Description of the Event:
On Monday, October 9, 1995, an earthquake of Mw 8.0 occurred offshore
of the states of Jalisco and Colima, along the Pacific coast of
central Mexico. Extensive shaking damage occurred in Manzanillo,
resulting in approximately 40 deaths and about 100 injured. The
earthquake set off a moderate tsunami that impacted about 200 km
of coastline. Tsunami runup ranged from 1 - 5 m, from Paraiso in
the south to Tenacatita Bay in the north, where the most severe
flooding was recorded. Severe damage also occurred in the Port of
Manzanillo, where a 12-knot tsunami induced current caused extreme
erosion on the banks of the port. This event was remarkable in two
ways: it provided data on the freefield offshore profile of a tsunami,
taken by two bottom mounted pressure transducers which had been
deployed offshore at 50 m water depth for a study of internal wave
motion, and photographs were taken by eyewitnesses as the wave climbed
This earthquake was the largest event in 60 years along the Northern
Middle America Subduction Zone. It exhibited extremely shallow faulting,
with maximum displacement of 5m at 15 km depth and 2m at 8 km depth.
The fault had a dip angle of 16ľ, and ruptured along 200 km of strike
length, causing widespread coastal subsidence. The quake was felt
in Mexico City, and the tsunami reached the shores of Puerta Vallarta,
300 km to the north.
A team from the University of Southern California arrived on the
scene first and conducted a survey, ranging from Paraiso in the
south to Punta Chalacate in the north, about 60 km north of Manzanillo.
From Manzanillo a long, wide, sandy beach extends southward for
more than 80 km to a river mouth known as Boca de Pacuales. Runup
heights decreased along the beach southward from Manzanillo from
4.75 to 1.75m. In Paraiso, witnesses reported strong shaking, followed
a few minutes later by a recession of the sea surface to about 50
m offshore, before advancing “slowly” up to a maximum elevation
of 1.76m above sea level.
Manzanillo Bay is split into two parts by a rock peninsula, the
southern part known as Manzanillo Bay and the northern part as Santiago
Bay. In Santiago Bay, two points were measured. The southern point
had a measured runup of 2.3m. The flooding was minor and caused
minimal damage. The point farther north showed slightly higher runup.
Here a sea-wall was overtopped, and a staircase down to the beach
provided a runup measurement (see picture). On the extreme north
end of Santiago Bay is an area of steep cliffs. Witnesses reported
seeing the water withdraw several hundred meters offshore and then
rush in and up the 10m cliff, violently splashing over the edge.
This was the point of highest measured runup, with a corrected value
of 10.9m. This can be attributed to the topography of the area.
The first of the series of five waves arrived about 15 minutes after
the initial shock.
The greatest damage occurred in the next bay to the north, the Bay
of Tenacatita. La Manzanilla is a small town that lies at the southern
end of the bay. The town was flooded to a depth of more than 2m
in most areas, advancing 200m inland to the base of a steep hill.
All witnesses reported a leading depression wave (Tadepalli and
Synolakis, 1994 and 1996), with water receding several hundred meters.
A family eating breakfast on their patio felt the quake and confirmed
that the bay began to empty 15 minutes later (see photo). Their
16 year old son reported the “normal” depth at the point of largest
withdrawal to be 5-6m, as he frequently dives for shellfish in the
vicinity, thereby giving the first-ever field observation of the
amplitude of the leading depression wave. When the water began to
return and advance, it did so “like a fast rising tide." One resident
of La Manzanilla, Pepe Martinez, was able to take a sequence of
photos as the water advanced through the town (see pictures).
Although this was not an extremely large event, it was interesting
for three reasons. First, it struck at midmorning and was observed
by many eyewitnesses, who were able to describe the sea-surface
motions well. Second, through extreme serendipity, two moorings
of bottom mounted pressure sensors had been installed offshore of
La Manzanilla the week before for a study of internal wave motions.
This data will prove invaluable for future tsunami modeling. Finally,
the inundation patterns confirmed the extreme vulnerability of flat
lands near river inlets, and they showed that even moderate tsunamis
can produce large currents inside harbors.