skip to main content

Title: Tephra glass chemistry provides storage and discharge details of five magma reservoirs which fed the 75 ka Youngest Toba Tuff eruption, northern Sumatra

The Youngest Toba Tuff contains five distinct glass populations, identified from Ba, Sr and Y compositions, termed PI (lowest Ba) – PV (highest Ba), representing five compositionally distinct pre‐eruptive magma batches that fed the eruption. The PI–PV compositions display systematic changes, with higher FeO, CaO, MgO, TiO2and lower incompatible element concentrations in the low‐SiO2PIV/PV, than the high‐SiO2PI–PIII compositions. Glass shard abundances indicate PIV and PV were the least voluminous magma batches, and PI and PIII the most voluminous. Pressure estimates using rhyolite‐MELTS indicate PV magma equilibrated at ~6 km, and PI magma at ~3.8 km. Glass population proportions in distal tephra and proximal (caldera‐wall) material describe an eruption which commenced by emptying the deepest PIV and PV reservoirs, this being preferentially deposited in a narrow band across southern India (possibly due to jet‐stream and/or plinian eruption transport), and as abundant pumice clasts in the lowermost proximal ignimbrites. Later, shallower magma reservoirs erupted, with PI being the most abundant as the eruption ended, sourcing the majority of distal ash from co‐ignimbrite clouds (PI‐ and PIII‐dominant), where associated ignimbrites isolated earlier (PIV‐ and PV‐rich) deposits. This study shows how analysis of tephra glass compositional data can yield pre‐eruption magma volume estimates, and enable aspects of magma storage conditions and eruption dynamics to be described.

more » « less
Author(s) / Creator(s):
 ;  ;  ;  ;  
Publisher / Repository:
Wiley Blackwell (John Wiley & Sons)
Date Published:
Journal Name:
Journal of Quaternary Science
Page Range / eLocation ID:
p. 256-271
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. null (Ed.)
    Abstract Volcán Quizapu, Chile, is an under-monitored volcano that was the site of two historical eruptions: an effusive eruption in 1846–1847 and a Plinian eruption in 1932, both of which discharged ∼5 km3 (dense rock equivalent) of lava and/or tephra. The majority of material erupted in both cases is trachydacite, nearly identical for each event. We present H2O-saturated, phase equilibrium experiments on this end-member dacite magma, using a pumice sample from the 1932 eruption as the main starting material. At an oxygen fugacity (fO2) of ∼NNO + 0·2 (where NNO is the nickel–nickel oxide buffer), the phase assemblage of An25–30 plagioclase + amphibole + orthopyroxene, without biotite, is stable at 865 ± 10 °C and 110 ± 20 MPa H2O pressure (PH2O), corresponding to ∼4 km depth. At these conditions, experiments also reproduce the quenched glass composition of the starting pumice. At slightly higher PH2O and below 860 °C, biotite joins the equilibrium assemblage. Because biotite is not part of the observed Quizapu phase assemblage, its presence places an upper limit on PH2O. At the determined storage PH2O of ∼110 MPa, H2O undersaturation of the magma with XH2Ofluid = 0·87 would align Ptotal to mineral-based geobarometry estimates of ∼130 MPa. However, XH2Ofluid < 1 is not required to reproduce the Quizapu dacite phase assemblage and compositions. A second suite of experiments at lower fO2 shows that the stability fields of the hydrous silicates (amphibole and biotite) are significantly restricted at NNO – 2 relative to NNO + 0·2. Additional observations of Quizapu lava and pumice samples support the existing hypothesis that rapid pre-eruptive heating drove the effusive 1846–1847 eruption, with important refinements. We demonstrate that microlites in the end-member dacite lavas are consistent with in situ crystallization (during ascent), rather than transfer from an andesite. In one end-member dacite lava, newly identified reverse zoning in orthopyroxene and incipient destabilization of amphibole are consistent with small degrees of heating. Our work articulates a clear direction for future Quizapu studies, which are warranted given the active nature of the Cerro Azul–Descabezado Grande volcanic axis. 
    more » « less
  2. Abstract Despite their global societal importance, the volumes of large-scale volcanic eruptions remain poorly constrained. Here, we integrate seismic reflection and P-wave tomography datasets with computed tomography-derived sedimentological analyses to estimate the volume of the iconic Minoan eruption. Our results reveal a total dense-rock equivalent eruption volume of 34.5 ± 6.8 km³, which encompasses 21.4 ± 3.6 km³ of tephra fall deposits, 6.9 ± 2 km³ of ignimbrites, and 6.1 ± 1.2 km³ of intra-caldera deposits. 2.8 ± 1.5 km³ of the total material consists of lithics. These volume estimates are in agreement with an independent caldera collapse reconstruction (33.1 ± 1.2 km³). Our results show that the Plinian phase contributed most to the distal tephra fall, and that the pyroclastic flow volume is significantly smaller than previously assumed. This benchmark reconstruction demonstrates that complementary geophysical and sedimentological datasets are required for reliable eruption volume estimates, which are necessary for regional and global volcanic hazard assessments. 
    more » « less
  3. Abstract The Lower Jurassic Ferrar Large Igneous Province consists predominantly of intrusive rocks, which crop out over a distance of 3500 km. In comparison, extrusive rocks are more restricted geographically. Geochemically, the province is divided into the Mount Fazio Chemical Type, forming more than 99% of the exposed province, and the Scarab Peak Chemical Type, which in the Ross Sea sector is restricted to the uppermost lava. The former exhibits a range of compositions (SiO 2 = 52–59%; MgO = 9.2–2.6%; Zr = 60–175 ppm; Sr i = 0.7081–0.7138; ε Nd = −6.0 to −3.8), whereas the latter has a restricted composition (SiO 2 = c. 58%; MgO = c. 2.3%; Zr = c. 230 ppm; Sr i = 0.7090–0.7097; ε Nd = −4.4 to −4.1). Both chemical types are characterized by enriched initial isotope compositions of neodymium and strontium, low abundances of high field strength elements, and crust-like trace element patterns. The most basic rocks, olivine-bearing dolerites, indicate that these geochemical characteristics were inherited from a mantle source modified by subduction processes, possibly the incorporation of sediment. In one model, magmas were derived from a linear source having multiple sites of generation each of which evolved to yield, in sum, the province-wide coherent geochemistry. The preferred interpretation is that the remarkably coherent geochemistry and short duration of emplacement demonstrate derivation from a single source inferred to have been located in the proto-Weddell Sea region. The spatial variation in geochemical characteristics of the lavas suggests distinct magma batches erupted at the surface, whereas no clear geographical pattern is evident for intrusive rocks. 
    more » « less
  4. null (Ed.)
    Abstract The Okataina Volcanic Centre (OVC), located in the Taupo Volcanic Zone, New Zealand, is a dominantly rhyolitic magmatic system in an arc setting, where eruptions are thought to be driven by mafic recharge. Here, Sr–Pb isotopes, and compositional and textural variations in plagioclase phenocrysts from 10 rhyolitic deposits (two caldera, one immediately post-caldera, four intra-caldera, and three extra-caldera) are used to investigate the OVC magmatic system and identify the sources and assimilants within this diverse mush zone. Plagioclase interiors exhibit normal and reverse zoning, and are commonly in disequilibrium with their accompanying glass, melt inclusions, and whole-rock compositions. This indicates that the crystals nucleated in melts that differed from their carrier magma. In contrast, the outermost rims of crystals exhibit normal zoning that is compositionally consistent with growth in cooling and fractionating melts just prior to eruption. At the intra-crystal scale, the total suite of 87Sr/86Sr ratios are highly variable (0·7042–0·7065 ± 0·0004 average 2SE); however, the majority (95 %) of the crystals are internally homogeneous within error. At whole-crystal scale (where better precision is obtained), 87Sr/86Sr ratios are much more homogeneous (0·70512–0·70543 ± 0·00001 average 2SE) and overlap with their host whole-rock Sr isotopic ratios. Whole-crystal Pb isotopic ratios also largely overlap with whole-rock Pb ratios. The plagioclase and whole-rock isotopic compositions indicate significant crustal assimilation (≥20 %) of Torlesse-like metasediments (local basement rock) by a depleted mid-ocean ridge mantle magma source, and Pb isotopes require variable fluid-dominant subduction flux. The new data support previous petrogenetic models for OVC magmas that require crystal growth in compositionally and thermally distinct magmas within a complex of disconnected melt-and-mush reservoirs. These reservoirs were rejuvenated by underplating basaltic magmas that serve as an eruption trigger. However, the outermost rims of the plagioclase imply that interaction between silicic melts and eruption-triggering mafic influx is largely limited to heat and volatile transfer, and results in rapid mobilization and syn-eruption mixing of rhyolitic melts. Finally, relatively uniform isotopic compositions of plagioclase indicate balanced contributions from the crust and mantle over the lifespan of the OVC magmatic system. 
    more » « less
  5. Abstract

    Continental flood basalts intruded and erupted millions of km3of magma over ∼1–5 Ma. Previous work proposed the presence of large (–106 km3) crustal magma reservoirs to feed these eruptions. However, in Paper I, we illustrated that this model is inconsistent with observations, by combining eruptive rate constraints with geochemical and geophysical observations from the Deccan Traps and other Continental flood basalt provinces (CFBs). Here, we use a new mechanical magma reservoir model to calculate the variation of eruptive fluxes (km3/year) and volumes for different magmatic architectures. We find that a single magma reservoir cannot explain the eruptive rate and duration constraints for CFBs. Using a 1D thermal model and characteristic timescales for magma reservoirs, we conclude that CFB eruptions were likely fed by a number of interconnected small‐medium (∼102–103 km3) magma reservoirs. It is unlikely that each individual magma reservoir participated in every eruption, thus permitting the occasional formation of large xenocrysts (e.g., megacrystic plagioclase). This magmatic architecture permits (a) large volume eruptive episodes with 10–100s of years duration, and (b) relatively short time‐periods separating eruptive episodes (1000s of years) since multiple mechanisms can trigger eruptions (via magma recharge or volatile exsolution, as opposed to long term (105–106 year) accumulation of buoyancy overpressure), and (c) lack of large upper‐crustal intrusive bodies in various geophysical datasets. Our new proposed magmatic architecture has significant implications for the tempo of CFB volatile release (CO2and SO2), potentially helping explain the pre‐K‐Pg warming associated with Deccan Traps.

    more » « less