6
Terror Waves: Tsunami

The ancient Polynesians believed that earthquakes were caused by the god Ruau-Moko, the youngest son of their ancient gods. Ruau-Moko had remained in the Earth Mother’s womb and it was his movements that caused earthquakes. Other ancient civilizations had similar legends; some believed the earth is supported on the backs of turtles or, as in Mongolia, on the back of a giant frog. How else could we explain how solid earth and rock could suddenly move?

THE RING OF FIRE

Today we know that earthquakes result from the movement of tectonic plates—massive layers of stone deep within the earth that over the ages slowly move and grind against each other. At some point, the enormous forces overcome the friction of rock against rock, and one plate will suddenly move relative to another, deep within the earth, or under the sea, where a tsunami can be produced. The Pacific Ocean is bordered on all sides by the intersection of such plates, giving rise to the numerous earthquakes that occur in Chile, California, Alaska, Japan, and other locations on the edge of this great ocean. Because of its fre-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 127
Extreme Waves 6 Terror Waves: Tsunami The ancient Polynesians believed that earthquakes were caused by the god Ruau-Moko, the youngest son of their ancient gods. Ruau-Moko had remained in the Earth Mother’s womb and it was his movements that caused earthquakes. Other ancient civilizations had similar legends; some believed the earth is supported on the backs of turtles or, as in Mongolia, on the back of a giant frog. How else could we explain how solid earth and rock could suddenly move? THE RING OF FIRE Today we know that earthquakes result from the movement of tectonic plates—massive layers of stone deep within the earth that over the ages slowly move and grind against each other. At some point, the enormous forces overcome the friction of rock against rock, and one plate will suddenly move relative to another, deep within the earth, or under the sea, where a tsunami can be produced. The Pacific Ocean is bordered on all sides by the intersection of such plates, giving rise to the numerous earthquakes that occur in Chile, California, Alaska, Japan, and other locations on the edge of this great ocean. Because of its fre-

OCR for page 127
Extreme Waves quent seismic and volcanic activity, this zone that circles the Pacific is known as the “Ring of Fire.” On the east, the Pacific Plate encounters the Nazca Plate in the south and the North American Plate in the north. On the west coast of South America, the Nazca and South American plates come together and are the source of considerable seismic activity there. In the southwestern Pacific, it is the intersection between the Pacific and Australian plates that gives rise to seismic activity. The strong earthquake belt that runs roughly east-west through the Mediterranean and Central Asia occurs along the intersection of the African and Eurasian plates. TSUNAMI CHARACTERISTICS In addition to shaking the land, earthquakes in or near the sea can create extreme waves. Scientists believe that some of the largest waves ever experienced on earth were generated by earthquakes in prehistoric times. In the decade from 1990 to 2000, 14 seismic sea waves hit somewhere in the world, resulting in extensive damage and considerable loss of life. In the Pacific region over the last 2,000 years, nearly 500,000 people have died from tsunami; in contrast, the 2004 Sumatra tsunami in the Indian Ocean alone exerted a death toll now estimated at more than 280,000 persons.1 Ironically, the Indian Ocean has had little tsunami activity in recent history, although the circular area south from Myanmar to Indonesia and east and north along the Philippines bounds the Eurasian Plate—a hotbed of seismicity over the last 100 years. There have been nine magnitude 8 earthquakes since 1900. The magnitude 9 Sumatra-Andaman Island earthquake of December 26, 2004, occurred along this fault zone. The lack of tsunami associated with these numerous earthquakes created a false sense of security in the region. Seismic sea waves—often referred to incorrectly as “tidal waves”—are called tsunami, which is Japanese for “large wave in harbor,” to distinguish them from tide waves. The same spelling is used for the singular or plural form. Tsunami are quite different from storm-generated waves in that they have long wavelengths between 6 and 300 miles. Recalling from Chapter 1 that “deep” water for a wave is defined as a

OCR for page 127
Extreme Waves water depth equal to or greater than half a wavelength, and recognizing that the oceans are generally no deeper than 3 miles with an average depth of around 2.5 miles, it is clear that the entire ocean is “shallow” to most tsunami waves. Hence the speed of a tsunami depends solely on water depth (as in the case of wind waves approaching the shore), but more significantly, the entire column of water—from surface to sea bottom—moves in the case of a tsunami. Since the speed of the wave depends only on depth, in the open ocean—where the average depth is 2.5 miles—the speed of the wave is 384 knots (442 miles per hour). Near an earthquake source in the open sea, the wave height may be only 3 to 6 feet and would not be noticeable to a large vessel. As the wave approaches the shore, things get more interesting. As the depth decreases, the wave slows, causing the wavelength to decrease and the moving mass of water to “pile up,” dramatically increasing the wave height. When this massive wave runs up on the shore, it sweeps all before it, carrying boats in the harbor onto dry land and destroying waterfront installations and buildings inland. Some buildings that survive the initial onslaught will be swept out to sea as the water floods back into the ocean. If the initiating earthquake caused the seafloor to rise, the crest of the tsunami will reach the shore first. If the floor sinks, a trough is created, followed by a crest. In this case, the first event on shore is that the water recedes a great distance. Spectators who rush to the seashore or harbor to look at the receding sea find a few minutes later that they are doomed by a great wall of water when the crest arrives (see Figure 13). FIGURE 13 Tsunami wave height and run-up.

OCR for page 127
Extreme Waves Geoscientists have recently uncovered evidence that huge sea waves—possibly caused by a massive submarine earthquake, an underwater landslide, or a meteorite or comet impact in the ocean—have swept inland for as much as 3 miles in Australia, inundating hills as high as 200 feet! The evidence for such events comes from places where the impact of giant wave-tossed boulders has left star-shaped fracture patterns in rock faces, where coastal hills have been carved by wave action, or where beach soils and shells have been deposited far inland.2 TSUNAMI IN ANCIENT TIMES Tsunami are known from ancient times—dating back almost 4,000 years in China. In the Mediterranean, perhaps the earliest written record comes from the ancient city of Ras Shamra (Syria). Near the modern city of Latakia, on Syria’s Mediterranean coast, was the ancient port of Ugarit. The city was a center of trade for the Minoans and had connections with ancient Egypt. At some time between 1400 and 1200 B.C. the city was destroyed, possibly by an earthquake followed by a tsunami. Archaeologists excavating the ancient ruins have uncovered several libraries containing hundreds of clay tablets written in four ancient languages. They detail trade transactions and legends, and mention the port’s destruction by giant waves.3 To the west of Ugarit, at nearly the same latitude, lies the island of Kríti (Crete), and about 80 miles to the north of Crete is a small group of five islands, collectively called the Santorini Islands, the largest of which is Thira (Thera). If you divide the Mediterranean into two halves separated by a line drawn south from the toe of Italy, the island of Crete would lie almost in the center of the eastern half. Within the Santorini Island group is a gigantic subsurface caldera that seismologists suspect is the remains of a huge volcano that exploded in ancient times—probably around 1600 B.C.—with a force far greater than the one that destroyed Krakatoa. The ancient island of Thera was demolished; the Santorini Islands are the remaining pieces. The Santorini Islands remain active; eruptions have continued—for example, in 1939, 1950, and 1956—ever since. Along the coast of Crete there is evidence that an ancient tsunami

OCR for page 127
Extreme Waves FIGURE 14 Map of Thira Island and vicinity. struck the island at an elevation of 300 feet. Similar evidence has been found on the coast of North Africa as far west as Tunis and to Ugarit on the east shore of Syria. However, the trace disappears along the North African coast from Banghazi, Libya, to Marsa (Matruh), Egypt—precisely in the “shadow” of Crete (see Figure 14). Such a cataclysmic event was sure to be noted by ancient people all around the Mediterranean, and indeed the ancient Egyptians were aware that something had happened, somewhere. It is possible that this event reappeared in ancient legends as a great flood. Historian Douglas Myles has speculated that the tsunami caused the parting of the waters referenced in the Old Testament when Moses led the Hebrews out of Egypt. On July 9, 551, in Lebanon, a very strong earthquake—its epicenter offshore from Byblos—devastated Beirut and caused the collapse of many buildings. It was followed by a tsunami in which the sea at first receded around 2 miles and then came roaring back to rush inland. A total of 30,000 persons died in the earthquake and tsunami.4

OCR for page 127
Extreme Waves Other early records of tsunami include a number that struck Japan; the disastrous Port Royal, Jamaica, tsunami of 1692; and the Lisbon earthquake and tsunami of 1755. Japan has a long history of earthquakes and tsunami, dating back at least 1,300 years. In A.D. 869 a tsunami killed 1,000 persons: another in 1293 killed 30,000. In 1611, an 80-foot-high wave entered Yamada Bay, killing thousands. Other notable Japanese tsunami include those of September 20, 1498, in Nankaido, Japan, which killed 26,000; October 28, 1707, in Nankaido, Japan, responsible for 30,000 deaths; and June 15, 1896, in Sanriku, Japan, where a wave 125 feet high claimed 27,000 victims. In the 400-year span from 1596 to 1996, more than 25 major tsunami hit the Japanese islands.5 TWO TSUNAMI OF HISTORICAL SIGNIFICANCE6 In 1691, William Dampier, the British seaman and explorer cited in Chapter 2, returned to England. Nine months later, he was already anxious to return to the sea, planning to go back to Port Royal, Jamaica, to engage in trade in the Caribbean. Before he could set sail, word reached England that Port Royal had been wiped out by an earthquake and subsequent tsunami. The major buildings had collapsed and some streets had dropped into the sea. Several thousand people had died; the report indicated that the living were too few to bury the dead. No records exist concerning the height of the tsunami that overran the city. However, the wave destroyed most of the shipping in the harbor, and a British frigate was reportedly carried over the tops of two- and four-story houses near the waterfront. Port Royal represents an important archaeological site because it is in effect a piece of seventeenth-century history frozen in time by the tsunami. In the late 1960s an expedition led by Robert Marx explored the site using accepted archaeological methods. Six months were spent mapping the site, and then the divers began their work. More than 50,000 items were recovered, among them household goods, including silver and pewter plates and cups, glass bottles, jewelry and silver coins, tools, ship’s fittings, and many other items. Still, only a fraction of the submerged city was explored and much more remains to be done.

OCR for page 127
Extreme Waves Lisboa (Lisbon), Portugal, was the hub of a great seafaring nation. Early in the morning of November 1, 1755, a huge earthquake occurred somewhere out in the Atlantic to the west of the city. Terrible destruction resulted—so much that although no instruments existed at that time, seismologists have estimated its magnitude as 9.0. Around an hour later, waters in the bay receded and the first wave hit with a height of 50 feet. It rushed inland, completing the destruction caused by the earthquake. It was followed by two more waves. An estimated 60,000 persons died and approximately 80 percent of the city was destroyed. EXAMPLES OF RECENT TSUNAMI Following an earthquake on August 13, 1868, a 70-foot-high tsunami swept over Arica, Peru (now part of Chile). The U.S. gunship Wateree, a side-wheel steamer, was one of several ships in the harbor that witnessed the town collapse as the earthquake struck. At first the sea receded, causing the Wateree to settle on her flat bottom. This perhaps saved her, for when the actual wave came rushing in, she was carried 3 miles up the coast and nearly 2 miles inland—over tall buildings, dams, and trees. As the waters receded, Wateree was deposited upright in the desert next to a Peruvian man-of-war, America. In this position, with no possibility of returning to the sea, the Wateree maintained operations in the desert for a while, assisting in relief efforts while waiting for a U.S. vessel to come retrieve the crew. Since the town was devastated—and much of its population dead—the crew lived on the boat. The crew planted a vegetable garden nearby, and when the captain was “piped ashore,” it was on a burro, rather than the captain’s gig.7 While tsunami are most commonly caused by submarine earthquakes, they can also be caused by landslides, volcanoes, or meteorite impacts in the ocean. Geoscientist Edward Bryant calculates that in the Pacific region during the last 2,000 years, 82 percent of tsunami have been caused by earthquakes, around 5 percent by volcanoes, 5 percent by landslides, and 8 percent by unknown means.8 The damage caused by tsunami, and the wave heights created, depend not only on the magnitude of the earthquake or other cause but also on the nearshore con-

OCR for page 127
Extreme Waves figuration of the land. Tragic and sometimes strange and miraculous events have occurred, as demonstrated by several major tsunami within the last 60 years. The April 1, 1946, magnitude 7.3 Alaska earthquake caused a tsunami that struck Unimak Island in Alaska and destroyed a lighthouse at Scotch Cap, killing the crew of five men. The wave also hit a coast guard station on the cliffs above the lighthouse (105 feet above sea level) and heavily damaged it. The lighthouse was a reinforced-concrete structure located on a shelf 46 feet above mean low-water level. The light was 98 feet above sea level. The earthquake occurred at 1:30 A.M. and was strongly felt at both the lighthouse and the station on the cliff above, but a telephone call to the lighthouse crew indicated that they’d had no damage. About 25 minutes later there was a second quake, followed by still more aftershocks throughout the night and next day. What the lighthouse crew did not—could not—know was that the earthquake had generated a tsunami that was now hurtling toward them at about 260 knots. The epicenter of the first quake was later found to be 93 nautical miles distant—the second, even closer. A little after 2:00 A.M. the crew in the upper station heard a roaring noise and then a loud crash described as a “sonic boom.” A wall of water rocked the coast guard station and flooded past it. The phones to the lighthouse went dead. Following the initial shock, the crew in the upper station evacuated and moved to higher ground. Looking back, there was no sign of the lighthouse or its powerful light. When daybreak came, they investigated and found the lighthouse destroyed—ripped from its foundation and smashed to pieces. The mangled remains of two crew members were found, the others presumably swept into the sea.9 Waves also traveled across the Pacific and about four and a half hours later struck the Hawaiian Islands and other locations, the greatest damage occurring at Hilo, on the island of Hawaii, where 150 people died. The run-up on the Hawaiian Islands ranged from a low of around 10 feet on the sheltered back side of the islands to highs of 52 to 56 feet on the northerly shores facing the oncoming waves.10 Why was the damage so great at Hilo? To find out, the Army Corps of Engineers built a scale model of Hilo’s triangular bay. The model—85 feet wide—

OCR for page 127
Extreme Waves faithfully reflects the bottom characteristics of the bay and its shoreline, including docks, rivers entering the bay, bridges over the rivers, and other important features. Incoming tsunami are simulated by releasing water from large tanks. The model tests have demonstrated that due to Hilo Bay’s unique configuration, almost any sizable wave that enters the bay will hit downtown Hilo or the coast north of town. Waves are reflected and refracted by the shoreline, and in some cases constructive interference of these waves creates very large waves in the center of the bay.11 Following the 1946 tsunami, tsunami warning and prediction centers were established in Hawaii and Alaska. Consequently, when the next Alaskan earthquake tsunami (caused by the March 9, 1957, magnitude 8.3 earthquake near the Aleutians, south of Andreanoff Island) hit the Hawaiian Islands, even though the waves were higher, there were no deaths because the warning enabled evacuation of low-lying areas. The tsunami from this earthquake—with a maximum wave height of 12 feet—caused flooding at Laie Point, Oahu, as shown in Plate 9. On Unimak Island, where the Scotch Cap lighthouse was destroyed in 1946, waves 39 to 49 feet high were reported.12 “The largest earthquake in the world,” as it is now being called, occurred on May 22, 1960, near the southern coast of Chile. The magnitude of the earthquake, originally reported as 8.3, has been recalculated as 9.5. It generated tsunami waves as high as 82 feet that hit shore 15 minutes later, killing thousands of people along the Peru-Chile coast. The wave also traveled westward 5,400 nautical miles across the Pacific and struck Hilo, on the big island of Hawaii, 15 hours later, killing an additional 61 persons and inflicting extensive damage (see Plate 2). The initial 35-foot-high wave was followed by seven more that arrived every 15 minutes or so. The wave caused damage and 138 deaths in Japan as well. The tsunami impacted the West Coast of the United States but did not cause serious damage or loss of life. Wave amplitudes of 4.6 feet were recorded at Santa Monica. Boats broke loose from moorings, some sinking in Los Angeles and Long Beach harbors. Other vessels were knocked around in San Diego harbor and caused damage. Crescent City recorded a wave height of 12 feet and boat damage.13 (See Figure 15.)

OCR for page 127
Extreme Waves FIGURE 15 Tsunami travel time in hours, Chile earthquake, May 22, 1960. Alaska was struck again on March 27, 1964. The epicenter of the magnitude 9.2 Prince William Sound earthquake was not far from the town of Seward. Anchorage, some 74 miles northwest of the epicenter, Valdez, 56 miles east of the epicenter, and other towns along the coast were hard hit by the intense shaking and by the tsunami that resulted. Seward was struck by a 33-foot-high tsunami, badly damaging docks and harbor installations; the force of the water was so great that box-cars were carried inland. In the source area, the average run-up elevation was 36 feet and the maximum was 220 feet! The earthquake caused the largest true tsunami, 219 feet high, which hit Valdez, Alaska. This wave was 10 feet higher than the tsunami that hit Russia’s Kamchatka Peninsula in 1737. A huge area of southern Alaska subsided. Years later, when I visited the area, there were vast areas of dead trees killed by saltwater still standing.14 The tsunami also damaged coastal areas in Canada—including Port Alberni—and in Washington and Oregon. Later, the wave struck the town of Crescent City in northern California, killing eight people and inflicting $11 million in damage. It finally reached Hawaii, where

OCR for page 127
Extreme Waves the wave height was measured at 14 feet at Waimea and additional damage occurred. There were 131 deaths, 122 of which were due to the tsunami, 5 in Hawaii, 16 in California and Oregon, and the balance in Alaska. Earthquakes in the Gulf of Alaska have caused tsunami and severe damage, but the magnitude 7.8 earthquake of March 6, 1988, had some strange effects. It was felt in communities all along Alaska’s Gulf Coast and produced a small tsunami that measured 1.25 feet at Yakutat, but caused no damage or injuries. The epicenter was in the Gulf of Alaska about 200 nautical miles south of Valdez. At the same time the 500,000-barrel crude oil tanker Sansinena II, under the command of Captain Bent Christiansen, was steaming from Portland, Oregon, to Valdez, Alaska, to pick up a load of crude. Captain Christiansen is now chief port pilot for the Port of Los Angeles. Here is his account of what happened. “We were on a northwest heading making 16 knots with the wind out of the southeast at 30 knots and a southwest swell of 15 feet. I was on the bridge. Suddenly, without warning, an extremely severe vibration started to shake the entire ship. My first thought was that we’d lost one or several propeller blades. I immediately pulled the throttle back to about 40 rpm, but there was no change in the intensity of the shaking, so I pulled the throttle to FULL STOP. I called the engine room and asked the engineer on watch if he knew what was causing the shaking. He did not know, so next I ran out on the bridge wing to look around. I could see the stack shaking so hard I thought it might collapse. I returned to the bridge and a few moments later the shaking subsided. “About this time I heard a call over the very high frequency (VHF) emergency Channel 16. It was the Exxon Boston calling the Exxon North Slope and reporting that she had encountered heavy vibrations, had lost power, and was experiencing some flooding. The Exxon North Slope also was without power and called a third Exxon ship, the Exxon New Orleans, which turned around and headed back to stand by the Exxon Boston. (The Sansinena had a steam turbine, no diesel engine in those days, and I believe the same was true of the Exxon tankers.) Meanwhile our radio operator heard about the earthquake from a station at

OCR for page 127
Extreme Waves Ketchikan. We proceeded to check the deck and the engine room and found no signs of damage, so I gradually resumed speed. I called the Exxon North Slope, gave our position, and offered to assist if needed. They responded that they were restoring power and did not need any help at this time. The Exxon Boston reported that the flooding was under control and the Exxon New Orleans was now standing by. While this was happening we felt the first of several aftershocks, leaving no doubt that it was an earthquake we’d felt. We resumed course to Valdez, where eventually all vessels arrived without further incident. “I plotted the position of the earthquake and the positions of the three Exxon tankers and found that they were about 35 nautical miles southwest of our position. That put them about halfway between us and the earthquake epicenter. I would never have believed that we could feel such a sharp, jarring motion transmitted through that great depth and distance of water.” I asked Christiansen if he had noticed any change in the sea state following the earthquake. “No,” he said. “If there was a wave, it was too small for us to notice. But that hasn’t always been the case. In the five years I was master on the Sansinena II, we ran into big waves several times. These were 75 feet high and covered the entire 800-plus-foot-long deck with green water. Waves hit the bridge as well. At such times about all you can do is heave to. You can’t help but have a number of concerns, foremost being not to lose power. I don’t know if those are what you would call extreme waves, but for me they were pretty extreme.” Coincidentally, as I thanked Captain Christiansen for talking with me, he mentioned that he was on his way to pilot the Lane Victory back into Los Angeles Harbor. I told him to be sure to say hello to Ernie Barker (see Chapter 8), since steam ships and big waves are something that he and Ernie have in common. The list could go on and on, but I will mention just a few others to indicate the geographical distribution of tsunami occurring in a short time period. On September 1, 1992, a magnitude 7.0 earthquake occurred off the coast of Nicaragua, causing a wave with a run-up of 33 feet. Three months later, on December 12, 1992, very nearly on the opposite side of the globe, a magnitude 7.5 earthquake struck near

OCR for page 127
Extreme Waves Flores Island, in the Sunda-Banda Island group (Indonesia), where the average run-up was 16 feet, with a maximum height of 65 feet, causing 2,080 deaths. On July 12, 1993, a magnitude 7.8 earthquake in the Sea of Japan and the resulting tsunami hit Okushiri Island with an average run-up of 33 to 49 feet and a maximum of 98 feet. It killed 185 people and caused extensive property damage. TSUNAMI CAUSED BY LANDSLIDES AND VOLCANOES Earthquakes can cause large waves by means other than displacement beneath the ocean’s surface. On July 10, 1958, a magnitude 7.9 earthquake occurred in southeast Alaska, not far from the current site of Glacier Bay National Park, causing landslides, submarine slides, and icefalls from glaciers and producing six separate wave events. Since the earthquake occurred in a remote area, there was little damage or loss of life, but it did result in several amazing survival stories.15 The earthquake dislodged a mammoth wall of rock and pieces of ice from the glacier in the headlands of nearby Lituya Bay, creating a huge wave that roared through the bay and out into the open sea of the Gulf of Alaska at a speed of 80 to 110 knots. Some would call this a “splash,” rather than a tsunami. Whatever you choose to call it, to the crews of the three boats anchored in the bay that day, it was the biggest wave they’d ever seen. The salmon troller Edrie, with two crew (Howard Ulrich and his six-year-old son, Howard Jr.) on board, was anchored inside the bay; two other boats were anchored near the entrance, at a place called Anchorage Cove. Ulrich heard a roaring noise and saw the wave coming as it broke around Cenotaph Island, a 320-foot-high island in the middle of the bay. The wave was steep and 66 to 98 feet high as it approached his boat. He frantically tried to maneuver, but the wave picked up Edrie, swept it up and over dry land, and then by a random chance of fate dropped it back into the bay. The other two boats (Badger, with Bill and Vivian Swanson aboard, and Sunmore, with Orville and Micki Wagner aboard) were swept out to sea over the tops of trees on a spit of land at the entrance to the bay; both sank, but the Swansons scrambled into a dinghy and survived. The Swansons reported that the wave first hit the southern side of the bay near Mudslide Creek, where

OCR for page 127
Extreme Waves the water height reached 656 feet; then as the wave passed Cenotaph Island, it cleared trees to a height of about 160 feet, before hitting them and tossing their boat over the sand spit at the mouth of the harbor. Aerial photographs taken after the earthquake show the mountains on the north and south sides of the bay swept clean of trees and vegetation. Water surged to a height of 1,700 feet as it poured over the fingers enclosing the bay, stripping off the trees and topsoil down to bare bedrock. Geological records and examination of tree-growth rings show that this was not the first time a giant wave has coursed through Lituya Bay; apparently large waves occurred in 1854, 1874, 1899, and 1936. Recently, scientists have developed various wave models for the bay to predict what wave height could occur under various scenarios. They conclude that a 1,700-foot-high wave is indeed possible.16 Volcanoes are another source of tsunami; at least 92 cases have been documented.17 Of these, the most famous is the eruption of Krakatau (Krakatoa) on August 27, 1883. Following a series of lesser eruptions and explosions for several days preceding the 27th, the volcano finally destroyed itself in an immense explosion that literally echoed around the globe. In the Sunda Straits nearby, ships were lost, 165 villages were destroyed, and more than 36,000 people were killed along the coasts of Java and Sumatra, mostly by two extreme waves that followed when millions of tons of debris were dislodged into the sea. These tsunami had periods of one to two hours. Many towns were only about 33 miles from Krakatoa, so there was no way a warning could have been issued. In addition, since the volcano had been erupting periodically for days, people became complacent. But when it finally blew up, within minutes the town of Merak on the island of Java was hit by a wave 100 feet high. In Merak the wave destroyed stone buildings on top of a hill that stood 115 feet high. Lighthouses toppled, a naval vessel was picked up in the harbor and carried several kilometers inland, and from the high water marks, it appears that the largest wave ran up at least 133 feet when it hit the shore.18 TSUNAMI WARNING SYSTEMS In its most elementary form, a tsunami warning center has three major components: seismographs to detect if an earthquake capable of caus-

OCR for page 127
Extreme Waves ing a tsunami has occurred, tide stations that monitor water level and indicate whether or not a tsunami wave has been created, and a means of disseminating the warning via a variety of different, robust communications routes to civil defense or disaster management authorities. A number of nations operate regional tsunami warning systems. However, of the three major oceans, only the Pacific Ocean has an integrated multinational tsunami warning system. Twenty-six nations in or bordering on the Pacific Ocean participate in an International Coordination Group formed under the auspices of the United Nations Educational, Scientific, and Cultural Organization’s (UNESCO’s) Intergovernmental Oceanographic Commission. The nerve center for the Pacific Ocean warning system is a small concrete block building located at Ewa Beach, a short distance west of Pearl Harbor on the south side of the island of Oahu. Following the 1946 Alaskan earthquake, the U.S. government established the first elements of a tsunami warning system that linked the mainland and the Hawaiian Islands. Then, after the Chile earthquake in 1960, the system was expanded to cover all of the countries along the edges of the Pacific Ocean, the so-called Ring of Fire. I visited the center to get a firsthand look at its operation. I was met by tsunami experts Barry Hirshorn and Stuart Weinstein, who kindly gave me a detailed briefing on how the system operates. Hirshorn and Weinstein live in houses next door to the center. They each carry not one, but two pagers, set to go off automatically if an earthquake capable of causing a tsunami occurs anywhere in the Pacific. Even if they are not already in the center, within 90 seconds they will be assembling the data to decide if a tsunami bulletin should be issued. There are four levels of bulletins issued by the center upon detection of an earthquake of magnitude 6.5 or greater: A tsunami information bulletin: This is a message to advise all participants in the warning system that a major earthquake has occurred in the Pacific. The bulletin will also advise that it does not appear that a tsunami has occurred or that the possibility of a tsunami is being investigated. The reason for this uncertainty is that depending on the location of the earthquake, more time and data may be needed to make the determination.

OCR for page 127
Extreme Waves A fixed tsunami warning for all coasts within 1,000 kilometers (620 miles) of the epicenter of a potentially tsunamigenic earthquake. A regional tsunami warning or watch bulletin: If a tsunami appears possible, all areas within a three-hour tsunami travel time will be placed on a warning status and areas within a three- to six-hour travel time will be alerted to a “watch” status. The purpose of the “warning-watch” system is to provide as much advance warning as possible to enable local areas to mobilize warning and evacuation systems, but at the same time to try to minimize the number of false alarms. A Pacific-wide tsunami warning bulletin: Once the presence of a tsunami with destructive potential that goes beyond a local region has been confirmed, a broad general warning is issued with hourly updates to all areas with coastal populations. While the Pacific Tsunami Warning Center issues tsunami bulletins for the entire Pacific region, there are a number of important regional systems that cover local areas and share data with the center. An example is the regional system of Japan. Due to Japan’s long history of devastating tsunami, this country has the most advanced system in the world, consisting of more than 1,500 seismometers and more than 500 water-level gauges. The Japanese system includes a number of underwater seismometers. The location of an earthquake can be pinpointed in Japanese waters and tsunami warnings can be issued within a minute. Timing is especially important for Japan because of the large number of destructive earthquakes in nearby waters that have caused tsunami. The travel time for these tsunami is so short that warnings must be nearly instantaneous to be effective. Other regional systems include those in French Polynesia (Tahiti), Russia, Chile, and Australia. For Alaska and the West Coast of the United States there is a regional warning center located in Palmer, Alaska, about 45 miles from Anchorage. Besides serving as the warning center for the entire Pacific Ocean, the Pacific Tsunami Warning Center performs a dual role as the regional center for the Hawaiian Islands and all other U.S. interests in the Pacific. A decade or so ago it took about 60 minutes for the center to evaluate the data, determine where the earthquake was located, and issue a

OCR for page 127
Extreme Waves warning for an event occurring anywhere in the Pacific Basin. Regional warnings for earthquakes occurring in the Hawaiian Islands could be issued in about 15 minutes at that time. Today, with more instruments sending data to the center, faster computers, improved computer software for earthquake analysis, and models for tsunami wave propagation, these times have been reduced to 20 to 25 percent of the previous values. A warning for a major tsunami from an earthquake occurring in the Pacific Basin now takes about 15 minutes; a regional warning can be issued in 3 minutes. The heart of the detection system is the Incorporated Research Institutions for Seismology (IRIS) global seismic network, an array of approximately 140 broadband seismometers located throughout the world. In addition to those installed and maintained by IRIS and additional instruments on the U.S. mainland installed by the United States Geological Survey (the USGS National Seismic Network), the center automatically receives data from seismometers operated by some of the other member countries. The heart of the detection system for locally generated tsunami in Hawaii is the center’s local network of seismometers and water-level instruments and the USGS Hawaiian Volcano Observatory seismic network on the island of Hawaii. When an earthquake occurs, the first step is to get a rapid estimate of its location by examining the readings from several different seismometers; then, by knowing how fast seismic waves travel through the earth, the distance to the earthquake can be calculated. By using several such readings and a process of triangulation, the location of the earthquake is established. The more seismometers, the more quickly an accurate determination can be made. Unfortunately, the Pacific Ocean encompasses a vast area, and the center would like to see more seismometers located within it. To provide better coverage, a number of these should be underwater instruments like those the Japanese have, but they are very expensive. The next step is to determine the earthquake’s magnitude. This is not a simple matter of reading a gauge; it requires a computational process. The initial estimate is usually refined later as more data are collected. Historically, tsunami are rarely caused by earthquakes with magnitudes less than 7.5. But to stay on the safe side, any earthquake

OCR for page 127
Extreme Waves with an apparent magnitude of 5.5 to 6.0 will trigger the evaluation process. The second element in the detection system is composed of water-level measuring devices. There are two basic types: shore-based systems and deep-water systems. Fortunately, there are a large number of automatic tide stations around the Pacific Rim. These stations transmit water-level information to the center via satellite. The warning system employs eight deep-water buoys to monitor deep ocean waters. These buoys consist of pressure sensors placed on the bottom of the ocean. Their sensitivity is such that they can detect a tsunami wave from as little as 0.4 inches in height to as much as 40 inches. (Remember, tsunami waves are not very high in deep water.) The bottom sensor (which may be in waters as deep as 3 miles) transmits data acoustically to a tethered buoy on the surface. The buoy in turn transmits the information to a satellite, which relays it to the center. The deep-water instruments constitute an important part of the detection system and several dozen more are needed. However, they are expensive to install and maintain and cost around $250,000 each. In addition to these instruments, the center receives sea level data from more than 100 stations operated by member countries. At the shoreline, the problems are different than in the deep ocean where the buoys are. First, it is difficult to get accurate readings, since local effects (run-up or seiches) can make a given wave 3 feet high in one location and 30 feet high in another. In a major tsunami, the shoreline instruments in the vicinity will be driven off-scale or destroyed, rendering their readings useless, though they may still be useful in helping coastal observers provide a warning to adjacent areas. Data from all of the seismographs are monitored continuously on the center’s computers. Data from ocean- and shore-based tide instruments are transmitted by satellite to the center. Sea level measurements are made every two seconds, and then averaged over three or four minutes and transmitted to the center every three or four hours. The center’s computers actuate pagers as soon as an earthquake occurs or if large water level amplitudes are observed. Once it has been determined from the seismometer data that a tsunami may have been formed, the center issues a bulletin and then

OCR for page 127
Extreme Waves scans the water level gauges in the vicinity of the earthquake to determine if a wave has been detected and, if so, how large it is. The bulletin is issued before a tsunami has been observed, by estimating potential tsunami travel times based on bathymetry and a tsunami propagation model developed by Paul Wessel at the University of Hawaii at Manoa. Countries that lie in the potential path of the tsunami now have adequate warning time to evacuate low-lying coastal areas, unless they are located very close to the epicenter. While this procedure does not eliminate property damage, it has drastically reduced the number of deaths attributed to this type of extreme wave. With this information—a combination of nearshore tidal data, deep-ocean data, and the size and type of the earthquake—accurate forecasts of a tsunami’s danger can be made. Here again, the speed and accuracy of this determination are influenced by how many instruments detect the wave and their location relative to the earthquake. Due to a lack of coverage, there are areas of the Pacific where warnings may be delayed or where the center is unable to determine that the danger is past. TSUNAMI INTENSITY SCALE Tsunami intensity and damage potential are measured by the Modified Sieberg Tsunami Intensity Scale.19 Very light.Wave so weak as to be perceptible only on tide gauge records. Light. Wave noticed by those living along the shore and familiar with the sea. On very flat shores generally noticed. Rather strong.Generally noticed. Flooding of gently sloping coasts. Light sailing vessels are carried away on shore. Slight damage to light structures situated near the coasts. In estuaries, reversal of the river flow some distance upstream. Strong. Flooding of the shore to some depth. Light scouring on man-made ground. Embankments and dikes damaged. Solid structures on coasts injured. Large sailing vessels and small ships drift inland or are carried out to sea. Coasts are littered with floating debris. Very strong. General flooding of the shore to some depth. Quay walls and solid structures near the sea are damaged. Light structures

OCR for page 127
Extreme Waves are destroyed. Severe scouring of cultivated land; coast littered with floating items and dead sea animals. With the exception of large ships, all other vessels are carried inland or out to sea. Big bores in estuary rivers. Harbor works damaged. People drown. Wave accompanied by strong roar. Disastrous. Partial or complete destruction of man-made structures for some distance from the shore. Flooding of coasts to great depths. Large ships severely damaged. Trees uprooted or broken. Many casualties. For major earthquakes that occur within a few hundred kilometers (100 miles or so) from a coast, if a tsunami is created, the travel time of the wave is such that it will hit the coast at any time from within a few minutes very near the event to as much as 40 minutes farther away from it—the Sumatra earthquake on December 26, 2004, is an example. In such places there is a high risk of loss of life unless immediate steps to evacuate are taken. Even if the center issues a warning, there is very little time to evacuate in this case. Once the tsunami warning is sounded, residents need to know they must immediately walk inland to higher ground. Coastal communities need to understand that if they feel the ground shake, there is probably a large offshore earthquake occurring and they should immediately evacuate and not wait for a tsunami warning. Barry Hirshorn is enthusiastic, articulate, and very committed to his work. We were in the Pacific Tsunami Warning Center control room, surrounded by racks of computers, monitors, plotters, electronic instrument racks, and large-scale maps, looking at displays that showed tidal measurements from around the Pacific, when he brought up the subject of evacuation. He told me: “If you feel the ground shake, or see the ocean recede or behave in an unusual way, get out of there. Don’t drive—leave the Mercedes or that brand new SUV behind! Immediately walk inland. A fast walk of only 10 to 15 minutes will probably be enough to save your life. Remember, there will be multiple waves, and it may not be safe to return for hours.” The Lituya Bay wave described earlier is another instance in which a warning might not be possible. If the earthquake is too small to trigger the Pacific Tsunami Warning Center’s pagers, but has caused a large

OCR for page 127
Extreme Waves submarine landslide, the center might not detect this as a tsunami source. The scientific community is working on this problem and hopes to use global positioning satellites to see these submarine landslides even if there is no earthquake to trigger them. To address this problem in the interim, the center has written software that triggers the pagers if large amplitudes are detected on certain coastal water level gauges. Finally, as Hirshorn pointed out to me, the Pacific Tsunami Warning Center is analogous to one leg of a three-legged structure. The other two legs are emergency management and public education and awareness. Without the other two “legs,” the center’s warnings are of little value. Once the center has sent out a warning, communities likely to be affected by the tsunami need to implement predetermined emergency procedures. Typically these will involve mobilization of police and other emergency services, beach clearance, sounding of sirens or alarms, and traffic control to permit prompt evacuation of low-lying areas. Public education and awareness are essential. The public needs to be informed in order to evacuate promptly in the event of a major offshore earthquake or if the ocean is observed to suddenly recede. If a warning is given, people need to know the shortest route to high ground—and, after a wave has hit, to stay clear of low-lying areas because there are very likely to be many more waves on the way. There can be waves for hours after the initial wave hits, and often the first wave is not the largest. The importance of public education was proven dramatically by the Papua New Guinea tsunami of July 17, 1998, which slaughtered 2,500 people—75 percent of the population in the coastal areas hit by the tsunami. When the next tsunami occurred, the death toll was drastically reduced—because a public education campaign had informed people of the warning signs and the steps to be taken to evacuate. Also, during the Southeast Asia tsunami of December 2004, fatalities were low on several Indian Ocean islands simply because the village elders saw the ocean waters withdrawing from the shore; recognized, thanks to village lore, that this phenomenon heralded a tsunami; and warned the populace to withdraw to high ground. Others were not so fortunate.