Investigating dynamic fracture processes in 'weak' rocks: applications to volcano seismology and forecasting
- Dr Philip Benson (PI)
- Pete Rowley (CoI)
- Chris Bean (CoI)
Volcano seismicity is an important tool in remotely monitoring and forecasting activity at volcanoes around the world. Volcanic earthquakes show diverse spectral characteristics, with shallow Long Period (Low Frequency) seismicity and long duration tremor generally interpreted as indicators of rapid fluid migration in fractures and faults, sometimes detected before eruption. This project investigates how low-cohesion volcanic sediment from Campi Flegrei caldera (Italy) produces Low Frequency and long duration seismicity whilst undergoing deformation in dry conditions. Correlated X-Ray tomography of samples before and after deformation constrain the source as distributed damage. Given the ubiquitous nature of slow edifice deformation, and the frequent occurrence of such low cohesion materials in the upper edifice of volcanoes, we suggest low frequency seismicity and tremor in volcanic settings do not require fluid movement. Instead, these characteristic signals can be an indicator that deformation within the edifice is being accommodated by weak volcaniclastic materials, and may relate to why low frequency seismicity is not always associated with eruption.
We employ acoustic-emission rock deformation experiments at a range of strain rates to produce events which are spectrally indistinguishable when normalised for scale from Long Period and tremor seismicity observed in natural volcanic settings. Generation of these signals is enhanced at lower strain rates. Our data suggest that deformation within weak, porous, volcanic materials may trigger similar signals to tremor and LP seismicity at low confining pressures, but without requiring the presence or interaction of fluids. Given the frequent observation of shallow ground deformation in volcanic settings, it seems likely the conditions for subjecting volcanic sediments to these types of conditions are widespread. We do not suggest that tremor and LP signals cannot be produced by hydrothermal fluid and magma migration, but we highlight a mechanism for shallow seismicity and tremor unrelated to magma movement, with the capacity to confound the current interpretation of volcanic LP seismicity as always fluid-derived.
|Effective start/end date||1/09/14 → …|