USGS - science for a changing world

Pacific Coastal and Marine Science Center

Tsunamis and Earthquakes

USGS PCMSC Tsunami and Earthquake Studies


Preliminary Analysis of the September 29, 2009 Samoa Tsunami, Southwest Pacific Ocean

The tsunami that was triggered by a magnitude 8.0 earthquake on September 29, 2009, caused significant damage and loss of life on Samoa, American Samoa, and Tonga. In the hopes that disasters such as this can be minimized in the future, we attempt to understand the mechanism and impact of this tsunami. The information presented here is focused on geologic aspects of the disaster.

For More Information on the Earthquake from the National Earthquake Information Center:

For More Information on the Tsunami:

Tectonic and Seismological Background

The September 2009 Samoa tsunami was generated by an unusual type of earthquake that occurs near ocean trenches. Unlike typical tsunamigenic earthquakes that occur on the thrust fault that separates tectonic plates in a subduction zone (termed the inter-plate thrust), outer-rise earthquakes, as they are called, occur within the subducting or downgoing plate before it enters the subduction zone (Figure 1). There have only been a few verified instances of tsunamis generated by outer-rise earthquakes, but those that have occurred have been devastating. The 1933 Sanriku tsunami generated from a magnitude 8.6 outer-rise earthquake resulted in over 3,000 deaths in Japan and significant damage on the Island of Hawai'i. The 1977 Sumba magnitude 8.2-8.3 outer-rise earthquake resulted in 189 deaths in Indonesia. The 2009 Samoa outer-rise earthquake may have resulted in comparable fatalities and was the fourth largest outer-rise earthquake that has been instrumentally recorded since 1900.

Outer-rise earthquakes are caused by stresses in the subducting, oceanic plate induced by bending as the plate enters the trench.

Schematic diagram of subduction zone showing location of outer-rise fault
Figure 1. Schematic diagram of a subduction zone, showing the location of the outer rise and tensional stresses within the subducting plate.

Flexure of the plate elevates the sea floor, creating an oceanic feature known as an "outer rise" that parallels the oceanic trench. As the plate flexes, tensional stress in the oceanic crust creates potentially large normal faults. Crustal stresses caused by earthquakes on the inter-plate thrust fault in subduction zones can also be transferred to the outer rise, triggering earthquakes on normal faults that are already close to failure.

At the Tonga trench, the Pacific plate entering the subduction zone is particularly old and dense, resulting in a steep angle of descent and many normal faults near the trench. The 2009 Samoa earthquake occurred east of the Tonga trench, near the northern terminus of the Tonga volcanic arc where the trench takes a sharp bend to the west (Figure 2).

Location of earthquake relative to Tonga and Samoa

Figure 2. Location of 2009 epicenter and that of a similar earthquake in 1917 (white circles) in relation to the bend in the Tonga trench (yellow line).

Normal faults in the outer-rise and trench slope change orientation from NNE-SSW trends near the main part of the Tonga trench to E-W trends near the part of the Tonga trench between the northern Tonga islands and Samoa. The fact that the normal faults are nearly parallel with the trench suggests that the faults occur primarily in response to bending stresses in the oceanic plate. Another possible outer-rise earthquake occurred in 1917 (see Figure 2) that resulted in a maximum runup of 12 m on Samoa. These and other tectonic characteristics similar to the Tonga outer-rise region are also found in many other subduction zones throughout the world's oceans.

When a fault ruptures beneath the sea floor, the rocks surrounding the fault are permanently uplifted and downdropped, with the ocean going along for the ride to generate the tsunami. Rupture of an inter-plate thrust at a subduction zone typically occurs below a substantial thickness of sediment, and the rupture does not reach the sea floor (see subduction-zone diagram). In contrast, outer-rise normal faults typically rupture brittle oceanic basalt in a region that usually has very little sediment cover, and so the rupture commonly reaches the sea floor. For this reason, it is likely that the fault that ruptured during the 2009 Samoa earthquake can be mapped using marine geophysical techniques.

Detailed images of the sea floor can be obtained using multibeam mapping techniques. Prior to the 2009 earthquake, two multibeam surveys were conducted of the Tonga Trench near the epicenter. The first was conducted in 1996 by Prof. Dawn Wright at Oregon State University, while the second was conducted in 1998 by Scripps Insitution of Oceanography (Figure 3).

Composite multibeam image of the sea floor near the Samoa 2009 epicenter

Figure 3. Map and perspective view of the Tonga Trench near the earthquake epicenter. Shaded relief bathymetry generated from multibeam data available from the National Geophysical Data Center (NGDC) (Survey KIWI11RR, Chief Scientist Nancy Kanjorski. XML metadata) and Oregon State Univ. (courtesy Prof. Dawn Wright), along with ETOPO-1 bathymetry data (metadata). The vertical exaggeration of the perspective view is 4x. Approximate location of epicenter (surface projection of earthquake’s origin) shown by orange circle. [Larger version]

A series of prominent normal faults are observed as the trench curves around the northern Tonga arc.

Preliminary analysis of how much the fault slipped during the September 29, 2009, earthquake is given by the finite-fault model on the NEIC event page. This analysis of seismic waveform data indicates that there was a large amount of slip near the sea floor (Figure 4), further suggesting that the earthquake may have produced a mappable step on the sea floor where the causative fault intersects the surface.

Slip distribution for the 2009 Samoa earthquake

Figure 4. Preliminary slip distribution from the finite-fault model on the NEIC event page. Vectors indicate magnitude and direction of slip within the rupture area.

Preliminary Simulation of Tsunami

Many of the aforementioned characteristics of outer-rise earthquakes can explain why the tsunami was so large. The maximum fault slip for this earthquake is much higher than for an inter-plate thrust earthquake of comparable magnitude. This translates to higher vertical movement of the sea floor, lifting the entire ocean above the rupture zone. Moreover, tsunami generation occurred in much deeper water than normal. When a tsunami travels from deep water to shallow water, the speed of the wave crest or trough slows, the wavelength decreases, and the amplitude increases. This process is sometimes referred to as “shoaling amplification”. (See Life of a Tsunami.) If a tsunami starts off in deeper water, then it will be amplified to a greater extent by the time it reaches shore than a comparable tsunami that starts off in shallow water. Preliminary field survey data of American Samoa indicates that the tsunami runups (height above ambient sea level) reached as much as 12 m, which is larger than most tsunamis generated by magnitude 8.0 earthquakes on the inter-plate thrust.

Figures 5 through 7 below show snapshots of a the preliminary tsunami simulation as viewed from different directions.

Initial tsunami wavefield - regional view to east
Figure 5. Initial tsunami wavefield. Regional view to the east.
See the animation (1 MB).

Tsunami wavefield at 2.9 minutes - local view to northwest
Figure 6. Tsunami wavefield at 2.9 minutes. Regional view to the northwest.
See the animation (1.2 MB).

Initial tsunami wavefield - local view to west
Figure 7. Initial tsunami wavefield. Regional view to the northwest.
See the animation (1.3 MB).

It is hoped that continued research on the nature and occurrence of outer-rise earthquakes around the world will help identify potential sites for future outer-rise earthquakes of this size and help mitigate the tsunami hazard associated with such rare but devastating events.


Accessibility FOIA Privacy Policies and Notices logo U.S. Department of the Interior | U.S. Geological Survey
Science questions to: Eric Geist
Site maintained by: Laura Zink Torresan
Page last modified: 2 June 2016 (lzt)