HR: 0830h
AN: V51E-0330    [PDF]
TI: Exploration of eruption mechanisms through parametric sensitivity analysis: Using a numerical model as a diagnostic tool
AU: * Proussevitch, A
EM: alex.proussevitch@unh.edu
AF: EOS & Earth Sci., University of New Hampshire, Durham, NH 03824
AU: Sahagian, D
EM: dork.sahagian@unh.edu
AF: EOS & Earth Sci., University of New Hampshire, Durham, NH 03824
AB: Numerical models can be useful for exploring the dependence of eruption timing and style to variations in magma chemical and physical characteristics, as well as to geometry of the system. We have developed a model that can be used for this purpose ("Bubbledrive") and have applied it to a number of scenarios involving rhyolitic systems. We have tested the sensitivity of eruption style to variations in conduit depth and radius, volatile diffusivity in the melt, melt viscosity, magma recharge from below, and decompression triggers. The results show that after the initiation of eruption, there is a gradual acceleration stage in case of cylindrical conduit shapes. Exit velocity is approximately a linear function of time, and discharge rate is close to constant. The difference between velocity and discharge reflects the increasing vesicularity of the erupting magma during degassing and bubble growth. After a climax in exit velocity, the eruptions quickly cease. As such, these results suggest that an empirical equation can be formulated that describes exit velocity in terms of eruption duration and an acceleration factor, which, in turn are functions of conduit radius, depth, diffusivity, viscosity, etc. The above relations do not apply to magma systems with more "realistic" or complex conduit geometries, such as those that include a magma chamber. In the latter case, initialization is followed by quasi-steady state eruption of different intensity over extended time (about 80% of eruption cycle). This relatively uniform stage of eruption is maintained by the degassing of magma within the chamber that feeds the upper conduit. The uniform stage of eruption ends abruptly when the magma chamber is degassed. An eruption spike appears as an explosion near the end of degassing of the chamber. This spike begins when the overlying conduit is filled with foam or gassy spray that provides little overburden; this leads to rapid decompressive degassing of the chamber with little resistance to the flow from the overlying conduit. The maximum explosion occurs as the foam layer moves into the chamber. These results of numerical modeling suggest that observations or monitoring systems could be used most fruitfully in the areas of conduit geometry. At present, it is simpler to develop numerical models of specific conduit geometries than it is to directly probe and map them in natural systems. This highlights the importance of monitoring systems for specific volcanic systems.
DE: 8414 Eruption mechanisms
DE: 8419 Eruption monitoring (7280)
DE: 8434 Magma migration
DE: 8439 Physics and chemistry of magma bodies
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting