Microstructural Constraints on Magmatic Mushes under Kīlauea Volcano, Hawai'i

Distorted olivines of enigmatic origin are ubiquitous in erupted products from a wide range of volcanic systems (e.g., Hawai'i, Iceland, Andean Southern Volcanic Zone). At Kīlauea volcano, distorted olivines are commonly attributed to ductile creep within dunitic bodies located around the central conduit, or within the deep rift zones (~5–9 km depth). However, a recent suggestion that lattice distortions are produced by an early phase of branching dendritic growth, followed by textural ripening and the merging of misoriented crystal buds, has gained considerable traction.

A quantitative examination of the microstructures in distorted olivines by electron backscatter diffraction (EBSD) reveals striking similarities to microstructures observed in deformed mantle peridotites, but significant differences to the crystallographic signatures of dendritic growth. This suggests that lattice distortions record the application of differential stresses at high temperatures within the magmatic plumbing system, rather than rapid crystal growth. Previous petrological work has suggested that differential stresses are produced by ductile creep within Kīlauea’s deep rift zones. Crucially, this has fuelled suggestions that significant quantities of magma must travel along these rift zones in order to acquire distorted olivines, despite the paucity of geophysical evidence for these magma transport paths. In contrast, we show that the spatial distribution of eruptions containing distorted olivines is consistent with their derivation from the main magma storage reservoir. This model not only aligns petrological and geophysical observations at Kīlauea, but also accounts for the occurrence of distorted olivines in a wide variety of basaltic systems worldwide (which lack deep rift zones).

Application of piezometers developed for mantle peridotites reveals that distorted olivines have experienced differential stresses of ~3–12 MPa. Assuming that mush piles behave as granular materials, and form force chains, these stresses can be generated within cumulate piles of ~180–720 m. Based on available constraints on the magma supply rate and the geometry of Kīlauea’s summit reservoir, these thicknesses accumulate in a few centuries (consistent with residence times inferred from melt inclusion records).

Overall, we demonstrate that microstructural investigations of erupted olivine crystals by EBSD generates rich datasets which provide quantitative insights into crystal storage within mush piles. Under the increasingly prevalent view that crustal magmatic systems are mush-dominated, constraining the geometry and dynamics of crystal storage regions is crucial to further our understanding of magmatic plumbing systems. The presence of distorted olivines in many different volcanic settings highlights the global applicability of the methods developed in this study. Furthermore, assessments of deformation conditions using EBSD need not be restricted to olivine-bearing lavas. Microstructural fabrics types in natural and experimental samples have been established for a wide variety of igneous phases (e.g. diopside, plagioclase, hornblende), so similar approaches may be utilized in more evolved volcanic systems.