Infrasonic resonance of volcanic craters Leighton Watson Hugo Ortiz - PowerPoint PPT Presentation

Infrasonic resonance of volcanic craters Leighton Watson Hugo Ortiz Eric Dunham Pontificia Universidad Catolica del Ecuador Stanford University Jeffrey Johnson Mario Ruiz Alex Miller Instituto Geofisico, Ecuador Boise State University

Harmonic peaks seen in infrasound from open-vent volcanoes amplitude amplitude amplitude amplitude Data courtesy of Ortiz and Ruiz

Harmonic peaks seen in infrasound from open-vent volcanoes Kilauea Kilauea amplitude amplitude amplitude amplitude Fee et al. (2010) Er Erebus ebus Station: RAY Station: SHK Amplitude 0 0.5 1 1.5 2 2.5 3 3.5 Frequency (Hz) Data courtesy of Ortiz and Ruiz Miller and Johnson (2017)

Harmonic peaks seen in infrasound from open-vent volcanoes amplitude amplitude amplitude amplitude amplitude amplitude amplitude Data courtesy of Marin, Cardoña and Palma Goto and Johnson (2011)

Harmonic peaks are due to resonance of crater Helmholtz resonator Resonant modes of horn Fee et al. (2010); Goto and Johnson (2011) Richardson et al. (2014) Resonating cavity Approximate volcanic crater as a • • horn open at one end and closed Resonant frequency • at the other is related to size of cavity Only discrete wavelengths fit in the • horn Resonant frequency is related to • length of horn After Fee et al. (2010)

Harmonic peaks are due to resonance of crater Helmholtz resonator Resonant modes of horn Fee et al. (2010); Goto and Johnson (2011) Richardson et al. (2014) Resonating cavity Approximate volcanic crater as a • • horn open at one end and closed Resonant frequency • at the other is related to size of cavity Only discrete wavelengths fit in the • horn Resonant frequency is related to • length of horn After Fee et al. (2010) Acoustic waves are excited by: Explosive bubbles bursts at lava lake surface • Unsteady gas flux through crater floor •

Overview of model Compute transfer function: transfer function: maps source (velocity at base of crater) to observable (pressure perturbation) recorded at receiver

Overview of model Compute transfer function: transfer function: maps source (velocity at base of crater) to observable (pressure perturbation) recorded at receiver (1) (1) Resonant modes of crater Resonant modes of crater (2) (2) Acoustic radiation Acoustic radiation

(1) Resonant modes of crater Model wave propagation inside crater Solve linear acoustics within crater • • Compute resonant modes • Quasi 1D axisymmetric crater • Frequency domain • Damping occurs through acoustic • radiation Controlled by impedance contrast • across crater outlet

(2) Acoustic radiation Model wave propagation from crater Radiation into a half space • • outlet to receiver Describe acoustic source as a baffled • piston embedded in an infinite plane Do not account for topographic • scattering (Kim et al., 2015) Observations are similar across all • stations in the network

Model outputs 0 Crater geometry Transfer Function Depth (m) 100 Amplitude -100 0 100 0 1 2 3 Radius (m) Frequency (Hz)

Model outputs 0 Crater geometry Transfer Source: velocity profile at base of crater Function Amplitude Amplitude Spectra Depth (m) 0 2 4 6 0 1 2 3 4 5 Time (s) Frequency (Hz) 100 Amplitude -100 0 100 0 1 2 3 Radius (m) Frequency (Hz)

Model outputs 0 Crater geometry Transfer Source: velocity profile at base of crater Function Amplitude Amplitude Spectra Depth (m) 0 2 4 6 0 1 2 3 4 5 Time (s) Frequency (Hz) 100 Amplitude Amplitude 0 10 20 20 30 30 Time (s) Time (s) -100 0 100 0 1 2 3 Radius (m) Frequency (Hz)

Infrasonic resonance depends on crater depth 0 Crater PE41B-1: Johnson et al., Forecasting Paroxysmal Eruptions at Volcán Villarrica (Chile) Using PE41B-1: geometry Infrasound . Thursday August 17 th , 10:30-10:45 am. Depth to lava lake: Source: velocity profile at base of crater 140 m Amplitude Amplitude Depth (m) 0 2 4 6 0 1 2 3 4 5 Time (s) Frequency (Hz) 100 Amplitude Amplitude 0 10 20 30 Time (s) -100 0 100 0 1 2 3 Radius (m) Frequency (Hz)

Infrasonic resonance depends on crater depth 0 Crater PE41B-1: PE41B-1: Johnson et al., Forecasting Paroxysmal Eruptions at Volcán Villarrica (Chile) Using geometry Infrasound . Thursday August 17 th , 10:30-10:45 am. Depth to lava lake: Source: velocity profile at base of crater 140 m Amplitude Amplitude 100 m Depth (m) 0 2 4 6 0 1 2 3 4 5 Time (s) Frequency (Hz) 100 Amplitude Amplitude 0 10 20 30 Time (s) -100 0 100 0 1 2 3 Radius (m) Frequency (Hz)

Infrasonic resonance depends on source 0 Crater function geometry Source function: Source: velocity profile at base of crater Gaussian Amplitude Amplitude Depth (m) 0 2 4 6 0 1 2 3 4 5 Time (s) Frequency (Hz) 100 Amplitude Amplitude 0 10 20 30 Time (s) -100 0 100 0 1 2 3 Radius (m) Frequency (Hz)

Infrasonic resonance depends on source 0 Crater function geometry Source function: Source: velocity profile at base of crater Gaussian Amplitude Amplitude Brune Depth (m) 0 2 4 6 0 1 2 3 4 5 Time (s) Frequency (Hz) 100 Amplitude Amplitude Amplitude 0 10 20 30 Time (s) -100 0 100 0 1 2 3 Radius (m) Frequency (Hz)

Infrasonic resonance depends on crater 0 Crater geometry geometry Geometry: Source: velocity profile at base of crater Johnson et al. (2017) in prep Amplitude Amplitude Depth (m) 0 2 4 6 0 1 2 3 4 5 Time (s) Frequency (Hz) 100 Amplitude Amplitude Amplitude Amplitude 0 0 10 10 20 20 30 30 Time (s) -100 0 100 0 0 1 1 2 2 3 3 Radius (m) Frequency (Hz) Frequency (Hz)

Infrasonic resonance depends on crater 0 Crater geometry geometry Geometry: Source: velocity profile at base of crater Johnson et al. (2017) in prep Amplitude Amplitude Richardson et al. (2014) Depth (m) 0 2 4 6 0 1 2 3 4 5 Time (s) Frequency (Hz) 100 Amplitude Amplitude Amplitude 0 10 10 20 20 30 30 Time (s) -100 0 100 0 1 2 3 Radius (m) Frequency (Hz)

Harmonic infrasound observations can be inverted for: • Depth of crater • Source function • Crater geometry Amplitude Amplitude 0 1 2 3 4 5 0 5 10 Frequency (Hz) Time (s) Villarrica, Chile Data courtesy of Marin, Cardoña and Palma

Invert harmonic infrasound Inversion: observations for: Fit frequency and quality factor of • harmonic peak • Depth of crater • Source function • Crater geometry MCMC inversion scheme • Amplitude Amplitude 0 1 2 3 4 5 0 5 10 Frequency (Hz) Time (s) Villarrica, Chile Data courtesy of Marin, Cardoña and Palma

Invert infrasound observations for crater geometry and source Amplitude Amplitude 0 5 10 Time (s) 0 1 2 Frequency (Hz) Data Data uncertainty

Invert infrasound observations for crater geometry and source 0 Amplitude 50 Depth (m) Amplitude 0 5 10 Time (s) 100 Amplitude Amplitude 150 200 0 1 2 0 5 0 1 2 -100 0 100 Time (s) Frequency (Hz) Frequency (Hz) Radius (m) Data Inversion input Data uncertainty

Invert infrasound observations for crater geometry and source 0 Amplitude 50 Depth (m) Amplitude 0 5 10 0 5 10 Time (s) Time (s) 100 Amplitude Amplitude 150 200 0 1 2 0 5 0 1 2 -100 0 100 Time (s) Frequency (Hz) Frequency (Hz) Radius (m) Data Inversion input Data uncertainty

Invert infrasound observations for crater geometry and source 0 Amplitude 50 Depth (m) Amplitude 0 5 10 0 5 10 Time (s) Time (s) 100 Amplitude Amplitude 150 200 0 1 2 0 5 0 1 2 -100 0 100 Time (s) Frequency (Hz) Frequency (Hz) Radius (m) Data Inversion output Inversion uncertainty Data uncertainty

Inverted geometry matches visible observations of Villarrica crater 0 Amplitude 50 Depth (m) Amplitude 0 5 10 0 5 10 Time (s) Time (s) 100 Amplitude Amplitude 150 200 0 1 2 0 5 0 1 2 -100 0 100 Time (s) Frequency (Hz) Frequency (Hz) Radius (m) Data Inversion output Inversion uncertainty Data uncertainty

Conclusion Infrasonic resonance is seen at multiple open-vent volcanoes The resonant amplitude spectra is dependent on: • Crater depth • Geometry • Source function Resonant infrasound observations can be inverted to determine these properties