Surf Infrasound from Oahu’s North Shore: Real-time Monitoring of the Seven Mile Miracle

M. Garces, D. Fee, J. ParkInfrasound Laboratory, University of Hawaii, Manoa

F. HamFlorida Institute of Technology

Surf Infrasound from Oahu’s North Shore: Real-time Monitoring of the Seven Mile Miracle

Background: HI deployments

BACKGROUND

Moorea Deployment: Temae Reef

• Wave gauge

• Infrasound array

• 3C broadband seismometer

• Video camera (GPS time on frames)

Previous Conclusions• Possible to acoustically track one wave breaking - Progressive

Wavefront Tracking• Sound appears to scale with breaking ocean wave intensity• Possible to determine swell direction and period• Source process identification

– Fluid impact– Bubble cloud oscillation– Gas ejection

• Want to apply and test methods in near real-time environment

N. Shore Deployment, Oahu• Expressed interest from NOAA/NWS Honolulu

• Deployed array ~2 km from shoreline (Shark’s Cove).

• Test real-time operational system to monitor broad coastal sections of North Shore

• Correlate observations with Waimea directional buoy and other available data

• Targeted GPS-timed video/IR camera measurements

N. Shore Deployment: Winter 2006-07

• Shark’s Cove = Lava Bell

• North Swell: Log Cabins

• NW-W: Pipeline

• Big NW-W: Waimea/Pinballs

N. Shore Deployment: Winter 2006-07• Note bubble cloud dimensions > 2 m

N. Shore Deployment

N. Shore Deployment

Neural Networks at North Shore (Ham)Confusion Matrix

Prediction

Actual

Pin-balls

Pipe-line

Shark’scove

Unk-nown

Total

Pin-balls

72 0 4(3) 4 80

Pipe-line

1 64 6(1) 9 80

Shark’scove

1(2) 0 114 5 120

F.M. Ham, R. Acharyya, Y-C. Lee, M. Garces, D. Fee, C. Whitten, and E. Rivera, "Classification of Infrasound Surf Events Using Parallel Neural Network Banks," In the Proceedings of the International Joint Conference of Neural Networks, August 12-17, 2007, Orlando, FL, pp. 720-725.

Diagonal numbers indicate the number of correctly classified signals. The off diagonal elements are associated with the number of misclassifications.

Bubble Oscillation Model at Polihale (Park)

Bubble Oscillation Model at Polihale (Park)

0.0 0.5 1.0 1.5 2.0P lum e R adius (m )

1

10

100

f (H

z)

1

10

100

Void Fraction

0.3

0.4

0.5

fc - Sem icylinder

fs - Spherica l

D epth 1m

0.0 0.2 0.4 0.6 0.8 1.0t / T

1.0

1.2

1.4

1.6

1.8

RB

t/RB

0 = 1

/2 .

(A*)

1/2

1.0

1.2

1.4

1.6

1.8

fR(ω) – Transformation from Breaking Wave Height to temporal evolution of Characteristic

Spatial Dimension of bubble plume.

fA(ω) – Transform from Characteristic Plume Dimensions to Spectrum of Radiation

Frequencies

Bubble Oscillation Model at Polihale (Park)

0

5

10

15

20

25

30

Am

plitu

de (

dB)

10 knotsSW H = 1.7 m = 0.099 H z

0

5

10

15

20

25

Am

plitu

de (

dB)

30 knotsSW H = 4.5 m = 0.069 H z

20 knotsSW H = 3.1 m = 0.079 H z

40 knotsSW H = 5.9 m = 0.062 H z

a) b)

c) d)

0 4 8 12 16 20Frequency (H z)

0 4 8 12 16 20Frequency (H z)

Using open ocean wave spectrum, compared modeled (blue) and observed (red) infrasound spectra from Polihale, Kauai.

Even with gross representation of geophysical parameters, the model seems to capture the essential characteristics of surf infrasound which include spectral shape, migration of dominant infrasonic energy to lower frequencies as ocean wave energy increases, and a broadening of the infrasonic energy distribution across the main lobe.

Concluding Remarks

Operational acoustic surf monitoring systems are

feasible and may be useful for nowcasting and shoreline hazard

assessment.

Ongoing progress on modeling of surf

infrasound signals would permit extraction of

oceanographic information from the

acoustic data.