[Coral-List] Tsunami on atolls from ITIC Tsunami Bulletin Board

Bruce Richmond brichmond at usgs.gov
Mon Mar 21 17:45:23 EDT 2011

Coral - listers,


FYI - Here is a recent posting on the tsunami hazard on atolls by Gerard
Fryer at the NOAA Pacific Tsunami Warning Center:


The question of tsunami hazard on atolls comes up a lot. But if you search
all the catalogs, you'll find very little data on tsunami runup on Pacific
atolls, presumably because there have been so few observations, despite the
five great ocean-crossing tsunamis of the 20th century. The highest value I
am aware of is at Midway in 1952, when runup reached 1.9 meters. That was
for an exceptionally large earthquake, however, magnitude 9, and Midway was
square in the center of the radiated beam and only 3000 km from the source.
Where on the atoll the 1.9 measurement was made I don't know, but from
photographs of cars driving through the flood, I suspect it was on the
lagoon side of the west island. The lagoon had been extensively dredged and
the pass to the NW (facing the approaching tsunami from Kamchatka) had been
deepened and broadened during WWII. I suspect the flooding was a consequence
of that modification. (BTW, Midway has a tide gauge, but it's pretty crummy;
its measurements of tsunamis are very erratic and vary tremendously with
azimuth). The only other island with measurements approaching Midway's is
Johnston Atoll---another atoll extensively modified by dredging---where the
Chile tsunami of 1960 had a runup of 0.7 meters. Of atolls not extensively
modified to handle shipping, the highest runup I am aware of is at
Kiritimati in 1960, which recorded a runup of 0.3 meters. So if your island
is 2 m high, you are safe.


Why should the hazard be so low? The answer is sea level rise since the last
glaciation. For the last 18k years, coral has been very happy and growing
almost vertically upwards. As a consequence, each atoll in the Pacific has
very steep upper slopes, making the island look pretty much like a spar buoy
to the approaching tsunami. The tsunami just doesn't interact with the
island much. Unless man has gone in and modified things, runup seems to be
little more than a factor of two from the height of the tsunami on the open
ocean. Compare that to a factor of five or six for open coasts on high
islands (and even more where there is focusing). Even high islands have a
tremendously reduced hazard if they have reefs. For example, the greatest
runup in Tahiti in 1960 was about a meter, and that was opposite a pass
through the reef. By contrast, in 1960 Hawaii saw about 3 meters on open
coasts without focusing and numerous locations where runup approached 10
meters (in Hilo runup reached 11 meters and 61 people died).


Given all the above, what happened in the Maldives in 2004? Runup at those
atolls averaged about 3 meters and many of the islands were completely
inundated. The Maldives, however, stood as a rampart across the path of the
2004 tsunami and were only 2000 km away from the source. The
Maldive-Laccadive Ridge is also shallower than the abyssal Pacific from
which most Pacific atolls rise, so the tsunami was influenced more by the
bathymetry (runup in the Maldives was still systematically less than at
mainland sites). The geography of the Maldives is exquisitely bad for
tsunamis coming from Sumatra. By contrast, there is no island group in the
Pacific lying squarely astride the path of any tsunami from a subduction
zone. Even the Tuamotus, whose closely-spaced islands would appear to make
the group a vulnerable target, are aligned parallel to rather than across
the path of tsunamis from South America.


Take tsunami warnings seriously, but don't overreact.




Gerard Fryer


NOAA Pacific Tsunami Warning Center

91-270 Fort Weaver Road

Ewa Beach, HI 96706-2928, USA

+1 808 689 8207





Bruce Richmond
U.S. Geological Survey

Pacific Coastal and Marine Science Center
400 Natural Bridges Drive
Santa Cruz, CA 95060
Phone: (831) 427-4731
Fax: (831) 427-4748


More information about the Coral-List mailing list