Seismic and Acoustic Observations of Lava Lake Explosions at Mount Erebus Volcano, Antarctica.


Rowe, C.A., R.C. Aster, P.R. Kyle, J.W. Schlue
New Mexico Institute of Mining & Technology and Geophysical Research Center,
Socorro, New Mexico

R.R. Dibble
Victoria University, Wellington, New Zealand

Continuous and triggered digital recordings of synchronous seismic and acoustic data have been collected at Erebus volcano since December, 1996. The presence of an acoustic sensor co-located with a seismometer 720 meters from the actively convecting phonolitic lava lake within Erebus' summit crater has provided data from hundreds of Strombolian explosions ranging in size by over two orders of magnitude. Despite the size variations, most explosions exhibit strong similarities in their acoustic spectra, including a consistent corner frequency of ~3 Hz, suggesting that the acoustic signature is strongly influenced by repeatable source and path characteristics which are independent of explosion size. Comparison of acoustic and seismic amplitudes reveals a well-defined linear trend suggesting that the ratio of seismic and acoustic energy partitioning is generally consistent among explosions and is not a function of bubble size; however, there is a population of explosion events which digresses significantly from this central trend.

Cumulative seismic and acoustic energy rates over a two-year period suggest an approximately steady rate of explosive degassing of the Erebus magmatic system, comprised of smaller cycles of activity which begin more energetically and decay in time. We interpret this behavior as a manifestation of cyclic volatile recharge resulting from a nearly constant rate of magmatic degassing at depth. Although most explosive activity is presumed to result from the introduction of magmatic volatiles, we have documented at least one instance of explosive activity which was triggered by the slumping of a large volume of snow and ice onto the lava lake. This event was followed by three days of vigorous bubbling and numerous small explosions which resulted directly from the surficial introduction of excess H2O in the uppermost part of the magma column. Despite this heightened activity, no large explosions occurred immediately following the avalanche, suggesting that the shallowest layer is convectively isolated from the remainder of the magma feeder system and that the added volatiles were not incorporated into the deeper regime.