5:15 PM - 6:45 PM
[SVC31-P05] Interstitial melt connectivity in the mushy gabbroic xenoliths of Fuji 1707 eruption
Keywords:crystal mush, interstitial melt, Fuji volcano, magma reservoir
Recent volcanology suggests that a large part of the magma plumbing system consists of crystal mush. Extraction of the interstitial melt from crystal mush is an important process for not only preparing the eruptible magma chamber but also for the formation of a plutonic rock. However, the process of how interstitial melt decreases in crystal mush to form plutonic rock is still poorly understood. Interstitial melt-bearing plutonic xenoliths are the key to investigate this. Mushy gabbroic xenoliths are often found in the volcanic deposits of the Fuji 1707 eruption; they include interstitial glass up to ~40 vol% and are thought to be fragments of crystal mush. In this study, we performed image analyses of the BSE images taken for the mushy gabbroic xenoliths of the Fuji 1707 eruption to investigate the spatial distribution and connectivity of the interstitial melt. The samples were already petrologically investigated by Otsuka et al. (2021VSJ meeting abstract) and the BSE images were taken using FE-EPMA (JEOL JXA-8530FPlus) at Earthquake Research Institute, the University of Tokyo.
~20% of the gabbroic xenoliths have interstitial melt and they are classified into 2 groups; Group-A and B characterized by relatively homogeneous rhyolitic and heterogeneous andesitic-dacitic interstitial glasses, respectively (Otsuka et al., 2021VSJ). Among these, we selected samples with different volume fractions of the interstitial phase (= intetstital glass + bubble). Because no textural evidence indicating significant deformation induced by bubble growth is observed, we assumed the total volume fraction of interstitial phase has been constant during ascent. We made interstitial phase maps by extracting only the interstitial glass and bubble from the BSE images taken of the entire thin sections. Then, the phase maps were analyzed using ImageJ software to quantify the areas and the ellipse-approximated major and minor axes lengths of the interstitial phase clusters, where we define “cluster” as a spatially continuous mass. In addition, the representative points of the interstitial phase were determined by shrinkage processing and the homogeneity of the spatial distribution of the interstitial phase was investigated from the distance frequency distribution of these points; we called here as representative point distribution (RPD) analyses.
We analyzed the following samples; FGB200, FGB213 (Group-A gabbronorite), FGB165, and FGB293 (Group-B norite). The interstitial phase abundance is 38.8% for FGB200, 36.2% for FGB213, 22.9% for FGB165, and 10.2% for FGB293, respectively. We define the areal fraction of the largest cluster to the total area of the interstitial phase as CM1 and the sum of the areal fractions of the first and second largest clusters as CM2. CM1 and CM2 are 47.7% and 66.7% for FGB200, 84.9% and 85.6% for FGB213, 30.3% and 36.7% for FGB165, and 19.3% and 22.2% for FGB293, respectively. CM1 of FGB200 is significantly lower than that of FGB213 although F is almost the same; we interpreted that the largest cluster happens to divide into two in the analyzed cross section of FGB200. Therefore, we adopt CM2 as an indicator of melt connectivity. CM2 decreases monotonously as F decreases, indicating that melt connectivity is decreased by solidification of crystal mush. The ratio of the ellipse-approximated major length/minor length of the largest cluster is 1.9 for FGB200, 1.1 for FGB213, 1.3 for FGB165, and 2.1 for FGB293, respectively. This indicates that the interstitial phase clusters are shrinking isotropically. Furthermore, the results of RPD analyses indicate that interstitial phase clusters are randomly distributed in FGB200, FGB213, and FGB165 and slightly localized in FGB293. Present results suggest that during the solidification of crystal mush, the interstitial phase is not localized, isotropically shrinks, and decreases melt connectivity.
~20% of the gabbroic xenoliths have interstitial melt and they are classified into 2 groups; Group-A and B characterized by relatively homogeneous rhyolitic and heterogeneous andesitic-dacitic interstitial glasses, respectively (Otsuka et al., 2021VSJ). Among these, we selected samples with different volume fractions of the interstitial phase (= intetstital glass + bubble). Because no textural evidence indicating significant deformation induced by bubble growth is observed, we assumed the total volume fraction of interstitial phase has been constant during ascent. We made interstitial phase maps by extracting only the interstitial glass and bubble from the BSE images taken of the entire thin sections. Then, the phase maps were analyzed using ImageJ software to quantify the areas and the ellipse-approximated major and minor axes lengths of the interstitial phase clusters, where we define “cluster” as a spatially continuous mass. In addition, the representative points of the interstitial phase were determined by shrinkage processing and the homogeneity of the spatial distribution of the interstitial phase was investigated from the distance frequency distribution of these points; we called here as representative point distribution (RPD) analyses.
We analyzed the following samples; FGB200, FGB213 (Group-A gabbronorite), FGB165, and FGB293 (Group-B norite). The interstitial phase abundance is 38.8% for FGB200, 36.2% for FGB213, 22.9% for FGB165, and 10.2% for FGB293, respectively. We define the areal fraction of the largest cluster to the total area of the interstitial phase as CM1 and the sum of the areal fractions of the first and second largest clusters as CM2. CM1 and CM2 are 47.7% and 66.7% for FGB200, 84.9% and 85.6% for FGB213, 30.3% and 36.7% for FGB165, and 19.3% and 22.2% for FGB293, respectively. CM1 of FGB200 is significantly lower than that of FGB213 although F is almost the same; we interpreted that the largest cluster happens to divide into two in the analyzed cross section of FGB200. Therefore, we adopt CM2 as an indicator of melt connectivity. CM2 decreases monotonously as F decreases, indicating that melt connectivity is decreased by solidification of crystal mush. The ratio of the ellipse-approximated major length/minor length of the largest cluster is 1.9 for FGB200, 1.1 for FGB213, 1.3 for FGB165, and 2.1 for FGB293, respectively. This indicates that the interstitial phase clusters are shrinking isotropically. Furthermore, the results of RPD analyses indicate that interstitial phase clusters are randomly distributed in FGB200, FGB213, and FGB165 and slightly localized in FGB293. Present results suggest that during the solidification of crystal mush, the interstitial phase is not localized, isotropically shrinks, and decreases melt connectivity.