Infrasounds and Background Free Oscillations

1
Infrasounds and Background Free Oscillations

• Naoki Kobayashi 1
• T. Kusumi and N. Suda 2
• 1 Tokyo Tech 2 Hiroshima Univ.

2
Free oscillations
spherical harmonics

• Normal modes of the solid earth
• Earthquakes with Magnitude gt 6
• Characteristic time

where
radial eigenfunction
3
What are the background free oscillations?

• 610-19 m2/s3 in the mHz band even on
  seismically quiet days
• Annual and/or semiannual variations in amplitudes
• Larger amplitudes at thebranch crossings with
  theinfrasound modes

Nawa et al. 1998
PSD of ground accelerations
4
PSD on seismically quiet days
PSD of ground accelerations

• 0Sl are observed on seismically quiet days
• Peaks lt 10-18 m2s-3
• Higher intensities in the summer season of the
  northern hemisphere
• Larger amplitudes of 0S29 and 0S37

19902006, IRIS 25 quiet stations,
90 days-average
5
Larger intensities at the branch crossings with
the infrasound modes

• Peak intensities are larger at the branch
  crossings with the infrasound modes!

all year
mHz
July
Angular degree
6
What is the excitation?

• Atmospheric turbulences
• Kobayashi Nishida (1998)
• Nishida Kobayashi (1999)
• Modes are excited independently one another.
• Fukao et al. (2002)
• Oceanic process
• Rhie Romanowicz (2004)
• Stronger wave radiations from northern and
  southern pacific ocean in winter season
• Tanimoto (2005), Webb (2007)
• Wave-wave interaction of ocean gravity waves

global source region
small source region
7
Atmospheric excitation
turbulent cells
cycles in life
N
Force
degeneracy
Mass (response)
210-12 m/s2
the earth
8
Observation and synthetic
acceleration
pressure
Fukao et al. (2002)
9
Well but

• Fukao et al. (2002) well explain the background
  free oscillations using observed pressure PSD.
• But it fails to explain the excesses of
  amplitudes of 0S29 and 0S37.
• We need the atmosphere!

Branch crossings
10
New method of normal mode calculation

• Anelasticity
• Open boundary condition
• Quick search for a complex eigenfrequency
• Numerically stable

Vertical displacement eigenfunction
Kobayashi (GJI 2007)
Both modes are calculated from the center of the
Earth to an altitude of 1000km.
11
Excitation by atmospheric turbulence
Power spectral densities of the ground
accelerations
Force
N
Response
where
From volumetric pressure forces
12
Comparison with observation
response
Obs./synthetic
residual
force
13
Seasonal variation due to thermal structure in
the atmosphere
Obs./synthetic
response
force
residual
14
Excitation of acoustic modes by atmospheric
turbulence
only
Too small to observe them!
15
Another estimate
Excess in amplitude
a contribution of acoustic mode pressure.
For a singlet of
(multiplet)
16
Schematic view
Acoustic waves
Boundary turbulence
Surface waves
17
conclusion

• The Earth is oscillating incessantly due to other
  mechanism than earthquakes. Their amplitudes are
  about 10-18 m2/s3 in the central mHz band and
  varies annually.
• Amplitudes of modes are explained by the
  atmospheric turbulence in the boundary layer.
• Excesses of amplitudes of modes at the branch
  crossings with the infrasound modes are also
  explained by the atmospheric turbulence.
• We also predict pressure signals of infrasound
  modes at 3.7 and 4.4 mHz are about 10-4 Pa2/Hz
  which may NOT be detectable.

But a broad band seismometer can be a good
detector for the acoustic modes!
18
Ground acceleration spectra
New Low Noise Model (Peterson 1993)
19
Model atmosphere
PREM
Globally averaged July atmosphere
NRLMSISE-00 (Picone et al. 2002)
20
Discussion on the dynamic pressure

• We use the same PSD as Fukao et al. (2002) for
  the dynamic pressure.
• This is not the pressure of the B.L. turbulence.
• However
• The values around 5 mHz are comparable with
  observed aero dynamic pressure.
• Correlation length is also comparable with a
  scale of boundary layers. (700m)

mesoscale
pressure
B. L.
winds
temperature
at Boso peninsula in Japan
21
Vertical displacement eigenfunctions
altitude