Thomas W. Sisson

Delayed, disequilibrium degassing in rhyolite magma: decompression experiments and implications for explosive volcanism

Recent numerical models and analog shock tube experiments show that disequilibrium degassing during magma ascent may lead to violent vesiculation very near the surface. In this study a series of decompression experiments using crystal-free, rhyolite melt were conducted to examine the development of large supersaturations due to delayed, homogenous (spontaneous) bubble nucleation. Melts were saturated at 900°C and 200 MPa with either 5.2 wt% dissolved H2O, or with 4.2 wt% H2O and 640 ppm CO2, and isothermally decompressed at linear rates of either 0.003, 0.025, or 8.5 MPa/s to final pressures between 25 and 175 MPa. Additional isobaric saturation experiments (900°C, 200–25 MPa) using pure H2O or mixed H2O–CO2 fluids establish reference equilibrium solubility curves/values. Homogenous nucleation is triggered in both H2O-only and H2O–CO2 experiments once the supersaturation pressure (ΔPss) reaches ∼120–150 MPa and the melt contains ∼two times its equilibrium water contents. Bubble number density and nucleation rate depend on the supersaturation pressure, with values on the order of 102/cm3 and <1/cm3/s for ΔPss∼120 MPa; 106/cm3 and 103–105/cm3/s for ΔPss∼130–150 MPa; and 107/cm3 and 106/cm3/s for ΔPss∼160–175 MPa. Nucleation rates are consistent with classical nucleation theory, and infer an activation energy for nucleation of 1.5×10−18 J/nucleus, a critical bubble radius of 2×10−9 m, and an effective surface tension for rhyolite at 5.2 wt% H2O and 900°C of 0.10–0.11 N/m. The long nucleation delay limits the time available for subsequent diffusion such that disequilibrium dissolved H2O and CO2 contents persist to the end of our runs. The disequilibrium degassing paths inferred from our experiments contrast markedly with the equilibrium or quasi-equilibrium paths found in other studies where bubble nucleation occurs heterogenously on crystals or other discontinuities in the melt at low ΔPss. Homogenous and heterogenous nucleation rates are comparable, however, as are bubble number densities, so that at a given decompression rate it appears that nucleation mechanism, rather than nucleation rate, determines degassing efficiency by fixing the pressure (depth) at which vesiculation commences and hence the time available for equilibration prior to eruption. Although real systems are probably never truly crystal-free, our results show that rhyolitic magmas containing up to 104 crystals/cm3, and perhaps as high as 106 crystals/cm3, are controlled by homogenous, rather than heterogenous, nucleation during ascent.

Gas-driven filter pressing in magmas

Most silicic and some mafic magmas expand via second boiling if they crystallize at depths of about 10 km or less. The buildup of gas pressure due to second boiling can be relieved by expulsion of melt out of the region of crystallization, and this process of gas-driven filter pressing assists the crystallization differentiation of magmas. For gas-driven filter pressing to be effective, the region of crystallization must inflate slowly relative to buildup of pressure and expulsion of melt. These conditions are satisfied in undercooled magmatic inclusions and in thin sheets of primitive magma underplating cooler magma reservoirs. Gas-driven filter pressing thereby adds fractionated melt to magma bodies. Gas-driven filter pressing is probably the dominant process by which highly evolved melts segregate from crystal mush to form aplitic dikes in granitic plutons; this process could also account for the production of voluminous, crystal-poor rhyolites.

Volcano collapse promoted by hydrothermal alteration and edifice shape, Mount Rainier, Washington

Catastrophic collapses of steep volcano flanks threaten many populated regions, and understanding factors that promote collapse could save lives and property. Large collapses of hydrothermally al- tered parts of Mount Rainier have generated far-traveled debris flows; future flows would threaten densely populated parts of the Puget Sound region. We evaluate edifice collapse hazards at Mount Rainier using a new three-dimensional slope stability method incorporating detailed geologic mapping and subsurface geophysical imaging to de- termine distributions of strong (fresh) and weak (altered) rock. Quan- titative three-dimensional slope stability calculations reveal that size- able flank collapse ( .0.1 km 3 ) is promoted by voluminous, weak, hydrothermally altered rock situated high on steep slopes. These con- ditions exist only on Mount Rainiers upper west slope, consistent with the Holocene debris-flow history. Widespread alteration on lower flanks or concealed in regions of gentle slope high on the edifice does not greatly facilitate collapse. Our quantitative stability assessment method can also provide useful hazard predictions using reconnais- sance geologic information and is a potentially rapid and inexpensive new tool for aiding volcano hazard assessments.

A mass proportion method for calculating melting reactions and application to melting of model upper mantle lherzolite

We present a method for calculating quantitative melting reactions in systems with multiple solid solutions that accounts for changes in the mass proportions of phases between two points at different temperatures along a melting curve. This method can be applied to any data set that defines the phase proportions along a melting curve. The method yields the net change in mass proportion of all phases for the chosen melting interval, and gives an average reaction for the melting path. Instantaneous melting reactions can be approximated closely by choosing sufficiently small melting intervals. As an application of the method, reactions for melting of model upper mantle peridotite are calculated using data from the system CaO-MgO-Al2O3-SiO2-Na2O (CMASN) over the pressure interval 0.7 – 3.5 GPa. Throughout almost this entire pressure range, melting of model lherzolite involves the crystallization of one or more solid phases, and is analogous to melting at a peritectic invariant point. In addition, we show that melting reactions for small melting intervals ( 10%), reaction stoichiometries calculated in CMASN are usually in good agreement with those available for melting of natural peridotite. The coefficients of melting reactions calculated from this method can be used in equations that describe the behavior of trace elements during melting. We compare results from near-fractional melting models using (1) melting reactions and rock modes from CMASN, and (2) constant reactions representative of those used in the literature. In modeling trace element abundances in melt, significant differences arise for some elements at low degrees of melting (< 10%). In modeling element abundances in the residue, differences increase with increase in degree of melting. Reactions calculated along the model lherzolite solidus in CMASN are the only ones available at present for small degrees of melting so we recommend them for accurate trace element modeling of natural lherzolite.

Aerogeophysical measurements of collapse-prone hydrothermally altered zones at Mount Rainier volcano

Hydrothermally altered rocks can weaken volcanoes, increasing the potential for catastrophic sector collapses that can lead to destructive debris flows. Evaluating the hazards associated with such alteration is difficult because alteration has been mapped on few active volcanoes and the distribution and severity of subsurface alteration is largely unknown on any active volcano. At Mount Rainier volcano (Washington, USA), collapses of hydrothermally altered edifice flanks have generated numerous extensive debris flows and future collapses could threaten areas that are now densely populated. Preliminary geological mapping and remote-sensing data indicated that exposed alteration is contained in a dyke-controlled belt trending east–west that passes through the volcanos summit. But here we present helicopter-borne electromagnetic and magnetic data, combined with detailed geological mapping, to show that appreciable thicknesses of mostly buried hydrothermally altered rock lie mainly in the upper west flank of Mount Rainier. We identify this as the likely source for future large debris flows. But as negligible amounts of highly altered rock lie in the volcanos core, this might impede collapse retrogression and so limit the volumes and inundation areas of future debris flows. Our results demonstrate that high-resolution geophysical and geological observations can yield unprecedented views of the three-dimensional distribution of altered rock.

Gas evolution in eruptive conduits: Combining insights from high temperature and pressure decompression experiments with steady-state flow modeling

Abstract In this paper we examine the consequences of bubble nucleation mechanism on eruptive degassing of rhyolite magma. We use the results of published high temperature and pressure decompression experiments as input to a modified version of CONFLOW, the numerical model of Mastin and Ghiorso [(2000) U.S.G.S. Open-File Rep. 00-209, 53 pp.] and Mastin [(2002) Geochem. Geophys. Geosyst. 3, 10.1029/2001GC000192] for steady, two-phase flow in vertical conduits. Synthesis of the available experimental data shows that heterogeneous nucleation is triggered at ΔP 120–150 MPa, and leads to disequilibrium degassing at extreme H2O supersaturation. In this latter case, nucleation is an ongoing process controlled by changing supersaturation conditions. Exponential bubble size distributions are often produced with number densities of 106–109 bubbles/cm3. Our numerical analysis adopts an end-member approach that specifically compares equilibrium degassing with delayed, disequilibrium degassing characteristic of homogeneously-nucleating systems. The disequilibrium simulations show that delaying nucleation until ΔP=150 MPa restricts degassing to within ∼1500 m of the surface. Fragmentation occurs at similar porosity in both the disequilibrium and equilibrium modes (∼80 vol%), but at the distinct depths of ∼500 m and ∼2300 m, respectively. The vesiculation delay leads to higher pressures at equivalent depths in the conduit, and the mass flux and exit pressure are each higher by a factor of ∼2.0. Residual water contents in the melt reaching the vent are between 0.5 and 1.0 wt%, roughly twice that of the equilibrium model.

Ridge-forming, ice-bounded lava flows at Mount Rainier, Washington

Large (0.3–4 km3) andesite and dacite lava flows at Mount Rainier, Washington, sit atop or are perched along the sides of high ridges separating deep valleys. Early researchers proposed that these ridge-forming lavas flowed into paleovalleys and displaced rivers to their margins; entrenchment of the rivers then left the lavas atop ridges. On the basis of exceptional flow thickness, ice-contact features, and eruption age measurements, we propose that the lavas flowed beside and between valley glaciers that filled the adjacent valleys in the Pleistocene. When the glaciers retreated, the flows were left high on the adjacent ridges. These lavas were never situated at valley floors and do not represent products of reversed topography. Instead, ridge-forming and perched lava flows at Mount Rainier and at many other high stratovolcanoes illustrate the ability of ice to dam, deflect, and confine flowing lava.

Young cumulate complex beneath Veniaminof caldera, Aleutian arc, dated by zircon in erupted plutonic blocks

Mount Veniaminof volcano, Alaska Peninsula, provides an opportunity to relate Quaternary volcanic rocks to a coeval intrusive complex. Veniaminof erupted tholeiitic basalt through dacite in the past ∼260 k.y. Gabbro, diorite, and miarolitic granodiorite blocks, ejected 3700 14 C yr B.P. in the most recent caldera-forming eruption, are fragments of a shallow intrusive complex of cumulate mush and segregated vapor-saturated residual melts. Sensitive high-resolution ion microprobe (SHRIMP) analyses define 238 U- 230 Th isochron ages of 17.6 ± 2.7 ka, 5 +11/–10 ka, and 10.2 ± 4.0 ka (2σ) for zircon in two granodiorites and a diorite, respectively. Sparse zircons from two gabbros give 238 U- 230 Th model ages of 36 ± 8 ka and 26 ± 7 ka. Zircons from granodiorite and diorite crystallized in the presence of late magmatic aqueous fluid. Although historic eruptions have been weakly explosive Strombolian fountaining and small lava effusions, the young ages of plutonic blocks, as well as late Holocene dacite pumice, are evidence that the intrusive complex remains active and that evolved magmas can segregate at shallow levels to fuel explosive eruptions.

Ultra-high chlorine in submarine Kı̄lauea glasses: evidence for direct assimilation of brine by magma

Basaltic glass grains from the submarine south flank of Kilauea, Hawai′i, have Cl concentrations of 0.01–1.68 wt%, the latter being the highest Cl content yet recorded for a Hawaiian glass. The high-Cl glass grains are products of brine assimilation by tholeiite magma. The glasses are grains in a sandstone clast from bedded breccias draping the southwestern margin of Kilauea’s submarine midslope bench. The clast contains two distinct suites of glass grains: abundant degassed tholeiites, perhaps derived from subaerial lavas of Mauna Loa that shattered upon ocean entry, and a smaller population of Kea-type tholeiite (n=17 analyzed) that erupted subaqueously, based on elevated S (780–1050 ppm), H2O (0.42–1.27 wt%), and CO2 ( 1000 ppm, six >5000 ppm, and two grains have >10 000 ppm dissolved Cl. Abundances of H2O, Na2O, K2O, and several trace elements increase regularly with Cl concentration, and we estimate that Cl enrichment was due to up to 13 wt% addition of a brine consisting of 78% H2O (wt), 13% Cl, 4.4% Na, 2.6% K, 2.6% Ca, 620 ppm Ba, 360 ppm Sr, 65 ppm Rb, and 7 ppm Pb. The large amounts of brine addition argue against bulk assimilation of low-porosity brine-bearing rock. The brine’s composition is appropriate for a seawater-derived hydrothermal fluid that reacted with basaltic wall rocks at T>100°C, losing Mg and S and gaining K, Ca, Rb, Ba, Sr, and Pb, followed by phase separation near 500°C and ∼50 MPa (5 km below sea level at hydrostatic pressure). Brine was assimilated at or near the depth it formed, as estimated on petrologic grounds, but under lithostatic conditions. The highest extents of assimilation either forced volatile saturation of the magma or enriched already coexisting magmatic vapor in H2O. Possible mechanisms for assimilation are: (1) forcible injection of brine into magma during bursting of overpressured pockets heated by new dikes, or (2) intrusion of magma into lenses or sills occupied by trapped brine.

Igneous phenocrystic origin of K-feldspar megacrysts in granitic rocks from the Sierra Nevada batholith

Study of four K-feldspar megacrystic granitic plutons and related dikes in the Sierra Nevada composite batholith indicates that the megacrysts are phenocrysts that grew in contact with granitic melt. Growth to megacrystic sizes was due to repeated replenishment of the magma bodies by fresh granitic melt that maintained temperatures above the solidus for extended time periods and that provided components necessary for K-feldspar growth. These intrusions cooled 89–83 Ma, are the youngest in the range, and represent the culminating magmatic phase of the Sierra Nevada batholith. They are the granodiorite of Topaz Lake, the Cathedral Peak Granodiorite, the Mono Creek Granite, the Whitney Granodiorite, the Johnson Granite Porphyry, and the Golden Bear Dike. Megacrysts in these igneous bodies attain 4–10 cm in length. All have sawtooth oscillatory zoning marked by varying concentration of BaO ranging generally from 3.5 to 0.5 wt%. Some of the more pronounced zones begin with resorption and channeling of the underlying zone. Layers of mineral inclusions, principally plagioclase, but also biotite, quartz, hornblende, titanite, and accessory minerals, are parallel to the BaO-delineated zones, are sorted by size along the boundaries, and have their long axes preferentially aligned parallel to the boundaries. These features indicate that the K-feldspar megacrysts grew while surrounded by melt, allowing the inclusion minerals to periodically attach themselves to the faces of the growing crystals. The temperature of growth of titanite included within the K-feldspar megacrysts is estimated by use of a Zr-in-titanite geothermometer. Megacryst-hosted titanite grains all yield temperatures typical of felsic magmas, mainly 735–760 °C. Titanite grains in the granodiorite hosts marginal to the megacrysts range to lower growth temperatures, in some instances into the subsolidus. The limited range and igneous values of growth temperatures for megacryst-hosted titanite grains support the interpretation that the megacrysts formed as igneous sanidine phenocrysts, that intrusion temperatures varied by only small amounts while the megacrysts grew, and that megacryst growth ceased before the intrusions cooled below the solidus. Individual Ba-enriched zones were apparently formed by repeated surges of new, hotter granitic melt that replenished these large magma chambers. Each recharge of hot magma offset cooling, maintained the partially molten or mushy character of the chamber, stirred up crystals, and induced convective currents that lofted, settling megacrysts back up into the chamber. Because of repeated reheating of the magma chamber and prolonged maintenance of the melt, this process apparently continued long enough to provide the ideal environment for the growth of these extraordinarily large K-feldspar phenocrysts.