Alfred T. Anderson

Alteration of oceanic crust and geologic cycling of chlorine and water

We report new estimates of transport rates for H2O and Cl between the mantle and surface reservoirs. Our estimates take into consideration alteration of oceanic crust, especially that of plutonic rocks, and possible subduction of sediments. The effect of (hydrothermal) alteration on the Cl budget seems to be negligible, but the effect on the H2O budget is significant. Altered oceanic crust (excluding sediments) contains about 10 times as much H2O as the unaltered crust, and its subduction may result in a net transport of H2O to the upper mantle in subduction zones. However, the rate of expulsion of H2O from the mantle by subduction-zone magmatism is comparable to the amount released by ridge magmatism, and is only about 10% of the amount subducted. Therefore, about 90% of the subducted H2O must be returned to the mantle or returned to the crust by other processes. In addition, subduction of oceanic sediments to mantle depths will result in 1. (1) a further increase in the return rate of H2O to the mantle reservoir, 2. (2) possible net transfer of Cl to the mantle, depending on the rate of pore water expulsion.

Chlorine, Sulfur, and Water in Magmas and Oceans

The concentrations of chlorine, sulfur, and water in basaltic and andesitic glasses trapped in large crystals are commonly 5 to 20 times as high as those found in vesicular glasses and bulk lavas. Chlorine and sulfur concentrations in trapped glasses range from 100 to 2,200 ppm and from 50 to 2,800 ppm, respectively. Vesicular glasses contain 100 to 600 ppm chlorine and 50 to 400 ppm sulfur. The chlorine-rich trapped glasses tend to have high water contents, from 1 to 7 percent by weight. In general, the basaltic trapped glasses have as high or higher concentrations of chlorine, sulfur, and water as do the andesitic glasses. For individual eruptions, water-rich magmas (Mount Shasta, California) lose an appreciable fraction of their chlorine prior to extrusion, whereas water-poor magmas (Kilauea, Hawaii) lose little if any chlorine, even after extrusion. Apparently, water acts as a carrier gas and distills chlorine out of magmas as it boils away. The behavior of sulfur is complex and appears to reflect pre-eruption saturation with respect to a sulfide melt and strong loss of sulfur as SO 2 during eruption (the pre-eruption fugacity of SO 2 is about 60 atm for Kilauean fractionated basalt). The potential chlorine, sulfur, and water contents of andesitic magmas have been estimated by assuming derivation from basaltic precursors and extrapolating the volatile contents of related basalts along lines of constant ratios to K 2 O to 1 percent K 2 O. The potential volatile concentrations are: 0.08 to 0.66 percent Cl, 0.10 to 0.67 percent S, and about 5 to 15 percent H 2 O. Assuming that the estimated rate of production of continental crust in Cenozoic island arcs has been constant throughout earth history, the total amounts of Cl, H 2 O, and continental crust produced are within factors of 0.3 to 4 times the presently existing masses of these substances in the surface reservoir of sea water, sediment, and crust. The amount of sulfur produced is larger than that in the reservoir. Sulfur probably is less efficiently transferred into the surface reservoir than are chlorine and water. The data can be reconciled with models of recycling of volatiles through the mantle, but the near coincidence of residence times for chlorine, water, and crust in the surface reservoir with the age of the earth is not explained. Also, the equality between the C1/H 2 O ratio of oceanic tholeiite, Kilauean basalt, and H 2 O-rich Mount Shasta basalts with the same ratio for the surface reservoir is consistent with the idea of igneous derivation and permanent storage of Cl and H 2 O in the surface reservoir.

H2O, CO2, CI, and gas in Plinian and ash-flow Bishop rhyolite

Infrared spectroscopic (H_2O, CO_2 and electron microprobe (CI) analyses of glass inclusions in Plinian and ash-flow quartz phenocrysts from the Bishop Tuff reveal the preeruption concentrations of volatiles in separate parts of the body of magma. There is an inverse relation between H_2O, CO_2 that can be explained (1) by closed-system, gas-saturated crystallization of parent magma to yield Plinian magma + (lost) crystals or (2) by the rise of CO_2-rich bubbles through water-rich magma. Our estimated pressures of gas-saturation range from about 1.6 kbar (Plinian) to 2.3 kbar (ash flow) and accord with geologic evidence for 3 km of magma withdrawal and caldera subsidence.

Gradients in H2O, CO2, and exsolved gas in a large‐volume silicic magma system: Interpreting the record preserved in melt inclusions from the Bishop Tuff

Infrared spectroscopic analyses of ∼140 melt inclusions in quartz phenocrysts from the zoned Bishop rhyolitic tuff demonstrate that systematic gradients in dissolved magmatic H2O and CO2 concentrations were present during preemptive crystallization of the magma body. Melt inclusions from the earliest erupted samples contain lower H2O (5.3±0.4 wt %) and CO2 (62±37 ppm) than inclusions from the middle of the eruption (5.7±0.2 wt % H2O; 120±60 ppm CO2). Melt inclusions from late erupted samples have much lower H2O (4.1±0.3 wt %) and higher and variable CO2 (150–1085 ppm). Trace element analyses of melt inclusions by ion microprobe show that inclusions within single pumice clasts from the early and middle Bishop Tuff have an inverse correlation between CO2 and incompatible elements. This pattern indicates that the magma was gas-saturated during crystallization, with CO2 partitioning into a coexisting gas phase. Quantitative modeling using H2O-CO2 solubility relations reveals a preeruptive gradient in exsolved gas, with gas contents varying from ∼1 wt % in the deeper regions of the magma body to nearly 6 wt % near the top. Dissolved Cl, B, Li, and Be in melt inclusions correlate negatively with CO2. Mass balance modeling of Cl loss to exsolving H2O-rich gas during crystallization provides strong corroborating evidence for the mass fractions of exsolved gas estimated from H2O, CO2, and trace element data. Pressures of quartz crystallization and melt inclusion entrapment calculated from inclusion H2O-CO2 data are consistent with progressive downward tapping of a zoned magma body during the eruption. Melt inclusion gas saturation pressures, magma volume estimates, and time-stratigraphic-compositional relations suggest that early erupted magma was stored at the top of a downward widening magma body. Melt inclusion data and the inferred gradients in dissolved H2O, CO2 and exsolved gas in the Bishop magma body suggest that gas saturation plays an important role in the formation and subsequent preservation of compositional gradients in silicic magma reservoirs.

Low-O18 basalts from Iceland

The volcanic rocks of Iceland are anomalous in their oxygen isotope content. Recent tholeiitic and transitional alkali basalts from Iceland range in (δO18 from 1·8 to 5δ7%. Most of the tholeiitic basalts and their phenocrysts are at least 1% lower in δO18 than unaltered basalts from other oceanic islands or oceanic ridges. The Icelandic basalts that resemble ridge basalts in δO18 also resemble them in major element chemistry. δO18 values of alkali olivine basalts are closest to those of other oceanic islands. Secondary alteration processes have lowered as well as raised the δO18 values of some Icelandic rocks, but such surface mechanisms cannot account for the distribution of oxygen isotopes in the Recent basalts of Iceland. Three mechanisms that could give rise to the low-O18 magmas are (1) exchange of oxygen between magma and low-O18 hydrothermally altered rock, (2) exchange with low-O18 meteoric water, or (3) an exceptional mantle under Iceland. None of the above models can satisfactorily account for all the observations.

Submarine metamorphism of gabbros from the Mid-Cayman Rise: Petrographic and mineralogic constraints on hydrothermal processes at slow-spreading ridges

The coarse-grained, submersible-collected rocks from the Mid-Cayman Rise in our study span a vertical section of about 700 m and range from amphibolite to gabbro. Petrographic and mineralogic studies showed that: (1) some deformation is present in all rocks; (2) the distribution of deformation is not uniform within a specimen; and (3) alteration has taken place preferentially in the deformed zones. The abundance of amphibole decreases with sample depth and argues that seawater flux into the oceanic crust decreased with depth. The compositional changes in rocks indicate that partially exchanged seawater was the metamorphic fluid that supplied Na, K, and H2O to and removed Ca from the rock. In all, about 15% of the original rock has been transformed to amphibolite in the 700 m vertical section. Our study indicates that deformation of oceanic crust is necessary for providing pathways for seawater penetration which in turn is necessary for the submarine metamorphism to occur. If deformation continues to take place during the spreading of seafloor, alteration may also continue to occur along newly formed fractures and cracks.

The before-eruption water content of some high-alumina magmas

An analysis by difference technique yields estimates of H2O in basaltic and andesitic glasses, which are sufficiently accurate (± 1.4 percent absolute) to be useful. Glass inclusions trapped in large olivine crystals from tephra-rich eruptions have 1 to 5 percent H2O. The highest H2O contents are found in basaltic inclusions in magnesium rich olivines from Mount Shasta, California. Andesitic inclusions have less H2O. It seems probable that tephra-rich high-alumina magmas evolve in a vapor saturated environment at fairly shallow depths (few kilometers). This depth appears to be less for Medicine Lake Highlands than for Mount Shasta. Vapor saturation probably inhibits the rise of magma, thus the initial vapor content of a magma may govern its stagnation level. Volatile-rich parental magmas like Mount Shasta basalt probably tend to stagnate at deeper levels, crystallize early amphibole and produce comparatively calcic differentiates.

Zoned quartz phenocrysts from the rhyolitic Bishop Tuff

Abstract Cathodoluminescence (CL) reveals growth zones in quartz phenocrysts from the rhyolitic Bishop Tuff. Melt inclusions occur in various zones and record the evolving melt composition during zonal growth. The zones form an oscillatory pattern between bright and dark CL quartz. There are three recognizable patterns of CL zoning in these crystals: (1) weakly zoned cores and bright CL rims; (2) weakly zoned cores and dark CL rims; and (3) no CL intensity difference from core to rim. Dark CL quartz generally occurs at crystal edges, contains most of the melt inclusions and is interpreted as fast-growing. Zones that occur along recognizable crystal edges (edge zones) are thicker than the same zone on adjacent faces, consistent with relatively fast growth of these zones. In each successive zone, these edge zones decrease in size toward the rim, while the zones along the crystal faces increase. Some of the melt inclusions have bright CL quartz locally associated with them. This is interpreted as the postentrapment crystallization of slow-growing quartz in the melt inclusions. Many crystals display zone discordance from the weakly zoned interiors to the rims. Most of the discordant surfaces are rational and probably are primary growth features. Pumice clasts from the southern vents are largely compositionally and texturally distinct from those from the northern vents, and this distinction is also evident in the quartz CL. The crystals that have bright CL rims are all associated with the late-erupted northern part of the Bishop Tuff. Melt inclusion compositions and CL zoning patterns suggest a common origin for early and middle-erupted quartz and the interior zones of late-erupted quartz; however, the bright CL rim on the late-erupted quartz indicates an additional stage of crystallization in late-erupted magma. Melt inclusions in individual early erupted crystals have small variations in Ba whereas inclusions in late-erupted crystals markedly increase in Ba toward the rim, which is opposite to the normal zoning of sequentially trapped melts expected during closed system crystallization differentiation. The quartz zoning features are consistent with the hypothesis of crystal settling in evolving magma that erupted late from northern vents.

Concentrations, sources, and losses of H2O, CO2, and S in Kilauean basalt

Basaltic glasses included in olivine phenocrysts from Kilauea volcano contain concentrations of H2O, CO2, and S similar to glassy Kilauean basalt dredged from the deep sea floor and greater than vesicular, subaerial Kilauean basalt. Our result contrasts with earlier reports that inclusions of basaltic glass in phenocrysts have little or no H2O and large ratios of CO2H2O. Our analysed inclusions of glass are larger than 100 micrometers thick and similar in chemical composition to the host glass surrounding the olivine crystals indicating that the trapped melts are representative of the bulk liquid from which the crystals grew. Crystallization of about 2–8% of olivine from the melts after they were trapped is indicated by slight departures from the experimentally established equilibrium distribution of Mg and Fe between olivine and liquid. The measured concentrations of CO2 correspond to phenocryst crystallization pressures of about 1.3 kbar for a subaerial basalt and about 5 kbar for a submarine basalt, consistent with geophysical models of Kilauea volcano. The compositions of volcanic gas predicted from our analyses are consistent with restored compositions of actual Kilauean gases. The rate of sulfur emission predicted from our analyses is greater than the sulfur dioxide emission rate observed during repose, but probably consistent with total degassing including eruptive episodes. The concentrations of H2O, K2O, Cl, and P in parental Kilauean basalt can be derived from upper mantle phlogopitic mica, pargasitic amphibole and apatite with compositions close to those of natural primary minerals in ultramafic xenoliths from continental kimberlites, or solely from apatite and phlogopitic mica with H2OK2O near 0.47 ± 0.03, slightly higher than the range of values reported. The amounts of phlogopitic mica and pargasitic amphibole contributing volatiles to Kilauean tholeiite is about 10 percent by mass of the parental liquid, or about 5% if the source does not include amphibole. In view of an estimated 20% of partial melting of mantle source rock to produce Kilauean tholeiites, there may be about 2 weight percent of mica plus amphibole in part of the mantle beneath Kilauea, or about 1 weight percent of phlogopitic mica if amphibole is absent.

Bubble coalescence in basalt flows: comparison of a numerical model with natural examples

Gas accumulation in magma may be aided by coalescence of bubbles because large coalesced bubbles rise faster than small bubbles. The observed size distribution of gas bubbles (vesicles) in lava flows supports the concept of post-eruptive coalescence. A numerical model predicts the effects of rise and coalescence consistent with observed features. The model uses given values for flow thickness, viscosity, volume percentage of gas bubbles, and an initial size distribution of bubbles together with a gravitational collection kernel to numerically integrate the stochastic collection equation and thereby compute a new size spectrum of bubbles after each time increment of conductive cooling of the flow. Bubbles rise and coalesce within a fluid interior sandwiched between fronts of solidification that advance inward with time from top and bottom. Bubbles that are overtaken by the solidification fronts cease to migrate. The model predicts the formation of upper and lower vesicle-rich zones separated by a vesicle-poor interior. The upper zone is broader, more vesicular, and has larger bubbles than the lower zone. Basaltic lava flows in northern California exhibit the predicted zonation of vesicularity and size distribution of vesicles as determined by an impregnation technique. In particular, the size distribution at the tops and bottoms of flows is essentially the same as the initial distribution, reflecting the rapid initial solidification at the bases and tops of the flows. Many large vesicles are present in the upper vesicular zones, consistent with expected formation as a result of bubble coalescence during solidification of the lava flows. Both the rocks and model show a bimodal or trimodal size distribution for the upper vesicular zone. This polymodality is explained by preferential coalescence of larger bubbles with subequal sizes. Vesicularity and vesicle size distribution are sensitive to atmospheric pressure because bubbles expand as they decompress during rise through the flow. The ratio of vesicularity in the upper to that in the lower part of a flow therefore depends not only on bubble rise and coalescence, but also on flow thickness and atmospheric pressure. Application of simple theory to the natural basalts suggests solidification of the basalts at 1.0±0.2 atm, consistent with the present atmospheric pressure. Paleobathymetry and paleoaltimetry are possible in view of the sensitivity of vesicle size distributions to atmospheric pressure. Thus, vesicular lava flows can be used to crudely estimate ancient elevations and/or sea level air pressure.