Overview of the Science of Tsunamis

1
Overview of the Science of Tsunamis

• Harry Yeh

2
Recent Major Tsunami Events

• Casualties runup height
• Nicaragua (Sept. 92) - Ms 7.2 Mw 7.6 II
  III 93 9.9 m
• Flores Island, Indonesia (Dec. 92) - Ms 7.5
  VIII IX 1712 26.0 m
• Okushiri, Japan (July 93) - Ms 7.2 233 32.0 m
• East Java, Indonesia (June 94) - Ms
  7.2 223 11.3 m
• Shikotan, S. Kuril Islands (Oct. 94) - Ms 8.1
  IX X 12 7.1 m
• Mindoro, Philippines (Nov. 94) - Ms 7.0 74 7.3
  m
• Skagway, Alaska (Nov. 94) - Landslide 1
• East Timor, Indonesia (May 95) - Ms 6.9 8
• La Manzanilla, Mexico (Oct. 95) - Mw 8.0 5.0 m
  ?
• Irian Jaya, Indonesia (Feb. 96) - Mw
  8.0 110 7.7 m
• Chimbote, Peru (Feb. 96) - Ms 6.8 Mw
  7.5 12 5.0 m
• Aitape, PNG (July 98) - Ms 7.1 Mw 7.0
  2000 15.0 m
• Vanuatu (Nov. 99) - Ms 7.3 1
• Southern Peru (June 01) - Mw 8.3 gt 26 4.0 m
• Stromboli, Italy (Dec. 02) - Landslide
• Tokachi-Oki, Japan (Sept. 03) - Mw 8.0 4.2 m
• Indian Ocean (Dec. 04) - Mw 9.0 9.3
  230,000 36.0m
• Northern Sumatra (Mar. 05) - Mw 8.7 1300 3.0 m

3
Deaths caused by tsunamis
http//www.ngdc.noaa.gov/seg/hazard/tsu.shtml

• USAID Reports for the 2004 Indian Ocean Tsunami
• Jan. 10 150,521 dead 24,172 missing 175,693
  total
• Jan. 31 152,561 dead 142,129 missing 294,690
  total
• Feb. 22 169,680 dead 127,369 missing 297,049
  total
• May 6 176,633 dead 50,321 missing 226,954
  total

4
Outline

• Spatial scale of tsunamis and their formation
• Directivity in tsunami propagation.
• Similarities and differences between the 2004
  Indian Ocean tsunami and the previous tsunamis
• Distinct behaviors and characteristics of
  tsunamis from other coastal hazards
• Tsunami can sneak around an island (Disaster of
  Babi Island)
• Extreme and local tsunami enhancement (32 m runup
  in Okushiri)
• Tsunami can be reflective and the attenuation
  process is slow.
• Tsunami can cause severe scour and transport
  sediments inland
• Geomorphological changes and ecological impact

5
From DMA Chart
6
Tsunami with the wavelength of 500 km
Bathymetry Profile along N13
7
This is NOT a typical tsunami
Source Earthquake by Bruce A. Bolt
8
at Ta Phao Noi, Thailand, showing the leading
depression wave
at Tuticorin, India, showing the leading
elevation wave
9
Directivity in tsunami propagation The Cordex
Leicester of Leonardo da Vinci Folio 14 v
10
Solutions at t 0.5, 0.75, 1.0, 1.5, 2.0, 5.0
h
r
11
Solutions at t 5, 10, 20, 30, 40, 50
h
r
12
Initial Displacement
L 10
L 20
L 40
13
L 20
L 40
L 10
y
x
h
14
Computed maximum tsunami height by Koshimura
1 st. segment (Southern part) (Strike, Dip,
Slip) (329, 15, 90) (L,W) (500 km, 150 km)
Dislocation 11 m Depth 10 km 2nd. segment
(Northern part) (Strike, Dip, Slip) (345, 15,
90) (L,W) (400 km, 150 km) Dislocation 11
m Depth 10 km
15
The 2004 Great Indian Ocean Tsunami
By David George Randy LeVeque
16
The 2004 Indian Ocean Tsunami and The Previous
Tsunamis
17
The 2004 Indian Ocean Tsunami Lhoknga
Photo by Jose Borerro
18
Trubean, Flores, 1992
19
The 2004 Indian Ocean Tsunami Vedaranniyan, India
(N 1023.597, E 7952.014)
20
El Popoyo, Nicaragua, 1992
21
The 2004 Indian Ocean Tsunami Banda Aceh
Photo by Jose Borerro
22
Aonae, Okushiri Island, Japan 1993
23

• Individual tsunami effects on the coastal areas
  were similar to the previous tsunami events.
• What makes the 26 December 2004 event distinct is
  the extent of the affected area.

24
The 1992 Flores Tsunami
Tsunami attack
Babi Island, Flores, Indonesia
25
(No Transcript)
26
Experiments by Costas Synolakis and Michael Briggs
27
Numerical simulation by Philip Liu
28

• Tsunami behaviors and characteristics are quite
  distinct from other coastal hazards (such as
  storm waves) the effects may not be inferred
  from common knowledge or intuition.

29
The 1993 Okushiri Tsunami Monai
30
1993 Okushiri Tsunami - CRIEPIby Matsuyama
31
(No Transcript)
32
Numerical simulation by George LeVaque
33
(No Transcript)
34
El Transito, Nicaragua -- 1992
35

• Accurate nearshore bathymetry and coastal
  topography are crucial for tsunami runup modeling.

36
The 2003 Tokachi-Oki Tsunami Numerical
Simulation by Koshimura
37
The 1983 Nihonkai-Chubu Tsunami

• Storm waves pound the shore.
• Tsunamis sweep the coastal zone.

38

• Storm waves pound the shore.
• Tsunamis sweep the coastal zone.

39
The 2004 Indian Ocean Tsunami
Structure Damage Scour
Devanaanpattinam (1144.629N,
7947.271E) Tsunami effects on a masonry
house on beach berm
40
The 2004 Indian Ocean Tsunami
School house at Kalapakkom, India (N1230.378
E8009.688)
Scour Structure Damage
41
The 2004 Indian Ocean Tsunami
Infrastructure Scour
Undermined sidewalk in Chennai, India
(N1302.061 E8016.792)
42
Tsunami Scour - East Java (1994)
43
Tsunami-Induced Scour around a Vertical
CylinderThe 1998 Papua New Guinea
44
Tsunami Tank at PWRI - 135 m long
45
(No Transcript)
46
(No Transcript)
47
(No Transcript)
48
Scour Mechanisms

• Shear stress due to water motion Shields model
• Low effective stress between sand particles
• Dependent on pore pressure gradient
• Sediment liquefies if effective stress disappears
• Smaller pore pressure gradients can enhance scour
  due to shear stress

49

• Maximum scour occurs during the (runup/drawdown)
  process.
• Traditional shear stress modeling (Shields) does
  not predict rapid scour at the end of drawdown.

50
Enlarged channel made by tsunami drawdown
Devanaanpattinam (N1144.576 E7947.230)
51
Sri Lanka, Kalutara Beach
52
Discolored Vegetation Nicobar Island
53
Summary

• A tsunami can propagate more than several
  thousand kilometers without losing its energy.
  Long wave components propagate faster than the
  shorter wave components therefore, a
  transoceanic tsunami is usually characterized by
  a long-period wave (several to tens of minutes).
  Shorter wave components are left behind and
  attenuated by radiation and dispersion.
• Tsunami energy propagation has strong
  directivity the majority of its energy will be
  emitted in the direction normal to the major axis
  of the tsunami source.
• For a locally generated tsunami, the leading
  initial tsunami is often a receding water level
  followed by an advancing positive heave (an
  elevation wave). On the other hand, the leading
  wave of a far-source-generated tsunami is often
  elevation.
• Tsunami effects often last for several hours and
  the first wave is not necessarily the largest.
  This is because tsunamis are highly reflective at
  the shore, and capable of sustaining their motion
  without rapidly dissipating energy.

54
Summary -- continue

• Tsunami runup height varies significantly in
  neighboring areas. This characteristic is caused
  by tsunamis reflective behavior as well as the
  effects of local bathymetry and coastal
  topography.
• Tsunami behaviors and characteristics are quite
  distinct from other coastal hazards (such as
  storm waves and flooding), and the effects cannot
  be inferred from common knowledge or intuition.
  The primary reason for the difference is the
  unique timescale and spatial scale associated
  with tsunami phenomena. For a typical tsunami,
  the water surface near the shore fluctuates with
  amplitude of several meters during a period of
  tens of minutes. This timescale is intermediate
  between the hours to days typical of river-flood
  problems, and the tens of seconds or less
  associated with cyclic loading of wind waves.