A critical parameter governing the style and vigor of volcanic eruptions is the rate at which magma ascends and degases from the storage region to the surface. The magma ascent rate, in turn, controls the time available for loss of volatile species from the magma, melts and crystals. This project aims to assess directly the volatile loss from magmas over timescales of minutes to months, as measured in volcanic crystals and glass, and from this provide estimates of magma ascent rates critical to a quantitative understanding eruptive processes. The focus here is on explosive, basaltic to basaltic andesite eruptions, for which very little is yet known about the magma ascent and degassing processes leading to eruption. The approach is motivated by new developments in analytical techniques that enable measurement of volatile species (including water and carbon dioxide) at fine spatial resolution.
Three different kinds of measurements will interrogate volcanic systems at different timescales: 1) volatile concentration profiles in melt embayments within crystals (by nanoSIMS ionmicroprobe), which may reflect diffusive re-equilibration with the exterior, degassing magma over the timescale of minutes (water) to hours (sulfur). 2) water contents within nominally anhydrous phenocrysts, which may preserve diffusive exchange over the timescale of hours (olivine) to months (clinopyroxene). 3) volatile contents in olivine-hosted melt inclusions, which may reflect timescales and processes intermediate between those that affect melt embayments and crystals, above.
Well-studied Central American volcanoes (Cerro Negro, Irazu, Arenal) will enable comparisons of volatile loss and ascent rates for different eruptive styles. The project has implications for the understanding of volcanic hazards associated with explosive volcanism, will support the training and development of two graduate students, and provide lecture and seminar material for a course at Columbia University (Frontiers of Science) that engages 500 freshmen each year.
One of the primary factors that is thought to relate to the vigor of volcanic eruption is the rate at which magma rises in the conduit prior to eruption, as this affects how bubbles form, merge and migrate with respect to the magmatic liquid. Despite the importance of ascent rate in physical models for eruption, there are very few such estimates for any eruption. This project aimed to develop novel magma ascent clocks using the chemical composition of volcanic crystals. The clocks involve gradients in volatile species, primarily H2O, CO2 and S, and their relatively rapid diffusivity in crystals and melt at magmatic temperatures (> 1000°C). Three different types of materials were used: 1) melt tubes, or embayments, in olivine crystals, 2) clinopyroxene crystals, and 3) melt inclusions inside olivine crystals (see images). All samples derived from the same day, October 17, of the 1974 eruption of Volcan Fuego in Guatemala, an explosive (Volcano Explosivity Index = 4) eruption of basalt to basaltic andesite (see photo). Ion microprobe, including nanoSIMS, measurements reveal gradients of decreasing H2O toward the outer rim of all materials, reflecting attempt to equilibrate with the degassing host magma during ascent and/or cooling. Our results bears on the fidelity of olivine and pyroxene to retain water during cooling and ascent. Crystals that cool slowly in volcanic bombs (> 6 cm diameter) and lava flows have lost significant H2O after eruption, and can't be used to infer the pre-eruptive volatile content of the magma. Embayments in well-quenched ash samples record gradients in H2O, CO2 and S, which owing to each species different diffusivity and solubility, provide overlapping constraints on decompression rates in the conduit, on the order of 0.4 MPa/s or 15 m/s. This is the equivalent of ascent from the magma chamber (~ 6 miles depth) to the surface in just 10 minutes! Such rapid run-ups to eruptions challenge most existing techniques for volcano monitoring and informing the public. The tools developed here can now be used to test whether less explosive eruptions involve longer rise times. Results of this work have been disseminated to the science community in the form of conference presentations and publications, and to the public through lectures and online videos, such as: www.nasonline.org/news-and-multimedia/video-gallery/151st-annual-meeting/research-briefings/terry-a-plank.html This project largely supported the Ph.D. dissertation research of a graduate student at Columbia University, Alexander Lloyd, who is now a postdoctoral Columbia University Science Fellow, teaching the freshman core curriculum class, Frontiers of Science.