HR: 10:20h
AN: V51L-01    [PDF]
TI: Decompression Pattern and Vesiculation Texture of Magma: Preliminary Results of Decompression Experiments Using IHPV With Decompression Speed Controller
AU: * Tomiya, A
EM: a.tomiya@aist.go.jp
AF: Geological Survey of Japan, AIST, Tsukuba Central 7, 1-1-1 Higashi, Tsukuba, 305-8567 Japan
AB: Decompression experiment is an important way to understand vesiculation processes during eruptions. We have installed an internally heated pressure vessel with "decompression speed controller". Our apparatus has two stronger points as follows: (1) high-temperature ($>$900 C) experiments are possible; (2) smooth decompression is possible at various decompression rates (ca. 10 MPa/s $\sim$ 0.0001 MPa/s). The latter is expected to overcome the difficulty on step-like decompression (e.g., Gardner et al., 1999) that pulsative change in decompression rate at each decompression step could cause another influence on kinetics of vesiculation. Using the apparatus, we are conducting a series of decompression experiments on the 2000 eruptive product (low-K dacite) of Usu volcano, Japan. Firstly, the influence of the decompression patterns was compared: (a) single-step, (b) multi-step (5 steps), (c) smooth. The pressure was decreased from 98 MPa to 50 MPa at a constant temperature 900 C after two days of homogenization. The initial water content is 4.0 wt.%, slightly above the saturation at the initial pressure. For the average decompression rate of 0.003 MPa/s, vesicularity is 31% for (a), 27% for (b), and 20% for (c), and the water content in matrix glass is 2.3$\pm$0.4% for (a), 2.4$\pm$0.5% for (b), and 2.7$\pm$0.6% for (c). These difference are probably caused by the difference of the duration at the final pressure (e.g., the longest in (a)). The BSD (bubble size distribution) plot shows exponential for (c), and power-law for (a) and (b). This can be interpreted as follows: at the smooth decompression, both nucleation and growth occurred "continuously"; at the step-like decompression, nucleation occurred pulsatively at the "step(s)", and growth occurred during the following "plateau". Secondly, the experimental results with various (smooth) decompression rates were compared to the natural product. The product is characterized by high water content (ca. 2.5 wt.%; Miyagi et al., 2001) in matrix glass, so that we have proposed that the water content was "frozen" (fixed) at a much deeper level (ca. 2$\sim$3km) than the proposed aquifer (several tens to hundreds meters in depth) (Tomiya et al., 2001). The product is also characterized by many small bubbles (less than several tens of microns in diameter). Our decompression experiments show that high decompression rate causes high water content in matrix glass and domination of small bubbles (the slope of BSD plot is steep). Thus, the characteristics of the 2000 product of Usu volcano can be explained by a high decompression rate during its ascent.
UR: http://staff.aist.go.jp/a.tomiya/tomiyae.html
DE: 8404 Ash deposits
DE: 8414 Eruption mechanisms
DE: 8434 Magma migration
DE: 8494 Instruments and techniques
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting