Risk and Avoidance - Tsunami - page 198

The International Tsunami Information Center ( 2013 )
also gathers and disseminates general information about
tsunami, provides technical advice on the equipment required
for an effective warning system, checks existing systems to
ensure that they are up to standard, aids the establishment of
national warning systems, fosters tsunami research, and
conducts post disaster surveys for the purpose of documen-
tation and understanding of tsunami disasters. As part of its
research mandate, the ITIC maintains a complete library of
publications and a database related to tsunami. Research also
involves the construction of mathematical models of tsunami
travel times, height information, and extent of expected
inundation for any coast. Planners and policy makers use
results from these models to assess risk and to establish cri-
teria for evacuation. The ITIC trains scientists of member
states who, upon returning to their respective countries, train
and educate others on tsunami programs and procedures, thus
ensuring the continuity and success of the program. The
Center also organizes and conducts scientific workshops and
educational seminars aimed towards tsunami disaster edu-
cation and preparedness. In recent years, emphasis has been
placed on the preparation of educational materials such as
textbooks for children, instruction manuals for teachers, and
videos for the lay public.

north of Japan, Pacific-wide tsunami warnings were issued
for tsunami that never eventuated (Walker 1995 ). Both
events cost 30 million dollars in lost salaries and business
revenues in Hawaii, where evacuations were ordered. The
people who distribute such warnings are only human. Each
time a false warning is issued, it weakens their confidence
in predicting future tsunami, especially if the tsunami have
originated from less well-known source regions. Worse than
a false alarm is one that is realistic, but where the time has
been underestimated. Tsunami travel charts have been
constructed for possible tsunami originating in many loca-
tions around the Pacific Ocean. Many of these charts are
inaccurate, fortunately, with tsunami traveling faster than
predicted. Before 1988, about 70 % of the Pacific Ocean did
not have publicly accessible bathymetry to permit accurate
tsunami travel-time forecasting. Fortunately, since the end
of the Cold War, these data have become more available.
Earthquakes do not cause all tsunami. A relatively small
earthquake can trigger a submarine landslide that then
generates a much bigger tsunami. Nor is the size of an
earthquake necessarily a good indicator of the size of the
resulting tsunami. The July 17, 1998 Tsunami along the
Aitape coast of Papua New Guinea illustrates this fact. The
earthquake that generated this event only registered a sur-
face wave magnitude of 7.1, yet the resulting tsunami at
shore was up to 15 m high (González 1999 ). As described
in Chap. 5 , such tsunami earthquakes are common. For
example, the April 1, 1946 Alaskan earthquake had a sur-
face wave magnitude of 7.2, but it generated run-ups of 16.
7 m as far away as Hawaii (Fig. 2.10 ). On June 15, 1896, an
earthquake that was scarcely felt along the Sanriku coast of
Japan generated the Meiji Tsunami that produced run-ups of
38.2 m above sea level and killed 27,132 people.
Finally, our knowledge of tsunami is rudimentary for
many countries and regions, not just in the Pacific Ocean,
but also other oceans. Not all of the coastline around the
Pacific Rim has been studied. This was made apparent on
March 25, 1998 when an earthquake with a magnitude, M s ,
of 8.8 occurred in the Balleny Islands region of the Ant-
arctic directly south of Tasmania, Australia (Bryant 2005 ).
Because of the size of the earthquake, a tsunami warning
was issued, but no one knew what the consequences would
be. The closest tide gauges were located on the south coast
of New Zealand and Australia. Forecasters at the PTWC in
Hawaii had to fly by the seat of their pants and wait to see if
any of these gauges reported a tsunami before they issued
warnings further afield. While that may have helped resi-
dents in the United States or Japan, it certainly was little
comfort to residents living along coastlines facing the
Antarctic in the Antipodes. In cities such as Adelaide,
Melbourne, Hobart, and Sydney, emergency hazard per-
sonnel knew they were the mine canaries in the warning
system.

10.3.2 Flaws in Regional Warning Systems

The Pacific Warning Tsunami System is not flawless. The
risk still exists in Japan and other island archipelagos along
the western rim of the Pacific for local earthquakes to
generate tsunami too close to shore to permit sufficient
advance warning. Of the 10 most destructive trans-oceanic
tsunami over the last 250 years, 84 % of fatalities occurred
within the first hour of generation (Gusiakov 2008 ). For
example, the 7.8 magnitude earthquake that struck in the
Moro Gulf on the southwest part of the island of Mindanao,
the Philippines, on August 17, 1976, generated a 3.0-4.5 m
high local tsunami. The event was virtually unpredictable
because the earthquake occurred within 20 km of a popu-
lated coastline (Bryant 2005 ). The Papua New Guinea
Tsunami of July 17, 1978 and the Indian Ocean Tsunami of
December 26, 2004 at Banda Aceh also arrived at shore
within 15 min of the earthquake (González 1999 ).
The accuracy of any warning system does not rely upon
the number of tsunami predicted, but upon the number that
are significant. False alarms weaken the credibility of any
warning system. Although tide gauges can detect tsunami
close to shore, they cannot predict run-up heights accu-
rately. Consequently, 75 % of tsunami warnings since 1950
have resulted in erroneous alarms (González 1999 ). For
example, on May 7, 1986, following an earthquake in the
Aleutian Islands, and again in 1994 after an earthquake Fortunately, the Antarctic earthquake was not