Storage and interaction of compositionally heterogeneous magmas from the 1986 eruption of Augustine Volcano, Alaska

Diana C. Roman; Katharine V. Cashman; Cynthia A. Gardner; Paul J. Wallace; John J. Donovan

doi:10.1007/s00445-005-0003-z

Home/Archive/USGS

USGS · 70030827

Storage and interaction of compositionally heterogeneous magmas from the 1986 eruption of Augustine Volcano, Alaska

Diana C. Roman·Katharine V. Cashman·Cynthia A. Gardner·Paul J. Wallace·John J. Donovan

View original source ↗Export metadata ↓

Abstract

Compositional heterogeneity (56–64 wt% SiO 2 whole-rock) in samples of tephra and lava from the 1986 eruption of Augustine Volcano, Alaska, raises questions about the physical nature of magma storage and interaction beneath this young and frequently active volcano. To determine conditions of magma storage and evolutionary histories of compositionally distinct magmas, we investigate physical and chemical characteristics of andesitic and dacitic magmas feeding the 1986 eruption. We calculate equilibrium temperatures and oxygen fugacities from Fe-Ti oxide compositions and find a continuous range in temperature from 877 to 947°C and high oxygen fugacities (ΔNNO=1–2) for all magmas. Melt inclusions in pyroxene phenocrysts analyzed by Fourier-transform infrared spectroscopy and electron probe microanalysis are dacitic to rhyolitic and have water contents ranging from <1 to ∼7 wt%. Matrix glass compositions are rhyolitic and remarkably similar (∼75.9–76.6 wt% SiO 2 ) in all samples. All samples have ∼25% phenocrysts, but lower-silica samples have much higher microlite contents than higher-silica samples. Continuous ranges in temperature and whole-rock composition, as well as linear trends in Harker diagrams and disequilibrium mineral textures, indicate that the 1986 magmas are the product of mixing between dacitic magma and a hotter, more mafic magma. The dacitic endmember is probably residual magma from the previous (1976) eruption of Augustine, and we interpret the mafic endmember to have been intruded from depth. Mixing appears to have continued as magmas ascended towards the vent. We suggest that the physical structure of the magma storage system beneath Augustine contributed to the sustained compositional heterogeneity of this eruption, which is best explained by magma storage and interaction in a vertically extensive system of interconnected dikes rather than a single coherent magma chamber and/or conduit. The typically short repose period (∼10 years) between Augustine's recent eruptive pulses may also inhibit homogenization, as short repose periods and chemically heterogeneous magmas are observed at several volcanoes in the Cook Inlet region of Alaska.

Explore related subjects

Bulletin of Volcanology Article

Source-reported bounding extent: 59.31076795603884° to 59.428315784042574° latitude; -153.60260009765625° to -153.31695556640625° longitude. This indicates report coverage, not an exact sampling location. View area on OpenStreetMap.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Diana C. Roman, Katharine V. Cashman, Cynthia A. Gardner, Paul J. Wallace, John J. Donovan. 2005-10-26. Storage and interaction of compositionally heterogeneous magmas from the 1986 eruption of Augustine Volcano, Alaska. https://doi.org/10.1007/s00445-005-0003-z

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Geophysical insights into Kīlauea’s eruptive reawakening after its 2018 eruption and caldera collapse

A little more than two years after the destructive 2018 lower East Rift Zone eruption and summit collapse ended, renewed volcanic activity began at Kīlauea volcano, Hawaiʻi. During 2020–2025, activity at the volcano included six eruptions in the summit caldera, one in the Southwest Rift Zone, and one in the middle East Rift Zone, as well as multiple non-eruptive intrusions in all three locations. We categorize the volcano’s activity post-2018 into four phases based on eruption locations and style, as well as patterns of magma storage and migration inferred from geophysical data. In some cases, these observations illuminated parts of the volcano’s magmatic system for the first time since the advent of modern geophysical monitoring—particularly deformation datasets that include tilt, continuous GNSS, and InSAR. Analyses of these geodetic data and complementary seismic, geological, and geochemical observations provide an opportunity to explore summit and rift zone connectivity, rift-zone magma storage conditions, the variable geophysical manifestations of eruption onsets, magma-fault interactions, and other long-standing questions associated with eruptions and intrusions at Kīlauea.

Hawaii↗

USGS2026

An extensive and short-lived episode of mid-Holocene fayalite- bearing high-silica rhyolite eruptions at Mono Craters, California

Having last erupted ~ 700 years ago, the Mono Craters volcanic chain is an active system with a history of explosive and effusive high-silica rhyolite eruptions. Approximately ~ 30 lava domes and flows range from ~ 42 to ~ 0.7 ka, form a ~ 15 km curvilinear trend, and change mineralogy (biotite-, orthopyroxene-, or fayalite-bearing) and crystallinity through time. However, the eruptive chronology of this system is incompletely understood. We report new 238 U- 230 Th isochron ages for nine fayalite-bearing domes from paired zircon and allanite crystal surfaces that reveal a major shift in eruptive frequency and volume over the history of the chain, with extrusion of > 0.45 km 3 of high-silica rhyolite lava between 7.1 and 5.2 ka. This contrasts with older biotite- (~ 42–20 ka) and orthopyroxene-bearing (~ 12–9 ka) domes, each of which erupted smaller volumes over a more protracted period. Whole-rock compositions suggest these older domes tapped separate domains within a shared magmatic system because the former are higher in Sr and Ba whereas the latter are lower in SiO 2 , higher in Zr, and more enriched in light rare-earth elements. Titanomagnetite chemistry and zircon saturation temperatures also reflect changing magma compositions and conditions associated with these domes. Fayalite-bearing domes, in contrast, have uniform whole-rock and titanomagnetite compositions which support rapid extrusion of magma over an interval short enough to prevent Fe-Ti oxides from recording any changes in magmatic conditions. The youngest lavas were also produced in a short-lived episode ~ 700 years ago, showing that multivent, geologically brief eruptive episodes occur repeatedly at Mono Craters.

California↗

USGS2026

Satellite observations, plume dynamics and volcanic lightning from the 2023 eruption of Shishaldin Volcano, Alaska

The 2023 eruption of Shishaldin Volcano, Alaska, provided an unusually well-monitored example of a “dry” (non-hydromagmatic) explosive basaltic eruption with sustained, buoyant plumes, and pyroclastic density currents. Here, we examine the 13 basaltic paroxysms that occurred between July and November 2023 using geostationary satellite observations, plume modeling, and volcanic lightning detection to characterize mass eruption rates and plume dynamics. We show that seven volcanic plumes reached 10–14 km above sea level, sustained for minutes to tens of minutes. Translating plume heights into mass eruption rates, using both empirical scaling and one-dimensional plume modeling, gives peak rates of 10 4 –10 7 kg·s −1 and a total erupted volume of ~ 0.05 km 3 (dense rock equivalent) across the eruption sequence. Despite the moderate intensity of this eruption, automated satellite detection algorithms generally failed to detect volcanic ash. We attribute this persistent challenge to the fines-poor grain size distribution of the eruption. Volcanic lightning provides an additional window into the eruption dynamics. A total of 491 volcanic lightning flashes were detected, concentrated during seven of the most energetic phases. The timing of this electrical activity suggests that lightning only became detectable by the global lightning monitoring network when plumes ascended to atmospheric levels favorable for ice formation. Based on the available data, we infer that the low fine ash content of the plumes, combined with minimal water and ice, led to the overall modest rates of lightning (1–13 flashes per minute) compared to “wet” eruptions of similar intensity. These results show that combining satellite observations, plume modeling, and lightning detection provides a valuable timeline of eruptive intensity, even for “dry” basaltic eruptions in remote locations.

Alaska↗