Record Details
Magma dynamics and evolution in continental arcs : insights from the Central Andes
ScholarsArchive at Oregon State University
| Field | Value |
|---|---|
| Title | Magma dynamics and evolution in continental arcs : insights from the Central Andes |
| Names |
Grocke, Stephanie B.
(creator) de Silva, Shanaka L. (advisor) |
| Date Issued | 2014-05-29 (iso8601) |
| Note | Graduation date: 2014 |
| Abstract | Constraining the magma evolution and dynamics that lead to the eruption of large volume continental arc systems is fundamental to our understanding of continental crust formation. An investigation into the magmagenesis that results in the formation of the Central Volcanic Zone (CVZ) in the Andes of South America, situated atop overthickened continental crust (<80 km thick), provides insights into large volume silicic magma reservoirs and how they evolve prior to their potentially catastrophic explosive eruption on the Earth’s surface. A focused case study of the Cerro Guacha Caldera Complex (CGCC), a nested volcanic system in the Altiplano-Puna Volcanic Complex of SW Bolivia, puts constraints on the progressive stages of development of the magmatic underpinnings of the caldera complex. Whole rock data, in conjunction with matrix glass, mineral compositions and melt inclusions, are used to infer processes that gave rise to the formation of the Guacha II Caldera, the younger of two main collapse features, formed from the supereruption of the Tara Ignimbrite (>800 km³ DRE) at 3.49 ± 0.01 Ma. The eruptive history of the Guacha II Caldera from pre-caldera to post-caldera is fully represented, allowing magma dynamics associated with a complete caldera cycle, from pre-climactic (catastrophic caldera-forming) magma accumulation through to post-climactic effusions that are part of the resurgent history of the caldera, to be examined. Analysis of the high-K, calcalkaline suite of andesite to high Si-rhyolite Tara pyroclastic deposits provides insights into the storage conditions and magma dynamics leading up to a supervolcanic eruption. The Tara eruptive products define a liquid line of descent from the basal andesite lava (62 wt % SiO₂) to the high-silica rhyolite post-collapse Chajnantor Dome lava (78 wt.% SiO₂), with major and trace element trends consistent with fractionation of quartz, plagioclase, orthopyroxene, hornblende, sanidine, biotite, and Fe-Ti oxides. Isotope ratios span a significant range in ⁸⁷Sr/⁸⁶Sr (0.709 to 0.713) and a relatively narrow range in ¹⁴³Nd/¹⁴⁴Nd (0.512179 to 0.512297) and δ¹⁸O[subscript (qtz)] (+8.68 to +8.43‰). These data require AFC processes to explain both the isotope and trace element compositions in the Tara magmas. Geothermobarometry reveals pre-eruptive temperatures (~800 - 950 °C), pressures (~200 MPa), and H₂O contents (~5 wt%) that suggest storage of a large-volume rhyodacite magma reservoir between 5 and 9 km depth in the upper crust. Analyses of quartz-hosted melt inclusions from pumices in the climactic plinian and ignimbrite phase of eruption reveal that pre-eruptive H₂O contents in the plinian pumice overlap with those in the ignimbrite pumice (2.2 to 6.0 and 2.1 to 5.4 wt.% H₂O, respectively). The ignimbrite magma, however, contains higher CO₂ (<630 versus <300 ppm) suggesting a vertically arranged pre-climactic magma column from 7.5 km to 4 km depth, which is in agreement with the geothermobarometry from mineral analysis. Andesite recharge can account for significant compositional heterogeneities identified in the Tara magmas. Variations in trace element concentrations from melt inclusions from each of the three eruptive phases also suggest that the large-volume pre-climactic reservoir, as well as the remnant magma, contained significant chemical heterogeneities due to variations in the degree of crystal fractionation and the influence of recharge magma. Based on Ti diffusion in quartz, we estimate that the recharge event occurred within <100 years of eruption and may signal the onset of the pressurization of the system that led to eruption. The following petrogenetic model is proposed for the Tara magmatic system: andesite recharge magma ascended into a large-volume rhyodacite magma reservoir in the upper crust, where it cooled and crystallized to form a small volume of rhyolite, crystal-poor, residual melt. Andesite erupted onto what is now the Guacha II caldera-floor, and was followed by an explosive plinian eruption that produced a rhyolite pumice fall deposit outside the caldera to the east and south. Based on melt inclusions, the plinian eruption tapped a less-evolved, crystal poor (~10 vol.%), more roofward liquid that syneruptively mixed with the highly crystalline (<30 vol.%), large volume rhyodacite magma reservoir. Caldera collapse was accompanied by an explosive eruption of the large-volume rhyodacite magma reservoir and subsequent column collapse, resulting in a pyroclastic flow that produced the >800 km³ Tara ignimbrite. Extrusion of three post-caldera domes followed caldera collapse. Subtle differences between these domes invoke separate coexisting "pods" of magma that evolved independently from one another from the remnant magma that is represented by the ignimbrite. Analyses of quartz-hosted melt inclusions from the post-collapse Chajnantor Dome suggest that the highly-differentiated remnant magma contained no detectable CO₂ and low H₂O contents, representative of a degassed magma. The eruptive transition during the climactic eruption was not controlled by the volatile budget of the melt but more likely by external factors such as vent geometry and conduit evolution. Post climactic effusive volcanism reflects the degassed nature of the remnant magma. Using a suite of volcanic rocks from the CVZ, we quantify the effect of assimilation of continental crust on magmatic oxygen fugacity (ƒO₂). We use several proxies to estimate the ƒO₂ recorded by lavas, pumice and scoria: 1) whole rock Fe³⁺/ΣFe ratios, 2) Fe³⁺/ΣFe ratios in quartz-hosted melt inclusions, and 3) Fe-Ti oxide oxybarometry. Samples span a range of crustal contribution, as indicated by their radiogenic isotope compositions (⁸⁷Sr/⁸⁶Sr = 0.705-0.713), and cover the full suite of magma compositions erupted during the Neogene history of the arc (52 - 74 wt.% SiO₂). Some samples show excellent agreement across multiple ƒO₂ proxies. In other cases, where pumices show evidence of alteration in hand-sample for example, the ƒO₂ recorded by bulk Fe³⁺/ΣFe ratios is two orders of magnitude more oxidized than corresponding ratios from melt inclusions or Fe-Ti oxides. This suggests modification of whole rock Fe³⁺/ΣFe ratios, but not melt inclusion Fe³⁺/ΣFe ratios, by syn- or post-eruptive processes and that care must be taken when relying on bulk techniques to determine magmatic ƒO₂. We cannot resolve any oxidation due to crystal fractionation in our sample suite. Crustal assimilation, however, can oxidize arc magmas. The increase in ƒO₂ due to crustal assimilation reaches, but does not exceed, ~1 log unit - even in the Andes, where crustal assimilation is extreme. |
| Genre | Thesis/Dissertation |
| Access Condition | http://creativecommons.org/licenses/by-nc-nd/3.0/us/ |
| Topic | Central Andes |
| Identifier | http://hdl.handle.net/1957/50609 |
