09:00 〜 09:15
[SCG48-01] Transcrustal magmatic mushes - implications for igneous differentiation and the heights of volcano edifices
★Invited Papers
キーワード：Volcanoes, Crustal structure, Chemical differentiation
Chemical differentiation of igneous rocks in Earth’s crust is viewed traditionally through the prism of melt-rich magma chambers, wherein physical separation of crystals drives change in melt composition. Such crystal fractionation is central to interpretation of igneous geochemistry and crustal structure. Given the centrality of magma chambers to the interpretation of igneous geology it is surprising that geophysical studies of the crust beneath volcanoes find extensive regions of low melt-fraction crustal mush, but no suitably large chambers. As a result petrologists are moving away from the long-standing magma chamber paradigm to a concept of vertically-extensive, transcrustal magmatic systems that for most of their lifetimes exist in a partially molten or mushy state Transcrustal magmatic mushes (or hot zones) have a number of important implications for magma chemistry, the structure of the crust and the heights of volcanic edifices.
Chemical differentiation in hot zones occurs primarily by percolative reactive flow whereby melt chemistry is controlled (or ‘buffered’) by the mineralogy of the volumetrically-dominant crystalline mush through which melts pass, and the associated mineral-melt chemical reactions. Erupted mush fragments (plutonic xenoliths) testify texturally and chemically to such processes. Amphibole, formed by reaction between hydrous melts and clinopyroxene + plagioclase, is a common reaction product in xenoliths from arc volcanoes, accounting for the widespread evidence of ‘cryptic’ amphibole fractionation in arc magmas lacking amphibole phenocrysts.
Melt chemistry becomes fixed at the point at which magma segregates from the hot zone and ascends; this point is termed the ‘launch depth’. Chemically speaking, launch depths represent discrete points in P-T-X space on a multiple-saturation surface that is defined by the hot zone mush mineralogy. Thus, melts do not follow conventional liquid lines of descent (LLD), although LLD and multiple-saturation surfaces are often hard to distinguish chemically. Some magmas ascend from their launch depths to erupt at the surface, whereas others stall en route to form plutons. In both cases, little chemical differentiation occurs following launch, due to the inefficiency of crystal-liquid separation during ascent. However, the ascent path and rate do exercise important controls on magma textures.
Magma launch depth also exercises a first-order control on the heights of volcanic edifices above the surrounding base level. This is due to the density difference between buoyant magma and surrounding crust integrated over the vertical distance from the launch depth to the surface. At 30 well-characterised, long-lived volcanoes from a wide range of locations and tectonic settings (including Japan), the calculated volcanic edifice height points to a launch depth of 6 to 50 km. At each volcano studied the launch depth matches petrological (thermobarometry) and geophysical (seismic tomography, MT) constraints on the depth to the upper surface of the underlying hot zone. Although local factors, such as regional stresses or faults, facilitate magma ascent, over the approximately Myr lifetime of a single volcano, the equilibrium (isostatic) edifice height appears to be maintained. Modification of the volcanic edifice and/or changes in magma launch depth and chemistry perturb the equilibrium volcano height, and may lead to renewed growth. Thus, periods of increased edifice construction may be ascribed to episodes of erosion, deglaciation or sector collapse, or changes to hot zone architecture.
It is observed that, in general, more silicic melts are launched from shallower depths than more mafic melts, consistent with differentiation via percolative flow and the overall layered structure of the continental crust. The greatest (sub-Moho) launch depths are found at ocean islands, where mantle xenoliths have been recorded.
Chemical differentiation in hot zones occurs primarily by percolative reactive flow whereby melt chemistry is controlled (or ‘buffered’) by the mineralogy of the volumetrically-dominant crystalline mush through which melts pass, and the associated mineral-melt chemical reactions. Erupted mush fragments (plutonic xenoliths) testify texturally and chemically to such processes. Amphibole, formed by reaction between hydrous melts and clinopyroxene + plagioclase, is a common reaction product in xenoliths from arc volcanoes, accounting for the widespread evidence of ‘cryptic’ amphibole fractionation in arc magmas lacking amphibole phenocrysts.
Melt chemistry becomes fixed at the point at which magma segregates from the hot zone and ascends; this point is termed the ‘launch depth’. Chemically speaking, launch depths represent discrete points in P-T-X space on a multiple-saturation surface that is defined by the hot zone mush mineralogy. Thus, melts do not follow conventional liquid lines of descent (LLD), although LLD and multiple-saturation surfaces are often hard to distinguish chemically. Some magmas ascend from their launch depths to erupt at the surface, whereas others stall en route to form plutons. In both cases, little chemical differentiation occurs following launch, due to the inefficiency of crystal-liquid separation during ascent. However, the ascent path and rate do exercise important controls on magma textures.
Magma launch depth also exercises a first-order control on the heights of volcanic edifices above the surrounding base level. This is due to the density difference between buoyant magma and surrounding crust integrated over the vertical distance from the launch depth to the surface. At 30 well-characterised, long-lived volcanoes from a wide range of locations and tectonic settings (including Japan), the calculated volcanic edifice height points to a launch depth of 6 to 50 km. At each volcano studied the launch depth matches petrological (thermobarometry) and geophysical (seismic tomography, MT) constraints on the depth to the upper surface of the underlying hot zone. Although local factors, such as regional stresses or faults, facilitate magma ascent, over the approximately Myr lifetime of a single volcano, the equilibrium (isostatic) edifice height appears to be maintained. Modification of the volcanic edifice and/or changes in magma launch depth and chemistry perturb the equilibrium volcano height, and may lead to renewed growth. Thus, periods of increased edifice construction may be ascribed to episodes of erosion, deglaciation or sector collapse, or changes to hot zone architecture.
It is observed that, in general, more silicic melts are launched from shallower depths than more mafic melts, consistent with differentiation via percolative flow and the overall layered structure of the continental crust. The greatest (sub-Moho) launch depths are found at ocean islands, where mantle xenoliths have been recorded.