor F of Wilson and Huang [1979], sphericity (R) of Riley et al. [2003], sphericity (A) of Aschenbrenner [1956] and sphericity divided by particle circularity (R and A) as described by Dellino et al. [2005], measured for each grain size class of the samples FL1 (a), KMU (b) and SHV (c).
18
Insights on tephra settling velocity from morphological observations Table 2.1. Average values and standard deviation of the morphological parameters for each grain size class () and for the whole sample of FL1, KMU and SHV. Morphologic Parameters F
R
A
R
A
>4
0.77 ± 0.09
0.83 ± 0.10
0.93 ± 0.03
0.77 ± 0.14
0.84 ± 0.07
4
0.79 ± 0.08
0.76 ± 0.10
0.93 ± 0.02
0.66 ± 0.12
0.81 ± 0.07
3
0.77 ± 0.08
0.88 ± 0.09
0.93 ± 0.03
0.82 ± 0.12
0.87 ± 0.07
2
0.77 ± 0.08
0.85 ± 0.09
0.93 ± 0.03
0.78 ± 0.12
0.85 ± 0.06
1
0.80 ± 0.08
0.77 ± 0.10
0.94 ± 0.02
0.69 ± 0.12
0.82 ± 0.06
0
0.79 ± 0.08
0.74 ± 0.10
0.93 ± 0.02
0.64 ± 0.12
0.80 ± 0.06
-1
0.78 ± 0.08
0.69 ± 0.10
0.93 ± 0.02
0.58 ± 0.12
0.77 ± 0.07
-2
0.79 ± 0.08
0.66 ± 0.09
0.93 ± 0.03
0.55 ± 0.11
0.76 ± 0.07
-3
0.75 ± 0.10
0.69 ± 0.08
0.92 ± 0.03
0.57 ± 0.10
0.77 ± 0.05
-4
0.67 ± 0.07
0.72 ± 0.06
0.90 ± 0.03
0.61 ± 0.08
0.76 ± 0.05
Whole Sample
0.78 ± 0.08
0.76 ± 0.11
0.93 ± 0.02
0.67 ± 0.14
0.81 ± 0.07
>2
0.77 ± 0.08
0.93 ± 0.06
0.93 ± 0.02
0.90 ± 0.09
0.90 ± 0.05
2
0.77 ± 0.08
0.87 ± 0.08
0.93 ± 0.02
0.81 ± 0.11
0.87 ± 0.06
1
0.81 ± 0.07
0.82 ± 0.09
0.94 ± 0.02
0.75 ± 0.12
0.85 ± 0.05
0
0.81 ± 0.07
0.81 ± 0.09
0.94 ± 0.02
0.73 ± 0.12
0.84 ± 0.06
-1
0.81 ± 0.07
0.82 ± 0.09
0.94 ± 0.01
0.74 ± 0.12
0.85 ± 0.06
-2
0.82 ± 0.07
0.72 ± 0.13
0.94 ± 0.01
0.61 ± 0.16
0.79 ± 0.08
-3
0.82 ± 0.06
0.70 ± 0.13
0.94 ± 0.01
0.59 ± 0.16
0.78 ± 0.08
-4
0.72 ± 0.10
0.86 ± 0.05
0.91 ± 0.03
0.80 ± 0.07
0.85 ± 0.04
Whole Sample
0.80 ± 0.07
0.81 ± 0.10
0.94 ± 0.02
0.74 ± 0.13
0.84 ± 0.06
>4
0.75 ± 0.09
0.86 ± 0.11
0.92 ± 0.03
0.81 ± 0.14
0.85 ± 0.08
4
0.79 ± 0.08
0.82 ± 0.09
0.93 ± 0.03
0.75 ± 0.12
0.84 ± 0.07
3
0.77 ± 0.08
0.88 ± 0.08
0.93 ± 0.03
0.83 ± 0.12
0.87 ± 0.06
2
0.78 ± 0.08
0.85 ± 0.08
0.93 ± 0.03
0.78 ± 0.11
0.86 ± 0.06
1
0.80 ± 0.07
0.85 ± 0.07
0.94 ± 0.08
0.79 ± 0.10
0.86 ± 0.05
0
0.81 ± 0.07
0.84 ± 0.06
0.94 ± 0.08
0.77 ± 0.08
0.86 ± 0.04
-1
0.78 ± 0.08
0.80 ± 0.07
0.93 ± 0.08
0.72 ± 0.09
0.83 ± 0.05
Whole Sample
0.78 ± 0.08
0.85 ± 0.09
0.93 ± 0.03
0.79 ± 0.12
0.86 ± 0.06
FL1
KMU
SHV
; where L, I, S are the three orthogonal axes of the particle [Wilson and Huang, 1979]. ; where Ap and Pp are the projected area and projected perimeter of the particle [Riley et al., 2003]. ; [Aschenbrenner, 1956].
R, A: R and A divided by particle circularity according to [Dellino et al., 2005].
19
Chapter 2 The diameter of the equivalent sphere (dV) was calculated based on the ratio between AP and PP as described by Riley et al. [2003]. The shape factor F depends only on the length of the three orthogonal axes of the particle. This factor has been calculated based on the feret diameters obtained through image analysis, assuming the maximum axis equal to the largest feret diameter, the minimum axis equal to the smallest feret diameter and the intermediate axis equal to the average of the 90 measured feret diameters [Riley et al., 2003]. Sphericity has been calculated applying different methods. A first value of sphericity (R) is calculated based on the ratio between AP and PP2 as described by Riley et al. [2003]. A second method consists in calculating the sphericity of an equivalent tetrakaidecahedron (i.e., 14-face solid) which is considered as the geometrical solid figure that best approximates a natural irregular particle [Aschenbrenner, 1956]. Particle sphericity (A) is calculated assuming the shape equivalent to a tetrakaidecahedron with the same orthogonal axes of the particle. The orthogonal axes of the particle were determined as described above for the calculation of the shape factor F. Finally, the shape factors R and A were calculated by dividing R and A for particle circularity (ratio between Pp and the perimeter of the circle with area equal to Ap), as described by Dellino et al. [2005]. Other values of sphericity have been calculated based on the direct surface area measurements carried out on single particles using the gas adsorption technique (GA) and the 3D scan analysis (3D).
2.5 – Results 2.5.1 – Morphological characterization The three samples considered (FL1, KMU, SHV) are characterized by a similar range of values and trends of morphological parameters as a function of grain size (Fig. 2.1). Samples show average values of particle sphericity and of the shape factor F for each size class ranging between 0.6 and 1.0, and a standard deviation () less than 0.14 (Table 2.1). The average values of the morphological parameters do not show any clear correlation with grain size, and there is no significant variability of their standard deviation within each individual grain-size class. For all samples, A (sphericity calculated according to Aschenbrenner [1956]) is characterized by the highest values (> 0.9) and the lowest standard deviation (0.01 ≤ ≤ 0.03) with respect to the average 20
Insights on tephra settling velocity from morphological observations values. In fact, F and R (sphericity calculated according to Riley et al. [2003]) are more sensitive to shape irregularity than A, showing values ranging from 0.6 to 1.0 and a higher standard deviation (0.05 ≤ ≤ 0.13). The morphological parameters R and A (i.e., sphericity divided by particle circularity according to Dellino et al. [2005]) follow the distribution of R and A, but show lower values due to the effect of the correction for the particle circularity.
Figure 2.2. Log-Log plot showing the BET as distributions on bulk individual fractions. Analyses carried out applying an outgassing protocol at 300 °C under vacuum, for samples FL1, FL2, KMU and SHV. All measurements are affected by a systematic error of 4 %.
Gas-adsorption-derived specific surface area, as, increases with decreasing size for all bulk size fractions analyzed (Fig. 2.2), with FL1 and FL2 tephra displaying higher as values (3.0–21 m2/g) than the KMU and SHV samples (< 3.3 m2/g). In addition, juvenile clasts are characterized by higher as (2.2–7.9 m2/g, Fig. 2.3a) and A (0.11–15.00 m2, Fig. 2.3b) than lithic clasts (as = 0.5–0.8 m2/g, Fig. 2.3a; A = 0.03–3.36 m2, Fig. 2.3b). Absolute surface area values derived from as measurements (AGA) carried out on single particles (0.7–15.0 m2) were compared with A calculated based on equivalent
simple
shapes
(i.e.,
equivalent
sphere,
AES
and
equivalent
tetrakaidecahedron, AT). The resulting AGA values are up to four orders of magnitude larger than the values based on the parameterization of simple shapes (Table 2.2). Figure 2.4 shows the results of a 3D scan analysis carried out on a lithic particle of KMU - 4 : values of surface area (A3D) and volume (V3D) increase with the resolution 21
Chapter 2 of the image (number of points of the polygon mesh), up to a plateau which represents the analytical limit of the technique. A3D values (1.3–1.4 x 10-3 m2) are close to the values calculated for the equivalent sphere and tetrakaidecahedron (Table 2.2), resulting in a sphericity value 3D equal to 0.9. On the other hand, AGA is three orders of magnitude larger than A3D at its maximum resolution for the lithic particle KMU - 4 .
Figure 2.3. a) Semi-Log plot showing the comparison between the BET a s values measured on sample from KMU of lithic and juvenile origin. b) Log-Log plot of absolute surface area values for individual particles (full symbols) and of average surface area (open symbols) values calculated for a known number of particles. The values are referred to juvenile and lithic clasts of the sample KMU. All measurements are affected by a systematic error of 4 %.
2.5.2 – Pore size distribution A pore size distribution analysis was carried out in order to better understand and constrain the origin of particle surface area. Results are summarized in Table 2.3. FL1 - 1 sample shows dav and Vmeso values of 4.3 nm and 1.9 x 10-3 cm3/g, respectively, associated with a total volume of micropores of 4.4 x 10-4 cc/g. A comparatively higher dav (7.2 nm), lower Vmeso (2.9 x 10-2 cm3/g ) values and lack of micropores were obtained for FL1 > 3 . The dav of the SHV tephra ranged from 7.7 to 22
Insights on tephra settling velocity from morphological observations 9.9 nm, and the Vmeso ranged from 0.3 to 0.6 x 10-2 cm3/g. The t-plot analysis did not reveal the presence of micropores in the SHV tephra. Finally, the juvenile clast from KMU showed dav and Vmeso values of 5.1 nm and 3.9 x 10-3 cm3/g respectively, associated with a total volumes of micropores of 7.6 x 10-4 cm3/g. A comparatively higher dav (7.8 nm), lower Vmeso (0.9 x 10-3 cm3/g ) and a total volumes of micropores of 1.1 x 10-4 cm3/g were obtained for the lithic clast from KMU.
Figure 2.4. Semi-Log plot of surface (m2) and volume (cm3) obtained through 3D scan of the tephra particle KMU - 4 lithic versus the resolution of the image (number of points). All measurements are affected by a systematic error of 4 %.
Table 2.2. Summary of absolute surface area (A) and sphericity () for individual lithic and juvenile particles from KMU sample determined through gas adsorption, 3D scan analyses and geometrical assumption (i.e., sphere, tetrakaidecahedron). Lithic -4 -3 Absolute Surface (A, m2) AGA AES (x 10-3) AA (x 10-3) A3D (x 10-3)
Juvenile -4 -3
6.0 1.2 1.6 1.4
0.7 0.5 6.2 -
15.0 0.9 1.1 -
5.1 0.4 4.8 -
0.9 0.7 0.8
-0.8 0.9
-0.8 0.9
-0.8 0.8
Sphericity ()
3D A GA (x 10-4)
GA: gas adsorption ES: equivalent sphere A: equivalent tetrakaidecahedron (Aschenbrenner, 1956) 3D: scan 3D.
23
Chapter 2 Table 2.3. Results of the pore size distribution analysis carried out using the gas adsorption technique on bulk sample of particles of different grain sizes and typologies. KMU - 1 juvenile lithic -1 >3 2 3 4 >4 2 as (m /g) 1.7 15.8 1.4 1.7 1.8 2.4 3.0 0.4 Vmeso (cm3/g) 1.9 x 10-3 2.9 x 10-2 3.2 x 10-3 3.2 x 10-3 4.2 x 10-3 5.9 x 10-3 3.9 x 10-3 0.9 x 10-3 Vmicro (cm3/g) 4.4 x 10-4 ------ 7.6 x 10-4 1.1 x 10-4 dav (nm) 4.3 7.2 9.4 7.7 9.1 9.9 5.1 7.8 2 a s: BET specific surface area (m /g). Vmeso: total volume of the mesopores (cm3/g). Vmicro: total volume of the micropores (cm3/g). dav: average pore diameter calculated assuming a cylindrical pore geometry (nm). FL1
SHV
2.5.3 – Terminal fall velocity characterization Particle TFV was calculated based on a range of morphological parameters and using the models of Wilson and Huang [1979] (TFVWH), Ganser [1993] (TFVG (R) and TFVG (A)), and Dellino et al. [2005], (TFVD (R) and TFVD (A)), at sea level. Resulting TFV values range from a minimum of 0.05 m/s, for the small particles (> 4 ) of the samples SHV, up to a maximum of about 35 m/s, for the coarse particles (- 4 ) of the KMU sample. Average values for each grain size class and the associated standard deviation are summarized in Table 2.4. TFVs increase with grain size classes, showing also an increase of the standard deviation. The distribution of the average TFV values for each class compared with the corresponding velocity calculated assuming a spherical shape [Haider and Levenspiel, 1989], shows a different behavior for the different models (Fig. 2.5). For small particles ( ≥ 3), TFVWH, TFVG (R) and TFVG (A) values are close to the TFV calculated for the equivalent sphere, whereas TFVD (R) and TFVD (A) values are higher, reflecting the limited applicability of the empirical model of Dellino et al. [2005] that was experimentally validated down to Re of about 100. For larger grain size classes ( < 2) all the considered models show the same trend, with values of TFV smaller than TFV calculated for equivalent spheres. TFV values calculated for individual particles of the sample KMU using GA are up to two orders of magnitude lower than TFV calculated based on image analysis and 3D scan derived sphericities (Fig. 2.5b). On the other hand, TFV of the KMU lithic particle based on3D falls in the range of values of TFV calculated based on image analysis derived sphericity (Fig. 2.5b).
24
Insights on tephra settling velocity from morphological observations Table 2.4. Average values and standard deviation of the calculated TFV for each grain size class () of the samples FL1, KMU and SHV. Terminal Fall Velocity (m/s) TFVHL
TFVWH
TFVG (R)
TFVG (A)
TFVD (R)
TFVD (A)
>4
0.10 ± 0.07
0.08 ± 0.06
0.09 ± 0.06
0.10 ± 0.07
0.81 ± 0.19
0.89 ± 0.20
4
0.53 ± 0.13
0.54 ± 0.16
0.46 ± 0.12
0.54 ± 0.15
1.28 ± 0.23
1.53 ± 0.18
3
1.17 ± 0.30
1.17 ± 0.33
1.08 ± 0.37
1.17 ± 0.34
1.97 ± 0.45
2.12 ± 0.34
2
2.09 ± 0.33
1.97 ± 0.30
1.90 ± 0.36
2.11 ± 0.36
2.60 ± 0.38
2.79 ± 0.29
1
4.09 ± 0.62
3.23 ± 0.44
3.08 ± 0.57
3.97 ± 0.61
3.35 ± 0.57
3.88 ± 0.45
0
5.84 ± 0.78
3.96 ± 0.54
3.67 ± 0.74
5.13 ± 0.67
3.87 ± 0.80
4.64 ± 0.64
-1
8.14 ± 0.78
4.90 ± 0.65
4.11 ± 0.70
6.29 ± 0.67
4.57 ± 0.87
5.73 ± 0.73
-2
11.23 ± 1.17
6.94 ± 0.96
5.11 ± 0.88
8.37 ± 0.90
6.23 ± 1.27
8.14 ± 1.17
-3
14.63 ± 1.07
9.23 ± 0.82
6.63 ± 1.06
10.71 ± 0.71
8.68 ± 1.57
11.30 ± 1.15
-4
20.39 ± 1.78
12.37 ± 1.19
10.40 ± 1.42
14.39 ± 1.49
14.53 ± 2.16
17.40 ± 1.95
FL1
KMU >2
1.69 ± 0.20
1.75 ± 0.23
1.82 ± 0.26
1.79 ± 0.22
2.88 ± 0.28
2.87 ± 0.21
2
2.56 ± 0.44
2.45 ± 0.37
2.51 ± 0.44
2.67 ± 0.46
3.30 ± 0.42
3.44 ± 0.34
1
5.86 ± 0.90
4.54 ± 0.59
4.68 ± 0.81
5.73 ± 0.77
4.95 ± 0.76
5.50 ± 0.60
0
8.65 ± 1.09
5.93 ± 0.70
6.02 ± 1.14
7.67 ± 0.76
6.34 ± 1.14
7.14 ± 0.85
-1
13.78 ± 1.58
8.64 ± 1.10
8.44 ± 1.35
10.65 ± 1.04
9.65 ± 1.60
10.74 ± 1.38
-2
18.35 ± 1.36
11.74 ± 1.28
9.66 ± 1.75
13.73 ± 1.12
12.04 ± 2.13
14.40 ± 1.30
-3
25.57 ± 2.17
17.33 ± 2.12
11.78 ± 3.49
19.47 ± 1.88
15.80 ± 5.20
20.95 ± 3.71
-4
35.24 ± 3.04
22.79 ± 1.82
22.25 ± 3.54
25.97 ± 2.04
31.22 ± 4.71
33.56 ± 2.49
>4
0.06 ± 0.05
0.05 ± 0.05
0.06 ± 0.05
0.06 ± 0.05
0.74 ± 0.19
0.78 ± 0.19
4
0.55 ± 0.14
0.57 ± 0.16
0.51 ± 0.13
0.57 ± 0.15
1.49 ± 0.23
1.66 ± 0.18
3
1.45 ± 0.33
1.48 ± 0.34
1.41 ± 0.37
1.49 ± 0.35
2.41 ± 0.44
2.53 ± 0.33
2
2.68 ± 0.54
2.51 ± 0.44
2.48 ± 0.51
2.75 ± 0.55
3.23 ± 0.47
3.44 ± 0.41
1
5.70 ± 0.92
4.32 ± 0.54
4.68 ± 0.84
5.46 ± 0.76
4.96 ± 0.75
5.37 ± 0.62
0
9.72 ± 1.28
6.40 ± 0.77
6.76 ± 0.93
8.26 ± 0.84
7.16 ± 0.94
7.86 ± 0.85
-1
13.99 ± 0.92
8.10 ± 0.75
7.98 ± 0.94
10.16 ± 0.66
8.98 ± 0.99
10.03 ± 0.78
SHV
TFVHL= TFV according to the model of Haider and Levenspiel [1989] assuming a spherical shape of the particles. TFVWH= TFV according to the drag prediction model of Wilson and Huang [1979] calculated using F. TFVG= TFV according to the drag prediction model of Ganser [1993] calculated using R and A. TFVD= TFV according to the model of Dellino et al. [2005] calculated using R and A. (see also caption of Table 2.1)
25
Chapter 2
Figure 2.5. Average TFV values of bulk grain size fraction of the samples FL1 (a), KMU (b) and SHV (c), calculated using the model of Wilson and Huang [1979], the model of Ganser [1993] based on R and A, and the model of Dellino et al. [2005] based on R and A, plotted vs. the terminal velocity of the equivalent sphere [Haider and Levenspiel, 1989]. TFVs of individual particle of the sample KMU calculated based on gas adsorption derived sphericity (GA) and 3D scan derived sphericity (3D) are included in the KMU plot.
26
Insights on tephra settling velocity from morphological observations The determination of TFV significantly depends on the morphological parameter and the model used in the calculation (Fig. 2.6). In particular, TFV based on R and R results in a range of values that cover a wider area than the values resulting using the other morphological parameter, and are characterized also by higher values of standard deviation (cf., Table 2.4), reflecting the high sensitivity to the shape of these morphological parameters. The model of Ganser [1993] appears to be more sensitive to the particle morphology than the other models, as it results in values of TFV reduced up to 50 % with respect to the TFV of equivalent spheres. The same behavior is shown by the model of Dellino et al. [2005] for grain size class larger than 0 .
Figure 2.6. Areas of dispersion of individual particle TFV for the sample FL1, calculated using the model of Wilson and Huang [1979], the model of Ganser [1993] based on R and A, and the model of Dellino et al. [2005] based on R and A in relation with the TFV of the equivalent sphere.
Figure 2.7a shows the relation between particle TFV, calculated using the method of Ganser [1993] and the sphericity of Riley et al. [2003] for the different grain size classes of samples FL1, KMU and SHV. The influence of shape on TFV increases with particle size as the ratio TFV / R (represented by the linear trends indicated in the plot) increases for the coarsest grain size classes. In fact, fine particles ( ≥ 3) show that for a sphericity varying in the range 0.4–1.0, TFV remains almost constant. For larger particles ( < 3) the variation of TFV increases progressively with the size for a same range of variation of sphericity. Particle TFV becomes progressively smaller than TFV of the equivalent sphere with increasing particle size. In fact, the difference between particle TFV and the TFV of the equivalent sphere ( TFV %; Fig. 2.7b and appendix 27
Chapter 2 A2) is < 10 % for small particles ( ≥ 3) and reaches 50 % for large particles ( ≤ - 1). The two breaks-in-slope A and B in Fig. 2.7b A range correspond to Reynolds number 7–30 and 600–2300, corresponding roughly to the transition of settling regime laminar to intermediate (0.4) and intermediate to turbulent (500) [Bonadonna et al., 1998; Kunii and Levenspiel, 1969].
Figure 2.7. a) Particle TFV of the sample FL1 plotted vs sphericity (R) showing the influence of the shape for each grain size class fraction . b) Difference ( %) between particle TFV, calculated using the model of Ganser [1993] and the sphericity of Riley et al. [2003], of the samples FL1, KMU and SHV and the TFV of the equivalent sphere plotted vs particle grain size showing the influence % of the shape on the calculated TFV. Red areas A and B indicate the break-in-slope of the trend of the points.
Individual size classes (e.g., whole- and half- classes) are associated with a range of particle sizes and, therefore, also with a range of TFV values, which increases with class size. As an example, the amplitude of the TFV ranges (expressed as 2) increases from a minimum of 0.1 m/s for the > 4 classes (SHV sample) to a maximum of 9.4 m/s for the – 4 class (KMU sample) (cf., table 2.4). Values of 2are smaller for the half- system, reaching a maximum 2of 8.2 m/s for the - 4.5 class of the KMU sample (Table 2.5). In Fig. 2.8 the standard deviation of TFVG (R) for the sample FL1, KMU and SHV is plotted in relation with the grain size classes for both the whole-and half- systems. Parallel linear trends of the two systems have been plotted, corresponding to an average reduction of equal to 0.24 m/s for samples FL1, 0.27 m/s for KMU and 0.31 m/s for SHV. Even though values in each system are larger for coarse-size fractions, the reduction of from the whole-to the half- systems is more relevant for small-size classes (> 20 % for ≥ 0) rather than for coarse-size classes (< 20 % for < 0). 28
Insights on tephra settling velocity from morphological observations Table 2.5. Average values and standard deviation of the calculated TFV for each grain size class (half-) of the samples FL1, KMU and SHV. TFVHL TFVWH FL1 0.10 ± 0.07 0.08 ± 0.06 >4 0.36 ± 0.05 0.34 ± 0.07 4 0.61 ± 0.07 0.63 ± 0.09 3.5 0.84 ± 0.10 0.82 ± 0.12 3 1.38 ± 0.16 1.39 ± 0.16 2.5 1.86 ± 0.14 1.79 ± 1.19 2 2.40 ± 0.24 2.21 ± 0.24 1.5 3.39 ± 0.27 2.82 ± 0.30 1 4.51 ± 0.33 3.48 ± 0.30 0.5 5.22 ± 0.26 3.59 ± 0.31 0 6.58 ± 0.50 4.41 ± 0.39 - 0.5 7.52 ± 0.39 4.44 ± 0.36 -1 8.81 ± 0.49 5.40 ± 0.49 - 1.5 6.35 ± 0.53 - 2 10.42 ± 0.48 7.93 ± 0.67 - 2.5 12.59 ± 0.58 9.11 ± 0.79 - 3 14.31 ± 0.61 - 3.5 16.87 ± 0.90 10.02 ± 0.49 - 4 19.50 ± 0.81 11.84 ± 0.74 - 4.5 22.93 ± 1.22 13.87 ± 0.89 KMU 1.69 ± 0.20 1.75 ± 0.23 >2 2.35 ± 0.22 2.30 ± 0.25 2 3.17 ± 0.32 2.87 ± 0.32 1.5 4.77 ± 0.43 3.93 ± 0.37 1 6.44 ± 0.43 4.86 ± 0.40 0.5 7.98 ± 0.51 5.59 ± 0.47 0 9.89 ± 0.72 6.56 ± 0.61 - 0.5 7.87 ± 0.68 - 1 12.45 ± 0.69 9.49 ± 0.81 - 1.5 15.26 ± 0.76 - 2 17.70 ± 0.75 11.26 ± 0.95 - 2.5 20.20 ± 0.90 13.10 ± 1.13 - 3 24.01 ± 1.18 15.96 ± 1.34 - 3.5 27.69 ± 1.18 19.19 ± 1.44 - 4 32.67 ± 1.62 22.25 ± 1.94 - 4.5 37.82 ± 1.53 23.33 ± 1.57 SHV 0.06 ± 0.05 0.05 ± 0.05 >4 0.39 ± 0.07 0.39 ± 0.08 4 0.63 ± 0.09 0.66 ± 0.10 3.5 1.03 ± 0.15 1.08 ± 0.19 3 1.65 ± 0.17 1.67 ± 0.19 2.5 2.31 ± 0.26 2.24 ± 0.26 2 3.25 ± 0.33 2.92 ± 0.34 1.5 4.94 ± 0.51 3.96 ± 0.38 1 6.50 ± 0.47 4.71 ± 0.39 0.5 8.40 ± 0.63 5.72 ± 0.49 0 6.86 ± 0.55 - 0.5 10.61 ± 0.70 8.01 ± 0.68 - 1 12.80 ± 0.66 9.09 ± 0.69 - 1.5 15.11 ± 0.52 (captions same as Table 4)
Terminal Fall Velocity (m/s) TFVG (R) TFVG (A) TFVD (R)
TFVD (A)
0.09 ± 0.06 0.31 ± 0.06 0.53 ± 0.07 0.71 ± 0.15 1.32 ± 0.21 1.72 ± 0.26 2.14 ± 0.33 2.73 ± 0.44 3.29 ± 0.53 3.29 ± 0.55 4.11 ± 0.69 3.93 ± 0.63 4.32 ± 0.72 4.76 ± 0.76 5.70 ± 0.76 6.45 ± 0.86 7.90 ± 1.48 9.95 ± 1.12 11.65 ± 1.49
0.10 ± 0.07 0.36 ± 0.06 0.63 ± 0.08 0.81 ± 0.12 1.40 ± 0.13 1.87 ± 0.20 2.44 ± 0.26 3.32 ± 0.36 4.36 ± 0.32 4.61 ± 0.34 5.74 ± 0.41 5.78 ± 0.35 6.83 ± 0.47 7.80 ± 0.48 9.33 ± 0.56 10.59 ± 0.64 11.53 ± 0.63 13.80 ± 1.00 16.05 ± 1.41
0.81 ± 0.19 1.11 ± 0.18 1.36 ± 0.21 1.60 ± 0.24 2.20 ± 0.31 2.48 ± 0.32 2.76 ± 0.40 3.10 ± 0.47 3.51 ± 0.56 3.46 ± 0.60 4.36 ± 0.74 4.25 ± 0.71 4.91 ± 0.89 5.66 ± 1.02 7.18 ± 1.07 8.39 ± 1.25 10.63 ± 2.19 13.82 ± 1.64 16.55 ± 2.25
0.89 ± 0.20 1.33 ± 0.11 1.62 ± 0.12 1.79 ± 0.14 2.33 ± 0.18 2.63 ± 0.20 3.00 ± 0.26 3.48 ±0.30 4.12 ± 0.33 4.18 ± 0.35 5.19 ± 0.47 5.21 ± 0.41 6.28 ± 0.57 7.40 ± 0.64 9.38 ± 0.74 11.02 ± 0.84 13.23 ± 1.14 16.54 ± 1.11 19.83 ± 1.78
1.82 ± 0.26 2.36 ± 0.33 2.94 ± 0.46 4.15 ± 0.59 4.97 ± 0.76 5.64 ± 0.95 6.73 ± 1.14 7.95 ± 1.01 8.98 ± 1.47 9.79 ± 1.68 9.27 ± 1.88 9.55 ± 2.20 14.80 ± 2.51 20.04 ± 3.30 24.47 ± 2.14
1.79 ± 0.22 2.46 ± 0.27 3.27 ± 0.37 4.82 ± 0.44 6.21 ± 0.38 7.25 ± 0.46 8.47 ± 0.55 9.86 ± 0.54 11.53 ± 0.69 13.25 ± 0.71 15.10 ± 0.91 18.13 ± 1.04 21.30 ± 1.04 24.97 ± 1.43 26.96 ± 2.11
2.88 ± 0.28 3.22 ± 0.34 3.57 ± 0.51 4.50 ± 0.59 5.19 ± 0.73 5.89 ± 0.89 7.16 ± 1.11 8.83 ± 0.98 10.57 ± 1.66 12.10 ± 1.99 11.86 ± 2.49 12.44 ± 3.40 20.38 ± 3.42 27.95 ± 4.12 34.50 ± 2.44
2.87 ± 0.21 3.32 ± 0.23 3.82 ± 0.34 4.87 ± 0.39 5.83 ± 0.39 6.68 ± 0.51 7.99 ± 0.70 9.71 ± 0.65 11.87 ± 1.04 14.00 ± 1.08 15.53 ± 1.20 18.32 ± 2.20 24.52 ± 1.94 30.98 ± 2.58 36.13 ± 2.08
0.06 ± 0.05 0.37 ± 0.07 0.59 ± 0.09 0.97 ± 0.22 1.62 ± 0.22 2.20 ± 0.32 2.91 ± 0.46 4.19 ± 0.62 5.20 ± 0.73 6.21 ± 0.76 7.13 ± 0.85 7.93 ± 0.90 8.51 ± 1.17
0.06 ± 0.05 0.41 ± 0.07 0.66 ± 0.09 1.06 ± 0.17 1.70 ± 0.18 2.38 ± 0.28 3.30 ± 0.37 4.87 ± 0.47 6.08 ± 0.46 7.44 ± 0.49 8.83 ± 0.50 10.07 ± 0.58 11.21 ± 0.58
0.74 ± 0.19 1.32 ± 0.19 1.58 ± 0.19 1.98 ± 0.35 2.61 ± 0.30 3.04 ± 0.36 3.52 ± 0.49 4.54 ± 0.57 5.41 ± 0.65 6.47 ± 0.67 7.64 ± 0.79 8.89 ± 0.91 10.04 ± 1.22
0.78 ± 0.19 1.47 ± 0.13 1.76 ± 0.12 2.14 ± 0.22 2.71 ± 0.19 3.20 ± 0.24 3.81 ± 0.32 4.92 ± 0.40 5.85 ± 0.43 7.04 ± 0.47 8.42 ± 0.55 9.89 ± 0.64 11.46 ± 0.70
29
Chapter 2
Figure 2.8. Plot showing the variation of the standard deviation (2; m/s) of TFV of the samples FL1, KMU and SHV, calculated using the model of Ganser [1993] based on the sphericity of Riley et al [2003], for different grain size categories and for both the whole- and half- system. Parallel linear trends are showed, with correlation coefficients R2 of the whole- and half- system equal to 0.96 and 0.97 for FL1, 0.99 and 0.96 for KMU and 0.95 and 0.97 for SHV. Empty circles indicate data points have not been taken into account in the calculation of the trend lines as they are obtained from limited number of particles and consequently are affected by a high scattering, which is not statistically representative.
30
Insights on tephra settling velocity from morphological observations
2.6 – Discussion 2.6.1 – Morphological characterization Several methodologies exist to describe particle morphology, with distinct advantages and disadvantages. Our analysis confirms the finding of previous studies [i.e., Riley et al., 2003] that 2D image analysis is a powerful tool that allows for a large number of data to be obtained over a relatively short time regardless of particle size. Even though it has the limitation to asses only the 2D shape, the resulting range of values of sphericity suggests that this method can capture most of the essential information of particle morphology as long as we assume that a generic projected image can be representative of the whole irregularity of the particle [Blott and Pye, 2008]. In fact, R is characterized by a higher variability for all size classes, and within individual classes, than F and A (cf., Table 2.1). A is characterized by a low variability and average values higher than 0.9 for all grain size classes. [Mele et al., 2011] had shown how particle irregularity (quantified by particle circularity) decreases with grain size for particles larger than 0.5 mm. Nonetheless, the morphological parameters of the samples analyzed in this work all fall in a narrow range (about 0.7–0.9) regardless of particle size. Specific surface area analyses carried out with gas adsorption on bulk size fractions show an inverse correlation of as with grain size (cf., Fig. 2.2) due to the increase of particle number in the measurement cell and therefore to the increase of exposed surface. In contrast, A derived from measurements on individual or known number of particles shows a direct correlation with size (cf., Fig. 2.3b). This implies that as measured on bulk samples is not representative of all sizes and cannot be directly related to A of individual particles. An average value of absolute surface area can be determined only if the number of particles in each bulk sample is known. However, the determination of the number of particles in a sample of fine ash can be cumbersome, as it requires to know the density and the mass of all individual particles. Another possible method has been suggested by Riley et al [2003] that propose a correction factor (Fa) for the surface area equal to the ratio between the specific surface area of the bulk sample and the calculated specific surface area of a particle assuming a spherical shape. They estimated correction factors in the range between 7 and 38. Nonetheless, correction factors based on individual bulk-sample analysis cannot be extrapolated to individual particles or to the whole particle population for the reasons described above and, 31
Chapter 2 therefore, cannot be used to correct particle TFV for the whole size range. Gasadsorption-derived values of A are several orders of magnitude larger than surface area values calculated for equivalent shapes or determined using the 3D scan analysis. These values produce extremely low GA values, which are outside the range of sphericity determined in this study (cf., Table 2.1 and 2.2), but also outside of the range of sphericity for natural silicic particles [0.3–0.9; Blott and Pye, 2008]. In fact, as values determined through gas adsorption describe the contribution of micro- and mesoporosity and, in general, of all features at the scale of the dimension of the cross sectional area of the adsorptive used in the measurement [i.e., 0.13–0.20 nm2 for N2; Gregg and Sing, 1982]. As demonstrated by gas adsorption and 3D scan analysis, the measurement of particle surface area is dependent on the scale of observation: the smaller the scale of observation, the larger the measured surface area. This trend is due to the fact that smaller irregularities can be detected when the observation scale is reduced resulting in larger surface area. Thus, it is not surprising that gas adsorption, which accounts for irregularities at the molecular scale, can produce such high values of particle surface area. PSD results show that all samples analyzed are mesoporous (4–50 nm). Microporosity (pores < 4 nm) is generally absent or represents a small fraction of the total porosity. Nonetheless, some samples are more porous than others, resulting in a larger as. As an example, the juvenile samples FL1 and FL2 show higher as than the lithic-rich samples SHV and KMU. Sample FL1 > 3 is characterized by a large as (15.8 m2/g) and a large volume of mesopores. The origin of mesoporosity in tephra particles is not known. The minimum size of gas vesicles in juvenile pyroclasts produced by strong explosive activity varies from a few microns in silicic material to 510 µm in basaltic Plinian scorias. Larger bubble size (20–25 µm) is recorded in pyroclasts from mild explosive basaltic activity [Adams et al., 2006; Carey et al., 2009; Costantini et al., 2010; Gurioli et al., 2005; Gurioli et al., 2008; Klug and Cashman, 1994; Klug et al., 2002; Lautze and Houghton, 2007; Polacci et al., 2003; Sable et al., 2006; Sable et al., 2009]. Thus, the gas vesiculation process during magma degassing is unlikely to be the source of the mesoporosity measured in volcanic tephra. Micron-scale angular voids between microlites in the clast groundmass have been observed in volcanic particles, but the origin is still unclear [Sable et al., 2006]. These voids may have been formed by contraction of the bubbles during microlite crystallization. Delmelle et al. [2005] suggested that high as values of fine volcanic ash (< 100 m) 32
Insights on tephra settling velocity from morphological observations were due to the presence of secondary minerals, including clays and sulfate salts. However, chemical analysis (XRF) and X-Ray Diffraction (XRD) indicated that the FL tephra were relatively fresh, lacking evidence of alteration mineralogy [Costantini et al., 2010]. Mesopores can also be associated with surface roughness [Carter et al., 2009; Riley et al., 2003].
2.6.2 – Terminal fall velocity of the particles The settling of a tephra particle is influenced by shape, which affects the drag forces acting against the volume forces. Several drag-prediction models have been developed during the last few decades that account for the irregularity of particle shape based on various morphological parameters. Results using different morphological parameters and different TFV models span over a wide range of values, up to ~ 50 % less than TFV of the equivalent sphere. TFV based on A and A does not vary significantly with particle shape and is up to 40% higher than TFV based on R and R (Fig. 2.1). As a result, we recommend the use of the sphericity of Riley et al. [2003] (R and R) as the sphericity of Aschenbrenner [1956] (A and A) might systematically overestimate TFV. Values of R and R show the same behavior in relation with the grain size of the samples (cf., Fig. 2.1). This shape descriptor is based on the assumption that a volcanic particle can be described approximating the shape to a scalene ellipsoid; the correction for particle circularity is then used to take into account particle roughness. The main difference in the results is the absolute values of the parameters that are lower for R. As a result, both parameters are able to catch roughly in the same way the irregularity of the shape, and it is not possible to understand a priori which method gives a more suitable shape description for TFV calculation. Our analysis and results show how gas-adsorption-derived surface area is not suitable for the determination of particle TFV, at least until complex correction factors are introduced. In fact, gas adsorption gives very high values of surface area resulting in very low values of TFV (cf., Fig. 2.5). This is due to the contribution of morphological features down to the range of nanometers. The influence of very small morphological features on drag forces is not totally understood. Specific studies have shown how small scale surface irregularities influence the drag in the intermediate flow regime by 33
Chapter 2 anticipating the transition to turbulent flow [Loth, 2008], but it is not clear how they affect particle TFV. More studies are required to characterize the scale of superficial roughness that significantly influences particle settling processes. The models of Ganser [1993] and Dellino et al. [2005], based on R and R, show a higher variability of the results than the model of Wilson and Huang [1979], as they are very sensitive to the effect of the shape (cf., Fig. 2.6). Nonetheless, all dragprediction models considered (i.e., Ganser [1993], Dellino et al. [2005] and Wilson and Huang [1979]) are empirical, so they should only be applied within experimental conditions. In particular, the model of Dellino et al. [2005] should only be applied to particles with Re >100 as it is based on large pumices ( < 1) with density ranging from 750 to 2000 kg/m3. In fact, TFV of small particles ( ≥ 1) determined with the model of Dellino et al [2005] are higher than TFV calculated for equivalent spheres. In addition, the model of Dellino et al. [2005] is originally based on the measurement of the three orthogonal axes of a particle, and, therefore, the morphologic characterization based on the approximation of the three orthogonal axes using the Feret diameters [Riley et al., 2003] can be inappropriate. However, for larger particles ( ≥ 1) values of TFV derived using the model of Dellino et al. [2005] are consistent with the values calculated using the model of Ganser [1993] and Wilson and Huang [1979]. The model presented by Wilson and Huang [1979] is validated for particles with grain size ≥ 1. TFV calculated for particle size ≤ 1 results in values up to 20 % higher than values obtained with the model of Ganser [1993] based on R. However the model of Ganser [1993] is validated in a wider range of Reynolds number (up to 105). We can conclude that, out of all the models considered in this work, the model of Ganser [1993] based on R represents the best method to calculate TFV of volcanic particles, as also suggested by Chhabra et al. [1999]. The influence of particle morphology on TFV depends on particle size. In particular, the discrepancy between TFV calculated for irregular particles and TFV calculated for the equivalent sphere increases with grain size (Fig. 2.5). This behavior is confirmed by the small ratio TFV / of small particles and the increasing trend of TFV % with grain size (cf., Fig. 2.7; i.e., sphericity between 0.4 and 1.0). Our results show how the error associated with the assumption of a spherical shape is low (< 10 %) for small particles ( ≥ 3), but it increases progressively up to 50 % for coarser grain size classes ( < 3). In addition, the variation of TFV % with grain size is also related 34
Insights on tephra settling velocity from morphological observations to flow-regime transition. In fact, the range of Reynolds number of the particles for the size and TFV range in which the trending points in Fig. 2.7b produce a break-in-slope are roughly coincident with the values of Reynolds number in which the flow regime passes from laminar, to intermediate to turbulent. Finally, our study shows how the range of variability (i.e., 2) of TFVG (R) varies in relation to the definition of the dimensional bins used to characterize grain size distributions (e.g., whole- versus half- system). We have demonstrated how the half- system can describe better the variability of particle TFV of a wide size population typical of explosive volcanic eruptions, as associated values of are reduced of approximately 0.2–0.3 m/s when the half- system is used as supposed to the whole- system. Nonetheless, for both systems it is more appropriate to indicate a range of settling velocities for each size class (e.g., average ± standard deviation) rather than giving a single value (e.g., as in Tables 2.4 and 2.5). Further statistical studies are needed in order to identify a grain size system that can better characterize the variability of particle size and TFV for large particles.
2.7 – Conclusions Our results provide important insights both on the morphological characteristics of volcanic particles and on the determination of terminal fall velocity.
Particle morphology: o Our dataset shows that there is no clear correlation between morphological parameters (i.e., sphericity and shape factor) and grain size. All particle morphological parameters are comprised in a narrow range of values (0.7–0.9) and are characterized by small standard deviations ( 0.1) among all class sizes; we can conclude that a mean value for each morphological parameter can be derived that is representative of the whole particle population. . o Sphericity calculated according to Riley et al. [2003] and the shape factor F of Wilson and Huang [1979] are more sensitive to shape variation than the sphericity of Aschenbrenner [1956]. The correction based on circularity suggested by Dellino et al. [2005] results in lower sphericity values but it does not modify the general trend with grain size. As a 35
Chapter 2 result the sphericity of Dellino et al. [2005] can be considered a good shape descriptor, but dedicated experiments on particle TFV are needed to understand if this shape parameter can be applied also to other models. o 2D
image
analysis
represents
an
easy
method
for
particle
characterization, allowing for the characterization of a large amount of particles in a relatively short time, with the only limitation of inferring a 3D characterization of the particles based on a 2D analysis. However, 2D analysis is able to catch morphological features of irregular particles and can be applied to calculate of TFV of volcanic particles. o There is an inverse correlation between values of gas-adsorption-derived specific surface area and mean diameter of the bulk samples analyzed related to the increase of particle number in a given sample volume. In contrast, measurements of absolute surface areas of individual particles show a positive correlation with particle diameter. As a result, as values derived for a given size fraction (e.g., a given class) represent the contribution of all the particles analyzed in the sample and cannot be considered representative of the surface area of individual particles or of the total particle population.
Terminal Fall Velocity: o The choice of the model used to derive specific morphological parameters (and F) and to calculate TFV significantly affects the resulting TFV values. TFV discrepancies with the spherical model are up to 50 % for the dataset analyzed. o The model of Ganser [1993] combined with the sphericity of Riley et al. [2003] is considered as the best model for the calculation of TFV of volcanic particles as it is highly sensitive to the effect of the morphology on TFV and can be applied to all particle dimensions and Reynolds number in the typical volcanic range. However, dedicated experimental studies of volcanic particle settling are required to evaluate the reliability of different models and different morphological parameters. o Gas-adsorption-derived surface area of volcanic particles is mainly due to porosity and surface irregularity on the scale of the nanometers. As a result sphericity based on gas-adsorption-derived surface area results in 36
Insights on tephra settling velocity from morphological observations very small values (~ 10-5), which are outside the range of sphericity typical for natural particles (0.3–0.9) and are associated with very low TFV (one or two orders of magnitude lower than the TFV calculated according to the other methods presented in this study) suggesting that this technique is not suitable for the determination of TFV of volcanic particles. o The application of 3D scan analysis represents a promising technique to be used in surface analysis and TFV determination for lapilli-size tephra particles (> 2 cm). o The influence of particle shape on TFV increases with particle grain size. For the smallest granulometric classes, this influence is negligible (< 0.3 m/s) and particles can be approximated as spherical ( ≥ 3; laminar flow regime). o Both the whole- and half- systems are associated with a range of TFV values that increases with class size. However, the use of the half- system is recommended as it provides a more accurate distribution of TFV values for the different grain size classes. For both systems it is more appropriate to indicate a range of settling velocities for each size class (e.g., average ± standard deviation) rather than giving a single value.
37
Chapter 2
Appendix A2.1 – List of symbols and abbreviations and their dimensions A = absolute surface area (L2) Ā = average surface area (L2) A3D = absolute surface area measured through 3D scan analysis (L2) AA = absolute surface area of the equivalent tetrakaidecahedron (Aschenbrenner, 1956) (L2) AES = absolute surface area of the equivalent sphere (L2) AGA = absolute surface area measured through gas adsorption (L2) Ap = Projected area a particle (L2) as = specific surface area (L2/M) BET = Brunauer, Emmet and Teller method for gas adsorption specific surface area determination CD = Drag coefficient DFT = Density Functional Theory d = diameter of a generic particle (L) dav = average diameter of the mesopores (L) dv = diameter of the equivalent sphere (L) F = shape factor (Wilson and Huang, 1979) FL = Fontana Lapilli g = gravity acceleration (L/T2) KMU = Kilauea Mystery Unit L, I, S = main orthogonal axes of the particle (Large, Intermediate, Small) (L) m = mass of a particle (M) Md, = Median and sorting [Inman, 1952] Pp = Projected perimeter of a particle (L) P/P0 = Relative pressure Re = Reynolds number 38
Insights on tephra settling velocity from morphological observations SHV = Soufriere Hills volcano TFV = terminal fall velocity (L/T) TFVKL = TFV calculated using the model of Kunii and Levenspiel [1969] (L/T) TFVWH = TFV calculated using the model of Wilson and Huang [1979] (L/T) TFVG = TFV calculated using the model of Ganser [1993] (L/T) TFVD = TFV calculated using the model of Dellino et al [2005] (L/T) V = Volume of a particle (L3) V3D = 3D scan derived volume of a particle (L3) VGA = Volume of gas adsorbed (L3/M) Vmeso = Volume of the mesopores (L3/M) Vmicro = Volume of the micropores (L3/M)
= standard deviation TFV = difference fraction between particle TFV and TFV of the equivalent sphere (%) = grain size class, with = - log2d, with d in mm M = Median grain size = Sorting, difference between the 16 and 84 percentile. = density of the air (M/L3) (r) = Pores density profile (L-3) s = density of a generic particle (M/L3) = sphericity = sphericity calculated according to Aschenbrenner [1956] 3D = 3D scan derived sphericity GA = gas-adsorption-derived sphericity R = sphericity calculated according to Riley et al [2003] = sphericity divided by particle circularity [Dellino et al, 2005]
39
Chapter 2 R = sphericity calculated according to Riley et al [2003] divided by particle circularity [Dellino et al, 2005] = sphericity calculated according to Aschenbrenner [1956] divided by particle circularity [Dellino et al, 2005].
A2.2 – Average values and standard deviation of the difference between the TFV of the equivalent sphere and the calculated TFV ( TFV %) for each grain size class () of the samples FL1, KMU and SHV (Captions as in Table 2.4).
TFVWH
TFVG (R)
TFVG (A)
TFVD (R)
TFVD (A)
FL1 >4
0.15 ± 0.09
0.03 ± 0.09
- 0.02 ± 0.03
- 12.79 ± 11.69
- 13.75 ± 12.63
4
- 0.01 ± 0.10
0.14 ± 0.09
- 0.02 ± 0.05
- 1.55 ± 0.62
- 2.03 ± 0.59
3
0.00 ± 0.08
0.09 ± 0.13
0.00 ± 0.06
- 0.72 ± 0.28
- 0.87 ± 0.27
2
0.06 ± 0.07
0.09 ± 0.11
- 0.01 ± 0.05
- 0.26 ± 0.19
- 0.35 ± 0.12
1
0.21 ± 0.07
0.24 ± 0.12
0.03 ± 0.05
0.17 ± 0.14
0.04 ± 0.08
0
0.32 ± 0.05
0.37 ± 0.10
0.12 ± 0.05
0.34 ± 0.11
0.20 ± 0.06
-1
0.40 ± 0.05
0.49 ± 0.08
0.23 ± 0.05
0.44 ± 0.09
0.30 ± 0.06
-2
0.38 ± 0.05
0.54 ± 0.07
0.25 ± 0.04
0.45 ± 0.09
0.28 ± 0.05
-3
0.37 ± 0.05
0.55 ± 0.06
0.27 ± 0.04
0.41 ± 0.09
0.23 ± 0.04
-4
0.39 ± 0.03
0.49 ± 0.06
0.29 ± 0.04
0.29 ± 0.08
0.15 ± 0.05
>2
- 0.04 ± 0.08
- 0.08 ± 0.11
- 0.06 ± 0.04
- 0.72 ± 0.20
- 0.72 ± 0.15
2
0.04 ± 0.07
0.01 ± 0.12
- 0.04 ± 0.04
- 0.31 ± 0.18
- 0.36 ± 0.13
1
0.22 ± 0.06
0.19 ± 0.13
0.02 ± 0.04
0.14 ± 0.14
0.05 ± 0.08
0
0.31 ± 0.05
0.30 ± 0.11
0.11 ± 0.04
0.27 ± 0.10
0.17 ± 0.05
-1
0.37 ± 0.04
0.38 ± 0.09
0.22 ± 0.03
0.30 ± 0.09
0.22 ± 0.04
-2
0.36 ± 0.04
0.47 ± 0.10
0.25 ± 0.02
0.34 ± 0.12
0.21 ± 0.05
-3
0.32 ± 0.04
0.55 ± 0.11
0.24 ± 0.02
0.39 ± 0.16
0.19 ± 0.09
-4
0.35 ± 0.05
0.37 ± 0.07
0.26 ± 0.04
0.11 ± 0.09
0.04 ± 0.04
>4
0.18 ± 0.09
0.02 ± 0.07
-0.01 ± 0.03
- 18.92 ± 13.42
- 19.57 ± 13.02
4
- 0.03 ± 0.09
0.07 ± 0.08
-0.04 ± 0.03
- 1.85 ± 0.64
- 2.16 ± 0.61
3
- 0.02 ± 0.08
0.03 ± 0.12
-0.03 ± 0.04
- 0.71 ± 0.29
- 0.80 ± 0.27
2
0.05 ± 0.08
0.07 ± 0.12
-0.02 ± 0.05
- 0.23 ± 0.20
- 0.31 ± 0.15
1
0.23 ± 0.07
0.17 ± 0.11
0.04 ± 0.05
0.12 ± 0.11
0.04 ± 0.08
0
0.33 ± 0.05
0.30 ± 0.09
0.15 ± 0.05
0.26 ± 0.08
0.19 ± 0.04
-1
0.37 ± 0.05
0.38 ± 0.07
0.22 ± 0.04
0.31 ± 0.07
0.23 ± 0.04
KMU
SHV
40
Insights on tephra settling velocity from morphological observations A2.3 – Gas adsorption on solid surfaces Background Gas adsorption is defined as the enrichment of one or more components in an interfacial layer [Gregg and Sing, 1982; IUPAC, 1972]. Interactions between the adsorbate and the adsorbent can be reversible (physisorption) or irreversible (chemisorption). Measurements of as is based on the general process of physisorption, which involves attractive dispersion forces and short-range repulsive forces. Lewis acid/base forces may come into play if either the solid or the gas is polar in nature. Physisorption is suitable for as determinations because this process does not induce structural changes in the solid surface, is achieved rapidly (generally no activation energy is required, except if small pores are present and diffusion limits the adsorption rate), is not restricted to specific sites, and may lead to filling of pores, thus allowing for pore volume measurements. At very low relative pressure (P/P0, where P and P0 are the equilibrium pressure and the saturation pressure of the gas, respectively) the first surface sites to be covered by the gas molecules are the most energetic ones. These are found on polar surface sites if the gaseous probe is polar or polarizable, within narrow pores and between the horizontal and vertical edges of surface steps. With increasing P/P0, the entire surface becomes progressively coated and the probability that a gas molecule will be adsorbed onto a previously bound molecule tends to increase. The quantity of gas adsorbed onto the solid (n) is proportional to the accessible surface area and the sample mass, and also depends on the temperature (T), P/P0 of the gas, and the nature of both the solid and the gas. For a given gas-solid system at a fixed temperature: n = f(P/P0)T, gas, solid
[A2.3.1]
Equation [A2.3.1] defines the adsorption isotherm, i.e., the relationship between the amount of gas adsorbed and P/P0 at a given T.
Gas adsorption measurements In order to remove any contaminating material which may alter the surface potential and block or fill pores, all samples were first outgassed at 300 °C until a residual pressure 4), individual particles and known number of particles. Due to the physical limit imposed by the dimensions of the sample container, only tephra with diameter between 4 and 32 mm could be analysed. BET analyses were repeated using GEMINI 2375 V4.01 gas adsorption instrument at the Laboratoire de Technologie des Poudres (LTP), Ecole Polytechnique Fédérale de Lausanne (EPFL). Sample conditioning was achieved by purging with N2 at atmospheric pressure and at 200°C. Finally, 5 samples were analysed by N2 adsorption using a step-by-step automatic adsorption apparatus built in the Laboratoire Environnement et Minéralurgie (Ecole Nationale Supérieure de Géologie, Nancy, France). Prior to each experiment, 1 g of tephra was outgassed in situ overnight at 120°C at a residual pressure 3), SHV (grain size classes: 234 and >4), and KMU (-1lithic and -1juvenile). The samples were outgassed at 300°C; the resulting isotherms are shown in figure A2.3.2.
45
Chapter 2
Figure A2.3.2. Complete cycle adsorption-desorption isotherms (-196 °C) for the samples FL2, KMU, juvenile and lithic, and SHV of different size classes. Total volume of the mesopores calculated for P/P0 = 0.95 and BET as are indicated for each sample.
46
Insights on tephra settling velocity from morphological observations Most of the samples show a normal type II isotherm [IUPAC, 1972; Sing et al., 1985] associated with an histeresys loop produced by capillary condensation of the gas into the mesopores. The samples KMU (Fig. A2.3.2 h and i) differ from the rest of the samples for the absence of the hysteresis suggesting that they can be considered as nonporous or macroporous samples. However, interpreting the pore size distribution directly from the isotherm plot can be difficult, so it is not excluded the possibility for this sample to present micro- or mesopores [Gregg and Sing, 1982]. Gas adsorption cannot asses the presence of macropores (d>50 nm). However, all plots of figure A2.3.2 do not show a plateau at saturation (P/P=1), which indicates the of macropores on the surface. Nonetheless, it is not possible to have quantitative measurements of the influence of these pores on the total specific surface area. According to the pore size distribution determined through the DFT method (Fig. A2.3.3), all the samples show a similar PSD with the only exception of sample FL1 >3 which is characterized by a larger pore volumes.
Figure A2.3.3. DFT Pore size distribution pore width (nm) vs. pore volume (cc/g) for the pore in the dimensional range of the mesopores [Rouquerol et al., 1999].
47
Chapter 2 A2.4 – Drag prediction models Wilson and Huang [1979] According to the model of Wilson and Huang [1979] the drag coefficient is expressed by the following equation:
[A2.4.1]
where F is the shape factor.
Haider and Levenspiel [1989] The model of Haider and Levenspiel [1989] give an estimation of the drag coefficient for particles with different shapes in relation of their Reynolds number. Here it is used for estimate the TFV of spherical particles. According to this model the drag coefficient of a spherical object is given by the following equation:
[A2.4.2]
Ganser [1993] The model of Ganser [1993] is an improvement of the model of Haider and Levelspiel [1989], as it gives a more accurate prediction of the drag coefficient for particle with irregular shape [Chhabra et al., 1999]. According to this model the drag coefficient is expressed by the following equation:
[A2.4.3]
48
Insights on tephra settling velocity from morphological observations The two coefficients K1 and K2 are respectively the Stokes’ shape factor ant the Newton’s shape factor, and they express the influence of the morphology on the drag coefficient. They are expressed according to the following equations:
[A2.4.4]
[A2.4.5]
where and I are the sphericity and the average diameter of the projected area of the particle respectively. The model of Ganser [1993] is recommended to be used in a range of values with Re K1 K2 ≤ 105.
Dellino et al. [2005] According to the model of Dellino et al [2005] the terminal fall velocity is expressed by the following equation:
[A2.4.6]
Where is the sphericity of the particle divided for the ratio between the projected perimeter of the particle and the perimeter of a circle with same projected area of the particle.
49
Chapter 2
References Adams, N. K., B. F. Houghton, and W. Hildreth (2006), Abrupt transitions during sustained explosive eruptions: examples from the 1912 eruption of Novarupta, Alaska, Bulletin of Volcanology, 69(2), 189-206. Alfano, F., C. Bonadonna, A. C. M. Volentik, C. B. Connor, S. F. L. Watt, D. M. Pyle, and L. J. Connor (2011), Tephra stratigraphy and eruptive volume of the May, 2008, Chait,n eruption, Chile, Bulletin of Volcanology, 73(5), 613-630. Arastoopour, H., C. H. Wang, and S. A. Weil (1982), Particle-particle interaction force in a dilute gas-solid system, Chemical Engineering Sciences, 37(9), 1379-1386. Aschenbrenner, B. C. (1956), A new method of expressing particle sphericity, J Sediment Petrol, 26(1), 15-31. Barbero, B. R., and E. S. Ureta (2011), Comparative study of different digitization techniques and their accuracy, Comput Aided Design, 43(2), 188-206. Bice, D. C. (1985), Quaternary volcanic stratigraphy of Managua, Nicaragua: correlation and source assignment for multiple overlapping plinian deposits, Geol Soc Am Bull, 96, 553-566. Blott, S. J., and K. Pye (2008), Particle shape: a review and new methods of characterization and classification, Sedimentology, 55, 31-63. Bonadonna, C., and J. C. Phillips (2003), Sedimentation from strong volcanic plumes, J Geophys Res, 108(B7), 2340-2368. Bonadonna, C., and A. Costa (in press), Modeling of tephra sedimentation from volcanic plumes, in Modeling Volcanic Processes: The Physics and Mathematics of Volcanism, edited by S. Fagents, T. Gregg and R. Lopes, Cambridge University Press. Bonadonna, C., G. G. J. Ernst, and R. S. J. Sparks (1998), Thickness variations and volume estimates of tephra fall deposits: the importance of particle Reynolds number, Journal of Volcanology and Geothermal Research, 81(3-4), 173-187. Brajlih, T., B. Valentan, I. Drstvensek, J. Balic, and V. Pogacar (2007), Testing the accuracy of ATOS (TM) 3d optical scanner measuring volumes, Annals of Daaam for 2007 & Proceedings of the 18th International Daaam Symposium, 111-112. Brunauer, S., P. H. Emmett, and E. Teller (1938), Adsorption of gases in multimulecular layers, Journal of American Chemistry Society, 60, 309-319.
50
Insights on tephra settling velocity from morphological observations Carey, R. J., B. F. Houghton, and T. Thordarson (2009), Abrupt shifts between wet and dry phases of the 1875 eruption of Askja Volcano: Microscopic evidence for macroscopic dynamics, Journal of Volcanology and Geothermal Research, 184(3-4), 256-270. Carter, A. J., M. S. Ramsey, A. J. Durant, I. P. Skilling, and A. Wolfe (2009), Micronscale roughness of volcanic surfaces from thermal infrared spectroscopy and scanning electron microscopy, J. Geophys. Res.-Solid Earth, 114. Chhabra, R. P., L. Agarwal, and N. K. Sinha (1999), Drag on non-spherical particles: an evaluation of available methods, Powder Technol, 101(3), 288-295. Costantini, L., B. F. Houghton, and C. Bonadonna (2010), Constraints on eruption dynamics of basaltic explosive activity derived from chemical and microtextural study: The example of the Fontana Lapilli Plinian eruption, Nicaragua, Journal of Volcanology and Geothermal Research, 189(3-4), 207-224. Costantini, L., C. Bonadonna, B. F. Houghton, and H. Wehrmann (2009), New physical characterization of the Fontana Lapilli basaltic Plinian eruption, Nicaragua, Bulletin of Volcanology, 71(19), 337-355. Dartevelle, S., G. G. J. Ernst, J. Stix, and A. Bernard (2002), Origin of the Mount Pinatubo climactic eruption cloud: Implications for volcanic hazards and atmospheric impacts, Geology, 30(7), 663-666. Deboer, J. H., B. C. Lippens, B. G. Linsen, Broekhof.Jc, Vandenhe.A, and T. J. Osinga (1966), T-Curve of Multimolecular N2-Adsorption, J Colloid Interf Sci, 21(4), 405-413. Dellino, P., D. Mele, R. Bonasia, G. Braia, L. La Volpe, and R. Sulpizio (2005), The analysis of the influence of pumice shape on its terminal velocity, Geophys Res Lett, 32(21), 4. Delmelle, P., F. Villieras, and M. Pelletier (2005), Surface area, porosity and water adsorption properties of fine volcanic ash particles, Bulletin of Volcanology, 67(2), 160-169. Durant, A. J., C. Bonadonna, and C. J. Horwell (2010), Atmospheric and Environmental Impacts of Volcanic Particulates, ELEMENTS, 6(4), 235-240. Ganser, G. H. (1993), A rational approach to drag prediction of spherical and nonspherical particles, Powder Technol, 77(2), 143-152. Gregg, S. J., and K. S. W. Sing (1982), Adsorption, surface area and porosity, 2ed, Academic Press, London.
51
Chapter 2 Gurioli, L., B. F. Houghton, K. V. Cashman, and R. Cioni (2005), Complex changes in eruption dynamics during the 79 AD eruption of Vesuvius, Bulletin of Volcanology, 67, 144-159. Gurioli, L., A. J. L. Harris, B. F. Houghton, M. Polacci, and M. Ripepe (2008), Textural and geophysical characterization of explosive basaltic activity at Villarrica volcano, J. Geophys. Res.-Solid Earth, 113(B8). Haider, A., and O. Levenspiel (1989), Drag Coefficient and Terminal Velocity of Spherical and Nonspherical Particles, Powder Technol, 58(1), 63-70. Holasek, R. E., and S. Self (1995), Goes Weather-Satellite Observations and Measurements of the May 18, 1980, Mount-St-Helens Eruption, J. Geophys. Res.-Solid Earth, 100(B5), 8469-8487. Inman, D. L. (1952), Measures for describing the size distribution of sediments, J Sediment Petrol, 22, 125-145. IUPAC (1972), Manual of symbols and terminology for physic-chemical quantities and units, Appendix 2, Definitions, Terminology and symbols in colloid and surface chemistry. Part. 1, Pure Applied Chemistry, 31, 578-638. Klug, C., and K. V. Cashman (1994), Vesiculation of May 18, 1980, Mount St-Helens Magma, Geology, 22(5), 468-472. Klug, C., K. V. Cashman, and C. R. Bacon (2002), Structure and physical characteristics of pumice from the climatic eruption of Mt Mazama (Crater Lake) Oregon, Bulletin of Volcanology, 64, 486-501. Kokelaar, B. P. (2002), Setting, chronology and consequences of the eruption of Soufrière Hills Volcano, Montserrat (1995-1999), in The eruption of Soufrière Hills Volcano, Montserrat, from 1995 to 1999, edited by T. H. Druitt and B. P. Kokelaar, pp. 1-43, Geological Society, London, Memoir. Kunii, D., and O. Levenspiel (1969), Fluidisation Engineering, Wiley and Sons, New York. Lautze, N. C., and B. F. Houghton (2007), Linking variable explosion style and magma textures during 2002 at Stromboli volcano, Italy, Bulletin of Volcanology, 69(4), 445-460. Loth, E. (2008), Drag of non-spherical solid particles of regular and irregular shape, Powder Technol, 182(3), 342-353. McPhie, J., G. P. L. Walker, and R. L. Christiansen (1990), Phreatomagmatic and phreatic fall and surge deposits from explosions at Kilauea volcano, Hawaii, 1790 A.D.: Keanakakoi Ash Member, Bulletin of Volcanology, 52, 334-354. 52
Insights on tephra settling velocity from morphological observations Mele, D., P. Dellino, R. Sulpizio, and G. Braia (2011), A systematic investigation on the aerodynamics of ash particles, Journal of Volcanology and Geothermal Research, 203(1-2), 1-11. Mills, O. P., W. I. Rose, and A. J. Durant (2007), Comparison of 3D Stereo SEM Shape Data With 2D Projections and BET Surface Area Data for Volcanic Ash: When 3D Might Be Advantageous, Eos Trans. AGU, 88(52), Fall Meet. Suppl., Abstract V31A-0298. Polacci, M., L. Pioli, and M. Rosi (2003), The plinian phase of the Campanian Ignimbrite eruption (Phlegrean Field, Italy): evidence from density measurements and textural characterization of pumice., Bulletin of Volcanology, 65, 418-432. Riley, C. M., W. I. Rose, and G. J. S. Bluth (2003), Quantitative shape measurements of distal volcanic ash, J. Geophys. Res.-Solid Earth, 108(B10). Rouquerol, F., J. Rouquerol, and K. Sing (1999), Adsorption by powders and porous solids. Principles, methodology and applications., 467 pp., Academic Press. Sable, J. E., B. F. Houghton, P. Del Carlo, and M. Coltelli (2006), Changing conditions of magma ascent and fragmentation during the Etna 122 BC basaltic Plinian eruption: evidence from clasts microtextures, Journal of Volcanology and Geothermal Research, 158, 333-354. Sable, J. E., B. F. Houghton, C. J. N. Wilson, and R. J. Carey (2009), Eruption mechanism during the climax of the Tarawera 1886 basaltic Plinian eruption inferred from microtextural characteristics of the deposit, in Studies in Volcanology: The Legacy of George Walker, edited by S. S. T. Thordarson, J. Larsen, K. Rowland and A. Hoskuldsson, Geological Society, London. Sing, K. S. W., D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquerol, and T. Siemieniewska (1985), Reporting physisorption data for gas/solid systems with Special Reference to the Determination of Surface Area and Porosity, Pure & Applied Chemistry, 57(4), 603-619. Sparks, R. S. J., and S. R. Young (2002), The eruption of Soufrière Hills Volcano, Montserrat (1995-1998): overview of scientific results, in The eruption of Soufrière Hills Volcano, Montserrat, from 1995 to 1999, edited by T. H. Druitt and B. P. Kokelaar, pp. 45-69, Geological Society, London, Memoir. Sugar, P., J. Sugarova, and L. Morovic (2008), Application of 3d Optical Scanning for the Shape Accuracy Analysis of Machine Parts Produced by Multi-Pass Metal Spinning, Ann Daaam, 1331-1332. Thorarinsson, S. (1944), Petrokronologista Studier pa Island, Geographes Annuales Stockholm, 26, 1-217.
53
Chapter 2 Wehrmann, H., C. Bonadonna, A. Freundt, B. F. Houghton, and S. Kutterolf (2006), A mafic plinian eruption revisited: case study of Fontana Tephra, Nicaragua, Geological Society of America Special Paper: Volcanic Hazards in Central America. Williams, S. N. (1983), Plinian airfall deposits of basaltic composition, Geology, 11, 211-214. Wilson, L., and T. C. Huang (1979), The influence of shape on the atmospheric settling velocity of volcanic ash particles, Earth and Planetary Sciences Letters, 44, 311324. Witham, C. S., C. Oppenheimer, and C. J. Horwell (2005), Volcanic ash-leachates: a review and recommendations for sampling methods, Journal of Volcanology and Geothermal Research, 141(3-4), 299-326.
54
Chapter 3
The May, 2008, Chaitén eruption1
3.1 – Introduction Before the 2008 eruption, Chaitén volcano, located in the northern part of Chilean Patagonia, to the west of the larger Michimahuida volcano, was considered as a long dormant volcanic complex [Naranjo and Stern, 2004]. It is widely believed that the last known explosive eruption of Chaitén was related to the formation of a 3-4 km diameter caldera about 9400
14
C years BP [Naranjo and Stern, 2004]. On the basis of
the collapse caldera’s volume, this event was estimated to have erupted about 4 km 3 of material. With the rejuvenation of Chaitén in 2008, new work has shown that Chaitén may have been the source of a major Holocene rhyolite pumice unit (Mic2) previously ascribed to Michimahuida, with an age 5600 years ago, on the basis of obsidian of the same composition having been found in nearby archaeological contexts [Stern et al., 2009]. In summary, there is growing evidence for previously unknown Holocene eruptions of Chaitén volcano, one of them possibly in the 17th century (Lara, personal commun.).
1
Published as: F. Alfano, C. Bonadonna, ACM Volentik, CB Connor, SFL Watt, DM Pyle and LJ Connor (2011). Tephra stratigraphy and eruptive volume of the May, 2008, Chaitén eruption, Chile. Bulletin of Volcanology 73:613–630.
55
Chapter 3 The first historical eruption of Chaitén volcano began in May 2008 [Lara, 2009]. Although the volcano was not monitored, the onset of activity appears to have been rapid, within about 36 hours of the first felt earthquake [Castro and Dingwell, 2009]. Deep, but unfelt, seismicity was recorded beneath Chaitén by a temporary network deployed in 2004-2005 [Cembrano and Lara, 2009; Lange et al., 2008]. The first major phase of explosive activity took place on May 2nd, but there is evidence that ash emissions began on May 1st, 2008 [Castro and Dingwell, 2009]. The erupted products are exclusively of a crystal-poor rhyolite, with a glass silica content of 73-76%; a composition which is rare in the Southern Andean volcanic arc [Naranjo and Stern, 2004]. At the time of writing the eruption continues, having shifted, after the first few days of explosive activity, into an extended dome-forming phase of eruption (Lara, 2009). This is believed to have been the first major explosive eruption of rhyolite since the eruption of Novarupta (Alaska) in 1912 [Carn et al., 2009; Houghton et al., 2004]. With so few historical examples of rhyolitic eruptions, it is particularly important to document the onset of Chaitén activity and the associated eruptive dynamics. Other examples of explosive young rhyolite eruptions include Askja 1875 [Sparks et al., 1981] and Taupo 186 AD [Walker, 1980]. The dispersal of tephra during the first week of the eruption affected a vast region (with ash deposited over an area > 2x105 km2), from Chile to the Atlantic coast of Argentina [Watt et al., 2009]. The eruption caused the evacuation of more than five thousand people from a 50 km radius area [Lara, 2009], led to the eventual abandonment and relocation of the town of Chaitén, and disrupted agriculture, tourism and aviation [Martin et al., 2009; Watt et al., 2009]. From the distribution of deposits in the distal area of Argentina presented in Watt et al. [2009], the May 2008 eruption of Chaitén is estimated to have generated more than 0.2 km3 of tephra in the period between May 1st and June 11th 2008. Our goal is to document the proximal–medial stratigraphy (3-25 km from the vent) of the tephra deposit from the early May activity, and relate these layers to the known sequence of explosive events known from direct observations and remotely sensed data. Moreover we want to integrate the data on distal deposits published by Watt et al. [2009] with field observations in the area between Chaitén town (10 km from the vent) and Futaleufù (75 km from the vent), in order to give a more complete description of the deposit produced by the eruption. 56
The May, 2008, Chaitén eruption The May 2008 tephra deposit consists of numerous layers, most of which can be correlated with specific explosions. The early stages of the eruption produced a complex stratigraphy characterized by at least 14 individual layers. These layers vary from extremely fine-grained ash to layers of lapilli and large blocks of both juvenile and lithic material. No clear deposits of pyroclastic density currents (PDC) were found in the area which was mapped, although patterns of vegetation damage and tree fall on the inaccessible slopes leading up to the dome appear to be consistent with the emplacement of local, damaging PDCs. All of the mapped deposits are interpreted as tephra fall from the main explosive-eruption plumes or from plumes associated with PDCs that flowed through the north-eastern valleys, which were not accessible due to the roughness of the terrain and safety reasons. Seismic data suggest that the extrusion of a new dome, which was first observed on May 21st, started on May 12th [Lara, 2009], so it can be inferred that PDCs generated during the onset of the May eruption were associated with column collapse rather than dome collapses. Although the May eruption was of only moderate volume, tephra fall seriously affected the proximal region by damaging seriously the vegetation of the surrounding forest, depositing very fine grained tephra that was remobilized, and serving as the source of lahars [Lara, 2009]. Distal areas were affected by fine ash in suspension in the atmosphere, which disrupted air traffic [Martin et al., 2009].
3.2 – The tephra fall deposit Our observations of the May tephra fall deposit were made during January 2009, about 8 months after the onset of the eruption, and were made at 69 stratigraphic stations (outcrops and pits) located between 3 and 25 km from the Chaitén dome (Fig. 3.1). These stations are distributed among four traverses, two located SE of the vent and two located N of the vent, and accessed by vehicle or on foot. Sample locations were limited by the rugged terrain and a dense temperate rainforest that surrounds Chaitén volcano. No sites within ~3 km of the vent were visited for safety reasons. Evidence of reworking was observed only for the top layers (N-O in the southeast sector, in the north sector), while the bottom and middle layers showed consistent values of thickness throughout the exposed deposit.
57
Chapter 3 Traverses of the southeast side of the volcano are located along the ~N-trending valleys of Rio Amarillo and Rio Michimahuida, about 15 km and 20 km from the volcano, respectively (Fig. 3.1). The station points along the Rio Amarillo traverse follow the “Ventisquero el Amarillo” trail. At the time of the fieldwork, access to this area was seriously compromised due to a flood that inundated the lowest part of the valley (area between RA01 and RA11) and from trees fallen across the trail, which had been topped by the weight of accumulated tephra (area between RA11 and RA02). Starting from the base camp located in RA08, where the “Sendero el Crater” trail begins, the trail was very easy to walk. The traverse stops at the station RA05 where it was not possible to go further due to the impassable Rio Amarillo. The station points along the Rio Michimahuida traverse follow a N-S dirt road which crosses N-S the valley and stops at the station point RM02. To reach the station points RM07 to RM03 it was necessary to cross the Rio Michimahuida. Tephra deposits in the SE sector are thickest along the Rio Amarillo, where they reach thicknesses of about 23 cm at stratigraphic station RA07 (16 km from the vent). The maximum thickness reached along the Rio Michimahuida is about 16 cm at stratigraphic station RM13 (21 km from the vent; Fig. 3.1). Stratigraphic stations located on the north side of the volcano are 3-15 km from the vent along a N-S trending road and along the “Sendero Michimahuida” trail, located south of the Rio Rayas (Fig. 1). The station points along the N-S trending traverse are mainly located next to the road, and are easily accessible by car. The points between CH73 and CH71 are located inside the area of the camp ground “El Volcan”. The points between F09 and F14 are located along the Sendero Michimahuida, which leads up to the northwest side of the Michimahuida glacier, in a forest area located south to the Rio Rayas. This forest, in the period of the fieldwork, was heavily covered by volcanic ash, which made it difficult to follow the trail. Total tephra deposits in this area reach a maximum thickness of 23 cm at location D19 along the N-S traverse, and 26 cm at location F12 on the E-W traverse (Fig. 3.1).
58
The May, 2008, Chaitén eruption
Figure 3.1. Map of the area around Chaitén volcano showing the four traverses in the southeastern sector (black and orange points) and on the northern sector (green and blue points). Dashed lines indicate the main roads and paths that provide access to the proximal and medial deposit.
59
Chapter 3 3.2.1 – Stratigraphy of the southeast sector A total of 16 layers, named from A to P are identified in the SE sector (RM08; Fig. 3.2a). The most complete sequence of layers in the southeastern sector is visible at locations RM08 and RM12 (Fig. 3.2b and c). The sequence lies conformably on soil and shows no evidence of reworking and contains no accretionary lapilli. The layers were distinguished from one another in the field by changes in color and grainsize. Occasionally these changes are abrupt, but other transitions are gradational. Layers J-O are missing in some sections, presumably due to erosion prior to deposition of P. A summary of the variation in thickness of the layers of the southeastern sector as measured in the most relevant stratigraphic stations, with polar coordinates, to the vent, is shown in Fig. 3.2d and 3.2e. The sections reported show the end points of the two traverses (RA05-RA12 and RM03-RM20; Fig. 3.1), the stations with maximum cumulative deposit thicknesses (RA07 and RM13; Fig. 3.1), the stations with the maximum cumulative thicknesses of the A-I layers (RA07 and RM13; Fig. 3.1) and the station with the maximum cumulative thicknesses of the layers K-M (RA05 and RM06; Fig. 3.1). The succession starts with Layer A, which rests directly on the pre-eruption soil (or the pre-eruption leaf litter) and consists of very well sorted coarse ash with ~60% pumice and ~40% lithic fragments and no matrix (all componentry information given in this paper is based on macroscopic field observations). Above this layer there is a sequence of alternating fine- and coarse-ash layers (B to O). These layers show in general gradational contacts, with the exception of layers E-F which have sharp contact. Thickness maxima could be identified for the layer-sequence A-J along both traverses: 17 cm along Rio Amarillo (RA06; Fig. 3.1) and 14 cm along Rio Michimahuida (RM13; Fig. 3.1). The variation in thickness, as clearly shown in Fig. 2d and 2e, seems to indicate that the sequence A-J was deposited under similar meteorological conditions. The stratigraphic stations where the maximum thickness is reached (RA06 and RM13; Fig. 3.1) are both located SE of the crater. A package of three massive ash layers lies above the A-J layers: a pink fine ash layer at the bottom, K; a layer of fine ash in the middle which varies from white to gray, L; a pink fine ash layer at the top, M. The sequence K-M has a clear maximum thickness of 7 cm along the Rio Michimahuida (RM06; Fig. 3.1). It was not possible to determine the thicknesses of K-M along the Rio Amarillo traverse as we could not sample further north due to the rough terrain in this 60
The May, 2008, Chaitén eruption area. The maximum thickness observed along the Rio Amarillo traverse is 3.5
RA07
1.0 1.7
2.2
>3.0
RA06
J
K
L
M
N
O
are missing. RA05
1.6
0.2
0.3
0.2
0.5
1.1
2.5
0.8
1.8
>2.6
RA11
2.5
0.1
0.4
1.4
0.8
0.9
1.3
>3.2
RA01
1.4
0.4
0.3
0.4
0.8
RA12
Table A3.3. Thicknesses (cm) of the layers as they outcrop in the stratigraphic when layers stations North of the N-S Rio Amarillo traverse of the southeast sector. The blank spaces indicate South
Chapter 3
1.0
K
85
0.3
2.0
1.4
3.4
>2.0
RM07
1.5
1.5
0.6
A
0.3
0.2
0.9
B
C
0.7
0.3
0.3
0.3
D
0.2
0.1
0.2
0.5
1.6
0.7
1.5
0.7
2.7
>4.0
RM02
E
0.4
F
0.4
0.2
G
0.2
0.6
0.1
2.8
2.0
>2.2
RM06
H
0.3
1.5
1.3
2.7
>3.5
RM05
0.7
0.2
2.0
1.2
2.3
>2.5
RM04
I
1.5
1.0
L
J
2.0
>2.5
M
N
O
RM03
North
1.8
0.2
0.6
0.3
0.4
0.1
1.2
1.2
0.4
0.4
0.5
0.3
2.7
>5.0
RM08
1.5
0.2
0.3
0.3
0.7
0.4
1.6
0.6
0.4
0.5
0.2
0.1
1.9
>2.0
RM09
1.2
0.2
0.6
0.2
1.2
1.0
0.8
1.2
1.5
>4.2
RM10
1.0
0.7
0.2
0.1
1.0
1.0
0.1
0.9
0.1
1.0
>3.4
RM01
1.0
0.6
0.2
0.4
0.8
0.7
1.8
0.6
0.8
0.8
0.5
>5.0
RM11
1.0
0.8
0.1
0.1
0.4
0.8
2.7
2.9
1.5
0.6
0.2
>2.0
RM12
1.5
0.9
5.3
3.6
2.8
0.1
0.1
>2.0
RM13
1.3
0.4
0.3
1.1
1.4
3.6
2.1
2.4
0.5
>0.5
RM14
1.0
0.3
0.5
2.7
0.8
1.1
>1.7
RM15
1.3
0.3
1.8
0.7
0.7
1.5
>1.1
RM16
1.3
0.1
1.6
0.8
0.6
0.8
>0.7
RM17
1.2
1.0
1.2
0.5
1.2
>0.1
RM18
1.5
0.8
0.6
0.4
1.2
>1.0
RM19
0.5
0.5
0.3
0.5
>1.4
RM20
South
Table A3.4. Thicknesses (cm) of the layers as they outcrop in the stratigraphic stations of the N-S Rio Michimahuida traverse of the southeast sector. The blank spaces indicate when layers are missing.
The May, 2008, Chaitén eruption
3.0
5.0
4.5
3.5
4.0
F19
1.5
2.0
2.0
2.0
4.0
3
2
3.5
2.4
F17
3.4
1.7
1.0
2.5
1.0
2.0
0.6
traces
1.2
0.5
1.0
1.5
1.3
0.7
1.5
0.5
2.0
2.5
F15
2.5
1.6
2.5
2.0
3.5
traces
F18
5.0
2.0
1.0
2.5
0.5
F13
1.0
1.0
2.5
4.0
1.5
F12
2.0
3.0
2.0
F42
1.5 3.2
6.7
F11
1.5
F10
2.5
3.0
F09
D18
3.6
CH71
1.5
D16
D17 0.1
CH72
CH73
0.2
0.5
0.5
1.0
0.8
1.5
0.2
1.5
0.9
1.6
1.0
0.7
0.2
1.1
F14
Table A3.5. Thicknesses (cm) of the layers as they outcrop in the stratigraphic stations of the E-W traverse of the north sector. The blank spaces Eastindicate when layers are missing. West
Chapter 3
F26
3.0
F25
F24
87
0.1
0.1
5 km, with an estimated average velocity of about 0.5 m/s and a brief magma ascent time. However, the viscosity (106-108 Pa s) is estimated to be at least one order of magnitude too small to produce a magma autobrecciation as a result of shear during its rise in the conduit [Castro and Dingwell, 2009].
97
Chapter 4
4.3 – Samples and methods 4.3.1 – Density and vesicularity of pumices Density distribution analysis has been carried out on a population of 100 pumices (samples from P1 to P100) of the climactic phase [i.e., Layer , May 6th; Alfano et al., 2011b] collected about 5-6 km north-east to the crater (cf., Fig. 3.4c for associated isopach maps). Pumices are roughly in the same dimensional range (2-6 cm diameter) with a relatively low volume (8.5 ± 4.9 cm3), small enough to reduce the possibility of a vescicle size distribution alterated by the presence of vescicles produced by post-fragmentation expansion. Density measurements have been carried out using a hydrostatic balance. In order to include all the superficial vesicles in the measurement, the pumices were wrapped using parafilm [Houghton and Wilson, 1989]. Results have been converted into bulk vesicularity based on the average bulk density measured on powdered samples using a water pycnometer.
4.3.2 – Textural analysis on pumices Pumice textures have been studied based on a total of 7 pumices selected from the most representative density classes of the pumice population (i.e., endmembers and intermediate; P02, P25, P26, P39, P38, P48, P70). Thin sections with generic orientation have been taken from 4 samples, and 2 oriented thin sections, orthogonal and parallel to vesicle elongation, have been taken from 3 samples (in order to highlight the presence of oriented structures within the pumices). For each of the 10 thin sections a set of 17 images were acquired at four different resolutions. An image of the entire thin section was taken using a Nikon Super Coolscan 4000 (resolution 157.5 pix/mm). 14 SEM backscattered images of limited portions of the thin section have been taken using the JEOL JSM7001F at the University of Geneva (resolutions 267, 1070 and 2670 pix/mm) following the nesting strategy descripted in Shea et al. [2010b]. Two additional images (resolution 1000x1000 pixels) have been extracted from the entire section image in order to analyse vesicles down to 0.5 mm equivalent diameter to cover the entire range of vesicle size. Images were cleaned and made binary using Adobe Photoshop CS3 and then processed using jmicrovision (www.jmicrovision.com). The image analysis was carried out in order to study the vesicle walls thickness, the morphology of the vesicles and the vesicle size distribution. 98
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Wall thickness has been studied by superimposing four grids of parallel lines (4 pixel thickness, 16 pixels spacing, oriented in 4 different directions rotated by 45°) on the images with highest magnification (2670 pix/mm). The wall thickness distribution has been determined by measuring the length of the segments obtained deleting the areas of the grid overlaying the vesicles. Morphology of the vesicles has been studied based on the frequency distribution of the aspect ratio (AR; ratio between width and length of the vesicle) and the solidity factor (SF; ratio between the area of the vesicle and the area of the convex hull of the vesicle) [Riley et al., 2003]. AR describes the elongation of the vesicle, varying from extremely elongated (0 < AR ≤ 2), very elongated (2 < AR ≤ 4), moderately elongated (4 < AR ≤ 6), slightly elongated (6 < AR ≤ 8) and not elongated (AR ≥ 8) [Blott and Pye, 2008]. SF describes the outline of the vesicle morphology, indicating a smooth and regular outline when SF is equal to 1. As SF is reduced, the irregularity of the outline becomes more irregular and the roughness increases [Riley et al., 2003]. Vesicle size distribution has been studied based on the determination of the vesicle number density per unit area (NA, mm-2) for each thin section, and a geometric size class distribution with constant ratio 10-0.1 [Sahagian and Proussevitch, 1998]. NA distributions have been calculated for each of the 17 images of each section, and a total NA distribution has been determined by convolution of the data obtained from each single image. NA distributions were converted to number of vesicles per unit volume (NV, mm-3) by dividing NA for the central value of diameter of each size class [Cheng and Lemlich, 1983].
4.3.3 – Fractal analysis on ash particles Morphologic analysis have been carried out based on the fractal analysis methodology of Carey et al.. [2000] and Maria and Carey [2007]. Analyses were carried out on 7 sets of 20 ash particles with diameter less than 1 mm obtained by sieving a sample with Md = -2.03 and = 1.35 [parameters from Inman, 1952] collected in a location situated 5-6 km north-east to the crater (cf., Fig. 3.4c for associated isopach maps). Electron backscattered images were taken at different resolutions in relation with the dimensions of the particles (1900, 3220, 4300, 6600, 99
Chapter 4 8600, 14000 and 17200 pix/mm). The resolution of the image sets were fixed in relation with the dimension of the selected particles. As a result, the population of particles analyzed is able to cover all the range of dimension down to 100 m diameter. The number of particles for each set was arbitrary chosen in order to obtain a population large enough to be statistically significant (140 ash particles in total). Images were processed using Adobe Photoshop CS3 in order to obtain an outline (1 pixel thickness) of the particles and elaborated using the dilation method [Berube and Jebrak, 1999; Maria and Carey, 2002; Maria and Carey, 2007]. The analysis has been performed using a dilation step-length of 2 pixels per interaction and measuring the relative pixel area of the outline for each interaction. The value of the perimeter (P) of the outline resulting after each interaction was calculated by dividing the area of the outline by the dilation diameter (d). Mandelbrot diagrams (P vs. d) were plotted to determine fractal dimension of the particles by analyzing the slope of the plotted curve [Mandelbrot, 1967]. Fractal spectrum plots, showing the variation of the fractal dimension in relation with the step-length, were obtained calculating the first derivative of the Mandelbrot plots [Carey et al., 2000; Maria and Carey, 2002; Maria and Carey, 2007]. Textural and structural fractal dimensions of each analyzed particle were determined from the Mandelbrot plots. An average for each set of particle was calculated in order to investigate how the morphology of particles change as a function of their dimension, and, in particular, how the vesicle size distribution can influence particle morphology in different size ranges based on the observations of the structural fractal dimension (which is related to the vesicularity of the particles) [Dellino and Liotino, 2002; Maria and Carey, 2002; Maria and Carey, 2007].
4.4 – Results 4.4.1 – Density measurements Pumice lapilli show unimodal density distribution (400-1300 kg/m3; main mode at 700 kg/m3) with a high-density tail representing 10 % of the distribution (Fig. 4.1). Vesicularity, calculated based on a bulk density of 2298 ± 28 kg m-3 (resulting from 5 measurements), ranges between 45% and 81%. No clear correlation between pumice 100
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
volume and vesicularity has been found. Analyzed samples have been selected according to the density and vesicularity distribution choosing the endmembers and the most representative classes (Fig. 4.1). Analyses on sections with generic orientation have been carried out on samples P48, P70, P38 and P26; analyses on orientated sections have been carried out on samples P02, P25 and P39.
Fig. 4.1. Density (blue bars) and vesicularity (red dashed line) distributions measured on a population of 100 pumices with diameter between 2 and 6 cm.
4.4.2 – Description of the thin sections Pumices erupted during the climactic phase of the May 2008 Chaitén eruption are characterized by highly deformed vesicles and an almost crystal free glass groundmass. A selection of images of all selected samples and for all magnifications is presented in Fig. 4.2. Vesicles tend to increase in dimension as the density is reduced, and coalescence appears more frequent for the samples with lower density (cf., P48, P02, P70 and P38). An increase of the glass fraction with particle density is also evident. Vesicle morphologies are highly irregular, with no particular difference in relation to the orientation of the sections (cf., P02, P25 and P39). Vesicles in pumices with low density show regular shapes with regular convex outlines (cf., P02, P25 and P39). High density pumices are characterized by stretched vesicles occasionally presenting indented walls characterized by concave portions of the outline (cf., P25 and P39).
101
Chapter 4 Image analysis on thin sections were carried out (Table 4.1). In order to investigate the distribution of vesicle walls, samples have been divided in two groups in relation to their density: low density samples (blue symbols; P48, P02, P38 and P70; density between 441 and 671 kg m-3) and high density samples (red symbols; P25, P39 and P26; density between 859 and 1271 kg m-3) (Fig. 4.3a). Median thickness of vesicle walls are in the range 4-8 m and 10-15 m respectively for low and high density samples, corresponding to a 2D glass fraction of 0.3-0.5 and 0.6-0.7 (Fig. 4.3b and c). Figure 4.4 shows the frequency distribution of AR measured for each thin section. AR shows unimodal distribution for all the analyzed sections with the only exception of P25 parallel section. Vesicles generally show a high degree of elongation, with median values of AR ranging between 0.4 and 0.6. A slightly higher degree of elongation is founded for samples P70 and P38 (cf., Table 4.1). Pumice sections orthogonal to vesicle elongation show AR values slightly higher than sections parallel to vesicle elongation, with a difference < 10 %. Vesicles present outlines with variable shape, with SF varying in a wide range (0.3-1.0). Outlines of the vesicles are more irregular for the most elongated vesicles (Fig. 4.5). SF varies from high values (about 0.9 = regular outline), for slightly and not elongated vesicles, and decreases progressively as elongation increases. This behavior is particularly evident in samples P25 and P39, where vesicles with high elongation can be very irregular (SF < 0.7).
Table 4.1. Summary of data for clast density, vesicularity and textural features for low (gray) and high (white) density samples. Section
P48 P02ort. P02par. P70 P38 P25ort. P25par. P39ort. P39par. P26 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
441
81
Md Vw
Md AR
Md SF
N A
N V
NVcorr,8
VSF9
4.9 0.56 0.91 11.0 7.2 9.2 8.0 0.56 0.91 7.8 5.5 6.5 582 75 7.5 0.51 0.90 11.1 8.8 11.0 670 71 6.0 0.46 0.87 20.2 16.7 23.6 671 71 4.5 0.41 0.89 14.2 11.5 15.4 14.8 0.58 0.86 9.8 7.9 12.7 859 63 13.8 0.49 0.88 7.6 5.5 7.9 12.0 0.56 0.87 12.0 10.1 14.2 1062 54 12.6 0.54 0.88 12.1 10.2 14.8 1271 45 10.1 0.49 0.90 11.0 9.7 13.6 Pumice density (kg m-3). Pumice vesicularity (%). Median thickness of vesicle walls (m). Median aspect ratio of the vesicles Median solidity factor of the vesicles Number vesicle per unit area (mm-2 x 102). Number vesicle per unit volume obtained converting NA values (mm-3 x 104). Number vesicle per unit volume corrected for the vesicularity (mm-3 x 104). Fraction % of the vesicle surface area measured on thin section (%). Fraction % of the vesicle volume calculated from NV values (%). Power law exponents of the cumulative NV distributions trends.
102
68 53 54 59 54 38 35 39 39 41
VVF10
E111
E211
63 63 60 50 53 39 44 38 37 32
1.0 1.2 1.4 1.5 1.5 1.5 1.3 1.6 1.7 1.9
3.7 3.5 3.5 4.2 3.9 3.6 3.6 4.2 3.9 4.1
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Fig. 4.2. Selection of images for all the analyzed samples with increasing density from top to bottom. First two colums of images on the left are acquired using a digital scanner Nikon Super Coolscan 4000 (resolution 157.5 pix/mm), the other three columns presents images acquired through ICM using the JEOL JSM7001F at the University of Geneva (resolutions 267, 1070 and 2670 pix/mm). Vesicles are in black, glass wall in white. Width of the images in mm is indicated for each magnification. Sample number is indicater in the right.
103
Chapter 4
Fig. 4.3. a) Thickness distribution of vesicle walls based on the measured length of the segments connecting adjacent vesicles. b) Median values of wall thickness plotted vs density showing the two different pumice cathegories of low and high density pumices. c) Fraction of glass plotted vs particle density.
Fig. 4.4. Frequency distribution of aspect ratio (AR) of the vesicles of all the low and high density pumices analyzed in this work.
104
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Fig. 4.5. Variation of the median value of the solidity factor (SF) for different degrees of elongation of the vesicles defined based on AR values: extremely elongated (AR ≤ 0.2), very elongated (0.2 < AR ≤ 0.4), moderately elongated (0.4 < AR ≤ 0.6), slightly elongated (0.6 < AR ≤ 0.8), not elongated (AR > 0.8) [Blott and Pye, 2008]. Symbols as in Fig. 4.3.
Fig. 4.6. Vesicle size distribution expressed in volume fraction corrected for the melt for all the low and high density pumices analyzed in this study.
105
Chapter 4 Despite these differences, only slight variations in the vesicle size distribution is observed. NA is similar for all samples, varying from 1.3 ± 0.5 x 103 mm-2, for the low density samples, to 1.0 ± 0.2 x 103 mm-2 for the high density samples. Average vesicle number per unit area is estimated to be 1.2 ± 0.4 x 103 mm-2. As a result, vesicle number density per unit volume (NV) gives similar values for the two density classes of samples, being equal to 9.9 ± 4.4 x 104 mm-3, for low density pumices, and to 8.7 ± 2.0 x 104 mm3
for high density pumices. These values correspond to different ranges of vesicle
volume fraction equal to 53-63 % and 32-44 % respectively for low and high density pumices.
4.4.3 – Vesicle size distribution Distribution of vesicle sizes are described plotting volume fractions (corrected for the melt) with vesicle sizes expressed as diameters of equivalent spheres (Fig. 4.6). Vesicle size shows unimodal distributions with consistent mode at 0.05-0.08 mm, with the only exception of the orthogonal section of P02 and the parallel section of P25 that show mode at 0.08-0.13 mm. Minimum size of the vesicles observed is 0.01 mm, whereas the maximum size is 3 mm. Generally, vesicles with equivalent diameter larger than 1 mm are present in the samples with the lowest density (P48 and P02). Vesicle size distribution is similar for all the analyzed samples and do not vary with the orientation of the sections nor with respect to the vesicle elongation. Volume fraction of the analyzed samples was
normalized to the average vesicle size distribution to
compare the textures of the different pumices (Fig. 4.7). The samples show roughly similar distributions. The most significant exceptions are represented by the content in vesicles < 0.03 mm for the sample P70, and the content in large vesicles (> 0.3 mm) for the low density samples (P48 and P02). Nonetheless all samples show similar values in the range 0.05-0.13 mm, in agreement with the modes of the volume fraction distributions. Converted cumulative number density (NVcorr > d; mm-3) was plotted versus particle diameter (Fig. 4.8). The distributions are characterized by two populations of vesicles following power-law trends with different slope. For small vesicles (d < 16 m), NVcorr > d follows a power-law trend with exponent (E1) in the range 1.0-1.9. For larger vesicles (d > 16 m) NVcorr > d follows a power-law trend with exponent (E2) 106
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
varying in the range 3.3-4.1. Cumulative number density distributions are characterized by a narrow range (Fig. 4.9a). Given these small differences between samples, an average distribution inferred for the bulk magma was calculated (Fig. 4.9b). As a result, the average number of vesicles per unit volume corrected for the melt (NVcorr) inferred for this explosive phase is estimated to be 1.3 ± 0.5 x 105 mm-3 and average power law coefficients E1 and E2 respectively equal to 1.4 and 3.6. Figure 4.10 shows the relation between Mass Eruption Rate [MER, kg s-1; Wilson and Walker, 1987] and the vesicle number density for the Chaitén explosion of May 6th 2008 and other studied subplinian and Plinian eruptions with basaltic, rhyolitic and phonolitic composition. Basaltic eruptions show a clear trend suggesting a relation linking MER and NV, whereas values of phonolitic eruptions seems to form a plateau. Rhyolitic eruptions seems to show an intermediate behavior. Values of the eruptive parameters of all the eruptions reported in Fig. 4.10 are collected in Table 4.2.
Fig. 4.7. Vesicle size distribution expressed in volume fraction normalized for the average volume fraction (Volume fraction / Average volume fraction) for all the analyzed samples showing the differences in vesicle populations. Symbols as in Fig. 4.3.
107
Chapter 4
Fig. 4.8. Cumulative size distribution (NVcorr > d) for all the analyzed pumices with indicated the power law trend exponents and the correlation coefficients.
108
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Fig. 4.9. a) Cumulative size distribution (NVcorr > d) for all analyzed pumices (symbols as in Fig. 4.3). b) Average cumulative size distribution corrected for the magma. Standard deviation is indicated for each point.
109
Chapter 4
Fig. 4.10. Log-Log plot of the mass eruption rate (MER) vs the vescicle number density corrected for the melt (NVcorr). Reference as in Table 4.2.
110
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Table 4.2. Summary table of main eruptive parameters of studied eruptions (HT, column height; MER, mass eruption rate; SiO2, silica wt %; ∆P/∆t, decompression rate). Values of MER in italic characters are calculated based on the column height [Wilson and Walker, 1987]. HT km
Eruption
MER kg s-1
SiO2 %
NV mm-3
∆P/∆t MPa s-1
Plinian eruptions Askja 1875a Cotopaxib Etna 122 BCc Fontana Lapilli 60 ka BPd Mt St Helense Novarupta 1912
f
Quilotoa 800 BP
g
Taupo 1.8 kah
Towada
i
Vesuvius 79 ADj
Unit C Unit D Layer 1 Layer 2 Layer 5 Unit C Unit E Microlite rich Microlite poor 1980 Ep. II Ep. III 800 BP Unit 2 Unit 3 Unit 4 Unit 5 Unit 6 A C E F EU1 fall EU2 fall EU3 max EU4 fall
23 26 33 28 29 26 26
6.8 x 107 3.5 x 107 1.6 x 108 8.1 x 107 9.3 x 107 8.5 x 107 8.5 x 107
32
1.4 x 108
53
19 22-25 17-23 35
7
66.0
55
30 32 25 23 15 26 29 23
1.9 x 10 2.0 x 108 4.0 x 107 1.7 x 108 1.0 x 107 1.0 x 106 1.0 x 106 1.0 x 108 1.0 x 1010 2.6 x 108 3.4 x 108 1.3 x 108 9.0 x 107 1.6 x 107 8 x 107 1 x 108 4 x 109
Subplinian eruptions Askja 1875a Chaiténk Etnal Izu Oshimai
a. b. c. d. e. f. g. h. i. j. k. l. m.
55-70 56.7 59.1 57.9 49 49
73-78 74-78
74.0
71.4 66.7 65.7 61.1 54.1 55.4 54.7 54.1
2.0 x 106 1.4 x 106 1.1 x 105 4.8 x 105 2.7 x 105 1.0 x 105 8.7 x 104 8.7 x 104 5.8 x 104 8.2 x 106 9.6 x 105 2.1 x 106 8.1-8.9 x 105 1.0 x 106 4.0 x 106 3.0 x 105 3.0 x 106 5.0 x 106 6.0 x 105 3.7 x 105 2.2 x 105 2.0 x 105 3.30 x 106 1.52 x 107 6.40 x 106 3.30 x 106
---------1.4 x 102 --------6.4 x 101 7.0 x 101 5.6 x 101 9.1 x 101 0.4 x 100 1.1 x 100 6.2 x 100 0.4 x 100
Unit B 8 2.6 x 106 55-70 9.0 x 105 -7 May 6th 2008 19 4.2 x 10 74.2 1.3 x 105 7.9 ± 2.8 x 100 July 22nd 1998 9 1.8 x 106 47.6 4.9 x 103 -1986 B 16 2.1 x 107 55.0 2.1 x 104 4.9 x 100 B 19 4.2 x 107 74.2 1.2 x 105 1.8 x 101 i 7 5 Towada D 16 2.1 x 10 67.8 1.9 x 10 4.1 x 101 G 12 6.7 x 106 66.1 9.0 x 103 6.3 x 100 6 5 U5 7-15 5.0 x 10 1.4-6.3 x 10 -Vesuvius 512 ADm 59.0 U7 6-9 1.0 x 106 1.2-8.9 x 105 -Carey et al [2009], Carey et al [2010]. Barberi et al [1995], Biass and Bonadonna [2011], Pistolesi et al [2011]. Coltelli et al [1998], Houghton et al [2004], Sable et al [2006a], Sable et al [2006b]. Costantini et al [2010]. Klug and Cashmann [1994]. Adams et al [2006a], Adams et al [2006b], Fierstein and Hildreth [1992]. Rosi et al [2004]. Houghton et al [2010], Sutton et al [1995], Sutton et al [2000], Wilson [1993], Wilson and Walker [1985]. Blower et al [2002], Toramaru [2006], Toramaru [1990]. Carey and Sigurdson [1987], Gurioli et al [2005], Shea et al [2011], Shea et al [2010a]. This work, Alfano et al [2011]. Bonadonna and Costa [in press], Corsaro and Pompilio [2004]. Cioni et al [2011].
111
Chapter 4
Fig. 4.11. a) textural and structural fractal dimensions plotted vs the particle diameter showing the influence of the vescicularity on particle morphology. b) average fractal spectrum for all the particle dimensions analyzed.
112
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption 4.4.4 – Fractal analisys on ash particles Average values of textural (D1) and structural (D2) fractal dimensions for each of the 7 sets of particles have been plotted versus average diameters of the analyzed particles (Fig. 4.11a). The textural fractal dimension does not show significant variation with size, being characterized by very similar values in the range 1.03-1.04 for all the particles. In contrast, the structural fractal dimension shows a weak dependence on particle size, increasing up to a maximum average value of 1.17 ± 0.03 for particles with diameter 0.14 ± 0.02 mm, to a minimum of 1.11 ± 0.02, for particles in the dimensional range 0.59 ± 0.09 mm. The same trend is confirmed by the average fractal spectrum plot (Fig. 4.11b) where the spectrum of fractal dimension is plotted versus the step-length for all the particle set. The step-length was converted from number of pixels to millimiters in order to make all the sets of particles directly comparable and have an immediate representation of the range of dimensions in which the structural features that characterize particle morphology have maximum influence. All the curves show a subhorizontal branch converging to a same value as the value of the step-lenght decreases. The second branch is characterized by an abrupt increase in value that reaches maximum values for the particles in the range of diameter 0.14 ± 0.02 mm.
Fig. 4.12. Decompression rate inferred for the explosion of May 6th 2008 of Chaitén volcano, calculated using the model of Toramaru [2005] based on NVf, in relation with the temperature and for different water content.
113
Chapter 4 4.4.5 – Decompression rate An estimation of the decompression rate that characterized the Chaitén explosion of May 6th has been calculated using the model of Toramaru [2005] defined for homogeneous nucleation of the vesicles. As input parameters, a rhyolitic magma (74 SiO2 wt %) rich in volatiles (2-5 H2O wt %) and a temperature interval in the range 775850 °C were considered [Alfano et al., 2011b; Castro and Dingwell, 2009]. Only the diffusion population vesicles [diameter < 0.01 mm; Shea et al., 2011], have been considered in the calculation assuming they correnspond to the last nucleation event before the fragmentation (NVf = 7.1 ± 2.8 x 104 mm-3). As a result, decompression rate is estimated to be in the range 7.9 ± 2.8 MPa s-1 (Fig. 12). If the total NVcorr is considered, the value of the estimated decompression rate increases up to 11.8 ± 4.2 MPa s-1 (depending on water content).
4.5 – Discussion 4.5.1 – Interpretation of the textures Pumices erupted during the climactic phase of May 6th 2008 of Chaitén volcano are characterized by a unimodal vesicle size distribution and no presence of microcrystals in the groundmass. Vesicle size distribution is characterized by a predominance of small vesicles, with modal diameter between 0.05 and 0.08 mm (cf., Fig. 4.6) that does not vary significantly in relation with the density of the clasts. Unimodal distribution and the high frequency of small vesicles suggest that magma vesiculation occurred in a very short time and relatively late during magma ascent [Klug et al., 2002], favoured by the absence of microcrystals and by the relatively low viscosity of the rhyolitic melt ( 106-108 Pa s-1) during its rapid rise into the conduit [Castro and Dingwell, 2009]. As a result, there was not enough time for magma vesicles to expand, resulting in a small amount of vesicles with diameter > 1 mm, that are only present in the pumices with low density (P48 and P02; cf, Fig. 4.2, 4.6 and 4.7). This lack of large vesicles in the dense pumices could be enhanced also by processes of collapse that produced the irregular vesicles observed in sections P25 and P39 (cf., Fig. 4.2 and 4.5), and that may also be responsible of creating the high density tail that characterizes the analyzed population of pumices.
114
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Vesicles show high elongated morphologies, with median AR in the range 0.40.6 (cf., Fig. 4.4), suggesting that the melt was affected by high degree of shear stress, yet not high enough for autobrecciation [Castro and Dingwell, 2009]. The high degree of elongation was observed in all the samples, regardless of their density. Given that the shear stress is produced by the viscous drag between the melt and the conduit walls, this suggests that the conduit was characterized by a small radius, allowing the propagation of the stress field to extend its influence to the middle of the conduit section. The high elongation could also be responsible for the developing of a permeable vesicle network within the magma and the consequent degassing of the system that is suggested by the presence of vesicles with morphology of collapse [Okumura et al., 2008; Okumura et al., 2009]. This observation is in agreement with the structural model proposed by Wicks et al. [2011]. According to their studies on the surface deformation data based on radar interferometry observations from the Japan Aerospace Exploration Agency (JAXA) DAICHI (ALOS) satellite, magma rise was produced through a dyking process controlled by the preexisting fault system [Wicks et al., 2011]. The two power law populations observed in the cumulative vesicle number density plots (cf., Fig. 4.8 and 4.9) suggest that the conditions of vesicle growth evolved through different regimes. For small vesicles, growth was dominated by diffusion, whereas expansion and coalescence regime was more important for larger vesicles [Gaonac'h et al., 1996; Klug et al., 2002]. Power law is interpreted as produced in non equilibrium conditions, which is characteristic of explosive eruptions especially when involving silica rich magmas [Blower et al., 2001; Blower et al., 2002; Mangan and Cashman, 1996]. In fact, similar trends are reported for the products of the 1875 eruption of Askja volcano, which show VSD characterized by two power-law trends with the branch representing intermediate and coarse vesicle size characterized by exponents in the range 2.3-5.1 [Carey et al., 2009]. The rapid rise and absence of microcrystals had the effect of delaying the beginning of the vesicle nucleation process, causing the magma to be in condition of high supersaturation with respect to the surrounding ambient. As a result, a mechanism of multiple and continuous nucleation of vesicles was induced, producing typical power-law trends characterized by high exponent values [Blower et al., 2001; Blower et al., 2002].
115
Chapter 4 4.5.2 – Influence of vesicularity on particle morphology The influence of vesicles on particle morphology is mainly controlled by the vesicle size distribution and the dimension of the clasts [Carey et al., 2000]. Our results show clearly this behavior, as it was possible to constrain dimensional range of particle size, between 120 and 160 m, where the vesicularity most influenced the shape. This range is roughly twice the value of the main mode of the vesicle diameters observed in the pumice samples (50-80 m). Outside of this range this influence results in lower values of the textural fractal dimension, as vesicles smaller than 120 m can only include a small range of vesicle size, whereas vesicles larger than 160 m have a size scale too big to get their morphology influenced by vesicles. This aspect suggests that for a given dimension of a volcanic fragment, varying the vesicle size distribution of a magma would produce a different effect on particle shape. As a result, the smaller the mode of the vesicle size distribution, the more fine ash particles are characterized by irregular morphology. This aspect represents an important insight into understanding the fragmentation mechanism and how it is influenced by the vesicle size distribution (see also Rust and Cashman [2011]). However, further investigations are necessary to increase the knowledge of the relation between vesicularity and morphology of the particles. in particular, vesicularity should be related to total grain size distribution of deposits. These data are unfortunately not available at the present time for the May 2008 Chaitén eruption.
4.5.3 – Insights into eruption dynamics from textural observations The multifractal morphology of ash particles indicates that the products of the May 6th 2008 explosion of Chaitén volcano is produced by magmatic fragmentation, and excludes the possibility of a magma-water-interaction-driven eruption. In fact, as vesiculatiry in phreatomagmatic fragmentation is not supposed to influence the morphology of the resulting ash [Carey et al., 2000; Dellino and Liotino, 2002; Maria and Carey, 2002; Maria and Carey, 2007]. The textural features highlighted studying Chaitén products suggest an eruption dynamics driven by a violent decompression of the magmatic system that triggered the homogeneous vesiculation of a super-saturated magma. Considering the qualitative componentry of this deposit, which is composed of 116
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption 80 % of lithic fragments [Alfano et al., 2011b], we can think that this subplinian event was generated by the disruption of the old obsidian dome and the consequent opening of a 800 m radius vent, as reported by the SERNAGEOMIN and Smithsonian institute. According to the calculated vesicle number density, for a temperature interval of 780800 °C and a water content of 2-5 wt%, the decompression rate estimated to have produced such a subplinian explosion is about 10 MPa s-1 [Toramaru, 2006]. This value of decompression rate is coherent with values calculated for past eruptions and presented by Toramaru [2006] using the decompression rate meter for homogeneous nucleation. As an example, the Chaitén explosion of May 6th shows values of column height, composition and NV (cf., Table 4.2) similar to the Subplinian episodes of the historical eruptions of Towada caldera [Blower et al., 2002; Toramaru, 1990; Toramaru, 2006], whose decompression rate is estimated to be in the range between 6.3-91.0 MPa s-1 (cf., Table 4.2). On the other hand, the decompression rate estimated for the Chaitén explosion of May 6th is lower than the estimated values for Izu-Oshima basaltic subplinian episode, which are characterized by a lower NV (cf., Table 4.2) and a decompression rate around 4.9 MPa s-1 [Toramaru, 2006]. This behavior can be related to the different magma composition and the different range of magma viscosity, which allows basaltic melts to diffuse more and makes coalescence more efficient, with the result that same decompression rates produce higher vesicle number density on rhyolitic melts. The decompression rate values calculated for the 79 AD eruption of Vesuvius [Shea et al., 2011] assuming an heterogeneous nucleation of the vesicles gives values in the range 0.4-6.2 MPa s-1, in the same order of magnitude of the values calculated for the Chaitén explosion of May 6th, even if the 79 AD Plinian eruption of Vesuvius [cf., Table 2; Gurioli et al., 2005; Shea et al., 2010a] is characterized by a NV one to two orders of magnitude higher than the NV calculated for the pumices of Chaitén. This aspect shows the important effect that the absence of microcrystals in Chaitén melt had on the eruption dynamics, as higher decompression rate are required to trigger vesiculation. As a result, the beginning of vesicle nucleation could be delayed and started only in the very latest phase of magma rising.
117
Chapter 4 The dynamics of highly explosive eruptions is controlled by many factors and one of the challenges of physical volcanology studies is to try to constrain the physical parameters that characterize an eruptive event based on the characteristics of the associated products. Volatiles represent a very important component of magmatic systems significantly influencing the dynamics of the eruption. However the relation between the evolution of the volatile fraction present into a melt and the explosivity of the eruption is not totally understood. Considering the values of MER and NV (cf. Table 4.2 and Fig. 4.10), Chaitén volcano May 6th explosion is characterized by values close to those estimated for layers 2 and 5 of Cotopaxi [Barberi et al., 1995; Biass and Bonadonna, 2011; Pistolesi et al., 2011], Etna 122 BC [Coltelli et al., 1998; Houghton et al., 2004; Sable et al., 2006a; Sable et al., 2006b] and units B, D and F of Towada volcano [Blower et al., 2002; Toramaru, 1990; Toramaru, 2006]. In all the other cases differences of at least one order of magnitude in the values of MER and NV are observed. As shown in Figure 4.10, the relation between the vesicle number density and the MER is complex, with no evident trend. The relation is also more difficult to interpret considering the high degree of uncertainty in the determination of both vesicle number density and mass eruption rate. In fact, the determination of the vesicle number density is obtained typically by the 2D analysis of pumice samples, and the density per unit volume is determined using statistical models. In addition, the obtained values of NV are valid in relation to the assumption that the analyzed pumices can be considered a population statistically representative of the whole magma involved in the eruptive process. In addition, the determination of the MER is affected by a large error which is estimated to be up to one order of magnitude [Mastin et al., 2009]. A relation between NV and MER seems to be present for basaltic systems, where higher MER are characterized by higher vesicle number densities. In phonolitic melts this trend disappears, as the values of the different eruptions shows a plateau. Considering that these two typologies of magmas show similar values of viscosity, the reason of this different behavior can be imputed to the volatile fraction, typically higher in phonolitic melts. As a result, basaltic eruptions show a vesicle number density increasing with the mass eruption rate, producing also a shifting in the eruptive style from Subplinian [Etna 1998 and Izu-Oshima 1986; Blower et al., 2002; Bonadonna and Costa, in press; Corsaro and Pompilio, 2004; Toramaru, 1990; Toramaru, 2006] to 118
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Plinian eruptions [Etna 122 BC, Fontana Lapilli and Quilotoa 800 BC; Coltelli et al., 1998; Costantini et al., 2010; Costantini et al., 2009; Houghton et al., 2004; Rosi et al., 2004; Sable et al., 2006a; Sable et al., 2006b]. Phonolithic eruptions do not show any dependence going from Subplinian [Vesuvius 512 AD; Cioni et al., 2011] to Plinian [Vesuvius 79 AD; Carey and Sigurdsson, 1987; Gurioli et al., 2005; Shea et al., 2011; Shea et al., 2010a] eruptive style. Nonetheless, more phonolitic eruptions should be studied to confirm this trend. More complex is the behavior of rhyolitic eruptions that seem to have an intermediate behavior between basaltic and phonolitic eruptions. According to the plot of Fig. 4.10, Subplinian rhyolitic eruptions [Chaitén May 6th 2008 and Towada volcano Units B, D and G; this work; Alfano et al., 2011b; Blower et al., 2002; Toramaru, 1990] seem to follow the trend of basaltic eruptions, with the exception of the Unit B of the 1875 Askja eruption [Carey et al., 2009; Carey et al., 2010]. Rhyolitic Plinian eruptions partially follow the basaltic trend [Cotopaxi layers 1, 2 and 5, Novarupta episode II and Towada units A, C, E and F; Adams et al., 2006a; Adams et al., 2006b; Barberi et al., 1995; Biass and Bonadonna, 2011; Blower et al., 2002; Fierstein and Hildreth, 1992; Pistolesi et al., 2011; Toramaru, 1990], and partially the phonolitic trend [Askja 1875 units C and D, Mt St Helens May 18 1980, Novarupta 1912 episode III and Taupo 1.8 ka; Adams et al., 2006a; Adams et al., 2006b; Carey et al., 2009; Carey et al., 2010; Fierstein and Hildreth, 1992; Houghton et al., 2010; Klug and Cashman, 1994; Sutton et al., 1995; Sutton et al., 2000; Wilson, 1993; Wilson and Walker, 1985]. This behavior suggests that rhyolitic eruptions are characterized by a more complex degassing process, depending also on the viscosity of the system. High viscosity melts favour the increasing of the vesicle number density as diffusion and coalescence is more difficult in comparison with less viscous magmas. As a result, the relation between NV and MER suggests that phonolitic melts are more able to develop a mature vesicularity due to the high volatile content and the low viscosity, whereas rhyolitic melts are limited by the high viscosity and basaltic melts by the lower content in volatiles.
4.6 – Conclusions
The juvenile products, in the size range 2-6 cm, of the climactic Subplinian explosion of May 6th 2008 of Chaitén volcano are 119
Chapter 4 characterized by a density in the range 400-1300 kg m-3 (vesicularity ranging between 54 % and 81 %) with vesicle diameters < 4mm, irregular vesicle morphologies, some vesicle collapse structures, unimodal vesicle size distributions with modal values in the range 0.050.08 mm and a total NV estimated to be 1.3 ± 0.5 x 105 mm-3.
Unimodal vesicle size distribution suggests that the magma was characterized by a non-equilibrium homogeneous nucleation that developed rapidly during the later phases of the magma rise through the conduit. Vesicle number density does not vary significantly for the different samples, producing cumulative vesicle number density that follow power-law trends indicating continuous vesicle nucleation processes. The rapid rise and the absence of microcrystals delayed the magma degassing that started at shallow levels.
Magma fragmentation was triggered by the nucleation of vesicles due to a sudden release of pressure estimated to be about 10 MPa s-1, produced by the failure of the preexisting obsidian dome during magma rise. As the conduit was opened, the eruptive activity decreased progressively with the starting of the extrusion of a new rhyolitic dome.
NV and MER values do not show a clear correlation. An increase of the MER with NV can be observed for basaltic plinian and subplinian eruptions, for rhyolitic subplinian eruptions and some rhyolitic plinian eruptions. NV seems to be constant with MER for phonolitic eruptions.
Ash particles show multifractal morphology suggesting a relation to magmatic fragmentation. The structural fractal dimension (D2), which is related to the influence of the vesicles on the morphology, varies with the diameter of the particle, reaching maximum values for particles with diameter in the range 120-160 m.
120
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Appendix A4.1 – Vesicle walls analysis Vesicle wall thickness was measured superimposing four grid of parallel lines oriented in 4 different orientations (0°, 45°, 90°, 135°, 180°) to the thin sectionimages at higher magnification (2670 pix/mm). The distribution was obtained by deleting the areas occupied by the vesicles from the grid and measuring the length of the remaining segments. The measurements taken in four different directions avoid the possibility to have a bias in the wall thickness distribution due to preferential orientation of the vesicles, resulting in a representative distribution of the vesicle wall thickness of the analyzed samples (Fig. A4.1.1). As a result, the main mode of the different distributions reproduce with a good accuracy the distribution of the glass fraction that characterizes each sample (cf., Fig. 4.3b and c).
Fig. A4.1.1. Example of analysis of the vesicle wall thickness through the different stages of the procedure.
A4.2 – Vesicle size distribution analysis For each thin section the vesicle number per unit area (NA, mm-2) were measured and a frequency distribution (NAi) was determined considering a diameter distribution following a geometric law. For each bin the central value (di; mm) was considered, and the volume of an equivalent sphere (Veq) with diameter di was calculated. The vesicle number per unit volume of each bin class (NVi; mm-3) was calculated by dividing NAi per di [Cheng and Lemlich, 1983], and a total vesicle number density (NV, mm-3) was obtained by summing the NVi of all bin class. The volume fraction of the sample (Vfi) for each bin size was calculated by multiplying NVi with the volume of the corresponding equivalent sphere and then corrected for the melt (Vficorr) by multiplying the value by a correction factor equal to the ratio between the measured and calculated vesicularity. The results for each thin section are collected in the following tables. 121
Chapter 4 Table A4.1. Section P02 orthogonal Bin mm 6.3 x 10-3 7.9 x 10-3 1.0 x 10-2 1.3 x 10-2 1.6 x 10-2 2.0 x 10-2 2.5 x 10-2 3.2 x 10-2 4.0 x 10-2 5.0 x 10-2 6.3 x 10-2 7.9 x 10-2 1.0 x 10-1 1.3 x 10-1 1.6 x 10-1 2.0 x 10-1 2.5 x 10-1 3.2 x 10-1 4.0 x 10-1 5.0 x 10-1 6.3 x 10-1 7.9 x 10-1 1.0 x 100 1.3 x 100 1.6 x 100 2.0 x 100 2.5 x 100 3.2 x 100
d mm 5.7 x 10-3 7.2 x 10-3 9.1 x 10-3 1.1 x 10-2 1.4 x 10-2 1.8 x 10-2 2.3 x 10-2 2.9 x 10-2 3.6 x 10-2 4.6 x 10-2 5.7 x 10-2 7.2 x 10-2 9.1 x 10-2 1.1 x 10-1 1.4 x 10-1 1.8 x 10-1 2.3 x 10-1 2.9 x 10-1 3.6 x 10-1 4.6 x 10-1 5.7 x 10-1 7.2 x 10-1 9.1 x 10-1 1.1 x 100 1.4 x 100 1.8 x 100 2.3 x 100 2.9 x 100
Veq mm3 9.9 x 10-8 2.0 x 10-7 3.9 x 10-7 7.8 x 10-7 1.6 x 10-6 3.1 x 10-6 6.2 x 10-6 1.2 x 10-5 2.5 x 10-5 5.0 x 10-5 9.9 x 10-5 2.0 x 10-4 3.9 x 10-4 7.8 x 10-4 1.6 x 10-3 3.1 x 10-3 6.2 x 10-3 1.2 x 10-2 2.5 x 10-2 5.0 x 10-2 9.9 x 10-2 2.0 x 10-1 3.9 x 10-1 7.8 x 10-1 1.6 x 100 3.1 x 100 6.2 x 100 1.2 x 101
NA mm-2 4.1 x 101 7.1 x 101 1.0 x 102 7.1 x 101 9.0 x 101 9.7 x 101 7.7 x 101 7.0 x 101 5.0 x 101 3.1 x 101 3.1 x 101 1.9 x 101 1.1 x 101 6.4 x 100 3.9 x 100 2.3 x 100 1.3 x 100 6.5 x 10-1 2.8 x 10-1 1.9 x 10-1 8.1 x 10-2 6.4 x 10-2 4.3 x 10-2 1.7 x 10-2 8.6 x 10-3 4.5 x 10-3 2.3 x 10-3 2.3 x 10-3
NV mm-3 7.2 x 103 9.9 x 103 1.1 x 104 6.2 x 103 6.3 x 103 5.4 x 103 3.4 x 103 2.4 x 103 1.4 x 103 6.8 x 102 5.3 x 102 2.7 x 102 1.2 x 102 5.6 x 102 2.7 x 102 1.3 x 101 5.6 x 100 2.3 x 100 7.8 x 10-1 4.1 x 10-1 1.4 x 10-1 8.9 x 10-2 4.7 x 10-2 1.5 x 10-2 6.0 x 10-3 2.5 x 10-3 1.0 x 10-3 7.9 x 10-4
Vf %
Table A4.2. Section P02 parallel Bin d Veq mm mm mm3 6.3 x 10-3 5.7 x 10-3 9.9 x 10-8 7.9 x 10-3 7.2 x 10-3 2.0 x 10-7 1.0 x 10-2 9.1 x 10-3 3.9 x 10-7 1.3 x 10-2 1.1 x 10-2 7.8 x 10-7 1.6 x 10-2 1.4 x 10-2 1.6 x 10-6 2.0 x 10-2 1.8 x 10-2 3.1 x 10-6 2.5 x 10-2 2.3 x 10-2 6.2 x 10-6 3.2 x 10-2 2.9 x 10-2 1.2 x 10-5 4.0 x 10-2 3.6 x 10-2 2.5 x 10-5 5.0 x 10-2 4.6 x 10-2 5.0 x 10-5 6.3 x 10-2 5.7 x 10-2 9.9 x 10-5 7.9 x 10-2 7.2 x 10-2 2.0 x 10-4 1.0 x 10-1 9.1 x 10-2 3.9 x 10-4 1.3 x 10-1 1.1 x 10-1 7.8 x 10-4 1.6 x 10-1 1.4 x 10-1 1.6 x 10-3 2.0 x 10-1 1.8 x 10-1 3.1 x 10-3 2.5 x 10-1 2.3 x 10-1 6.2 x 10-3 3.2 x 10-1 2.9 x 10-1 1.2 x 10-2 4.0 x 10-1 3.6 x 10-1 2.5 x 10-2 5.0 x 10-1 4.6 x 10-1 5.0 x 10-2 6.3 x 10-1 5.7 x 10-1 9.9 x 10-2 7.9 x 10-1 7.2 x 10-1 2.0 x 10-1 1.0 x 100 9.1 x 10-1 3.9 x 10-1 1.3 x 100 1.1 x 100 7.8 x 10-1 1.6 x 100 1.4 x 100 1.6 x 100 2.0 x 100 1.8 x 100 3.1 x 100 2.5 x 100 2.3 x 100 6.2 x 100 3.2 x 100 2.9 x 100 1.2 x 101
NA mm-2 7.1 x 101 1.5 x 102 1.5 x 102 1.4 x 102 1.4 x 102 1.2 x 102 1.1 x 102 7.6 x 101 6.3 x 101 4.3 x 101 2.0 x 101 1.7 x 101 8.9 x 100 5.3 x 100 3.8 x 100 2.0 x 100 1.1 x 100 4.3 x 10-1 3.6 x 10-1 9.3 x 10-2 6.6 x 10-2 3.8 x 10-2 1.8 x 10-2 1.4 x 10-2 1.8 x 10-2 4.5 x 10-3 2.3 x 10-3 --
NV mm-3 1.2 x 104 2.0 x 104 1.6 x 104 1.2 x 104 9.9 x 103 6.5 x 103 4.9 x 103 2.6 x 103 1.7 x 103 9.5 x 102 3.5 x 102 2.3 x 102 9.8 x 101 4.6 x 101 2.7 x 101 1.1 x 101 5.0 x 100 1.5 x 100 9.9 x 10-1 2.0 x 10-1 1.2 x 10-1 5.3 x 10-2 2.0 x 10-2 1.2 x 10-2 1.3 x 10-2 2.5 x 10-3 1.0 x 10-3 --
Vf %
122
0.1 0.2 0.4 0.5 1.0 1.7 2.1 3.0 3.4 3.3 5.3 5.3 4.8 4.4 4.2 4.0 3.5 2.8 1.9 2.0 1.4 1.8 1.9 1.2 0.9 0.8 0.6 1.0
Vfcorr % 0.1 0.2 0.5 0.6 1.2 2.0 2.5 3.6 4.0 3.9 6.2 6.2 5.7 5.1 4.9 4.7 4.1 3.3 2.3 2.4 1.6 2.1 2.2 1.4 1.1 0.9 0.7 1.2
0.1 0.4 0.6 1.0 1.5 2.0 3.0 3.3 4.3 4.7 3.5 4.6 3.8 3.6 4.1 3.5 3.1 1.9 2.5 1.0 1.1 1.0 0.8 0.9 2.0 0.8 0.6 --
Vfcorr % 0.2 0.5 0.8 1.2 1.9 2.5 3.8 4.1 5.4 5.8 4.4 5.8 4.8 4.5 5.2 4.4 3.9 2.3 3.1 1.3 1.4 1.3 1.0 1.2 2.5 1.0 0.8 --
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Table A4.3. Section P25 orthogonal Bin mm 6.3 x 10-3 7.9 x 10-3 1.0 x 10-2 1.3 x 10-2 1.6 x 10-2 2.0 x 10-2 2.5 x 10-2 3.2 x 10-2 4.0 x 10-2 5.0 x 10-2 6.3 x 10-2 7.9 x 10-2 1.0 x 10-1 1.3 x 10-1 1.6 x 10-1 2.0 x 10-1 2.5 x 10-1 3.2 x 10-1 4.0 x 10-1 5.0 x 10-1 6.3 x 10-1 7.9 x 10-1 1.0 x 100 1.3 x 100 1.6 x 100 2.0 x 100 2.5 x 100 3.2 x 100
d mm 5.7 x 10-3 7.2 x 10-3 9.1 x 10-3 1.1 x 10-2 1.4 x 10-2 1.8 x 10-2 2.3 x 10-2 2.9 x 10-2 3.6 x 10-2 4.6 x 10-2 5.7 x 10-2 7.2 x 10-2 9.1 x 10-2 1.1 x 10-1 1.4 x 10-1 1.8 x 10-1 2.3 x 10-1 2.9 x 10-1 3.6 x 10-1 4.6 x 10-1 5.7 x 10-1 7.2 x 10-1 9.1 x 10-1 1.1 x 100 1.4 x 100 1.8 x 100 2.3 x 100 2.9 x 100
Veq mm3 9.9 x 10-8 2.0 x 10-7 3.9 x 10-7 7.8 x 10-7 1.6 x 10-6 3.1 x 10-6 6.2 x 10-6 1.2 x 10-5 2.5 x 10-5 5.0 x 10-5 9.9 x 10-5 2.0 x 10-4 3.9 x 10-4 7.8 x 10-4 1.6 x 10-3 3.1 x 10-3 6.2 x 10-3 1.2 x 10-2 2.5 x 10-2 5.0 x 10-2 9.9 x 10-2 2.0 x 10-1 3.9 x 10-1 7.8 x 10-1 1.6 x 100 3.1 x 100 6.2 x 100 1.2 x 101
NA mm-2 6.1 x 101 1.3 x 102 1.4 x 102 1.4 x 102 1.2 x 102 1.1 x 102 9.2 x 101 7.5 x 101 4.7 x 101 3.3 x 101 1.7 x 101 1.3 x 101 6.2 x 100 2.9 x 100 1.7 x 100 1.1 x 100 5.8 x 10-1 2.5 x 10-1 1.0 x 10-1 9.3 x 10-2 4.7 x 10-2 2.3 x 10-2 1.2 x 10-2 ------
NV mm-3 1.1 x 104 1.7 x 104 1.5 x 104 1.2 x 104 8.3 x 103 6.0 x 103 4.0 x 103 2.6 x 103 1.3 x 103 7.2 x 102 2.9 x 102 1.8 x 102 6.8 x 101 2.5 x 101 1.2 x 101 6.2 x 100 2.6 x 100 8.8 x 10-1 2.8 x 10-1 2.0 x 10-1 8.1 x 10-2 3.2 x 10-2 1.3 x 10-2 ------
Vf %
Table A4.4. Section P25 parallel Bin d Veq mm mm mm3 6.3 x 10-3 5.7 x 10-3 9.9 x 10-8 7.9 x 10-3 7.2 x 10-3 2.0 x 10-7 1.0 x 10-2 9.1 x 10-3 3.9 x 10-7 1.3 x 10-2 1.1 x 10-2 7.8 x 10-7 1.6 x 10-2 1.4 x 10-2 1.6 x 10-6 2.0 x 10-2 1.8 x 10-2 3.1 x 10-6 2.5 x 10-2 2.3 x 10-2 6.2 x 10-6 3.2 x 10-2 2.9 x 10-2 1.2 x 10-5 4.0 x 10-2 3.6 x 10-2 2.5 x 10-5 5.0 x 10-2 4.6 x 10-2 5.0 x 10-5 6.3 x 10-2 5.7 x 10-2 9.9 x 10-5 7.9 x 10-2 7.2 x 10-2 2.0 x 10-4 1.0 x 10-1 9.1 x 10-2 3.9 x 10-4 1.3 x 10-1 1.1 x 10-1 7.8 x 10-4 1.6 x 10-1 1.4 x 10-1 1.6 x 10-3 2.0 x 10-1 1.8 x 10-1 3.1 x 10-3 2.5 x 10-1 2.3 x 10-1 6.2 x 10-3 3.2 x 10-1 2.9 x 10-1 1.2 x 10-2 4.0 x 10-1 3.6 x 10-1 2.5 x 10-2 5.0 x 10-1 4.6 x 10-1 5.0 x 10-2 6.3 x 10-1 5.7 x 10-1 9.9 x 10-2 7.9 x 10-1 7.2 x 10-1 2.0 x 10-1 1.0 x 100 9.1 x 10-1 3.9 x 10-1 1.3 x 100 1.1 x 100 7.8 x 10-1 1.6 x 100 1.4 x 100 1.6 x 100 2.0 x 100 1.8 x 100 3.1 x 100 2.5 x 100 2.3 x 100 6.2 x 100 3.2 x 100 2.9 x 100 1.2 x 101
NA mm-2 3.5 x 101 1.0 x 102 8.7 x 101 6.9 x 101 9.2 x 101 8.5 x 101 8.0 x 101 6.2 x 101 5.5 x 101 3.5 x 101 2.3 x 101 1.4 x 101 7.3 x 100 5.0 x 100 3.0 x 100 1.5 x 100 6.9 x 10-1 1.9 x 10-1 8.9 x 10-2 7.5 x 10-2 3.9 x 10-2 2.3 x 10-2 1.3 x 10-2 8.2 x 10-3 4.9 x 10-3 3.3 x 10-3 ---
NV mm-3 6.2 x 103 1.4 x 104 9.5 x 103 6.0 x 103 6.4 x 103 4.7 x 103 3.5 x 103 2.1 x 103 1.5 x 103 7.7 x 102 4.0 x 102 1.9 x 102 8.1 x 101 4.3 x 101 2.1 x 101 8.4 x 100 3.0 x 100 6.7 x 10-1 2.5 x 10-1 1.7 x 10-1 6.9 x 10-2 3.2 x 10-2 1.4 x 10-2 7.2 x 10-3 3.4 x 10-3 1.8 x 10-3 ---
Vf %
123
0.1 0.3 0.6 0.9 1.3 1.9 2.5 3.3 3.2 3.6 2.8 3.5 2.7 2.0 1.9 1.9 1.6 1.1 0.7 1.0 0.8 0.6 0.5 ------
Vfcorr % 0.2 0.6 1.0 1.5 2.1 3.0 4.0 5.3 5.2 5.8 4.6 5.6 4.3 3.2 3.0 3.1 2.6 1.8 1.1 1.6 1.3 1.0 0.8 ------
0.1 0.3 0.4 0.5 1.0 1.5 2.2 2.7 3.8 3.8 3.9 3.8 3.2 3.4 3.2 2.6 1.9 0.8 0.6 0.8 0.7 0.6 0.6 0.6 0.5 0.6 ---
Vfcorr % 0.1 0.4 0.5 0.7 1.4 2.1 3.1 3.8 5.4 5.5 5.6 5.4 4.5 4.8 4.6 3.7 2.7 1.2 0.9 1.2 1.0 0.9 0.8 0.8 0.8 0.8 ---
Chapter 4
Table A4.5. Section P26 Bin mm 6.3 x 10-3 7.9 x 10-3 1.0 x 10-2 1.3 x 10-2 1.6 x 10-2 2.0 x 10-2 2.5 x 10-2 3.2 x 10-2 4.0 x 10-2 5.0 x 10-2 6.3 x 10-2 7.9 x 10-2 1.0 x 10-1 1.3 x 10-1 1.6 x 10-1 2.0 x 10-1 2.5 x 10-1 3.2 x 10-1 4.0 x 10-1 5.0 x 10-1 6.3 x 10-1 7.9 x 10-1 1.0 x 100 1.3 x 100 1.6 x 100 2.0 x 100 2.5 x 100 3.2 x 100
d mm 5.7 x 10-3 7.2 x 10-3 9.1 x 10-3 1.1 x 10-2 1.4 x 10-2 1.8 x 10-2 2.3 x 10-2 2.9 x 10-2 3.6 x 10-2 4.6 x 10-2 5.7 x 10-2 7.2 x 10-2 9.1 x 10-2 1.1 x 10-1 1.4 x 10-1 1.8 x 10-1 2.3 x 10-1 2.9 x 10-1 3.6 x 10-1 4.6 x 10-1 5.7 x 10-1 7.2 x 10-1 9.1 x 10-1 1.1 x 100 1.4 x 100 1.8 x 100 2.3 x 100 2.9 x 100
Veq mm3 9.9 x 10-8 2.0 x 10-7 3.9 x 10-7 7.8 x 10-7 1.6 x 10-6 3.1 x 10-6 6.2 x 10-6 1.2 x 10-5 2.5 x 10-5 5.0 x 10-5 9.9 x 10-5 2.0 x 10-4 3.9 x 10-4 7.8 x 10-4 1.6 x 10-3 3.1 x 10-3 6.2 x 10-3 1.2 x 10-2 2.5 x 10-2 5.0 x 10-2 9.9 x 10-2 2.0 x 10-1 3.9 x 10-1 7.8 x 10-1 1.6 x 100 3.1 x 100 6.2 x 100 1.2 x 101
NA mm-2 8.2 x 101 1.9 x 102 1.8 x 102 1.6 x 102 1.3 x 102 1.1 x 102 7.5 x 101 6.0 x 101 4.5 x 101 2.7 x 101 1.9 x 101 9.2 x 100 5.8 x 100 3.5 x 100 2.0 x 100 5.5 x 10-1 1.7 x 10-1 1.1 x 10-1 4.2 x 10-2 3.6 x 10-2 1.2 x 10-2 --------
NV mm-3 1.4 x 104 2.6 x 104 2.0 x 104 1.4 x 104 8.8 x 103 6.2 x 103 3.3 x 103 2.1 x 103 1.2 x 103 5.9 x 102 3.3 x 102 1.3 x 102 6.4 x 101 3.1 x 101 1.4 x 101 3.1 x 100 7.5 x 10-1 3.7 x 10-1 1.2 x 10-1 8.0 x 10-2 2.1 x 10-2 --------
Vf %
Table A4.6. Section P38 Bin d Veq mm mm mm3 6.3 x 10-3 5.7 x 10-3 9.9 x 10-8 7.9 x 10-3 7.2 x 10-3 2.0 x 10-7 1.0 x 10-2 9.1 x 10-3 3.9 x 10-7 1.3 x 10-2 1.1 x 10-2 7.8 x 10-7 1.6 x 10-2 1.4 x 10-2 1.6 x 10-6 2.0 x 10-2 1.8 x 10-2 3.1 x 10-6 2.5 x 10-2 2.3 x 10-2 6.2 x 10-6 3.2 x 10-2 2.9 x 10-2 1.2 x 10-5 4.0 x 10-2 3.6 x 10-2 2.5 x 10-5 5.0 x 10-2 4.6 x 10-2 5.0 x 10-5 6.3 x 10-2 5.7 x 10-2 9.9 x 10-5 7.9 x 10-2 7.2 x 10-2 2.0 x 10-4 1.0 x 10-1 9.1 x 10-2 3.9 x 10-4 1.3 x 10-1 1.1 x 10-1 7.8 x 10-4 1.6 x 10-1 1.4 x 10-1 1.6 x 10-3 2.0 x 10-1 1.8 x 10-1 3.1 x 10-3 2.5 x 10-1 2.3 x 10-1 6.2 x 10-3 3.2 x 10-1 2.9 x 10-1 1.2 x 10-2 4.0 x 10-1 3.6 x 10-1 2.5 x 10-2 5.0 x 10-1 4.6 x 10-1 5.0 x 10-2 6.3 x 10-1 5.7 x 10-1 9.9 x 10-2 7.9 x 10-1 7.2 x 10-1 2.0 x 10-1 1.0 x 100 9.1 x 10-1 3.9 x 10-1 1.3 x 100 1.1 x 100 7.8 x 10-1 1.6 x 100 1.4 x 100 1.6 x 100 2.0 x 100 1.8 x 100 3.1 x 100 2.5 x 100 2.3 x 100 6.2 x 100 3.2 x 100 2.9 x 100 1.2 x 101
NA mm-2 1.0 x 102 1.9 x 102 1.8 x 102 2.0 x 102 1.8 x 102 1.5 x 102 1.2 x 102 1.0 x 102 7.4 x 101 5.0 x 101 3.1 x 101 1.8 x 101 8.6 x 100 3.7 x 100 2.6 x 100 9.8 x 10-1 6.4 x 10-1 3.1 x 10-1 1.3 x 10-1 6.7 x 10-2 3.4 x 10-2 1.5 x 10-2 6.8 x 10-3 ------
NV mm-3 1.7 x 104 2.6 x 104 2.0 x 104 1.8 x 104 1.2 x 104 8.4 x 103 5.4 x 103 3.6 x 103 2.1 x 103 1.1 x 103 5.3 x 102 2.5 x 102 9.5 x 101 3.2 x 101 1.8 x 101 5.4 x 100 2.8 x 100 1.1 x 100 3.6 x 10-1 1.5 x 10-1 5.9 x 10-2 2.1 x 10-2 7.5 x 10-3 ------
Vf %
124
0.1 0.5 0.8 1.1 1.4 1.9 2.0 2.6 3.1 2.9 3.3 2.5 2.5 2.4 2.1 1.0 0.5 0.5 0.3 0.4 0.2 --------
Vfcorr % 0.2 0.7 1.1 1.5 1.9 2.7 2.8 3.6 4.3 4.1 4.6 3.5 3.5 3.3 3.0 1.3 0.7 0.6 0.4 0.6 0.3 --------
0.2 0.5 0.8 1.4 1.9 2.6 3.4 4.5 5.1 5.5 5.3 4.8 3.7 2.5 2.8 1.7 1.7 1.3 0.9 0.7 0.6 0.4 0.3 ------
Vfcorr % 0.2 0.7 1.1 1.9 2.6 3.5 4.5 6.1 6.8 7.4 7.1 6.5 5.0 3.4 3.7 2.3 2.3 1.8 1.2 1.0 0.8 0.6 0.4 ------
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Table A4.7. Section P39 orthogonal Veq mm3 9.9 x 10-8 2.0 x 10-7 3.9 x 10-7 7.8 x 10-7 1.6 x 10-6 3.1 x 10-6 6.2 x 10-6 1.2 x 10-5 2.5 x 10-5 5.0 x 10-5 9.9 x 10-5 2.0 x 10-4 3.9 x 10-4 7.8 x 10-4 1.6 x 10-3 3.1 x 10-3 6.2 x 10-3 1.2 x 10-2 2.5 x 10-2 5.0 x 10-2 9.9 x 10-2 2.0 x 10-1 3.9 x 10-1 7.8 x 10-1 1.6 x 100 3.1 x 100 6.2 x 100 1.2 x 101
NA mm-2 9.8 x 101 1.7 x 102 1.7 x 102 1.5 x 102 1.5 x 102 1.3 x 102 1.1 x 102 8.8 x 101 5.8 x 101 3.9 x 101 1.9 x 101 1.1 x 101 6.6 x 100 3.2 x 100 1.4 x 100 6.3 x 10-1 3.8 x 10-1 2.0 x 10-1 7.2 x 10-2 3.0 x 10-2 7.6 x 10-3 7.6 x 10-3 -------
NV mm-3 1.7 x 104 2.4 x 104 1.9 x 104 1.3 x 104 1.0 x 104 7.2 x 103 4.7 x 103 3.1 x 103 1.6 x 103 8.6 x 102 3.4 x 102 1.5 x 102 7.3 x 101 2.8 x 101 1.0 x 101 3.5 x 100 1.6 x 100 6.9 x 10-1 2.0 x 10-1 6.6 x 10-2 1.3 x 10-2 1.0 x 10-2 -------
Vf % 0.17 0.47 0.73 1.00 1.64 2.25 2.91 3.82 3.99 4.27 3.33 2.88 2.86 2.23 1.56 1.08 1.02 0.86 0.49 0.33 0.13 0.21 -------
Vfcorr % 0.24 0.66 1.03 1.40 2.31 3.17 4.10 5.38 5.61 6.01 4.69 4.05 4.02 3.13 2.20 1.52 1.44 1.22 0.69 0.46 0.18 0.29 -------
Table A4.8. Section P39 parallel Bin d Veq mm mm mm3 6.3 x 10-3 5.7 x 10-3 9.9 x 10-8 7.9 x 10-3 7.2 x 10-3 2.0 x 10-7 1.0 x 10-2 9.1 x 10-3 3.9 x 10-7 1.3 x 10-2 1.1 x 10-2 7.8 x 10-7 1.6 x 10-2 1.4 x 10-2 1.6 x 10-6 2.0 x 10-2 1.8 x 10-2 3.1 x 10-6 2.5 x 10-2 2.3 x 10-2 6.2 x 10-6 3.2 x 10-2 2.9 x 10-2 1.2 x 10-5 4.0 x 10-2 3.6 x 10-2 2.5 x 10-5 5.0 x 10-2 4.6 x 10-2 5.0 x 10-5 6.3 x 10-2 5.7 x 10-2 9.9 x 10-5 7.9 x 10-2 7.2 x 10-2 2.0 x 10-4 1.0 x 10-1 9.1 x 10-2 3.9 x 10-4 1.3 x 10-1 1.1 x 10-1 7.8 x 10-4 1.6 x 10-1 1.4 x 10-1 1.6 x 10-3 2.0 x 10-1 1.8 x 10-1 3.1 x 10-3 2.5 x 10-1 2.3 x 10-1 6.2 x 10-3 3.2 x 10-1 2.9 x 10-1 1.2 x 10-2 4.0 x 10-1 3.6 x 10-1 2.5 x 10-2 5.0 x 10-1 4.6 x 10-1 5.0 x 10-2 6.3 x 10-1 5.7 x 10-1 9.9 x 10-2 7.9 x 10-1 7.2 x 10-1 2.0 x 10-1 1.0 x 100 9.1 x 10-1 3.9 x 10-1 1.3 x 100 1.1 x 100 7.8 x 10-1 1.6 x 100 1.4 x 100 1.6 x 100 2.0 x 100 1.8 x 100 3.1 x 100 2.5 x 100 2.3 x 100 6.2 x 100 3.2 x 100 2.9 x 100 1.2 x 101
NA mm-2 8.5 x 101 1.8 x 102 1.7 x 102 1.7 x 102 1.7 x 102 1.2 x 102 1.0 x 102 8.5 x 101 5.2 x 101 3.8 x 101 1.9 x 101 1.1 x 101 4.0 x 100 2.6 x 100 1.0 x 100 6.9 x 10-1 4.8 x 10-1 2.5 x 10-1 7.4 x 10-2 5.6 x 10-2 2.5 x 10-2 1.6 x 10-2 -------
NV mm-3 1.5 x 104 2.5 x 104 1.8 x 104 1.5 x 104 1.2 x 104 6.7 x 103 4.4 x 103 3.0 x 103 1.4 x 103 8.4 x 102 3.3 x 102 1.5 x 102 4.4 x 101 2.3 x 101 7.2 x 100 3.8 x 100 2.1 x 100 8.8 x 10-1 2.0 x 10-1 1.2 x 10-1 4.3 x 10-2 2.2 x 10-2 -------
Vf %
Vfcorr % 0.2 0.7 1.1 1.7 2.6 3.0 4.0 5.4 5.2 6.0 4.8 4.2 2.5 2.6 1.6 1.7 1.9 1.6 0.7 0.9 0.6 0.6 -------
Bin mm 6.3 x 10-3 7.9 x 10-3 1.0 x 10-2 1.3 x 10-2 1.6 x 10-2 2.0 x 10-2 2.5 x 10-2 3.2 x 10-2 4.0 x 10-2 5.0 x 10-2 6.3 x 10-2 7.9 x 10-2 1.0 x 10-1 1.3 x 10-1 1.6 x 10-1 2.0 x 10-1 2.5 x 10-1 3.2 x 10-1 4.0 x 10-1 5.0 x 10-1 6.3 x 10-1 7.9 x 10-1 1.0 x 100 1.3 x 100 1.6 x 100 2.0 x 100 2.5 x 100 3.2 x 100
d mm 5.7 x 10-3 7.2 x 10-3 9.1 x 10-3 1.1 x 10-2 1.4 x 10-2 1.8 x 10-2 2.3 x 10-2 2.9 x 10-2 3.6 x 10-2 4.6 x 10-2 5.7 x 10-2 7.2 x 10-2 9.1 x 10-2 1.1 x 10-1 1.4 x 10-1 1.8 x 10-1 2.3 x 10-1 2.9 x 10-1 3.6 x 10-1 4.6 x 10-1 5.7 x 10-1 7.2 x 10-1 9.1 x 10-1 1.1 x 100 1.4 x 100 1.8 x 100 2.3 x 100 2.9 x 100
125
0.1 0.5 0.7 1.2 1.8 2.1 2.8 3.7 3.5 4.1 3.3 2.9 1.7 1.8 1.1 1.2 1.3 1.1 0.5 0.6 0.4 0.4 -------
Chapter 4 Table A4.9. Section P48 Bin mm 6.3 x 10-3 7.9 x 10-3 1.0 x 10-2 1.3 x 10-2 1.6 x 10-2 2.0 x 10-2 2.5 x 10-2 3.2 x 10-2 4.0 x 10-2 5.0 x 10-2 6.3 x 10-2 7.9 x 10-2 1.0 x 10-1 1.3 x 10-1 1.6 x 10-1 2.0 x 10-1 2.5 x 10-1 3.2 x 10-1 4.0 x 10-1 5.0 x 10-1 6.3 x 10-1 7.9 x 10-1 1.0 x 100 1.3 x 100 1.6 x 100 2.0 x 100 2.5 x 100 3.2 x 100
d mm 5.7 x 10-3 7.2 x 10-3 9.1 x 10-3 1.1 x 10-2 1.4 x 10-2 1.8 x 10-2 2.3 x 10-2 2.9 x 10-2 3.6 x 10-2 4.6 x 10-2 5.7 x 10-2 7.2 x 10-2 9.1 x 10-2 1.1 x 10-1 1.4 x 10-1 1.8 x 10-1 2.3 x 10-1 2.9 x 10-1 3.6 x 10-1 4.6 x 10-1 5.7 x 10-1 7.2 x 10-1 9.1 x 10-1 1.1 x 100 1.4 x 100 1.8 x 100 2.3 x 100 2.9 x 100
Veq mm3 9.9 x 10-8 2.0 x 10-7 3.9 x 10-7 7.8 x 10-7 1.6 x 10-6 3.1 x 10-6 6.2 x 10-6 1.2 x 10-5 2.5 x 10-5 5.0 x 10-5 9.9 x 10-5 2.0 x 10-4 3.9 x 10-4 7.8 x 10-4 1.6 x 10-3 3.1 x 10-3 6.2 x 10-3 1.2 x 10-2 2.5 x 10-2 5.0 x 10-2 9.9 x 10-2 2.0 x 10-1 3.9 x 10-1 7.8 x 10-1 1.6 x 100 3.1 x 100 6.2 x 100 1.2 x 101
NA mm-2 3.0 x 101 8.6 x 101 1.1 x 102 1.2 x 102 1.4 x 102 1.5 x 102 1.3 x 102 1.1 x 102 8.7 x 101 6.6 x 101 3.7 x 101 2.1 x 101 1.1 x 101 4.9 x 100 2.5 x 100 1.5 x 100 7.7 x 10-1 2.1 x 10-1 2.1 x 10-1 8.5 x 10-2 5.5 x 10-2 2.1 x 10-2 2.2 x 10-2 5.5 x 10-3 5.5 x 10-3 3.7 x 10-3 -1.8 x 10-3
NV mm-3 5.2 x 103 1.2 x 104 1.2 x 104 1.0 x 104 9.6 x 103 8.4 x 103 5.5 x 103 3.9 x 103 2.4 x 103 1.4 x 103 6.5 x 102 2.9 x 102 1.3 x 102 4.3 x 101 1.7 x 101 8.5 x 100 3.4 x 100 7.4 x 10-1 5.7 x 10-1 1.9 x 10-1 9.7 x 10-2 3.0 x 10-2 2.4 x 10-2 4.8 x 10-3 3.8 x 10-3 2.0 x 10-3 -6.4 x 10-4
Vf %
Table A4.10. Section P70 Bin d Veq mm mm mm3 6.3 x 10-3 5.7 x 10-3 9.9 x 10-8 7.9 x 10-3 7.2 x 10-3 2.0 x 10-7 1.0 x 10-2 9.1 x 10-3 3.9 x 10-7 1.3 x 10-2 1.1 x 10-2 7.8 x 10-7 1.6 x 10-2 1.4 x 10-2 1.6 x 10-6 2.0 x 10-2 1.8 x 10-2 3.1 x 10-6 2.5 x 10-2 2.3 x 10-2 6.2 x 10-6 3.2 x 10-2 2.9 x 10-2 1.2 x 10-5 4.0 x 10-2 3.6 x 10-2 2.5 x 10-5 5.0 x 10-2 4.6 x 10-2 5.0 x 10-5 6.3 x 10-2 5.7 x 10-2 9.9 x 10-5 7.9 x 10-2 7.2 x 10-2 2.0 x 10-4 1.0 x 10-1 9.1 x 10-2 3.9 x 10-4 1.3 x 10-1 1.1 x 10-1 7.8 x 10-4 1.6 x 10-1 1.4 x 10-1 1.6 x 10-3 2.0 x 10-1 1.8 x 10-1 3.1 x 10-3 2.5 x 10-1 2.3 x 10-1 6.2 x 10-3 3.2 x 10-1 2.9 x 10-1 1.2 x 10-2 4.0 x 10-1 3.6 x 10-1 2.5 x 10-2 5.0 x 10-1 4.6 x 10-1 5.0 x 10-2 6.3 x 10-1 5.7 x 10-1 9.9 x 10-2 7.9 x 10-1 7.2 x 10-1 2.0 x 10-1 1.0 x 100 9.1 x 10-1 3.9 x 10-1 1.3 x 100 1.1 x 100 7.8 x 10-1 1.6 x 100 1.4 x 100 1.6 x 100 2.0 x 100 1.8 x 100 3.1 x 100 2.5 x 100 2.3 x 100 6.2 x 100 3.2 x 100 2.9 x 100 1.2 x 101
NA mm-2 1.1 x 102 2.7 x 102 2.8 x 102 3.0 x 102 2.9 x 102 2.8 x 102 1.9 x 102 1.2 x 102 7.5 x 101 4.8 x 101 2.5 x 101 1.2 x 101 4.9 x 100 3.3 x 100 1.4 x 100 8.1 x 10-1 3.1 x 10-1 1.0 x 10-1 5.6 x 10-2 2.1 x 10-2 1.4 x 10-2 6.2 x 10-3 -------
NV mm-3 2.0 x 104 3.8 x 104 3.1 x 104 2.6 x 104 2.0 x 104 1.5 x 104 8.5 x 103 4.1 x 103 2.1 x 103 1.1 x 103 4.3 x 102 1.6 x 102 5.4 x 101 2.8 x 101 9.5 x 100 4.5 x 100 1.4 x 100 3.5 x 10-1 1.6 x 10-1 4.6 x 10-2 2.4 x 10-2 8.6 x 10-3 -------
Vf %
126
0.1 0.2 0.5 0.8 1.5 2.6 3.4 4.8 5.9 7.2 6.4 5.8 4.9 3.4 2.7 2.6 2.1 0.9 1.4 0.9 1.0 0.6 0.9 0.4 0.6 0.6 -0.8
Vfcorr % 0.1 0.3 0.6 1.0 1.9 3.4 4.4 6.2 7.6 9.2 8.2 7.4 6.3 4.3 3.5 3.4 2.7 1.2 1.8 1.2 1.2 0.7 1.2 0.5 0.8 0.8 -1.0
0.2 0.7 1.2 2.0 3.1 4.8 5.3 5.1 5.2 5.2 4.2 3.2 2.1 2.2 1.5 1.4 0.8 0.4 0.4 0.2 0.2 0.2 -------
Vfcorr % 0.3 1.1 1.7 2.9 4.4 6.9 7.5 7.3 7.3 7.4 6.0 4.5 3.0 3.2 2.1 2.0 1.2 0.6 0.5 0.3 0.3 0.2 -------
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption A4.3 – Fractal analysis on ash particles The fractal analysis on ash particles has been carried out using the dilation method. The method consists in obtaining a binary figure where an outline (1 pixel thickness) of the boundary of the ash particle is drawn. The outline is progressively widened through dilation steps each increasing the widht of the outline of 2 pixels. For each step, the area of the outline is measured. An example of the dilation process is showed in Fig. A4.3.1. Based on the values of the area of the outline for each step of the process, a correspondent value of the perimeter is calculating by dividing the area for the dilation diameter of the correspondent interaction step (dilation diameter = 1 + 2·n° interactions; pix), and a Log-plot of the perimeter vs the dilation diameter is obtained [Mandelbrot, 1967]. From each of these plots, the values of textural and structural fractal dimension were obtained from the slope of the main branches present in the plot, as indicated in Maria and Carey [2002]. The fractal spectrum plots were then derived calculating the first derivative of the Mandlebrot plots.
Fig. A4.3.1. Snapshot of a dilation process applied to a fractal outline. As the dilation diameter increases, the shape is progressively smoothed. Fractal dimension is a function of the relationship between the diameter and area of the dilating boundary.
The results of the fractal analysis with the relative Mandelbrot plot and fractal spectrum plot of each selected ash particles is shown in the following pages, indicating the diameter of the equivalent circle (circle with same area as the projected area of the particle; d), the textural (D1) and structural (D2) fractal dimension as obtained from the Mandelbrot plots.
127
Chapter 4 Set 1: 1900 pix/mm
128
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 1: 1900 pix/mm (follows)
129
Chapter 4 Set 1: 1900 pix/mm (follows)
130
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 1: 1900 pix/mm (follows)
131
Chapter 4 Set 2: 3220 pix/mm
132
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 2: 3220 pix/mm (follows)
133
Chapter 4 Set 2: 3220 pix/mm (follows)
134
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 2: 3220 pix/mm (follows)
135
Chapter 4 Set 3: 4300 pix/mm
136
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 3: 4300 pix/mm (follows)
137
Chapter 4 Set 3: 4300 pix/mm (follows)
138
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 3: 4300 pix/mm (follows)
139
Chapter 4 Set 4: 6600 pix/mm
140
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 4: 6600 pix/mm (follows)
141
Chapter 4 Set 4: 6600 pix/mm (follows)
142
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 4: 6600 pix/mm (follows)
143
Chapter 4 Set 5: 8600 pix/mm
144
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 5: 8600 pix/mm (follows)
145
Chapter 4 Set 5: 8600 pix/mm (follows)
146
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 5: 8600 pix/mm (follows)
147
Chapter 4 Set 6: 14000 pix/mm
148
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 6: 14000 pix/mm (follows)
149
Chapter 4 Set 6: 14000 pix/mm (follows)
150
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 6: 14000 pix/mm (follows)
151
Chapter 4 Set 7: 17200 pix/mm
152
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 7: 17200 pix/mm (follows)
153
Chapter 4 Set 7: 17200 pix/mm (follows)
154
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Set 7: 17200 pix/mm (follows)
155
Chapter 4
References Adams, N. K., B. F. Houghton, and W. Hildreth (2006a), Abrupt transitions during sustained explosive eruptions: examples from the 1912 eruption of Novarupta, Alaska, Bulletin of Volcanology, 69(2), 189-206. Adams, N. K., B. F. Houghton, S. A. Fagents, and W. Hildreth (2006b), The transition from explosive to effusive eruptive regime: The example of the 1912 Novarupta eruption, Alaska, Geol Soc Am Bull, 118(5-6), 620-634. Alfano, F., C. Bonadonna, P. Delmelle, and L. Costantini (2011a), Insights on tephra settling velocity from morphological observations, Journal of Volcanology and Geothermal Research, 208(3-4), 86-98. Alfano, F., C. Bonadonna, A. C. M. Volentik, C. B. Connor, S. F. L. Watt, D. M. Pyle, and L. J. Connor (2011b), Tephra stratigraphy and eruptive volume of the May, 2008, Chait,n eruption, Chile, Bulletin of Volcanology, 73(5), 613-630. Barberi, F., M. Coltelli, A. Frullani, M. Rosi, and E. Almeida (1995), Chronology and dispersal characteristics of recently (last 5000 years) erupted tephra of Cotopaxi (Ecuador): implications for long-term eruptive forecasting, Journal of volcanology and geothermal research, 69, 217-239. Berube, D., and M. Jebrak (1999), High precision boundary fractal analysis for shape characterization, Computers & Geosciences, 25(9), 1059-1071. Biass, S., and C. Bonadonna (2011), A quantitative uncertainty assessment of eruptive parameters derived from tephra deposits: the example of two large eruptions of Cotopaxi volcano, Ecuador, Bulletin of Volcanology, 73(1), 73-90. Blott, S. J., and K. Pye (2008), Particle shape: a review and new methods of characterization and classification, Sedimentology, 55, 31-63. Blower, J. D., J. P. Keating, H. M. Mader, and J. C. Phillips (2001), Inferring volcanic degassing processes from vesicle size distributions, Geophys Res Lett, 28(2), 347-350. Blower, J. D., J. P. Keating, H. M. Mader, and J. C. Phillips (2002), The evolution of bubble size distributions in volcanic eruptions, Journal of Volcanology and Geothermal Research, 120(1-2), 1-23. Bonadonna, C., and A. Costa (in press), Modeling of tephra sedimentation from volcanic plumes, in Modeling Volcanic Processes: The Physics and Mathematics of Volcanism, edited by S. Fagents, T. Gregg and R. Lopes, Cambridge University Press.
156
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Carey, R. J., B. F. Houghton, and T. Thordarson (2009), Abrupt shifts between wet and dry phases of the 1875 eruption of Askja Volcano: Microscopic evidence for macroscopic dynamics, Journal of Volcanology and Geothermal Research, 184(3-4), 256-270. Carey, R. J., B. F. Houghton, and T. Thordarson (2010), Tephra dispersal and eruption dynamics of wet and dry phases of the 1875 eruption of Askja Volcano, Iceland, Bulletin of Volcanology, 72(3), 259-278. Carey, S., and H. Sigurdsson (1987), Temporal Variations in Column Height and Magma Discharge Rate During the 79 Ad Eruption of Vesuvius, Geol Soc Am Bull, 99(2), 303-314. Carey, S., A. Maria, and H. Sigurdsson (2000), Use of fractal analysis for discrimination of particles from primary and reworked jokulhlaup deposits in SE Iceland, Journal of Volcanology and Geothermal Research, 104(1-4), 65-80. Cas, R., and J. Wright (1987), Volcanic successions: modern and ancient: a geological approach to processes, products and successions, 1 edition ed., 544 pp., Springer. Cashman, K. V., and M. T. Mangan (1994), Physical Aspects of Magmatic Degassing .2. Constraints On Vesiculation Processes From Textural Studies of Eruptive Products, in Volatiles in Magmas, edited, pp. 447-478. Cashman, K. V., B. Sturtevant, P. Papale, and O. Navon (2000), Magmatic Fragmentation, in Encyclopedia of volcanoes, edited by H. Sigurdsson, Academic Press, New York. Castro, J. M., and D. B. Dingwell (2009), Rapid ascent of rhyolitic magma at Chaiten volcano, Chile, Nature, 461(7265), 780-U729. Cheng, H. C., and R. Lemlich (1983), Errors in the Measurement of Bubble-Size Distribution in Foam, Ind Eng Chem Fund, 22(1), 105-109. Cioni, R., A. Bertagnini, D. Andronico, P. D. Cole, and F. Mundula (2011), The 512 AD eruption of Vesuvius: complex dynamics of a small scale subplinian event, Bulletin of Volcanology, 73(7), 789-810. Coltelli, M., P. Del Carlo, and L. Vezzoli (1998), Discovery of a Plinian basaltic eruption of Roman age at Etna volcano, Italy, Geology, 26(12), 1095-1098. Corsaro, R. A., and M. Pompilio (2004), Dynamics of magma at Mount Etna, in Mt Etna: Volcano Laboratory, edited by B. A. e. al., pp. 91-110, American Geophysical Union, Washinghton DC.
157
Chapter 4 Costantini, L., B. F. Houghton, and C. Bonadonna (2010), Constraints on eruption dynamics of basaltic explosive activity derived from chemical and microtextural study: The example of the Fontana Lapilli Plinian eruption, Nicaragua, Journal of Volcanology and Geothermal Research, 189(3-4), 207-224. Costantini, L., C. Bonadonna, B. F. Houghton, and H. Wehrmann (2009), New physical characterization of the Fontana Lapilli basaltic Plinian eruption, Nicaragua, Bulletin of Volcanology, 71(19), 337-355. Dellino, P., and G. Liotino (2002), The fractal and multifractal dimension of volcanic ash particles contour: a test study on the utility and volcanological relevance, Journal of Volcanology and Geothermal Research, 113(1-2), 1-18. Dellino, P., D. Mele, R. Bonasia, G. Braia, L. La Volpe, and R. Sulpizio (2005), The analysis of the influence of pumice shape on its terminal velocity, Geophys Res Lett, 32(21), 4. Dingwell, D. B. (1996), Volcanic dilemma: Flow or blow?, Science, 273(5278), 10541055. Dingwell, D. B., and S. L. Webb (1989), Structural Relaxation in Silicate Melts and Non-Newtonian Melt Rheology in Geologic Processes, Phys Chem Miner, 16(5), 508-516. Fierstein, J., and W. Hildreth (1992), The plinian eruptions of 1912 at Novarupta, Katmai National Park, Alaska, Bulletin of Volcanology, 54, 646-684. Gaonac'h, H., S. Lovejoy, and D. Schertzer (2005), Scaling vesicle distributions and volcanic eruptions, Bulletin of Volcanology, 67(4), 350-357. Gaonac'h, H., S. Lovejoy, J. Stix, and D. Scherzter (1996), A scaling growth model for bubbles in basaltic lava flows, Earth Planet Sc Lett, 139(3-4), 395-409. Gardner, J. E., M. Hilton, and M. R. Carroll (1999), Experimental constraints on degassing of magma: isothermal bubble growth during continuous decompression from high pressure, Earth Planet Sc Lett, 168(1-2), 201-218. Gardner, J. E., R. M. E. Thomas, C. Jaupart, and S. Tait (1996), Fragmentation of magma during Plinian volcanic eruptions, Bulletin of Volcanology, 58(2-3), 144162. Gonnermann, H. M., and M. Manga (2007), The fluid mechanics inside a volcano, Annu Rev Fluid Mech, 39, 321-356. Gurioli, L., B. F. Houghton, K. V. Cashman, and R. Cioni (2005), Complex changes in eruption dynamics during the 79 AD eruption of Vesuvius, Bulletin of Volcanology, 67, 144-159. 158
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Houghton, B. F., and C. J. N. Wilson (1989), A vesicularity index for pyroclastic deposits, Bulletin of Volcanology, 51(6), 451-462. Houghton, B. F., C. J. N. Wilson, P. Del Carlo, M. Coltelli, J. E. Sable, and R. Carey (2004), The influence of conduit processes on changes in style of basaltic Plinian eruptions: Tarawera 1886 and Etna 122 BC, Journal of Volcanology and Geothermal Research, 137(1-3), 1-14. Houghton, B. F., R. J. Carey, K. V. Cashman, C. J. N. Wilson, B. J. Hobden, and J. E. Hammer (2010), Diverse patterns of ascent, degassing, and eruption of rhyolite magma during the 1.8 ka Taupo eruption, New Zealand: Evidence from clast vesicularity, Journal of Volcanology and Geothermal Research, 195(1), 31-47. Inman, D. L. (1952), Measures for describing the size distribution of sediments, J Sediment Petrol, 22, 125-145. Kaminski, E., and C. Jaupart (1997), Expansion and quenching of vesicular magma fragments in Plinian eruptions, J. Geophys. Res.-Solid Earth, 102(B6), 1218712203. Klug, C., and K. V. Cashman (1994), Vesiculation of May 18, 1980, Mount St-Helens Magma, Geology, 22(5), 468-472. Klug, C., and K. V. Cashman (1996), Permeability development in vesiculating magmas: Implications for fragmentation, Bulletin of Volcanology, 58(2-3), 87100. Klug, C., K. V. Cashman, and C. R. Bacon (2002), Structure and physical characteristics of pumice from the climatic eruption of Mt Mazama (Crater Lake) Oregon, Bulletin of Volcanology, 64, 486-501. Lara, L. E. (2009), The 2008 eruption of the Chaiten Volcano, Chile: a preliminary report, Andean Geol, 36(1), 125-129. Mandelbrot, B. B. (1967), How Long Is Coast of Britain - Statistical Self-Similarity and Fractional Dimension, Science, 156(3775), 636-&. Mangan, M., and T. Sisson (2000), Delayed, disequilibrium degassing in rhyolite magma: decompression experiments and implications for explosive volcanism, Earth Planet Sc Lett, 183(3-4), 441-455. Mangan, M. T., and K. V. Cashman (1996), The structure of basaltic scoria and reticulite and inferences for vesiculation, foam formation, and fragmentation in lava fountains, Journal of Volcanology and Geothermal Research, 73(1-2), 1-18.
159
Chapter 4 Maria, A., and S. Carey (2002), Using fractal analysis to quantitatively characterize the shapes of volcanic particles, J. Geophys. Res.-Solid Earth, 107(B11). Maria, A., and S. Carey (2007), Quantitative discrimination of magma fragmentation and pyroclastic transport processes using the fractal spectrum technique, Journal of Volcanology and Geothermal Research, 161(3), 234-246. Mastin, L. G., et al. (2009), A multidisciplinary effort to assign realistic source parameters to models of volcanic ash-cloud transport and dispersion during eruptions, Journal of Volcanology and Geothermal Research, 186(1-2), 10-21. Naranjo, J. A., and C. R. Stern (2004), Holocene tephrochronology of the southernmost part (42 degrees 30'-45 degrees S) of the Andean Southern Volcanic Zone, Rev Geol Chile, 31(2), 225-240. Okumura, S., M. Nakamura, A. Tsuchiyama, T. Nakano, and K. Uesugi (2008), Evolution of bubble microstructure in sheared rhyolite: Formation of a channellike bubble network, J. Geophys. Res.-Solid Earth, 113(B7). Okumura, S., M. Nakamura, S. Takeuchi, A. Tsuchiyama, T. Nakano, and K. Uesugi (2009), Magma deformation may induce non-explosive volcanism via degassing through bubble networks, Earth Planet Sc Lett, 281(3-4), 267-274. Papale, P. (1999), Strain-induced magma fragmentation in explosive eruptions, Nature, 397(6718), 425-428. Papale, P., A. Neri, and G. Macedonio (1998), The role of magma composition and water content in explosive eruptions I: conduit ascent dynamics, Journal of Volcanology and Geothermal Resarch, 87, 75-93. Pistolesi, M., M. Rosi, R. Cioni, K. V. Cashman, A. Rossotti, and E. Aguilera (2011), Physical volcanology of the post-twelfth-century activity at Cotopaxi volcano, Ecuador: Behavior of an andesitic central volcano, Geol Soc Am Bull, 123(5-6), 1193-1215. Proussevitch, A. A., D. L. Sahagian, and V. A. Kutolin (1993), Stability of foams in silicate melts, Journal of Volcanology and Geothermal Research, 59(1-2), 161178. Riley, C. M., W. I. Rose, and G. J. S. Bluth (2003), Quantitative shape measurements of distal volcanic ash, J. Geophys. Res.-Solid Earth, 108(B10). Rosi, M., P. Landi, M. Polacci, A. Di Muro, and D. Zandomeneghi (2004), Role of conduit shear on ascent of the crystal-rich magma feeding the 800-year-BP Plinian eruption of Quilotoa Volcano (Ecuador), Bulletin of Volcanology, 66(4), 307-321.
160
Insights into eruption dynamics from textural analysis: the case of the May, 2008, Chaitén eruption
Rust, A. C., and K. V. Cashman (2011), Permeability controls on expansion and size distributions of pyroclasts, J. Geophys. Res.-Solid Earth, 116. Sable, J. E., B. F. Houghton, P. Del Carlo, and M. Coltelli (2006a), Changing conditions of magma ascent and fragmentation during the Etna 122 BC basaltic Plinian eruption: evidence from clasts microtextures, Journal of Volcanology and Geothermal Research, 158, 333-354. Sable, J. E., B. F. Houghton, P. Del Carlo, and M. Coltelli (2006b), Changing conditions of magma ascent and fragmentation during the Etna 122 BC basaltic Plinian eruption: Evidence from clast microtextures, Journal of Volcanology and Geothermal Research, 158(3-4), 333-354. Sahagian, D. L., and A. A. Proussevitch (1998), 3D particle size distribution from 2D observations: stereology for natural applications, Journal of Volcanology and Geothermal Resarch, 84, 173-196. Simakin, A. G., and T. P. Salova (2001), Evolution of bubble size distribution during the gradual degassing of granitic melt: Experimental data, Geochem Int., 39(3), 258-267. Shea, T., L. Gurioli, B. F. Houghton, R. Cioni, and K. V. Cashman (2011), Column collapse and generation of pyroclastic density currents during the AD 79 eruption of Vesuvius: The role of pyroclast density, Geology, 39(7), 695-698. Shea, T., L. Gurioli, J. F. Larsen, B. F. Houghton, J. E. Hammer, and K. V. Cashman (2010a), Linking experimental and natural vesicle textures in Vesuvius 79AD white pumice, Journal of Volcanology and Geothermal Research, 192(1-2), 6984. Shea, T., B. F. Houghton, L. Gurioli, K. V. Cashman, J. E. Hammer, and B. J. Hobden (2010b), Textural studies of vesicles in volcanic rocks: An integrated methodology, Journal of Volcanology and Geothermal Research, 190(3-4), 271289. Sparks, R. S. J. (1978), Dynamics of Bubble Formation and Growth in Magmas Review and Analysis, Journal of Volcanology and Geothermal Research, 3(1-2), 1-37. Sutton, A. N., S. Blake, and C. J. N. Wilson (1995), An Outline Geochemistry of Rhyolite Eruptives from Taupo Volcanic Center, New-Zealand, Journal of Volcanology and Geothermal Research, 68(1-3), 153-175. Sutton, A. N., S. Blake, C. J. N. Wilson, and B. L. A. Charlier (2000), Late Quaternary evolution of a hyperactive rhyolite magmatic system: Taupo volcanic centre, New Zealand, J Geol Soc London, 157, 537-552. 161
Chapter 4 Toramaru, A. (1990), Measurement of Bubble-Size Distributions in Vesiculated Rocks with Implications for Quantitative Estimation of Eruption Processes, Journal of Volcanology and Geothermal Research, 43(1-4), 71-90. Toramaru, A. (2006), BND (bubble number density) decompression rate meter for explosive volcanic eruptions, Journal of Volcanology and Geothermal Research, 154(3-4), 303-316. Verhoogen, J. (1951), Mechanics of Ash Formation, Am J Sci, 249(10), 729-739. Walker, G. P. L. (1973), Explosive volcanic eruptions - a new classification scheme, Geologische Rundschau, 62, 431-446. Watt, S. F. L., D. M. Pyle, T. A. Mather, R. S. Martin, and N. E. Matthews (2009), Fallout and distribution of volcanic ash over Argentina following the May 2008 explosive eruption of Chaiten, Chile, J. Geophys. Res.-Solid Earth, 114, 11. Wicks, C., J. C. de la Llera, L. E. Lara, and J. Lowenstern (2011), The role of dyking and fault control in the rapid onset of eruption at Chaiten volcano, Chile, Nature, 478(7369), 374-+. Wilson, C. J. N. (1993), Stratigraphy, Chronology, Styles and Dynamics of Late Quaternary Eruptions from Taupo Volcano, New-Zealand, Philos T Roy Soc A, 343(1668), 205-306. Wilson, C. J. N., and G. P. L. Walker (1985), The Taupo Eruption, New-Zealand .1. General-Aspects, Philos T Roy Soc A, 314(1529), 199-&. Wilson, L., and T. C. Huang (1979), The influence of shape on the atmospheric settling velocity of volcanic ash particles, Earth and Planetary Sciences Letters, 44, 311324. Wilson, L., and G. P. L. Walker (1987), Explosive volcanic-eruptions .6. Ejecta dispersal in plinian eruptions - the control of eruption conditions and atmospheric properties, Geophys J Roy Astr S, 89(2), 657-679. Zhang, Y. X. (1999), A criterion for the fragmentation of bubbly magma based on brittle failure theory, Nature, 402(6762), 648-650.
162
Chapter 5
Concluding remarks
Pyroclastic fragments are the main products of explosive eruptions as they represent the result of all the processes that develop within the volcanic system, starting from the storage within the magmatic chamber, to the ascent through the volcanic conduit, the fragmentation and eventually the transport and deposition to the ground. In particular, tephra deposits are very important to the characterization of explosive eruptions as they can provide a large amount of information about several processes that characterize an explosive eruption. First, tephra particles are directly linked to the evolution of the physical conditions immediately before the fragmentation occurs. Thus from the study of their morphology and their textures it is possible to infer information about the developing of the magma rise and the conditions that triggered the fragmentation [Blower et al., 2001; Blower et al., 2002; Cheng and Lemlich, 1980; Klug and Cashman, 1994; Klug et al., 2002; Mangan and Cashman, 1996; Sahagian and Proussevitch, 1998; Shea et al., 2010; Toramaru, 2006]. In addition, as highlighted in chapter 2 and 4, these textural features have an important influence on the dispersal and sedimentation process, as they affect the settling velocity of the particles [Dellino et al., 2005; Ganser, 1993; Wilson and Huang, 1979]. Second, tephra deposits are the result of plume dynamics and atmospheric condition, and their geometry of dispersion and their volume reflect the energy involved in the explosive process. Thus, studying tephra deposits it is possible to infer specific information about the parameters that characterize
163
Chapter 5 the eruption, as the volume of material erupted [Bonadonna and Houghton, 2005; Pyle, 1989], the height of the column [Carey and Sparks, 1986], the mass eruption rate [Carey and Sparks, 1986; Wilson and Walker, 1987] and the VEI and magnitude of the eruption [Newhall and Self, 1982; Pyle, 2000]. Explosive eruptions can also be classified based on specific features of tephra deposits [Pyle, 1989; Walker, 1973]. As a result, it is clear how all the processes that characterize an explosive eruption, from the start of the ascent, to the deposition of the final products, are linked together. So it is important to study these relations in order to increase our knowledge of volcanic processes and, at the same time, to be able to obtain detailed information on the eruption and mitigate associated risk. Currently, the majority of the methods for the study of pyroclastic textures and morphology is based on 2D analysis. This is the main limitation of textural and morphological studies as it is difficult to extrapolate the third dimension. However, there is a rising interest in the application of micro-tomography techniques on volcanic particles in order to determine the vesicle size distribution of tephra samples based on 3D analysis of the particles. Only a few studies of the 3D textures of pumice samples have been carried out so far [i.e., Degruyter et al., 2010a; Degruyter et al., 2010b; Polacci et al., 2006; Song et al., 2001; Voltolini et al., 2011; Zandomeneghi et al., 2010]. The same considerations can be done for the morphological characterization of particles, as the measurement of the surface area of an object is not easy to determine. Measurements of the surface area of a particle, as highlighted in chapter 2, depend on the scale of observation, so very different results can be obtained in relation to the method used. Gas adsorption provides a very useful method for surface analysis, and it can give very useful insights when applied to the analysis of the adsorption by ash particle of volcanic gases within the plume [Delmelle et al., 2005] or trying to characterize the potential impact ash can have on human health [Horwell et al., 2003a; Horwell et al., 2003b]. However, this method has been proved to be not suitable when applied to the characterization of the settling velocity of the particles. On the other hand, 3D scan technology is rapidly developing, providing useful devices able to capture the real tridimensional features of the studied object in relation to the acquisition resolution, allowing to build digital models of real particles that can be extensively studied to determine their morphologic parameter and their fractal dimensions. These digital models can also be used in numerical simulations to study how the morphology influences the behavior of the particle in certain conditions, such as the falling motion 164
Concluding remarks
through the atmosphere. In the study of the settling process, these methods can have a very important impact especially on the investigation of another important aspects of the process itself, which is the scale of the morphologic features that influence the drag of the particle. In fact, as pointed out in chapter 2, the surface area of a particle, which is dependent on the shape, depends on the resolution of the method used in its determination. As a result, all these aspects can represent an important improvement in studying particle morphology and textures, and they can give the possibility to obtain always more precise and detailed information on the eruption characteristics and dynamics, and a better understanding of volcanic processes. A important aspect that needs to be further investigated is how to link together different eruptive parameters (e.g., volume of erupted material, mass eruption rate) with the conduit dynamics and the fragmentation of the magma. It is not possible to directly observe what happens within the conduit during an eruption, but it is possible to obtain informations through indirect measurements using geophysical monitoring systems (i.e., seismic signals, gas monitoring, ground deformation, infrasound measurements). However, this is possible only for volcanic areas that are preventively monitored. This is not the case of long dormant volcanoes that renew suddenly their activity. The May 2008 Chaitén eruption is a good example of this, as the only data available on the processes that anticipated the eruption are few data collected by seismic monitoring stations located more than 300 km far from the vent and the satellite images that could capture the various explosive phases of the eruption [Carn et al., 2009; Lara, 2009]. In these cases, the study of the pyroclastic deposit, the dispersion of the products and their characteristics are critical to understand and reconstruct the eruption development. This becomes even more important if considering that the study of recent eruption gives the possibility to observe a almost fresh, where reworking and erosion are less significant. Increasing our understanding of volcanic process and the relation between characteristics of the products and their dispersal can be very important also in relation with the study of past eruption. The disruption of volcanic deposits through alteration and erosion can make difficult to reconstruct historical eruption that were not observed during their occurrence. In these cases, the analysis of the products can be critical to correlate residual outcrops produced after a same volcanic event, individuating the vent from where they were erupted and reconstructing the whole eruption dynamics. This can have important insights in terms of having always more complete dataset of the 165
Chapter 5 volcanic activity of a region, in relation with statistical studies of frequency of the volcanic events, which is particularly important when studying the risk assessment of a region, but it can also have important insights in sedimentology, as tephra layers are very useful markers used in datation of sedimentary sequences (i.e., tephrochronology). The study of pyroclastic deposits, as shown in chapter 3, can provide a large amount of information on the eruptive event. The MER, whose estimation is based on the height of the column produced during an explosive event and on the dispersal of the pyroclastic fragments [Carey and Sparks, 1986; Wilson and Walker, 1987], can give a direct measure of the size of the eruption, as it is a quantification of the rate of emission of the eruptive products. As a result, the MER is a measure of the energy involved in the eruption as a result of the fragmentation process that triggers the explosion. On the other hand, the vescicle number density of the products is a direct consequence of the physical conditions within the conduit immediately before fragmentation occurs, as it is directly related to the decompression rate [Toramaru, 2006]. Ideally, relating together these two parameters can show how the fragmentation process in the conduit influences the eruptive dynamics into the atmosphere. As shown by the relation between MER and NV (cf. Fig. 4.10), these two parameters show a complex level of interconnection, as for some cases a relation seems to exist (i.e., basaltic eruption and in some case rhyolitic eruption), but for many other cases these relations are not obvious. Further studies on explosive processes are needed in order to increase the available dataset in relation to phonolitic eruptions, for whom very few cases are available. This relation is made even more difficult to understand also in relation with the high degree of uncertainty that affects the determination of these parameters. Some final considerations can be made also on the studies of settling processes of tephra particles. The drag of irregular particles has been widely studied in the past years [e.g., Bonadonna et al., 1998; Dellino et al., 2005; Ganser, 1993; Haider and Levenspiel, 1989; Kunii and Levenspiel, 1969; Wilson and Huang, 1979]. From these studies several analytical models have been developed, giving the possibility to estimate the terminal velocity of irregular particles for different ranges of Reynolds number values and using different shape descriptors. In general all these models are considered to give estimations of the settling velocity with a good grade of accuracy. But the main problem with these models is the characterization of the shape, as the morphologic characterization method applied can influence substantially the final results. A detailed 166
Concluding remarks
study of the dynamic of the process is then required in order to define a comprehensive methodology able to give an accurate estimation of the settling velocity, based on a morphologic characterization able to capture those features that have an influence on the settling dynamics. In addition it has to be taken into account that all these studies of the settling velocity are basically focused on a dynamic involving isolated particles, but the process of deposition actually develops in a more complex context, where a multitude of particles fall interacting one with another. As an example, tephra dispersal and sedimentation can be described using advection-diffusion models [e.g., Suzuki, 1983]. According to these models, the main parameters controlling the sedimentation of a particle are the settling velocity, diffusion and wind advection (cf., Chapter 2), and a sedimentation mechanism grain-by-grain is considered. Possible interactions between particles and the influence these interactions can have on the settling process are not taken into account. This aspect tan be particularly important for the sedimentation in the medial area, where the volcanic cloud is not yet diluted by the air, and the probability of particles to interact is higher. These interactions can have important effects on the dynamics of sedimentation. As an example, an important and widely studied process influencing the settling of tephra particle is represented by particle aggregation [Bonadonna and Phillips, 2003; Carey and Sigurdsson, 1982; Durant and Shaw, 2005; Durant and Rose, 2009; Durant et al., 2008; Durant et al., 2009; Gilbert and Lane, 1994; James et al., 2002; James et al., 2003; Rose and Durant, 2011; Rose et al., 2008; Schumacher and Schmincke, 1995], that makes ash particles aggregate together and fall closer to the vent than expected. Other studies focusing on the interaction between different particles, as the reduction of drag produced by the wake effect [Sethian et al., 2010] or the production of convective instabilities within the umbrella cloud [Bonadonna et al., 2002; Carey, 1997]. As a final remark, we can conclude that explosive eruption dynamics and dispersal and sedimentation of pyroclastic products represent very complex processes, and yet many aspects need further investigation. But the technological progress is providing always more opportunity to increase our understanding. It has been a long journey since the day in which an apple fell hitting Sir Isaac Newton’s head. And now we are able to see that things would have been different if he had been hit by a ballistic fragment.
167
Chapter 5
References Blower, J. D., J. P. Keating, H. M. Mader, and J. C. Phillips (2001), Inferring volcanic degassing processes from vesicle size distributions, Geophys Res Lett, 28(2), 347-350. Blower, J. D., J. P. Keating, H. M. Mader, and J. C. Phillips (2002), The evolution of bubble size distributions in volcanic eruptions, Journal of Volcanology and Geothermal Research, 120(1-2), 1-23. Bonadonna, C., and J. C. Phillips (2003), Sedimentation from strong volcanic plumes, J Geophys Res, 108(B7), 2340-2368. Bonadonna, C., and B. F. Houghton (2005), Total grainsize distribution and volume of tephra-fall deposits, Bulletin of Volcanology, 67, 441-456. Bonadonna, C., G. G. J. Ernst, and R. S. J. Sparks (1998), Thickness variations and volume estimates of tephra fall deposits: the importance of particle Reynolds number, Journal of Volcanology and Geothermal Research, 81(3-4), 173-187. Bonadonna, C., et al. (2002), Tephra fallout in the eruption of Soufriere Hills Volcano, Montserrat, Geological Society, London, Memoirs, 21(1), 483-516. Carey, S. N. (1997), Influence of convective sedimentation on the formation of widespread tephra fall layers in the deep sea, Geology, 25(9), 839-842. Carey, S. N., and H. Sigurdsson (1982), Influence of particle aggregation on deposition of distal tephra from the May 18, 1980, eruption of Mount St-Helens volcano, J Geophys Res, 87(B8), 7061-7072. Carey, S. N., and R. S. J. Sparks (1986), Quantitative models of the fallout and dispersal of tephra from volcanic eruption columns, Bulletin of Volcanology, 48, 109-125. Carn, S. A., J. S. Pallister, L. Lara, J. W. Ewert, S. Watt, A. J. Prata, R. J. Thomas, and G. Villarosa (2009), The Unexpected Awakening of Chaitén Volcano, Chile, EOS, Transactions American Geophysical Union, 90(24). Cheng, H. C., and R. Lemlich (1980), Prediction of Changes in Bubble-Size Distribution Due to Interbubble Gas-Diffusion in Foam - Comment, Ind Eng Chem Fund, 19(1), 133-133. Degruyter, W., O. Bachmann, and A. Burgisser (2010a), Controls on magma permeability in the volcanic conduit during the climactic phase of the Kos Plateau Tuff eruption (Aegean Arc), Bulletin of Volcanology, 72(1), 63-74.
168
Concluding remarks
Degruyter, W., A. Burgisser, O. Bachmann, and O. Malaspinas (2010b), Synchrotron X-ray microtomography and lattice Boltzmann simulations of gas flow through volcanic pumices, Geosphere, 6(5), 470-481. Dellino, P., D. Mele, R. Bonasia, G. Braia, L. La Volpe, and R. Sulpizio (2005), The analysis of the influence of pumice shape on its terminal velocity, Geophys Res Lett, 32(21), 4. Delmelle, P., F. Villieras, and M. Pelletier (2005), Surface area, porosity and water adsorption properties of fine volcanic ash particles, Bulletin of Volcanology, 67(2), 160-169. Durant, A. J., and R. A. Shaw (2005), Evaporation freezing by contact nucleation inside-out, Geophys Res Lett, 32(20). Durant, A. J., and W. I. Rose (2009), Sedimentological constraints on hydrometeorenhanced particle deposition: 1992 Eruptions of Crater Peak, Alaska, Journal of Volcanology and Geothermal Research, 186(1-2), 40-59. Durant, A. J., R. A. Shaw, W. I. Rose, Y. Mi, and G. G. J. Ernst (2008), Ice nucleation and overseeding of ice in volcanic clouds, J Geophys Res-Atmos, 113(D9). Durant, A. J., W. I. Rose, A. M. Sarna-Wojcicki, S. Carey, and A. C. M. Volentik (2009), Hydrometeor-enhanced tephra sedimentation: Constraints from the 18 May 1980 eruption of Mount St. Helens, J. Geophys. Res.-Solid Earth, 114. Ganser, G. H. (1993), A rational approach to drag prediction of spherical and nonspherical particles, Powder Technol, 77(2), 143-152. Gilbert, J. S., and S. J. Lane (1994), The origin of accretionary lapilli, Bulletin of Volcanology, 56(5), 398-411. Haider, A., and O. Levenspiel (1989), Drag Coefficient and Terminal Velocity of Spherical and Nonspherical Particles, Powder Technol, 58(1), 63-70. Horwell, C. J., I. Fenoglio, K. V. Ragnarsdottir, R. S. J. Sparks, and B. Fubini (2003a), Surface reactivity of volcanic ash from the eruption of Soufriere Hills volcano, Montserrat, West Indies with implications for health hazards, Environ Res, 93(2), 202-215. Horwell, C. J., R. S. J. Sparks, T. S. Brewer, E. W. Llewellin, and B. J. Williamson (2003b), Characterization of respirable volcanic ash from the Soufriere Hills volcano, Montserrat, with implications for human health hazards, Bulletin of Volcanology, 65(5), 346-362.
169
Chapter 5 James, M. R., J. S. Gilbert, and S. J. Lane (2002), Experimental investigation of volcanic particle aggregation in the absence of a liquid phase, J. Geophys. Res.Solid Earth, 107(B9). James, M. R., S. J. Lane, and J. S. Gilbert (2003), Density, construction, and drag coefficient of electrostatic volcanic ash aggregates, J. Geophys. Res.-Solid Earth, 108(B9). Klug, C., and K. V. Cashman (1994), Vesiculation of May 18, 1980, Mount St-Helens Magma, Geology, 22(5), 468-472. Klug, C., K. V. Cashman, and C. R. Bacon (2002), Structure and physical characteristics of pumice from the climatic eruption of Mt Mazama (Crater Lake) Oregon, Bulletin of Volcanology, 64, 486-501. Kunii, D., and O. Levenspiel (1969), Fluidisation Engineering, Wiley and Sons, New York. Lara, L. E. (2009), The 2008 eruption of the Chaiten Volcano, Chile: a preliminary report, Andean Geology, 36(1), 125-129. Mangan, M. T., and K. V. Cashman (1996), The structure of basaltic scoria and reticulite and inferences for vesiculation, foam formation, and fragmentation in lava fountains, Journal of Volcanology and Geothermal Research, 73(1-2), 1-18. Newhall, C. G., and S. Self (1982), The Volcanic Explosivity Index (Vei) - an Estimate of Explosive Magnitude for Historical Volcanism, Journal of Geophysical Research-Oceans and Atmospheres, 87(NC2), 1231-1238. Polacci, M., D. R. Baker, L. Mancini, G. Tromba, and F. Zanini (2006), Threedimensional investigation of volcanic textures by X-ray microtomography and implications for conduit processes, Geophys Res Lett, 33(13). Pyle, D. M. (1989), The thickness, volume and grainsize of tephra fall deposits, Bulletin of Volcanology, 51(1), 1-15. Pyle, D. M. (2000), Sizes of volcanic eruptions, in Encyclopedia of Volcanoes, edited by H. Sigurdsson, B. Houghton, H. Rymer, J. Stix and S. McNutt, pp. 263-270, Academic Press, San Diego. Rose, W. I., and A. J. Durant (2011), Fate of volcanic ash: Aggregation and fallout, Geology, 39(9), 895-896. Rose, W. I., S. Self, P. J. Murrow, C. Bonadonna, A. J. Durant, and G. G. J. Ernst (2008), Nature and significance of small volume fall deposits at composite volcanoes: Insights from the October 14, 1974 Fuego eruption, Guatemala, Bulletin of Volcanology, 70(9), 1043-1067. 170
Concluding remarks
Sahagian, D. L., and A. A. Proussevitch (1998), 3D particle size distribution from 2D observations: stereology for natural applications, Journal of Volcanology and Geothermal Resarch, 84, 173-196. Schumacher, R., and H. U. Schmincke (1995), Models for the Origin of Accretionary Lapilli, Bulletin of Volcanology, 56(8), 626-639. Sethian, J., J. Suckale, and L. T. Elkins-Tanton (2010), Direct numerical simulations of magmatic differentiation at the microscopic scale, American Geophysical Union, Fall Meeting 2010, abstract #NG23A-1374. Shea, T., B. F. Houghton, L. Gurioli, K. V. Cashman, J. E. Hammer, and B. J. Hobden (2010), Textural studies of vesicles in volcanic rocks: An integrated methodology, Journal of Volcanology and Geothermal Research, 190(3-4), 271289. Song, S. R., K. W. Jones, W. B. Lindquist, B. A. Dowd, and D. L. Sahagian (2001), Synchrotron X-ray computed microtomography: studies on vesiculated basaltic rocks, Bulletin of Volcanology, 63(4), 252-263. Suzuki, T. (1983), A theoretical model for dispersion of tephra, in Arc Volcanism, Physics and Tectonics, edited by D. Shimozuru and I. Yokoyama, pp. 95-113, Terra Scientific Publishing Company (TERRAPUB), Tokyo. Toramaru, A. (2006), BND (bubble number density) decompression rate meter for explosive volcanic eruptions, Journal of Volcanology and Geothermal Research, 154(3-4), 303-316. Voltolini, M., D. Zandomeneghi, L. Mancini, and M. Polacci (2011), Texture analysis of volcanic rock samples: Quantitative study of crystals and vesicles shape preferred orientation from X-ray microtomography data, Journal of Volcanology and Geothermal Research, 202(1-2), 83-95. Walker, G. P. L. (1973), Explosive volcanic eruptions - a new classification scheme, Geologische Rundschau, 62, 431-446. Wilson, L., and T. C. Huang (1979), The influence of shape on the atmospheric settling velocity of volcanic ash particles, Earth and Planetary Sciences Letters, 44, 311324. Wilson, L., and G. P. L. Walker (1987), Explosive volcanic-eruptions .6. Ejecta dispersal in plinian eruptions - the control of eruption conditions and atmospheric properties, Geophysical Journal of the Royal Astronomical Society, 89(2), 657-679.
171
Chapter 5 Zandomeneghi, D., M. Voltolini, L. Mancini, F. Brun, D. Dreossi, and M. Polacci (2010), Quantitative analysis of X-ray microtomography images of geomaterials: Application to volcanic rocks, Geosphere, 6(6), 793-804.
172
