9:30 AM - 9:45 AM
[SVC49-03] Sawtooth wave-like pressure changes (STW) appeared in a slug flow experiment: Toward understanding of volcanic oscillation systems
Keywords:Volcano, Laboratory experiment, Mathematical model
An apparatus for a slug-flow experiment was designed based on the syrup eruption experiment. This apparatus was equipped with a gas chamber (volume, Vc) and a vertical pipe for a slug flow. Initially the pipe was partially filled with the syrup to the height of Hs from the end. Then, gas was injected at a constant mass flux (Qin) to the chamber to flow into the pipe pushing up the syrup in the pipe. Two representative flow patterns were observed in the pipe. One was characterized by alternate layers of syrup slugs and gas slugs ascending in the pipe, which we called a slug flow. The other was characterized by repetitive transitions between the slug flow and an annular flow, which we called a slug-annular flow oscillation. The STW was observed during the slug-annular flow oscillation.
Pressure change in the chamber and acoustic waves at the vent of the pipe were measured. These measurements were assumed to correspond to geodetic and infrasonic observations at actual active volcanoes. In the experiment, the flow patterns were also constrained by image analyses. The occurrences and features of the STW in the chamber pressure were investigated with taking Vc, Qin, and Hs as the experimental parameters. The results showed that the STWs were observed if there were sufficiently large Vc and Qin, and that the STW changed from periodic to non-periodic cycles with increasing Qin.
A mathematical model was constructed based on the experimental results of the pressure changes and the flow behaviors in the pipe. The model took account of the compressibility of the gas in the chamber, and the nonlinearity of the pressure loss in the pipe flow due to the interaction between the ascending liquid slugs and falling liquid film along the pipe wall. The dependence of the occurrence, the period, and the amplitude of the periodic STW on the experimental parameters were well explained by the model. The model has a mathematically similar aspect compared to existing models for the volcanic oscillation.
Moreover, not only the periodic STW but also the non-periodic STW was observed in this experiment. The non-periodic STW behavior has not been captured by the present model. According to the image analyses, we inferred that the non-periodic behaviors were caused by the interaction between the ascending liquid slugs and surface disturbances of the falling liquid film. From these results, we obtain an insight that irregularity of actual eruptions can be caused not only by fluctuations in ascending flow but also by influences of descending flow such as a fall back, a drain back and a magma convection of magma in the conduit.