Pierre M. Vincent

The off-shore continuation of an active basaltic volcano: Piton de la Fournaise (Réunion Island, Indian Ocean); structural and geomorphological interpretation from sea beam mapping

Abstract In 1984 a Sea Beam survey of the submarine east flank of Piton de la Fournaise was performed with R/V “Jean Charcot”. Three main types of volcanic or volcano-tectonic features have been identified: 1. (1) The subaerial NE and SE volcanic rift zones of Piton de la Fournaise do not extend more than about five kilometers offshore. Unlike typical Hawaiian rift zones, which form narrow (2–4 km) ridges extending tens of kilometers from the summit, the active rift zones of Piton de la Fournaise widen downslope, attaining more than 10 km at their front. 2. (2) The submarine extension of the Grand Brule slide is larger than the subaerial portion. The entire slide forms a 7 × 24 km scar bounded by two ramparts to the north and south. The slumped material may have moved as a debris flow, forming a large talus downslope of the slide. However, the submarine counterpart of the south area of the Grand Brule seems to be composed by a slumped block whose structure is apparently not disturbed. 3. (3) Another prominent feature is a conspicuous topographic high that occupies nearly all the center of the surveyed zone. This “cast flank submarine plateau” cannot be associated with any active structure of Piton de la Fournaise. Its surface generally dips gently (2–3°), and its northern and southern flanks are extensively cut by landslides. Cones of variable dimensions are observed on the plateau and farther to the east. Three hypotheses are examined to account for the origin of this morphology: (a) remnant flank of an ancestral Fournaise volcano associated with a large buried intrusion found by drilling beneath the Grand Brule; (b) distinct volcanic massif or (c) material of a huge and ancient landslide. The geophysical data show that the western part of this submarine plateau is reversely magnetized and associated with a moderate positive gravity anomaly. This confirms that, as suggested by the bathymetric analysis, this part of the plateau is relatively coherent. Conversely, the eastern portion of the plateau appears to be poorly magnetized and composed of low-density material, probably chaotic materials derived from landslides. The reversed magnetization of the western part shows that the whole structure is older than 0.7 Ma. These results show that the volcanic history of Piton de la Fournaise is more complex than previously thought.

Le volcanisme ignimbritique du Tibesti Occidental (Sahara tchadien). Essai d’interprétation dynamique

RésuméCe volcanisme concerne 500 km3 de produits rhyolitiques émis au Tertiaire et au début du Quaternaire. On étudie d’abord les différents types d’ignimbrites et leurs relations à l’intérieur d’une même nappe. On essaie ensuite de dégager les conditions dans lesquelles la « lave vivante » initiale peut se fractionner (de façon irréversible) en une « coulée de tufs de néogenèse » progressivement ou brutalement. Etude de l’appareil externe (« bouclier-nappe » à pentes de 2°) et de l’appareil interne (faisceaux de dykes entrecroisés). L’éruption est précédée d’un bombement du sol et elle peut s’achever par un affaissement au centre du bouclier-nappe (caldeira ou dépression moins régulière).Le volcanisme ignimbritique est interprété comme un volcanisme de type central particulier, présentant des anologies assez inattendues avec le volcanisme basaltique de type hawaïen.

Les grandes étapes d’évolution d’un volcan andésitique composite: Exemple du Nevado de Toluca (Méxique)

AbstractThe Nevado de Toluca, in the middle of the Mexican volcanic belt, has been built by two very dissimilar phases. The first one that lasted more than one million years is mainly andesitic. Numerous massive and autobrecciated lava flows of this phase pass outwards into thick conglomeratic formations. The volume of this primitive volcano represents the essential part of the Nevado.After an intense periode of erosion, the second phase is of very short duration (about 100.000 years) and is dacitic in nature. Three main episode can be distinguished:1.Eruption of important ash and pumice pyroclastic flows related to caldera collapse above a shallow magmatic reservoir.2.Extrusions of several dacitic domes within and outside the caldera with numerous associated «nuées ardentes» surrounding the volcano.3.Plinian eruption leading to widespread pumiceous air-fall and to the opening of the present crater inside the caldera. Extrusion of a new small dacitic dome and late phreatic explosions. This second sequence of events can be interpreted as the progressive emptying of the crustal magmatic chamber without refilling by a new magma supply. The most recent activity in the area is represented by monogenic cones and flows of basic andesites outside the central vent system of the Nevado.

Flank failure–directed blast eruption at Soufrière, Guadeloupe, French West Indies: A 3,000-yr-old Mt. St. Helens?

An extensive, apparently monogenic, debris-flow deposit in the Basse-Terre-St. Claude area of southern Guadeloupe is contemporaneous with pyroclastic surge deposits that cover an area of at least 60 km2. These 3,000-yr-old deposits were emplaced by the avalanching of a sector of Soufriere volcano and an associated directed blast similar to the 1956 Bezymianni and the May 18, 1980, Mt. St. Helens eruptions.

The primitive volcano of Mount Pelée: its construction and partial destruction by flank collapse

Abstract The geology of Mount Pelee, the active volcano of Martinique (Lesser Antilles), has been actively studied during the past decade. This paper reviews the current knowledge about the oldest, least well-known part of the volcano (here called Paleo-Pelee) and the relations between Paleo-Pelee and the Mont Conil volcanic complex. Paleo-Pelee consists mostly of volcanic breccias with a few lava flows. It differs in this respect from neighbouring active volcanoes such as the Grande Decouverte (Soufriere) in Guadeloupe and the Soufriere in St. Vincent, the oldest products of which are mainly lava flows. Paleo-Pelee initially formed a regular cone which was comparable in size with the present edifice. It is now largely eroded and its products form prominent radial ridges on the western flank of Mount Pelee. These are buried under younger deposits on the eastern flank. The uniformly andesitic breccias of Paleo-Pelee are, for the most part, brecciated lava flows. It is inferred that the activity of Mount Pelee was less explosive at this stage than during following periods (Neo-Pelee). The Paleo-Pelee formations overlap those of Mont Conil. The geology of Mont Conil is poorly known. However, its lowest exposed products are submarine andesitic breccias cut by many dikes. This contrasts with Paleo-Pelee where all breccias are of subaerial origin. Pleistocene limestones are interbedded within the Mont Conil breccias between 120 and 150 m above sea level. Though Mont Conil seems to be closely linked to Mount Pelee (on the basis of available petrological data), a major uplift must have occurred sometime during or after the construction of Mont Conil and forms a recognizable geological boundary between the two volcanic systems. The Paleo-Pelee edifice was partially destroyed by a flank-collapse event that occurred on the southwest flank during the early stage (> 40,000-20,000 y.B.P.) of Neo-Pelee. This event is evidenced by a large (6 × 2.5 km), horseshoe-shaped structure which is limited by well-preserved morphological scarps on land and by a 300-m offshore scarp, as shown by Seabeam bathymetry. Paleo-Pelee breccias and lava flows are truncated by the boundary of the collapse structure. Lava domes occupied the upslope part of the structure at a later time. These domes are intersected by the present crater (“Etang Sec”), which has probably been the active vent for at least 2000 years. The active crater is located at the intersection between the flank-failure depression and an older caldera formed at the Paleo-Pelee stage but probably enlarged later. The flank-failure structure appears as a major structure of the volcano and creates a strong structural asymmetry in the edifice. This fact should be considered when dealing with such various aspects as the hydrogeology of the volcano, the structural control of past and future eruptions, and the volcanic hazards.

Observations, stratigraphy and eruptive processes of the 1990 eruption of Kelut volcano, Indonesia

Abstract The February 10, 1990 eruption of Kelut volcano (eastern Java) reportedly began with seven discrete, short-lived explosions between 11.41 and 12.35 local times. Deposits of this initial, phreatomagmatic stage include a basal ash-fall layer (unit A1), widespread pumice surge deposits (unit S) and related pisolitic ash layer (unit A2). The main, plinian phase of the eruption lasted about 4 hours from 12.35 and produced pumice-flow deposits (unit PF) overlain by a pumice fallout layer distributed mainly to the southwest (unit P), and intra-plinian scoria-flow deposits (unit SF). Uppermost scoria-rich ash fall layers (unit A3) likely relate to late, discrete eruptive pulses. A few small explosions resumed on February 11 and 12 leaving no recognizable deposit. An embryonic lava dome had formed in the crater bottom by April, then was submerged by the new crater lake. Destruction of the summit area resulted from emplacement of the pre-plinian pumice surge up to 4–5 km on the south and west flanks, and of the early plinian pumice flows up to 1–2 km radially from the crater, before these were channelized in the main valleys to further travel 3 km. Most of the 32 human deaths resulted from roof collapse under the load of fallout tephra beyond the devastated area, which had been evacuated before the eruption began. The eruption produced 0.13 km 3 of tephra, of which 0.12 km 3 represent the products of the plinian phase. The average eruptive rate of the plinian phase is estimated to have been ∼7.5×10 6 kg/s magma DRE. The pumice flows are interpreted to have been formed due to unsteadiness and low velocity of the eruptive column at the beginning of the plinian phase. The intra-plinian scoria flows incorporate either more degassed or colder juvenile magma; they were presumably erupted at the edge of the column, due to fluctuations in the mass flux and in pressure in the conduit.

Magma mixing in a main stage of formation of Montagne Pelée: the Saint Vincent-type scoria flow sequence (Martinique, F.W.I.)

Contrasting with the first stage of activity at Montagne Pelee, which built an andesitic volcano chiefly formed of brecciated lava flows and pyroclastic breccias, the second main stage (> 40,000 to 20,000 years B.P.) is characterized by the prevalence of scoria flows of more mafic composition. Many cross-sections of the deposits of stage 2 show an upward magmatic evolution from early acid andesites (plinian air-fall and pumice flow deposits) to late basaltic andesites (scoria flow deposits). Heterogeneous “banded” rocks, complex mineralogical associations and mineralogical disequilibrium features in the whole series are explained by mechanical mixing of contrasting magmas. Reversed zoning characterized plagioclase phenocrysts of both the acid and basic terms and also pyroxene phenocrysts of the mafic terms. Grossly uniform composition of residual glass in most macroscopically homogeneous rocks implies that chemical exchanges between liquids and local hybridization occurred. Plutonic rocks with cumulate textures are found in upper units of the sequence. Their phenocrysts (anorthite, salitic clinopyroxene, pargasitic hornblende and Ti-magnetite) are in equilibrium with a subalkaline basaltic interstitial liquid. The same phases are commonly present as xenocrysts in the basaltic andesites. It is inferred that basaltic andesites of stage 2 were firstly derived from a subalkaline basalt by crystal fractionation. Mixing of the basaltic andesites (52% SiO2) with acid andesites (62% SiO2) occurred in late stage of evolution of the series, producing heterogeneous rocks of intermediate composition. Successive eruption of acid up to mafic magmas, dominant amount of mafic rocks in the sequence and short duration of the magma mixing event shown by narrow rims of the reversely zoned phenocrysts, suggest that mixing was produced by intrusion of mafic magma into a differentiated magma body. We propose that hydrostatic overpressure induced by a high rate of intrusion triggered the eruption. Thus, magma mixing would be due to a large volume of mafic magma passing through the resident magma. Rapid admixture of the two batches increased gas exsolution, resulting in a highly explosive eruption. Magma mixing can thus explain the explosive dynamics of the basaltic andesites eruptions, which surprisingly contrast with the relatively effusive style of the preceding andesitic stage.

Le Capelinhos (Faïal, Açores) vingt ans après son éruption: le modèle éruptif «surtseyen» et les anneaux de tufs hyaloclastiques

Phreatomagmatic structures are of two kinds: maars and tuff-rings. Data given by records of Capelinhos activity (Faïal, Açores, 1957–1958), by structures at west point of Faïal island, and by palagonitic breccias of Velay and Cantal areas (France) lead to relate hyaloclastic tuff-rings and shallow subaquatic («surtseyan») eruptions. It is possible to precise causes, characteristics, and mechanism of formation of tuff-rings.ResumeLes structures phréatomagmatiques sont de deux types, maars et anneaux de tufs. Les données fournies par les descriptions de l’éruption du Capelinhos (Faïal, Açores, 1957–1958), par les structures de la pointe ouest de l’île de Faïal, et par les brèches palagonitiques du Velay et du Cantal (France), conduisent à relier anneaux de tufs hyaloclastiques et éruptions sub-aquatiques à faible profondeur («surt-seyennes»). Il est possible de préciser les causes, les caractéristiques, et le mécanisme de formation des anneaux de tufs.

Influence de la Nature des Magmas sur l’Activité Phréatomagmatique: Approche Volcanologique et Thermodynamique

Many volcanic forms resulting from phreatomagmatic eruptions of differentiated magmas have been studied in the Massif Central (France), in the Phlegrean Fields (Italy), and on Saõ Miguel island (Azores). They show a continuous series between explosion crater maar type — and the hyaoloclastic tuff-cone. An essential feature of this morphological series is the preponderance of tuff-rings resulting from subaerial eruptions. Subaerial tuff-rings of basic compositions are less common than maars. A thermodynamic approach shows that the quantity of heat supplied by the different kinds of magmas and the water / magma ratio are the essential parameters controlling the activity, and the resulting morohology of these volcanoes.ResumeDe nombreux volcans résultant d’éruptions phréatomagmatiques engendrées par des magmas différenciés ont été étudiés dans le Massif Central (France), dans les Champs Phlégréens (Italie), et l’ile de Sao miguel (Açores). Ils constituent une gamme de formes qui va du cratère d’explosion de type maar au cône de tufs hyaloclastiques. Un des traits essentiels de cette série morphologique est la prépondérance d’anneaux de tufs résultant d’éruptions subaériennes. Les anneaux de tufs subaériens liés aux magmas basiques sont beaucoup plus rare que les maars. Une approche thermodynamique montre que la quantité de chaleur fournie par les differents types de magmas et le rapport eau/magma sont le paramètres principaux controlant l’activité, done la morphologie des appareils phréatomagmatiques.

Découverte de coulées pyroclastiques sous-marines dans la zone de l’Iran Central — Mécanisme de mise en place

Study of these submarine pyroclastic flows is important for two reasons:1)From a regional point of view, submarine pyroclastic flows partially explain the origin of widespread argillized glasses in the Eocene sedimentary series of Central Iran, which have diversely been described by earlier workers as Green Tuffs, Tuffaceous Muds, Green Series.2)More generally, in contrast to basaltic volcanism, submarine acid volcanism is still very poorly-known. The mechanism here proposed is as follows: starting with a submarine, highly vesiculated lava containing idiomorphic phenocrysts, gas expansion gives rise to a vitroclastic shardy facies, whereas phenocrysts are fragmented. The whole makes up a gravitational flow that is rich in pumice, glassy shards, broken crystals, and gases; it becomes increasingly turbulent and able to tear-up sea-bottom sediments in its down-stream course. This flow evolves into turbidity currents at its distal reaches.ResumeL’étude des coulées pyroclastiques sousmarines découvertes en Iran revêt un double intérêt. Tout d’abord, un intérêt régional car elles permettent d’expliquerpro parte l’origine des débris de verre argilisés largement répandus dans les séries sédimentaires éocènes de la zone de l’Iran Central, que les anciens auteurs appelaient tufs verts, boues tuffitiques ou, le plus souvent, séries vertes. Un intérêt plus général ensuite car, contrairement au volcanisme basaltique, le volcanisme sous-marin acide est encore très peu connu. Le dynamisme que nous proposons implique qu’à partir d’une lave acide sous-marine, fortement vésiculée et à phénocristaux automorphes, l’expansion des gaz engendre un faciès vitroclastique à échardes, tandis que les phénocristaux sont fragmentés. L’ensemble forme un écoulement gravitaire, riche en ponces, échardes, cristaux brisés et gaz, de plus en plus turbulent vers l’aval et capable d’arracher au fond marin des lambeaux de sédiments. Il évolue par la suite à la manière d’un courant de turbidité.