HR: 17:45h
AN: V34B-08    [Abstracts]
TI: Inversion is the Solution to Dispersion: Modeling Tephra Fallout
AU: * Connor, C
EM: cconnor@chuma1.cas.usf.edu
AF: Department of Geology, Univerisity of South Florida, Tampa, FL 33620 United States
AB: Volcanologists increasingly rely on numerical simulations to understand the dynamics of erupting volcanoes. Mathematical models are often used to explain the geologic processes responsible for eruption deposits found in the geologic record, and to better characterize possible hazards from future volcanic activity. We wish to estimate parameters related to the dynamics of volcanic activity directly from field observations. For example, how well can we estimate the magnitude of an eruption from measurements of tephra deposits? One solution lies in coupling our numerical simulations of volcanic eruption phenomena to inversion methods that search for an optimal set of parameters that explains our observations. Here we use observations of tephra thickness and granulometry from the 1992 eruption of Cerro Negro volcano, Nicaragua, to test the performance of a numerical simulation of tephra fallout. The downhill simplex inversion method is used to search for optimal parameters, including the eruption column height, eruption mass, and wind velocity as a function of elevation about the volcanic vent, that produce deposits that best fit the thickness and grainsize variations observed on the tephra deposit. The computational efficiency of the model is greatly enhanced by parallelizing the numerical model. Through inversion, we estimate the column height and total mass of the eruption as 6500m +/- 750m and 3.1 x 10¹⁰ kg +/- 2.9 x 10⁹ kg respectively. These parameter ranges agree well with observations made during the 1992 Cerro Negro eruption: 7000-7500 m maximum column height and 2.3 x 10¹⁰ kg mass erupted. Parameter uncertainty, reported as one standard deviation from the mean, is estimated using a Monte Carlo method. Inversion techniques such as the downhill simplex method provide an unbiased method for utilizing volcanological observations to evaluate and improve numerical simulations of volcanic activity. Such an approach is essential for evaluating numerical models used in volcanic hazard assessments.
UR: http://www.cas.usf.edu/~cconnor/vg@usf/tephra.html
DE: 8414 Eruption mechanisms and flow emplacement
DE: 8428 Explosive volcanism
DE: 8455 Tephrochronology (1145)
DE: 8488 Volcanic hazards and risks
SC: Volcanology, Geochemistry, Petrology [V]
MN: Fall Meeting 2005