Tsunami Study

What is a Tsunami?

Fig. 1

Tsunamis are waves in the ocean or lakes generated by abrupt vertical movement of the seafloor or lake floor by earthquakes (Fig. 1), volcanic eruptions, landslides, and, rarely, asteroid impact. Tsunamis generated by submarine earthquakes travel at subsonic speed across the ocean surface.

In the open ocean tsunami waves are generally only a few tens of centimetres high, but as they approach shore they may grow to heights of 10 m or more (Fig. 2, 3). Because of this, they may cause fatalities and destroy property around the entire margin of the ocean basin.

Fig. 2

Unlike earthquake-triggered tsunamis, the waves created by landslides, whether triggered by earthquakes or not, have only a local impact. The tsunami hazard at a particular location along a coast depends on the tsunami potential of the local area, the exposure of the shoreline and the offshore and onshore topography of the coastal zone.

Fig. 3

Potential Sources of Tsunamis at the Fraser River Delta
Tsunamis that might reach the Richmond foreshore have three possible sources: (1) great earthquakes beneath the seafloor at subduction zones bordering the Pacific Ocean; (2) smaller, although still large earthquakes at shallow crustal depths beneath the Strait of Georgia or Juan de Fuca Strait; and (3) a large collapse of the front of the Fraser River delta into deep waters of the Strait of Georgia west of Richmond.

Subduction Zone Earthquakes
The tectonic plates that underlie the Pacific Ocean are being subducted beneath continental margins around the perimeter of the ocean basin. At each subduction zone the interface between the converging oceanic and continental plates is locked for long periods of time, and the accumulating strain compresses and deforms the continental margin. The accumulated strain is released every few decades or centuries in great (magnitude (M)=8) plate-boundary earthquakes. During these earthquakes the sea floor above the locked zone abruptly rises, generating a tsunami, and the coastal zone suddenly subsides, with flooding of low-lying areas. After the earthquake the plate interface relocks, and the cycle begins again. The great (M=9.3) earthquake that occurred off the coast of Sumatra in 2004 is an example of this rupture process.

Cascadia Subduction Zone

Fig. 4

Prior to 1987, assessments of the tsunami hazard on the west coast of North America were based on the impact of distant Pacific-wide tsunamis, but in the last two decades scientists have recognized that the west coast of North America is also vulnerable to tsunamis generated at the Cascadia subduction zone, where the Juan de Fuca plate is being thrust beneath the North America plate, from northern California to central Vancouver Island (Fig. 4).Studies of tidal marshes on the Pacific coast of North America that suddenly subsided during great (M >8) earthquakes show that the Cascadia subduction zone ruptured five times in the last 2600 years. Sheets of tsunami-deposited sand that record these, and earlier, plate-boundary earthquakes are widespread beneath marshes along the Pacific coasts of Oregon, Washington and British Columbia, at sites at the eastern end of Juan de Fuca Strait, and in some low-lying coastal lakes.

Fig. 5

Are tsunami waves triggered by a great plate-boundary earthquake at the Cascadia subduction zone a threat to lives and property in the coastal lowlands of the Strait of Georgia? The last great earthquake at this plate boundary occurred in AD 1700 (Fig. 5), and although there are no written records of the size or impacts of these waves, their devastating impact on coastal settlements on the outer coast of Vancouver Island is recorded in Native oral histories. As far as we are aware, however, there are no equivalent oral traditions of tsunamis from this event in the Strait of Georgia.

Fig. 6

Although direct evidence of tsunami impacts is lacking, a computer model  showing the propagation of tsunami waves from a great earthquake releasing 500 years of accumulated strain on the Cascadia subduction zone has recently been developed by a group of oceanographers at the Institute of Ocean Sciences in Sidney, British Columbia. Their simulation model predicts that a great earthquake at the Cascadia subduction zone will generate tsunami waves about 5–10 m high on the outer coast (Fig. 6). These large waves gradually diminish in height as they move through Juan de Fuca Strait and the narrows between the San Juan and Gulf Islands. The leading edge of the first wave is forecast to reach Boundary Bay on the southern foreshore of the Fraser River delta about 2 hours and 5 minutes after the earthquake. Because Boundary Bay is oriented at right angles to the direction of wave travel, this wave grows to a height of about 1 metre. The second wave, which is approximately the same size as the first, arrives at about three hours and 30 minutes. A third, slightly smaller, wave arrives at about five hours. By six hours this wave grows to almost 2 metres in height. Unlike Boundary Bay, the western foreshore of the Fraser delta lies parallel to the direction of wave travel, and the maximum wave heights on this foreshore are forecast to be much smaller (less than 0.5 metre).

If these model predictions are valid, we conclude that the sea dikes of the southern margin of the Fraser delta may be overtopped by tsunami waves about two hours after a great earthquake if the tsunami arrival coincides with a high tide and a strong onshore wind. At low tide the sea dikes would have more than two metres of freeboard and thus would restrain a tsunami wave. If the sea dikes had been weakened, or had collapsed, as a result of earthquake shaking, they might be locally breached by the tsunami even at low tide.

Other Subduction Zones Around the Pacific Rim
Of the subduction zones that surround the North Pacific, only the Alaska-Aleutian margin represents a significant tsunami threat to the west coast of Canada. Great earthquakes have ruptured this subduction zone six times in the last 4000 years. On the last occasion (27 March 1964; M = 9.2) tsunami waves up to six metres high devastated several communities on the outer coast of Vancouver Island. The islands at the north and south ends of the Strait of Georgia, however, effectively blocked the tsunami from reaching inland waters. At the east end of Juan de Fuca Strait, the largest wave from the 1964 Alaska tsunami was one to two metres high, whereas in the southern Strait of Georgia it diminished to less than 0.5 metre. We conclude therefore that tsunamis triggered by distant plate-boundary earthquakes do not constitute a significant source of hazard at the Fraser River delta.

Local Earthquakes at Shallow Crustal Depths
Earthquakes on faults within the North America plate represent an additional tsunami hazard to coastal communities of the Pacific Northwest. Faults in central and northern Puget Sound are known to have ruptured in large (M 7) earthquakes in the last few millennia. Some low-lying areas around Puget Sound were flooded by the tsunamis generated by these earthquakes. The narrow, winding passages of this inland sea, however, cause rapid loss of tsunami wave energy, and we consider it highly unlikely that tsunamis generated by submarine earthquakes in Puget Sound or Juan de Fuca Strait have ever inundated the lowlands of the Fraser River delta.

What is less certain, however, is the tsunami potential posed by submarine faults beneath the Strait of Georgia? If any of these faults are active, and if the seafloor is displaced during a future earthquake, then the ensuing tsunami would likely present a hazard to coastal areas in the Strait that are orientated parallel to the fault zone. The southern foreshore of the Fraser River delta is therefore likely at greater risk than the western foreshore from seismic tsunamis generated by east-west trending fault zones beneath the southern Strait of Georgia.

Collapse of the Front of the Fraser River Delta
A strong local earthquake could conceivable cause a landslide at the front of the Fraser River delta, where it slopes into deep waters of the Strait of Georgia. The landslide, if large and sudden enough, might generate a tsunami. The unconsolidated sediments forming the delta front may also fail without a seismic trigger, and yet still produce a tsunami.

Landslide-induced tsunamis are particularly dangerous because the waves may locally be very large and the warning time very short. For example, in November 1994 a submarine slide in Taiya Inlet created a wave that reached a height of nine to eleven metres at the shoreline in Skagway, Alaska, causing one fatality and over $20 million of damage.

It has long been recognized that the western front of the Fraser delta is at risk from submarine landslides. The Fraser River discharges about 17 million tons of sediment into the Strait of Georgia each year, and much of this sediment accumulates on the steep frontal slope of the delta. Small slides are common in this unconsolidated material; however, these slides are shallow-seated and move down the delta front over a period of hours, consequently they do not produce tsunami waves.

Researchers have investigated the ability of large submarine slope failure at the southwestern delta front. They conclude that a large slide could generate tsunami waves up to 18 m high on the east shores of Galiano and Mayne Island, but that the tidal flats of the delta foreshore would deflect and reduce wave energy, meaning that waves at the shoreline in Tsawwassen would likely not exceed two metres.

Documenting Past Tsunamis from the Geological Record
Has the Richmond area ever experienced a catastrophic tsunami? No tsunami has affected this area in the historic period (the last 100 years), but what about further back in time. Scientists can answer this question by searching for telltale tsunami deposits in the sediments of the Fraser River delta.

Fig. 7

This approach is based on the observation that when tsunami waves recede, the inundated area can be determined by mapping the sediments they have left behind. If new soils are quickly deposited on top of the tsunami sediment, the tsunami event is archived in the geological record (Fig. 7). If these tsunami deposits can be identified and dated, then it becomes possible to reconstruct the history of tsunamis at that location. This approach has been used to determine the incidence of tsunamis on the outer coast of Vancouver Island, in Puget Sound, and Juan de Fuca Strait. Tsunami deposits, characteristically, are:

• widespread in areas near the shore,
• distinctly different in character from the surrounding material and have a sharp base, 
• thin landward (because the wave loses energy and drops the sediment it carries),
• fine landward (e.g. pebbly sand becomes sandy mud), and
• contain the remains of marine organisms not found in the surrounding material.

2005 Richmond-Delta Study

Fig. 8

Professors John Clague (Department of Earth Sciences, Simon Fraser University) and Ian Hutchinson (Department of Geography, SFU) completed a study of near-surface sediments beneath the plain of the Fraser delta in the summer of 2005. Their objective was to determine whether or not any prehistoric tsunamis had crossed what are now Richmond and Delta. A total of 33 cores were retrieved from sites (Fig. 8) where the likelihood of tsunami deposit preservation was highest.

Fig. 9

The cores likely preserve sediments dating as far back as 4000 years ago. The investigation revealed no evidence of tsunami deposits in the Fraser River delta. Drs. Clague and Hutchinson could not completely rule out the possibility that tsunamis have inundated portions of the Fraser River delta in the past, but that the tsunami threat to Richmond and Delta is very small (Fig. 9).