Atlantic Tsunamis: "Like a River Returning"
by Alan Ruffman, P.Geo.
Illustrated talk in the Small Craft Gallery
Maritime Museum of the Atlantic
Tuesday, January 18, 2005, 7:30 p.m.
in support of Tsunami Relief efforts in the Indian Ocean by Oxfam Canada
January 12, 2005
revised January 15 and 18 and February 18, 2005
What's in a Name
A 'tsunami' is the ocean wave, or waves, created in the ocean, or in a body of fresh water, when something abruptly moves a large volume of water (Martitia Tuttle, New York Times, Long Island Edition, January 16, 2005). The sudden displacement can be caused by a large earthquake rupturing the ocean floor, of by something plunging into the body of water such as a landslide, a calving glacier, a meteorite, or by an explosion in the water.
The name 'tsunami' comes from the Japanese and means 'harbour wave'. Clearly tsunamis are not just confined to harbours. However, it was generally around harbours where the earliest Japanese coastal communities were located. By its very nature, a harbour is often in a coastal indentation that narrows and shallows landward -- hence focusing the energy of an incoming tsunami wave. It has nowhere to go but to build in height, and often begins to lose energy closer to shore by becoming a breaking wave as it rolls into shore.
The Japanese written historical record is very long and documents many Pacific tsunamis; it is only fitting that their word be adapted in other languages as the name of this natural phenomenon. A tsunami often comprises several significant waves and a myriad of smaller, complex, often derived, reflected and refracted waves. A tsunami should never be called a 'tidal wave'; it has nothing to do with the lunar or solar tides.
Nature of the Wave
In the deep, open ocean, the height, or amplitude, of a tsunami wave may be from a fraction of a meter, up to two or three meters above sealevel, with a period of about one hour, a wavelength of about 100 km, and a velocity of up to 600 to 700 km/hr. Thus if you were able to measure your absolute motion in an anchored vessel over an hour, you would rise perhaps 2 m above mean sealevel and sink 2 m below mean sealevel. The higher the amplitude and the higher the velocity of the tsunami, the more energy it has, and the more dangerous it is.
As a tsunami passes into shallower water such as coming onto a continental shelf, or passing over a mid-ocean ridge, it begins to "feel" the bottom -- i.e. it begins to experience frictional losses of energy. Thus a broad continental shelf, a wide coastal expanse of shallow coral reefs or a broad shallow delta front can sap energy from the tsunami waves and offer the coastal area some degree of protection. In the last stages of its approach to the shoreline, the shallowing depths will cause a tsunami to rise in height and to eventually become a breaking wave -- analogous to an ocean swell coming into a beach like Lawrencetown, Nova Scotia. The water falling down the front of the breaking wave is mechanism to loose energy.
The speed and long wavelength of the tsunami means that it possesses tremendous energy, thus as it crashes onto a shoreline there is a great deal of momentum in the water behind the breaking wave so the sea appears to rise as a sudden influx of the "tide". That excess height and momentum can propel the water inland to even greater heights -- hence the runup height of a tsunami on the shore, or at the head of a bay, can be much higher than the tsunami wave amplitude. Thus in the 1929 Newfoundland tsunami the tsunami wave heights were 3 m in St. Lawrence Bay and 7 m in Taylor's Bay, but the runup heights at the heads of the bays were about 13 m.
As the tsunami begins to "feel" the bottom, it progressively interacts with the seafloor sediment, disturbing it and moving it. In the end as the wave propels itself onto shore, the landfalling tsunami, "like a river returning" as one of my 1929 Burin witnesses put it, exercises all the erosive power of moving water entraining sediment and carrying it inland along with any other debris, pieces of broken human infrastructure, vehicles, trees, and humans and livestock all fighting for their lives. The riveting images from video cameras during the December 26, 2004 Indian Ocean event has brought home to all viewers, geologists included, the tremendous power of such an event.
Eventually the shoreward moving water rises over the land to such a height that all its forward kinetic energy is converted to potential energy. As it flows onto higher ground, it slows, stops its forward motion, then begins to drain back to the sea under the influence of gravity. The arrival of a second, or third pulse, or more, makes the survival of trapped victims in the stream even more problematic.
Origins of Tsunamis
Tsunamis can be caused in a number of ways. Large magnitude earthquakes greater than circa 6.5 on the open-ended Richter Scale can break the ground at the earth's surface. When this occurs underwater it is like hitting the outside of an above-ground swimming pool with a sledgehammer; a fast-moving wave is created that begins to propagate in all directions from a rupture. A strike-slip rupture may not create much of a tsunami, but a normal, or vertical, fault may lift one part of the ocean floor and drop another.
The December 26, 2004 Sumatra offshore earthquake was a 'subduction' earthquake exactly analogous to the massive earthquake and tsunami that occurred on January 26, 1700 along the Cascadia Subduction Zone that lies off California, Oregon, Washington, and Vancouver Island. In this area not far offshore, an east -moving oceanic tectonic plate is slowly passing down under a continental tectonic plate off Vancouver Island, or, in the case of the December 26th event, the Indian Ocean tectonic plate is 'subducting' to the east down under the Burma-Indonesian Plate. If stress is constantly released with a series of little earthquakes, there is little, or less, concern. However, when the subducting plate becomes locked with the overlying plate stress begins to accumulate and the eventual break and stress release gives a much more powerful seismic event. Subduction events are often in the range of 8.0 to 9.5 magnitude, and can be 'tsunamigenic' if they occur offshore. Not all large earthquakes centred under the ocean create tsunamis. On December 24, 2004 a large marine earthquake occurred in the Pacific Ocean near MacQuarrie Island off New Zealand, and no tsunami resulted. This earthquake was on a different geological structure, and was not related to the event two days later off Banda Aceh, on the island of Sumatra in the Indian Ocean.
Tsunamis can be generated when a large volume of ocean floor sediment moves either because it is shaken loose by an earthquake, or slumps spontaneously and moves downslope as an underwater landslide; this is what happened south of Newfoundland in 1929. The November 1, 1755 Lisbon Earthquake and tsunami may have had the same cause; in this case a significant tsunami flowed up the Tagus River into downtown Lisbon and 20,000 persons died in the city, with up to 80,000 elsewhere along the west coast of Europe and Africa.
If a volcanic island becomes over-steepened as the volcanic cone grows upward, a large slab may slough off and slide precipitously into the sea creating huge local tsunamis that have the potential to propagate across oceans. In the Atlantic the Canary Islands, The Azores, and the Cape Verde Islands all represent real but rare threats. The seafloor below the Hawaiian Islands shows ample evidence of slide scars; the problem is we do not know whether these were catastrophic slides, or slow creep over a long period of time.
Earthquakes, or simple freeze-thaw cycles, can trigger onshore landslides down the slopes of mountains directly into the sea and do the same thing. Thus on July 7, 1958, an 8.0 earthquake shook loose a massive landslide that plunged directly into Gilbert Bay at the head of an Alaskan panhandle fjord. This caused a huge local tsunami of up to 551 m in height above sealevel which sheared off the meter-thick west coast rain forest over a 16 km length as it ran down the length of Lityua Bay and over the baymouth bar into the Pacific Ocean; the tsunami was still 30 m high as it exited the fjord. In 2001 in northwest Greenland, a landslide into the sea created a local tsunami of 50 m height, and in Western Brook Pond of Gros Morne National Park, a 30 m tsunami was created when 'Broke Off Cliff' let go in the Fall of about 1905-1910.
If a volcanic island simply explodes or collapses, a huge tsunami can be created as was the case on August 26, 1883 when Krakatoa vaporized at the other end of Sumatra from Banda Aceh; an estimated 36,000 persons died. The Santorini volcano erupted and collapsed in 1490 off southwest Greece and created a tsunami that caused great loss of life all around the Aegean Sea. The detonation of the vessel MONT-BLANC in Halifax Harbour on December 6, 1917 created a short-lived local tsunami that locally rose above Campbell Road (Barrington Street) in Halifax's north end. Autopsies were not done on the Explosion's 1,950 victims, so we have no idea how many died of drowning as opposed to from the explosion shock and being hit by shrapnel.
Meteorological events with significant rapid atmospheric changes can give rise to 'tsunami-like waves' called by some 'meteorological tsunamis' or 'risaggas'. Eastern Newfoundland experienced two such daylight events in 1999 and 2000 in bright, clear weather as a result of hurricanes José and Hélène passing over the Tail of the Banks several hundred kilometres offshore. In 1938, the serious Category 3 New England Hurricane came ashore in Rhode Island, but rather destructive reflected and refracted tsunami-like waves were seen along the New Jersey coast from New York southwest to Cape May, New Jersey, well after the hurricane had gone by offshore.
A large bolide, or meteorite, hitting an ocean will create a very, very large tsunami that probably none of us even want to think about!
November 18, 1929; Canada's most tragic known, historic earthquake
The earth's continents are constantly eroding, and the erosional detritus is carried mainly by rivers (but on occasion by wind and by glacial ice) to settle in the sea on the continental shelves. Ocean waves and tidal currents sort the material over time and move it to, and over, the continental shelf edge. Until the early 1950s, the processes of moving these erosional sediments down into the deep ocean was not really known or understood by earth scientists.
It was an event quite unknown in the lives of most who felt it in Atlantic Canada. The surface wave magnitude (Ms) 7.2 earthquake of Monday, November 18, 1929 struck at 5:02 p.m. N.S.T. in the late afternoon -- seventy-five years ago. The hypocentre was some 18 km below the seafloor of the northwest Atlantic Ocean, at the mouth of the Laurentian Channel in 2 km of water depth on the continental slope south of the Burin Peninsula on the south coast of what was then the British colony of Newfoundland. It was felt as far away as Montréal, in the New England states as far south as New York City, and there is even a serendipitous felt report in Bermuda of a probable seismic 'surface wave'; it registered on seismographs around the world. It is still remembered by older residents of the Atlantic Provinces as the only felt earthquake experienced in their lives. Onshore the damage from the earthquake's shaking was restricted to some slumping and minor building damage in Cape Breton Island; some chimneys were dislocating resulting in subsequent chimney fires in the next few days. Newfoundland, despite its proximity to the epicentre, experienced virtually no physical damage onshore.
Two-and-a-half hours after the event, on a dead calm, bright, moonlit night, on a rising high tide, three main pulses of a tsunami arrived, quite unexpectedly, along the coast of the Burin Peninsula, with amplitudes of 2 to 7 m. There was an initial slow withdrawal of the sea to expose ocean floor in places never before seen by local inhabitants, then the water returned in three positive pulses that rose 2 to 7 m above sealevel. The height and forward momentum of the arriving tsunami caused the runup to rise to as much as 13 m above sealevel at the ends of the long narrow harbours such as Port au Bras, St. Lawrence, Little Lawn Harbour, Lawn, Lord's Cove, Taylor's Bay, and Lamaline. Twenty-eight persons lost their lives, and the fishing capability of the coastal communities was devastated.
There was as yet no road to connect the communities to each other or to link the Burin Peninsula to the rest of Newfoundland to the north. Landline telegraph communications with the rest of the Island had been broken by a storm two days earlier, and the tsunami took out the land lines between the coastal communities. In St. Lawrence the telegraph station ended up floating in the harbour. The Burin had to cope on its own for two-and-a-half days before a coastal ferry named the PORTIA, which had a working wireless radio, arrived on the scene. Despite the success of wireless 17 years earlier during the TITANIC disaster, the local communities had no radio sets, and while a wireless was available on the DAISY situated in Burin harbour, no-one knew how to operate it to get a message out!
The tsunami was seen in Cape Breton Island, Nova Scotia, at about 8:00 p.m. A.S.T. on November 18th, where it did minor damage. The one possible death in Nova Scotia has been shown to be false and was based on incomplete information. The tsunami refracted counterclockwise around the Avalon Peninsula to arrive in the Bonavista area about 1:30 a.m. N.S.T. the next morning. The tsunami was physically seen along the coast of Nova Scotia as far southwest as Lunenburg, and in Bermuda at about 8:00 p.m. local time in the evening. It rose in Halifax Harbour, where it flowed over the gates of the commercial drydock at Halifax Shipyards for five minutes and is recorded on the tide gauge record. The only tide gauge operating in Atlantic Canada to record the tsunami was in Halifax; the British had not yet installed a tide gauge anywhere in Newfoundland (or in Bermuda).
The tsunami travelled at about 615 km/hr south and eastwards in the deep ocean; the tsunami travelled at about 105 km/hr over the shallower continental shelf of Canada north and westwards. The tsunami was recorded on tide gauges as far south as Charleston, South Carolina, in the United States, in the Azores, and on the west coast of Portugal. The tide gauge records for the United Kingdom were destroyed during WW II bombings. The rather high water recalled by many Newfoundlanders as the "tidal wave" on the next morning of Tuesday, November 19, 1929 was not the tsunami. It was a significant storm surge of an early winter storm that had tracked up the Atlantic coast from New England and the Maritimes over the past day. It snowed that day on the Burin and turned bitterly cold, making life even more miserable for people affected by the tsunami.
At the instant of the earthquake, five transAtlantic telegraph cables broke in numerous places near the top of the continental slope as the underwater landslides began to move down into deeper water. Over the next two hours, seven more cables parted progressively in deeper and deeper water, and more distant from the initial breaks. The repairs to the twenty-eight breaks in the twelve transAtlantic telegraph cables required all available cable ships, and repairs stretched well into 1930. At the time, the mechanism of the seafloor disruption was not understood, and was not successfully worked out for some 23 years. It is now known that the earthquake's strong vibrations shook loose and mobilized up to 200 cubic kilometres of ocean floor sediments on the continental slope. The underwater slump, or landslide, travelled downslope, initially at speeds of up to 50 to 70 km/hr, as a slurry of water and sediment, now called a "turbidity current".
The turbidity current then slowed and eventually travelled over 1,200 km from its source out across the Sohm Abyssal Plain, laying down a thin layer of graded sediment -- material that had initially been deposited over thousands of years on the upper Continental Slope, and was now in 3,000 to 4,000 m water depth. This process of filling the ocean basins by an ongoing series of turbidity currents is now recognised as a very important final step of a process that moves sediments from the rivers and coasts of the continent, out onto the continental shelves, over the shelf break, and then down into the ocean basins as turbidity currents.
There is still an ongoing debate as to whether other earthquakes, in what is now known as the Laurentian Slope seismic source zone, could cause other slumps and tsunamis. One other apparent slump was reported nearby in October of 1884 when three transAtlantic cables all broke in one area at about the same time, south of the Tail of the Banks. The continental slope of Atlantic Canada, when mapped by deep ocean sidescan sonar, or other swath-mapping multi-beam sonar techniques, shows substantial evidence of other downslope mass movements, though the ages and frequency of these events is not known. A number of the apparent slump scars may reflect slow creep events rather than catastrophic landslides such as occurred 75 years ago south of Newfoundland. Only the rapid underwater landslides will create a tsunami. Much like moving one's leg rapidly to one side in a bathtub, the rapid movement of the ocean floor creates a very long wavelength, low amplitude, fast-moving, gravity wave on the ocean's surface -- a 'tsunami'.
It is popularly believed by many Newfoundlanders that the collapse of the fisheries and the loss of the eel grass in the early 1930s were a direct result of the 'tidal wave'; this observation appears to be unfounded. The onshore geological signature of the 1929 tsunami has been found in many of the harbours along the south coast of the Burin. At Taylor's Bay the tsunami's signature clearly shows as a band of white sand about 10 cm down in the brown peat (see photograph at the Dalhousie University Department of Earth Sciences website ). A case study at St. Lawrence, and a careful mapping survey in Taylor's Bay, have allowed the zone of tsunami runup to be mapped. In St. Lawrence, community growth has gone forward without regard to the 1929 runup zone or a possible recurrence. In contrast, the village of Taylor's Bay has never recovered from its losses on that fateful November 18th evening. Documenting of community folklore has allowed a rich oral history of the event, songs, stories, poems, photographs, and myths surrounding the event, to be documented throughout the Burin Peninsula.
The Storegga Slide and Tsunami
Since 1965 scientists in Scotland, England, Norway, Denmark, and Iceland have found the onshore signature of a massive pre-historical tsunami that is believed to have been caused by the Storegga Slide on the Continental slope of midwestern Norway. This slide moved in the order of 2,300 m3 of material into the abyssal depths. The resulting tsunami has left its onshore signature all around the Norwegian Sea, as far south as northern England, as far west as the Hebrides, in the Shetlands, in the Faroe Islands, on the north coast of Iceland, and all along the western coast of Norway, where post-glacial rebound of the coastal regions has lifted the deposits many meters above present-day sea-level.
The event is thought to have been a Fall event. It is known to have terminated human habitation at an archaeological site in downtown Inverness, Scotland. This tsunami certainly reached the coast of Canada and the U.S., though no modelling has yet been done to show its potential amplitude on this side of the Atlantic. There is a good chance that this event will have taken the lives of first nations peoples camped along the east coast of Canada, and that the Storegga tsunami's onshore signature can still be found by geologists.
Warning Systems and Canada's Response to the Banda Aceh Event
As we now know, deadly tsunamis are not confined to the Pacific. There has been a tsunami warning network in the Pacific for the past 56 years. The impetus for the Pacific Tsunami Warning System came from the April 1, 1946 Unimak Island, Alaskan earthquake and tsunami which caused the loss of 165 lives in the Pacific Ocean; 96 were lost in Hilo and another 25 in Laupahoehoe, Hawaii, 22 on Maui, 10 on Kuaui and 6 on Oahu. Five lives were lost in Alaska and another in Santa Cruz, California. A tsunami warning system is now being considered for the Indian Ocean, and in recent days Stephen Ward of the University of California has advocated a similar system in the Atlantic because of the hazard posed by possible flank collapses on the Canary Island volcano Cumbre Vieja. The Americans have just announced a $37.5 million plan to strengthen the Pacific Tsunami Warning System from 6 to 31 tsunami sensing buoys and bottom pressure sensors. also planned are the first seven bottom pressure gauges and floating satellite communication buoys for the Atlantic (5) and the Gulf of Mexico and Caribbean area (2). The Mediterranean Sea has in the historic past experienced very serious tragic tsunamis; it too has no tsunami warning system.
In eastern Canada we have historical evidence of at least five natural tsunamis, the November 18, 1929 Laurentian Slope-Burin Peninsula event, the Fall 1905-1910 rockslide into Western Brook Pond, an 1864 local earthquake and tsunami seen at St. Shotts on the southwest corner of the Avalon Peninsula of Newfoundland, the September 24, 1848 tsunami seen from St. John's Harbour to Fishing Ships Harbour in southern Labrador, and the November 1, 1755 Lisbon Tsunami seen in Bonavista, Newfoundland; only the 1929 event is known to have cost human lives. Clearly it would be prudent for Canada and its Provincial emergency measures organizations to carefully review east coast Canadian concerns in light of the lessons brought to bear in the Indian Ocean on December 26, 2004.
Selected References re 1929 Tsunami on the Burin Peninsula, Newfoundland
Ruffman, Alan. 1995. Tsunami Runup Maps as an Emergency Preparedness Planning Tool: The November 18, 1929 tsunami in St. Lawrence, Newfoundland as a case study.
Geomarine Associates Ltd., Halifax, Nova Scotia, Project 94-14,
Report for Emergency Preparedness Canada, Evaluation and Analysis, Ottawa, Ontario,
Contract No. 94-D025, March 31 (revised August 9, 1995), 399 pp.
Tuttle, Martitia P., Alan Ruffman, Thane Anderson and Hewitt Jeter. 2004. Distinguishing Tsunami from Storm Deposits in eastern North America: The 1929 Grand Banks Tsunami versus the 1991 Halloween Storm.
Seismological Research Letters, Vol. 75, No. 1, January/February, cover photo, pp. 117-131.
Research Associate, Maritime Museum of the Atlantic
President, Geomarine Associates Ltd.
P.O. Box 41, Station M
Halifax, Nova Scotia B3J 2L4
phone/fax (902) 477-5415