 |
| Titel |
Forecasting volcanic eruptions using Voight's relation |
| VerfasserIn |
Andrew Bell, John Greenhough, Christopher Kilburn, Michael Heap, Ian Main |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250051732
|
|
|
|
|
|
| Zusammenfassung |
| Voight’s relation d^2Ω/dt^2 = K(dΩ/dt)α between the acceleration and rate of earthquakes or
other geophysical parameters Ω has been proposed as a falsifiable hypothesis for precursory
signals prior to volcanic eruptions (K and α being empirical constants). Importantly, models
based on Voight’s relation may permit forecasting of eruption time. However, in
existing retrospective analyses it is common to find examples of (1) inappropriate
techniques for fitting these models to data and (2) inappropriate data selection that
fails to account for the complexity of pre-eruptive processes. Here we test the two
main competing hypotheses based on Voight’s relation - exponential and power-law
acceleration – using maximum likelihood techniques and an information criterion for
model choice, based on a Poisson process with variable rate. For examples from Mt
Etna and a laboratory brittle creep-to-failure experiment, the power-law is clearly
the best model, both in terms of the fit and the resulting error structure, which is
consistent with the Poisson approximation. At Kilauea, a spectrum of precursory
behaviour is observed indicating different mechanisms controlling the approach to
volcanic events. When these are accounted for, an exponential is clearly the best
model for all 7 of the appropriate sequences. Deviations from the models most likely
reflect local interactions and/or non-stationary loading processes not captured by the
mean-field approach inherent in Voight’s relation. In addition, we use simulations
to demonstrate an inherent problem with model preference, in that a power-law
model will only be preferred if failure or eruption occurs close to the singularity.
Consequently, even under ideal conditions, a power-law trend may only be observed a short
time ahead of an eruption. We conclude that prospective mode model testing is
essential for quantifying the predictability of volcanic eruptions using Voight’s relation. |
|
|
|
|
|
|