11:15 AM - 11:30 AM
[SVC29-07] Experimental constraints on the crystallization temperature and pressure during magma ascent in the Izu-Oshima 1986 sub-Plinian eruption
Keywords:Decompression-induced crystallization, magma temperature, crystal texture
During the Izu-Oshima 1986 eruption, the sub-Plinian eruption occurred from B vents and erupted basaltic andesite magma. In this eruption, the magma temperature was estimated to be 1070–1100°C (Fujii et al., 1988) and the depth of the magma chamber was 4–5 km (Meteorological Agency, 2008). The high number density of plagioclase microlite was observed in scorias from the Izu-Oshima 1986B eruption, which seems to increase magma viscosity, resulting in magma fragmentation (Ishibashi and Oida, 2018). However, a quantitative understanding of crystallization kinetics in mafic magmas remains poor because of experimental difficulty on continuous decompression (CD) in high temperature magma; hence, the conditions at which the high number density of plagioclase microlite formed is unclear. Here, we conducted decompression experiments and phase equilibrium experiments to investigate the crystallization process during magma ascent for Izu-Oshima 1986B sub-Plinian eruption.
In this study, we performed CD experiments and phase equilibrium experiments using internally heated pressure vessels (IHPV; SMC-8600, SMC-5000) at the Geological Survey of Japan. We used powdered scoria of the Izu-Oshima 1986B sub-Plinian eruption as a starting material. For the experiments, a starting hydrous glass was prepared by melting powdered scorias with 6 wt% water at a pressure of 130 MPa and a temperature of 1040–1250°C. In the starting glasses, no plagioclase was found. Decompression experiments were conducted isothermally at a temperature of 1040 and 1080°C using IHPV (SMC-8600). The pressure was continuously decompressed from 130 to 35 or 10 MPa at decompression rates of 20 and 100 MPa h-1. Phase equilibrium experiments were performed at temperatures of 1000–1120°C under 10–130 MPa pressure. After the experiments, run charges were investigated by using field-emission type scanning electron microscopes with an energy-dispersive X-ray spectrometry.
The number density and crystallinity of plagioclase microlite obtained from decompression experiments were much lower than those in natural scorias. In particular, no plagioclase was found in the decompression experiments at 1080°C and in the experiment decompressed to 35 MPa at a rate of 100 MPa h-1 at 1040°C. On the other hand, plagioclase compositions (An content) obtained from the decompression experiment to 10 MPa at 100 MPa h-1 agreed with those of natural scorias. Moreover, the An contents of plagioclase obtained from the decompression experiments also agree with those obtained from phase equilibrium experiments and calculated using rhyolite-MELTS. This means that even if the crystallinity does not reach equilibrium during CD experiments, the plagioclase composition becomes equilibrium. Based on this result, we estimated the magma crystallization temperature and pressure during the ascent, by comparing the An contents of plagioclase microlites in natural scorias with those calculated using rhyolite-MELTS. Under the isothermal ascent of magma to the surface (~1100°C), magma crystallization can occur at a shallow depth (<5 MPa). However, the vesicularity becomes approximately 60 vol% at 10 MPa under closed-system degassing, resulting in magma fragmentation. In addition to high number density and crystallinity of plagioclase microlite, these observations imply that magma temperature may have decreased during magma ascent, enhancing the crystallization in deep conduit. It should be noted that crystal texture (number density and crystallinity) was not reproduced even if magma temperature decreased to 1040°C. Hence, other mechanisms might influence crystallization processes during magma ascent in addition to the decrease in magma temperature.
In this study, we performed CD experiments and phase equilibrium experiments using internally heated pressure vessels (IHPV; SMC-8600, SMC-5000) at the Geological Survey of Japan. We used powdered scoria of the Izu-Oshima 1986B sub-Plinian eruption as a starting material. For the experiments, a starting hydrous glass was prepared by melting powdered scorias with 6 wt% water at a pressure of 130 MPa and a temperature of 1040–1250°C. In the starting glasses, no plagioclase was found. Decompression experiments were conducted isothermally at a temperature of 1040 and 1080°C using IHPV (SMC-8600). The pressure was continuously decompressed from 130 to 35 or 10 MPa at decompression rates of 20 and 100 MPa h-1. Phase equilibrium experiments were performed at temperatures of 1000–1120°C under 10–130 MPa pressure. After the experiments, run charges were investigated by using field-emission type scanning electron microscopes with an energy-dispersive X-ray spectrometry.
The number density and crystallinity of plagioclase microlite obtained from decompression experiments were much lower than those in natural scorias. In particular, no plagioclase was found in the decompression experiments at 1080°C and in the experiment decompressed to 35 MPa at a rate of 100 MPa h-1 at 1040°C. On the other hand, plagioclase compositions (An content) obtained from the decompression experiment to 10 MPa at 100 MPa h-1 agreed with those of natural scorias. Moreover, the An contents of plagioclase obtained from the decompression experiments also agree with those obtained from phase equilibrium experiments and calculated using rhyolite-MELTS. This means that even if the crystallinity does not reach equilibrium during CD experiments, the plagioclase composition becomes equilibrium. Based on this result, we estimated the magma crystallization temperature and pressure during the ascent, by comparing the An contents of plagioclase microlites in natural scorias with those calculated using rhyolite-MELTS. Under the isothermal ascent of magma to the surface (~1100°C), magma crystallization can occur at a shallow depth (<5 MPa). However, the vesicularity becomes approximately 60 vol% at 10 MPa under closed-system degassing, resulting in magma fragmentation. In addition to high number density and crystallinity of plagioclase microlite, these observations imply that magma temperature may have decreased during magma ascent, enhancing the crystallization in deep conduit. It should be noted that crystal texture (number density and crystallinity) was not reproduced even if magma temperature decreased to 1040°C. Hence, other mechanisms might influence crystallization processes during magma ascent in addition to the decrease in magma temperature.