R.S.J. Sparks

The dynamics of bubble formation and growth in magmas: A review and analysis

One portion of a constant current is applied to one semiconductor device and the remainder to a second semiconductor device. When the temperature difference between the devices changes, the division of current also changes. An operational amplifier responsive to this division of current may be employed to drive an indicator of the temperature difference and to reestablish the original temperature difference between the devices.

Thermal and mechanical constraints on mixing between mafic and silicic magmas

When magmas of different temperature and composition are intimately mingled together, transfer of heat results in substantial changes in the rheological properties of the magmas. Since thermal diffusion rates are orders of magnitude faster than chemical diffusion rates, mixing magmas will come nearly to the same temperature before complete homogenization of the magmas can occur by diffusion and shearing. The ability of magmas to mix thus depends on their physical properties after thermal equilibration. Calculations are presented on how the viscosity and crystal content of mafic and silicic magmas vary as a function of their initial temperatures and the proportion of mafic magma in the mixture. Three physical situations can be identified: (a) where the mafic magma remains less viscous than the silicic magma; (b) where the mafic magma becomes more viscous than the silicic magma due to crystallization; and (c) where the mafic magma is effectively solid due to its high crystal content. In the last situation it is proposed that complete mixing cannot take place, but the mafic magma is dispersed as solid xenoliths or inclusions within the silicic magma. Xenolith or inclusion formation occurs when there is a large temperature difference between the magmas or a large proportion of silicic magma. Complete hybridization can only occur when the magmas both behave as liquids at the same temperature. A diagram is constructed that shows the fields where the mafic magma becomes a solid or remains a fluid on a plot of the proportion of mafic magma against the composition of the mafic magma. Where there is a large proportion of silicic magma, complete hybridization can only occur with evolved mafic magmas (andesitic magmas). An example of this compositional selectivity is described from St. Kilda, Scotland where silicic magmas have only hybridized with highly evolved theoleiitic andesite magmas, although the silicic magma is intimately intermingled with more mafic magmas in net-veined complexes. When basaltic magma is intermingled with silicic magma, high proportions (typically 50% or greater) of basalt are necessary to enable mixing to occur. Hybrid magmas involving small proportions of basalr magma with large proportions of silicic magma are notably absent from St. Kilda hybrid rocks and other areas. Many mafic xenoliths may represent less evolved basaltic magma which solidified when commingled with a much larger volume of silicic magma.

Thickness variations and volume estimates of tephra fall deposits: the importance of particle Reynolds number

Abstract Well-preserved tephra fall deposits display thickness variations which are more complex than simple exponential thinning. On plots of log thickness against square root of area enclosed by an isopach contour, many deposits show two or more approximately straight-line segments and in some cases regions of curvature. We show that major changes in thinning rate occur as the particle size decreases with distance from the vent, as a consequence of the change of settling behaviour from high to low Reynolds number as predicted by W.I. Rose. Computer models of sedimentation from laterally spreading plumes predict a steep proximal segment with exponential thinning for coarse ejecta (lapilli and coarse ash) with high Reynolds number (Re>500). At greater distance finer ejecta are predicted to show power-law thinning. Two distal segments are identified. The most distal segment is composed of low Reynolds number particles and can be approximated by an exponential thinning law, but is better described by a power law. The distal and proximal segments are connected by a curved segment containing mixed populations of intermediate (0.4

Experimental studies of collapse calderas

We present experiments to investigate the formation of calderas using inflatable balloons in a medium of fused alumina powder. This system provides a model of a magma chamber in an homogeneous elastic media which scales to volcanic systems in terms of geometry and strength of the media. Three kinds of experiment were performed: (i) the balloon was inflated and then deflated simulating pre-emption tumescence and caldera collapse: (ii) the balloon was deflated without prior inflation, simulating caldera formation without pre-eruption tumescence: (iii) three balloons were inflated and deflated in a line, simulating multiple magma chambers. The influence of magma chamber shape was investigated using different shaped balloons. Balloon inflation formed a dome with a surface pattern of shallow polygonal fractures centrally, and radial fractures peripherally. Uplift occurs along deep inward-dipping concentric reverse faults. Deflation generated depressions with an outer set of concentric ring fractures surrounding a central funnel formed by sagging along numerous minor faults, some of which formed during doming. During collapse the concentric reverse faults were converted to normal faults and the radial normally-faulted fractures were converted to reverse faults. Without prior tumescence, an outer set of near vertical ring faults surround a flat central depression due to collapse of a coherent block. An inner set of outward-dipping reverse ring faults also develop. The experimental caldera area increased with balloon size and increased as the balloon depth decreased. For a given balloon volume and depth, collapse area is significantly smaller if doming had occurred prior to collapse. Arcuate fractures can provide the pathways for magma ascent during pre-emption tumescence (Smith & Bailey 1968), but the shallow radial and polygonal fracture systems may also influence vent location. Changes in fault geometry and sense of movement (extension to compression and vice versa) occur when subsidence initiates, and can account for changes in vent location that occur in many eruptions. Multiple overlapping collapse calderas are associated with magma chamber migration, whereas nested calderas can result from the activity of a single magma chamber.

A proximal ignimbrite breccia facies on santorini , Greece

Abstract Prominent among the pyroclastic deposits of Santorini are several thick, widespread lithic breccia deposits, which are found in intimate association with ignimbrite. At least three of these breccias are interpreted, on the basis of field and grain-size criteria, as having originated by the segregation of lithic clasts from active pyroclastic flows. They therefore record the occurrence of three large, previously unrecognized ignimbrite-forming eruptions of the volcano. The breccias of the 18,500 yr. B.P. Cape Riva eruption include two types. The first type is a thin, basal ground breccia, which overlies a strong erosion surface. This breccia shows pinch and swell structures and is strongly enriched in lithic and crystal components. It is considered to have formed by strong fluidization due to incorporation of air into the head of an active pyroclastic flow. The second, and predominant, type consists of thick co-ignimbrite lag breccias (up to 25 m), which overlie the ground breccia. These deposits are generally clast-supported, poorly sorted breccias which in places grade both vertically and laterally into non-welded pumiceous ignimbrite. They consist of well-defined, normally graded units which show coarse tail grading of lithic and pumice clasts. Each breccia unit is underlain by a thin, inversely graded ignimbrite basal layer, and correlates laterally with a flow unit of the associated ignimbrite. The lag breccias are therefore thick equivalents of the 2b lithic concentration zones of Sparks et al. The lag breccias and ignimbrite contain abundant lithic segregation structures that are characteristic of strong gas fluidization. These structures, the presence of basal layers, and the gradation into normal ignimbrite, suggest that the lag breccias originated by the segregation of lithic clasts within the bodies of dense, but strongly fluidized pyroclastic flows. The Cape Riva breccias occur within a few kilometers of their source vent and are interpreted as proximal facies of their associated ignimbrite. The presence of the ground breccia indicates that within this distance, the pyroclastic flows had developed the head and body regions characteristic of gravity currents. The deflation of the pyroclastic flow bodies, within a few kilometers from source, to particle concentrations sufficient to permit the generation of basal layers and coarse tail grading, is incompatible with present theories of column collapse. It is postulated that high pressures at the base of the collapsing Cape Riva eruption column were sufficient to significantly compress the dilute particle-gas mixture of the column close to the source vent. Subsequent sedimentation, as the pyroclastic flows moved laterally, increased the density further to the point where the observed sedimentary features could form. Simultaneous decompression of the gas phase resulted in strong fluidization, and the segregation of the lag breccias.

The initial giant umbrella cloud of the May 18th, 1980, explosive eruption of Mount St. Helens

Abstract The initial eruption column of May 18th, 1980 reached nearly 30 km altitude and released 10 17 joules of thermal energy into the atmosphere in only a few minutes. Ascent of the cloud resulted in forced intrusion of a giant umbrella-shaped cloud between altitudes of 10 and 20 km at radial horizontal velocities initially in excess of 50 m/s. The mushroom cloud expanded 15 km upwind, forming a stagnation point where the radial expansion velocity and wind velocity were equal. The cloud was initiated when the pyroclastic blast flow became buoyant. The flow reduced its density as it moved away from the volcano by decompression, by sedimentation, and by mixing with and heating the surrounding air. Observations indicate that much of the flow, covering an area of 600 km 2 , became buoyant within 1.5 minutes and abruptly ascended to form the giant cloud. Calculations are presented for the amount of air that must have been entrained into the flow to make it buoyant. Assuming an initial temperature of 450°C and a magmatic origin for the explosion, these calculations indicate that the flow became buoyant when its temperature was approximately 150°C and the flow consisted of a mixture of 3.25 × 10 11 kg of pyroclasts and 5.0 × 10 11 kg of air. If sedimentation is considered, these figures reduce to 1.1 × 10 11 kg of pyroclasts and 1.0 × 10 11 kg of air.

The 1975 sub-terminal lavas, mount etna: a case history of the formation of a compound lava field

The 1975 sub-terminal activity was characterised by low effusion rates (0.3–0.5 m3 s−1) and the formation of a compound lava field composed of many thousands of flow units. Several boccas were active simultaneously and effusion rates from individual boccas varied from about 10−4 to 0.25 m3s−1. The morphology of lava flows was determined by effusion rate (E): aa flows with well-developed channels and levees formed when E > 2 × 10−3 m3 s−1, small pahoehoe flows formed when 2 × 10−3 m3 s−1 >E > 5 > 10−4 m3 s−1 and pahoehoe toes formed when E < 5 × 10−4 m3 s−1. There was very little variation with time in the effusion temperature, composition or phenocryst content of the lava. New boccas were commonly formed at the fronts of mature lava flows which had either ceased to flow or were moving slowly. These secondary boccas developed when fluid lava in the interior of mature aa flows either found a weakness in the flow front or was exposed by avalanching of the moving flow front. The resulting release of fluid lava was accompanied by either partial drainage of the mature flow or by the formation of a lava tube in the parent flow. The temperature of the lava forming the new bocca decreased with increasing distance from the source bocca (0.035°C m−1). It is demonstrated from the rate of temperature decrease and from theoretical considerations that many of the Etna lavas still contained a substantial proportion of uncooled material in their interior as they came to rest. The formation of secondary boccas is postulated to be one reason why direct measurements of effusion rates tend, in general, to overestimate the total effusion rates of sub-terminal Etna lava fields.

Ignimbrites of the Cerro Galan caldera, NW Argentina

Abstract The 35 × 20 km Cerro Galan resurgent caldera is the largest post-Miocene caldera so far identified in the Andes. The Cerro Galan complex developed on a late pre-Cambrian to late Palaeozoic basement of gneisses, amphibolites, mica schists and deformed phyllites and quartzites. The basement was uplifted in the early Miocene along large north-south reverse faults, producing a horst-and-graben topography. Volcanism began in the area prior to 15 Ma with the formation of several andesite to dacite composite volcanoes. The Cerro Galan complex developed along two prominent north-south regional faults about 20 km apart. Dacitic to rhyodacitic magma ascended along these faults and caused at least nine ignimbrite eruptions in the period 7-4 Ma (K-Ar determinations). These ignimbrites are named the Toconquis Ignimbrite Formation. They are characterised by the presence of basal plinian deposits, many individual flow units and proximal co-ignimbrite lag breccias. The ignimbrites also have moderate to high macroscopic pumice and lithic contents and moderate to low crystal contents. Compositionally banded pumice occurs near the top of some units. Many of the Toconquis eruptions occurred from vents along a north-south line on the western rim of the young caldera. However, two of the ignimbrites erupted from vents on the eastern margin. Lava extrusions occurred contemporaneously along these north-south lines. The total D.R.E. volume of Toconquis ignimbrite exceeds 500 km 3 . Following a 2-Ma dormant period a single major eruption of rhyodacitic magma formed the 1000-km 3 Cerro Galan ignimbrite and the caldera. The ignimbrite (age 2.1 Ma on Rb-Sr determination) forms a 30–200-m-thick outflow sheet extending up to 100 km in all directions from the caldera rim. At least 1.4 km of welded intracaldera ignimbrite also accumulated. The ignimbrite is a pumice-poor, crystal-rich deposit which contains few lithic clasts. No basal plinian deposit has been identified and proximal lag breccias are absent. The composition of pumice clasts is a very uniform rhyodacite which has a higher SiO 2 content but a lower K 2 O content than the Toconquis ignimbrites. Preliminary data indicate no evidence for compositional zonation in the magma chamber. The eruption is considered to have been caused by the catastrophic foundering of a cauldron block into the magma chamber. Post-caldera extrusions occurred shortly after eruption along both the northern extension of the eastern boundary fault and the western caldera margin. Resurgence also occurred, doming up the intracaldera ignimbrite and sedimentary fill to form the central mountain range. Resurgent doming was centred along the eastern fault and resulted in radial tilting of the ignimbrite and overlying lake sediments.

Peridotite sills and metasomatic gabbros in the Eastern Layered Series of the Rhum complex

Mapping of the Eastern Layered Series (ELS) of the Rhum ultrabasic complex on the northern flank of Hallival shows that peridotite and allivalite (troctolite or gabbro) layers are laterally discontinuous and vary both in thickness and lithology. Peridotite generally has sharp upper and lower contacts against the allivalites, which sometimes cut across the layering in the allivalite. Reaction, dissolution and hybridization effects between peridotite and allivalite are developed locally. Some troctolite layers terminate as isolated, fingered blocks in peridotite. There are many small peridotite bodies which are clearly intrusive into allivalite and have previously been identified as distinct peridotite sheets and plugs. They are petrographically almost identical to the major stratiform peridotites and in some cases are apophyses from them. We propose that many of the peridotite layers in the ELS formed as thick sills of picritic magma emplaced into a partly solidified, layered troctolite complex. The stratiform gabbros of the ELS are heterogeneous, layered rocks that commonly contain relicts of troctolite and anorthosite. Wavy (metre-scale) contacts between gabbro and troctolite cut across pre-existing grain-size, modal and rhythmic layering with little disruption. These metasomatic gabbros mimic the textures, grain-size and rhythmic layering of their troctolitic protoliths. We propose that many of the ELS gabbros formed as a result of interaction between porous troctolites and a low-temperature basaltic melt. Residual basaltic melt segregated from solidifying peridotite may have caused this metasomatism.

Fall-out and deposition of volcanic ash during the 1979 explosive eruption of the soufriere of St. Vincent

Abstract The 1979 explosive activity of Soufriere Volcano, St. Vincent, commenced on April 13th and continued for two weeks, producing eleven eruption columns. The activity produced air-fall ash with abundant accretionary lapilli, plus minor pyroclastic flows and base surge deposits. The dispersal of the columns was observed by an SMS-1 geostationary satellite. All but two of the plumes were dispersed entirely to the east. The last event on April 26th formed a minor lobe to the east and a major lobe to the south-southeast. The April 26th column was observed to rise to 8 km within five minutes and ascended to a maximum of 14 km in the next hour. The magma discharge rate was greater than 3000 m3/s. The plume expanded at a rate of 4000 km2/hr and was dispersed at a mean velocity of 15 m/s. The ash deposits show bimodal grain-size distribution, with 95% of the ash collected being sub 1 mm diameter. Modelling of the ash fall-out has allowed interpretation of grainsize parameters in a quantitative manner. The fall velocities of ash particles appear to have been reduced by turbulence and downwind ascent of the plume which is attributed to fallout of ash reducing the plume density. The coarse population of the April 26th deposit displayed a systematic change in Mdθand σθ away from source. The fine mode displays uniform Mdθ and σθ everywhere and is interpreted as having been deposited in the form of accretionary lapilli.