PubTalk 1/2017 — Unusual sources of tsunamis

Video Transcript
Download Video
Right-click and save to download

Detailed Description

A presentation on "Unusual Sources of Tsunamis From Krakatoa to Monterey Bay" by Eric Geist, USGS Research Geophysicist

- Not all tsunamis are generated by earthquakes.
- Tsunamis can be caused by volcanoes, landslides, and even atmospheric disturbances
- Data from tide gauges can help unravel the complex physics of these sources

Videographers: Mitch Adelson, William Seelig, USGS


Date Taken:

Length: 01:06:14

Location Taken: USGS Menlo Park, CA, US


[ Silence ]

[inaudible background conversations]

Hello? Okay. We'll go ahead
and get started now.

Good evening.
My name is William Seelig.

Welcome to the U.S. Geological Survey.
I work for Science Information Services.

Glad to see everybody here
or our January public lecture.

Before I introduce this evening's
speaker, I’d like to remind everyone

about our February lecture on
scanning electron microscopes.

You can pick up a flier
at the back table there.

And I hope to see everyone here
on February 23rd for that lecture.

For tonight's lecture, the title is
Unusual Sources of Tsunamis

from Krakatoa to Monterey Bay,
presented by Eric Geist.

And a little bit about Eric.

He is a research geophysicist
with the Pacific Coastal and

Marine Science Center of USGS
where he's worked for 32 years.

Throughout his career, he has focused
on computer modeling of

geophysical phenomena, including
the physics of tsunami generation.

Eric has authored numerous journal
articles including a Scientific American

article on the devastating
2004 Indian Ocean tsunami.

Eric received his bachelor's degree in
geophysics from Colorado School of

Mines and his master's degree in
geophysics from Stanford University.

And as always, after the presentation,
we'll have questions from the audience.

Just please hold your
questions until afterward.

And please
welcome Eric Geist.

[ Applause ]

- Thank you, William.
Can everybody hear me okay?

All right.

Thanks for coming out tonight
and listening about tsunamis here.

As probably
everybody knows,

is that most tsunamis
are caused by earthquakes.

We've had a devastating 2004 tsunami
in the Indian Ocean, and of course the

2011 tsunami in Japan that destroyed
the Fukushima nuclear power plant.

What I thought I’d do for the
public lecture today, though,

is look at tsunamis not generated
by earthquakes – by all these

other sources, which I think is kind of
an interesting line of research.

And what we're going to do is look at
probably the most famous of these

non-seismic sources –
Krakatoa, that erupted in 1883.

And I'm going to frame
the talk around Krakatoa.

We're going to pose two problems.
The first problem is looking at the

tsunami that hit Jakarta very close
to where the volcano erupted.

And then we're going to take a step
back and look at basics of tsunamis,

how they're generated, the type
of waves tsunami waves are,

and importantly,
tide gauge measurements.

These are – we're going to mostly
be looking at case studies.

So tide gauges actually
give us the most –

the most important instrument
of data for tsunamis.

Then we're going to look at different
types of non-seismic tsunamis – volcanic

tsunamis, where we're going to try to
answer our Krakatoa problem number 1.

And then pose Krakatoa
problem number 2.

That is the San Francisco tsunami
and what's that all about.

We're going to take a break
from problems and look at –

briefly at asteroid tsunamis,
landslide tsunamis, and then

atmospheric tsunamis –
meteorological-induced tsunamis.

And then pose an answer to
Krakatoa problem number 2.

And end up with a kind of
interesting summary of all this.

So Krakatoa.

The activity in Krakatoa actually
started months before the big eruption.

It started in May of 1883
and culminated in this

huge explosion
in August of 1883.

It destroyed two-thirds of the island.
This is a map of the island before the

eruption shown here, and after the
eruption, two-thirds of this was gone.

The dashed line represents the crater
of what was left of the island

under the water and activity –
how Krakatoa continues today.

Now, Krakatoa is
known for a lot of things.

One is it generated a
pressure wave that was

recorded all the way
around the world.

The pressure wave actually traveled
3-1/2 times around the world.

It was recorded on virtually
every barometer in the world.

It produced the loudest
sound ever recorded.

The sound was heard 4,800
kilometers away, believe it or not.

And it produced one of the
most destructive and deadliest

non-seismic tsunamis.
And it turns out, the most complex,

and I had no idea how
complex it was before I started

putting this talk together. [laughs]
So there's a lot of things going on

with Krakatoa, so we'll start
from the beginning.

Where is Krakatoa? Krakatoa sits in this
body of water between Sumatra –

the island of Sumatra in Indonesia
and Java here in the Indian Ocean.

So we’re going to
zoom up in this place.

This body of water is
called the Sunda Strait.

This is a map not up to USGS
standards, but this is land here,

Sumatra and Java here,
and what’s contoured is the seafloor.

Krakatoa – what’s left of
Krakatoa sits right here.

This little passage –
it’s kind of like the Golden Gate.

The tsunami amplitudes die off pretty
quickly as it gets into this body of water.

And this is where Jakarta sits here, and
that’s where our tide gauge station is.

That’s what we really want right there.
But these numbers are tsunami run-up –

how high the waves got
on shore following this event.

These numbers are from a
very comprehensive report

soon after the event published by
the Royal Society of London.

But since then, some other surveys
came out and determined that the

highest run-up of a tsunami
from Krakatoa is 41 meters.

So it’s definitely a huge tsunami.
Could probably consider it a

mega tsunami, although we’ll see
bigger tsunamis later on in the talk.

So that’s where we’re at.
So what we’re looking at is

what is the exact mechanism
that caused this tsunami?

Volcano tsunamis are kind of
interesting because there’s a

number of different mechanisms
that can cause a tsunami.

Listed here, we’ll be looking
at a sampling of these.

And then we’ll add to this a number of
other non-seismic sources for tsunamis.

Some may be familiar,
and some may be unfamiliar to you.

We can group these into
tsunamis occurring in

the atmosphere,
or meteorological tsunamis.

And landslides, of course,
can cause tsunamis.

What we’re going to be
talking about here mostly

in this talk is the physics
of tsunami generation.

And what’s important is
knowing where these sources

are relative to
the water’s surface.

And then of course we have
earthquakes which are most common.

They generate about 70 to 80%
of the tsunamis that are in the catalog

or that have been recorded.
So we can't forget earthquakes.

So we’re all on the same page,
I’m going to go through the

life of a tsunami using
an earthquake source.

So what do you need
to get a tsunami generated?

Something that generates –
that’s able to push up and down

the entire water column.

The other thing that you need
is something that’s very wide.

It has to be wider
than the ocean is deep.

And optimally, it has to be
three times wider to generate

the very long waves necessary
for a tsunami to propagate

all the way across the
Pacific Ocean, for example.

So that’s what you need.
Once you generate this massive

instability in the water column, it’s not
going to want to stay in one place.

If you’re looking in cross-section,
you’re going to have the transoceanic

tsunami propagating out into the deep
open ocean, and then you’re going to

have the local tsunami, of course,
propagating in the opposite direction.

These waves gradually evolve.
When it gets into shallow water,

the tsunami actually slows down,
the wavelength decrease, and in order to

conserve this massive volume of water,
the amplitude has to increase.

This is termed
shoaling amplification, so –

and this is a
pretty dramatic effect.

You can have a 1-meter tsunami
in the deep ocean transform to

a 30-meter tsunami on shore,
as we’ve seen with the 2004 event.

So when it hits the shore,
that’s called tsunami run-up.

And it comes in
like a fast-moving tide.

Run-up also specifically refers to
the vertical height of the wave –

the maximum vertical height.
And inundation – another term I’ll use –

is the horizontal extent of flooding.
So it’s a horizontal measure.

Important thing to remember
is that tsunami waves –

there’s not just one tsunami wave,
but they last for many hours.

There are many
tsunami waves.

So to summarize here,
the three things you need to

keep in mind for a tsunami are
you need a source that generates

a motion of the entire water column
up and down – vertical motion.

You need a very wide source
relative to the water depth.

So if your water depth is very shallow,
you don't need as wide of a source.

And then another thing you need –
a fast source.

Something faster than the speed
that tsunami waves move.

But, as we’ll see, not too fast.
This is kind of new research.

If it’s too fast, it also
doesn't generate a tsunami.

So our data.

Our data comes, instrumentally,
primarily from tide gauge records.

Now, this is a very old tide
gauge record from San Diego.

Tide gauge records show the height of
the sea surface as a function of time.

So these, I think, are hour marks
on here – time along here.

It’s attached to a clock.
This drum is attached to a clock.

And in here you can see the high tide
and the low tide occurring twice a day.

But then you see these interesting
fluctuations on top of the tide.

So the people that installed
this were wondering,

what the heck are
these fluctuations?

We filtered out the wind waves.
So I don't know what these are.

Remember, this is 30 years before the
invention of the modern seismometer.

So in the weeks following the
installation of the West Coast

tide gauges, which was San Diego;
San Francisco; Astoria, Oregon.

And this was all brought about,
as you can imagine, by the Gold Rush.

They needed to know
where the tides are.

A person by the – by the name of
Lieutenant Trowbridge heard about an

entire fleet of the Russian Navy being
sunk by a tsunami in Japan in 1854.

So he thought, okay, this tsunami might
have been recorded in San Francisco.

They knew the speed
of tsunamis back then.

It’s by this formula I’ll refer to –
square root of g-h.

G is gravitational acceleration –
9.8 meters per second squared.

H is the water depth
in the ocean in meters.

So when you multiply them together,
take the square root,

you get a velocity of
meters per second.

So they knew how far Japan was
from San Francisco and San Diego.

And sure enough,
the times matched up perfectly.

So this was the first recording of a
tsunami on the U.S. West Coast.

Now, San Francisco has the fame
for having the longest record of tides

in the – in the entire United States and
has recorded 57 tsunamis since then.

These were the
first ones, though.

So why all the fuss
about tide gauge records?

Scientifically, there’s a lot of
information in tide gauge records.

So here we’re going to look at
more recent tide gauges

from the 2004
Indian Ocean tsunami.

There's two of them here.
So again, time along the horizontal

axis and sea surface amplitude here.
In this case, here’s high tide.

It’s coming in Phuket, Thailand.
And low tide over here.

So the Maldives are located to the west
of the big magnitude 9.2 earthquake

in the Indian Ocean that
caused that large tsunami.

Phuket, Thailand, is to the east.
Okay, just keeping that in mind.

So if you know the time
of the earthquake, the arrival time

of the tsunami is going to tell you
where the source of the tsunami

is located because
we know the velocity.

If you know the – well, we know we can
read off the amplitude of the first waves.

That’ll give you an idea of how much
the seafloor moved up and down.

Then the polarity of the waves,
importantly, tell us which way

the seafloor moved –
up or down.

Now, if you take this slice from life
of a tsunami, if you recall that it’s

moving up initially in the transoceanic
tsunami towards the Maldives.

And that’s what
we’re seeing over here.

In Phuket, Thailand,
it’s moving down first,

and that’s what we’re seeing over here.
So you get some idea of what

the seafloor was doing
right from the start there.

And finally, the period of these first
waves tells you how wide the source is,

or the horizontal
dimensions of the source.

So we can use a lot of information
from tide gauge records.

If you back out the tide, you get
something that looks like a seismogram.

We’re call them
marograms, actually.

So on to our four categories
of non-seismic tsunamis.

Volcanic tsunamis.
We’re going to just look at Krakatoa.

So unfortunately, I don't have time
to look at some of these others,

so we’re going to take flank failures
and volcanic earthquakes off.

We’re going to look at, first,
volcanic explosions,

then pyroclastic flows –
I’ll explain what those are –

and then caldera collapses.

So this is kind of how
a volcanic eruption evolves.

First we have the explosion,
in the case of Krakatoa.

Explosions are powered by the
upper thrust of expanding gases

in the magma
chamber here.

It sends off a huge column of ash
and gases into the upper atmosphere.

And, importantly,
and maybe not as well known,

is a shockwave coming off
the volcanic explosion.

I’ll turn this either as
a shockwave or pressure wave

or maybe acoustic wave
if you can hear it.

A lot of times, the higher
frequencies will fall off,

so when you’re recording this
around the world, it’s actually

termed an infrasonic wave.
And a lot of people are recording

these type of waves around the
world with infrasonic sensors.

And if you cover this explosion
up with water, the initial motion

of this would be in the up direction.
Hopefully that makes sense.

And they're rated similar to a
magnitude scale for earthquakes.

It’s called the Volcanic
Explosivity Index.

The person that came up
with this index is – one of the authors

is from the USGS –
Chris Newhall.

So Krakatoa is over here – a VEI of 6.
It’s not the largest explosion in history.

That occurred in Mount Tambora
just to the – to the east of Java.

But it’s certainly one of the largest,
and you can – you can see some of

the other historic eruptions up here.
But then you look in geologic time here.

You know, we have – close to us
we have Crater Lake, of course,

and Long Valley Caldera with
much larger explosions going on.

So that’s volcanic explosions.
Pyroclastic flows – pyroclastic flows

are to be distinguished
from lava flows.

Pyroclastic – it’s kind of a broad
term that includes rock, gas, ashes.

But here in this nice cartoon from
the USGS volcanoes group,

you can see they can move
very quickly from the top

of the volcano and down and
do a lot of destruction in their path.

And they can enter in the water.
So we’ve had, in particular,

two scientists at our volcano
observatories studying how

pyroclastic flows enter into various
water bodies – lakes and oceans.

Chris Waythomas at the
Alaska Volcano Observatory

and Joe Walder at the
Cascades Volcano Observatory.

And pyroclastic flows are really
interesting because you have this

really light, low-density layer here
that isn't going to go in the ocean.

It floats right on
top of the ocean.

And this was actually
observed from Krakatoa.

But then you have this heavier material
that’s splashing into this zone and then

eventually causing this coherent wave,
which is what we're interested in.

It’s going to propagate
outward from the volcano.

And that also is going to
start in the up direction.

And then finally we have our
caldera collapse when large volumes

of magma are erupted
over a short period of time.

Structural support of the crust
above the magma chamber is lost,

and then the ground
surface collapses downward.

So if you have water over this,
the initial motion of the water

is going to be in
the downward direction.

Now, that’s the volcano.
In terms of hydrodynamics,

something else is –
interesting is going on here.

So this is termed in
physics a monopole source.

So most of these unusual
sources are monopole sources.

It’s just going down
in one direction only.

The earthquake, if you remember,
is going down and up.

This one’s just
going down.

But as the waves propagate
away from this monopole source,

they bounce back up.
And they'll continue to do this.

They'll oscillate in the source.
So you have these waves

propagating outward, but then you have
this oscillation going on right at the

source zone far in time after these
waves have left the source region.

So to show you what
this might look – in a simulation.

This is a caldera collapse from a volcano
off the southern tip of Kyushu Island.

So here’s Japan.
Here’s the island of Kyushu.

KiKai Volcano erupted 5300 BCE.
So what we’re showing here is a

map view – again, not a great map,
but the red is coastline of Kyushu here.

And these are
snapshots in time.

So we’re starting at 12 minutes and
going along here to 96 minutes.

Here is our
caldera collapse.

It generates this big hole in the
ocean and very strong negative wave.

But then, soon after,
we get this positive wave here.

That’s the wave that’s going to
run ashore and cause some inundation

to the islands around the volcano.
So those are caldera collapses.

So let’s go back
to Krakatoa.

And to get oriented again,
here’s Krakatoa.

We’re interested in the
tsunami that’s going towards

the Jakarta tide
gauge station.

All right, here’s where we can use some
of that information from the tide gauge.

Now, here’s the original – or a copy
of the Jakarta tide gauge record.

It’s clipped here.
These tide gauge stations

were not designed to record the full
extent of some of these big tsunamis.

But you can see that the
initial motion from the

tide gauge record is
in the up direction.

So that means we can exclude
the caldera collapse mechanism.

It’s also coming in
a little late there.

So we’re left with the pyroclastic
flow and the volcanic explosion.

Now, to get these larger run-ups,
especially that 41-meter run-up here,

it turns out that the pyroclastic
flow is not going to do it.

You need a volcanic explosion.
And there’s still some controversy

about this, but I think that’s more or less
the scientific consensus right now.

So in the near field,
near the volcano,

you needed the volcanic explosion
to generate the tsunami.

So as I mentioned, the air pressure
wave from the volcanic explosion

was recorded all
around the world.

But interestingly, we have
these tide gauge records

from all around the world
recording tsunami-like waves.

Okay, so these tide gauge –
so these are – the ones that recorded

the tsunami-like waves from Krakatoa
are shown by the black circles here.

The ones in the Indian Ocean are
probably from that volcanic explosion,

just typical tsunami waves.

But then we have the interesting
records from the English Channel,

from Panama, from Alaska,
from Hawaii, and San Francisco.

And here’s that record
from 1883 in Sausalito.

The San Francisco tide gauge
station kind of moved around.

It went from Presidio to Sausalito –
now it’s in Fort Point.

But here are these waves.

Now, here is when we’re expecting the
tsunami to come in from Krakatoa.

We know this pretty well because it
just depends on the water depth.

So we can calculate that
with pretty good accuracy.

What it has to do is – and we know this
from the 2004 Indian Ocean tsunami –

it has to come around Australia,
go into the South Pacific,

and then come up all the way into
the Pacific to San Francisco.

These waves right here, which are
the biggest waves from Krakatoa

recorded at the San Francisco, or
Sausalito, tide gauge station,

are coming in 5-1/2 hours earlier,
so way earlier than what we were

expecting the tsunami to come in at.
So that’s our problem number 2.

So we’re going to leave that there and
look a little bit on asteroid tsunamis.

And probably what comes to mind
of asteroid tsunamis is Chicxulub

impact crater, which occurred
right off the Yucatan Peninsula

on the Campeche Shelf.

This occurred at the Cretaceous
and Paleogene boundary –

a period boundary in geologic time.
And it might – it was probably the

triggering event for the die-off
of the dinosaurs, of course.

But what we’re interested in here is the
tsunamis generated from this impact.

Geologic deposits that look an
awful lot like tsunami deposits

have been found all around,
from Mexico into Texas.

This is probably the most famous
one near the Brazos River

around the Gulf Coast –
in the Atlantic, in the Caribbean.

So the question here is, did the impact
itself generate the tsunami, or did it –

it must have caused a huge
amount of ground shaking.

Did that let loose a lot of landslides
in the Gulf of Mexico, and did those

landslides generate the tsunami?
So that’s our question.

We didn't have a tide gauge station,
unfortunately, in the Cretaceous.

[laughter] But we did have
these geologic deposits.

We can glean some
information from those.

And we can look at model simulations.
So this is – we are fortunate to have this.

This is the latest type
of asteroid simulation.

This is coming from – courtesy of
Dr. Darrel Robertson at the

NASA Ames Research Center
down in Mountain View.

This is brand-new simulations, and these
are pretty sophisticated simulations,

so I’m going to have to do
a little bit of explaining for these.

So we’re – what we have here
is an asteroid of very large size

coming directly vertical.

So it’s coming straight up and
down into the – into the ocean.

Now, because that’s – see if I can –
I think I messed this up here.

Okay, so here the
asteroid hits now.

So we’re only seeing
one half of the simulation.

Because of symmetry of the problem,
we only have to compute one half

of the simulation.
So in your minds, picture a mirror

image of this on the left-hand side,
and try to piece them together.

So just realize we’re only looking at
one half of the simulation.

This gray is – this is solid earth.
The blue is the water.

The black is the atmosphere, okay?
So the boundary between the black and

the blue is our tsunami wave height –
or wave height – I’m trying to hold off

on the tsunami part of it.
This is the wave height.

So here we’re seeing this –
already this huge splash

in the ocean
from the impact.

The other information that’s
conveyed in these simulations

is this color
gradation here.

Now these are essentially the
pressure waves caused by the impact.

So we have pressure
waves in the ocean.

There's pressure waves in the
solid earth that you can't really see.

And a lot of air flow
associated with the impact.

So there’s a lot of information in this.
So I’ll go ahead and continue it.

Recall from our – this is
another monopole source.

So you’re going to get this rebound
right in the source region as waves

start to propagate outwards. And it’s
going to kind of oscillate along here.

See, all these pressure waves
bouncing back and – these are

basically acoustic waves bouncing
back and forth in the ocean.

They're acting like a
wave guide in the ocean.

And a lot of this complex
flow in the atmosphere.

Now we’re going to speed this animation
on a little bit and see – okay, here are

waves, but remember from life of
a tsunami, these waves are short.

They're less than – the wavelength
of the waves from peak to peak

is much less than the ocean is deep.
So these are not tsunami waves.

These are deep ocean waves akin to –
they're very large amplitude, but they're

akin to a deep water swell you'd get on
a fishing boat or something like that.

The problem for a deep ocean impact –
it’s going to probably break at the

continental shelf edge and
really not do that much.

So this is probably an anticlimactic
part of the talk, especially for

fans of the Deep Impact movie.

Asteroids are not efficient tsunami
generators, believe it or not.

And I was talking to my postdoc after
the practice talk, and he said, yeah,

it’s almost – the water is almost –
it’s happening too fast.

The water is almost behaving
as a solid in this case instead of a –

instead of a liquid.
So that was one of the new things,

I think, coming out in the
world of asteroid tsunamis.

So now we’re going to
go on to landslide tsunamis.

Here it’s important to distinguish
between landslides occurring

above the water and those occurring
below the water, and I’ll show you why.

Most of our case studies –
all except one –

occur in Alaska
for a good reason.

Because we have the glaciers
retreating, leaving these

unstable hillsides of the fjords
such that you can have these

massive piles of rock fall into
the deep water of the fjords.

So a lot of these case studies
are going to be from Alaska.

First one we’re going to
talk about is Lituya Bay,

which people here
probably have heard of.

There’s, I think, a Discovery Channel
program on mega tsunamis on this.

This is the – on record for having the
highest tsunami run-up in Lituya Bay.

But what people are probably
less aware of is a similar event

occurred just a year and a half ago
in the Taan Fjord in Icy Bay here.

So to get oriented,
here’s Skagway, Juneau.

Glacier Bay National Park is here,
and Anchorage sits way over here.

We’ll look at a case that hasn't
happened yet in Tidal Inlet

in Glacier Bay National Park.
And then in Valdez – this is going to be

our case of a submarine landslide that
was triggered by the magnitude 9.2

big earthquake off of Alaska that
generated the tsunami in Crescent City.

But we’re just going to be looking at
the landslide tsunami in Port Valdez.

So let’s start with
Lituya Bay and see –

this is a famous USGS photo looking
from the ocean into the bay.

Common with these type of
mega tsunamis is you see that all

the vegetation and trees just sheared off
right through here and inside the bay

right here from the tsunami and
the aftermath of the tsunami.

The rockfall actually occurred
just around the corner here.

So this view is looking up towards
this direction where the glacier –

Lituya Glacier sits here.
So this big pile of rocks

fell into the ocean, and this
nice simulation from Professor

Juan Horrillo – let’s see if
I can get that started –

shows exactly what happened.

So you have this big splash going in
here, and then this wave go up to the –

and he’s able to exactly match that
run-up. I don't know how he does that.

But 524 meters, so that's
a huge wave, of course.

Now, it swept over the nose of this ridge,
and this ridge is right here.

So that’s that
kind of tsunami.

Again, something very similar
just happened in the Taan Fjord.

Captured the attention of a lot of tsunami
scientists that have gone up here.

These photos are courtesy of
Dr. Bretwood Higman,

who has done a lot
of work up there.

Here is our big landslide occurring next
to a glacier here, fell into the water,

stripped a lot of the vegetation,
even far – the landslide is

way back here, so the vegetation is
still affected way this far into the fjord.

And satellite views of before and after
of what it looked like after the tsunami.

This tsunami was 192 meters.
Still a pretty big tsunami.

In fact, the fourth-largest
tsunami ever recorded.

So now we’re on to Tidal Inlet.
Tidal Inlet is this body –

well, you've probably never heard of.
It’s this really narrow body of water

that sits off the western arm of Glacier
Bay in Glacier Bay National Park.

So if you venture in – if you’re brave
enough to venture into Tidal Inlet,

you see this
feature here.

And this is a landslide that has only
slipped down the slope part way,

[chuckles] not quite all the way yet.
And it’s been like that for a while.

So our concern, of course, what if
this huge body of rock were to

fall into Tidal Inlet, what kind of
wave is that going to generate?

And this is that simulation.
It’s going to probably generate

very large waves inside Tidal Inlet.
What we are more concerned with

are the cruise ships that go up and
down in Glacier Bay and what

effect these kind of waves are
going to have on a cruise ship.

So you can see the waves are quite
different as it goes into the deep water.

They're longer waves, smaller amplitude
waves, but still of concern, obviously.

So now
submarine landslides.

I hope to give you the impression
that submarine landslides are a

different beast altogether than
landslides occurring above the water.

So submarine landslides are also –
are triggered by earthquakes.

So what happens in a submarine
landslide is that you have this

big pile of rocks that fails and
starts moving down the slope.

Where it fails, we call this
the excavation region.

It’s going to suck the
water right down with it.

So we get a
negative-amplitude wave.

And then, as these rocks move down
the slope, it pushes the water up.

So here we’re going to
get a positive wave.

So we’re – this is a dipole wave
just like our earthquake –

positive and
negative here.

Unlike earthquakes, though, this is
occurring at much slower speeds than –

earthquakes happen a lot
faster than tsunami speeds.

This is happening
at an unknown speed.

This is the frustrating thing about
submarine landslides is we have no –

virtually no idea of how fast
these rocks are moving underwater.

We can tell above water, but underwater,
there’s very little information.

So we need other data.

And often, there's just not enough data
to really understand these waves.

So we’re really good at mapping these –
the aftermath of these landslides.

This is off Santa Barbara
from the Monterey Bay

Aquarium Research Institute.

But – and for tsunami purposes, we need
to know the speed of these waves.

And to demonstrate that point,
we’re going to have two experiments.

One is we’re going to
have a landslide go off

the continental shelf edge
at 50 meters per second.

The tsunami waves want to
move at 200 meters per second.

So we’re going to start this
and stop this a little bit here.

So landslide is underwater.
We’re seeing the surface of the water.

Here’s our landslide.

It pushes up waves, but no more than –
it starts to push them up.

The tsunami waves start to
zip off really quickly.

So they never build up
to large amplitude.

Now, if we have an experiment
set up so that our landslide

is moving at the exact right velocity –
200 meters per second,

we’re going to get very large
amplitude waves moving

in the direction of landslide motion.
The wave coming back is still small,

but it’s complexly related to landslide
motion. But concentrate on this.

So this is what I’m going to
call a coupled source.

That means that the source of the
tsunami has to be coupled, in terms of

at the same speed that tsunami
waves want to move in the ocean.

That's square root of g-h.
Remember that.

That’s the only formula you’re
ever going to hear in this talk.

Square root of g-h.

Okay, so here’s our case for a submarine
landslide in Port Valdez here.

So, like our caldera
collapse case study,

this is a case study for the
1964 tsunami in Port Valdez.

This is map view of Port Valdez here –
the eastern part of Port Valdez.

One minute – these are snapshots,
again – two minutes,

and then down
to six minutes.

Now, here we have a really
interesting point of observation.

We have a ship – a freighter called
the S.S. Chena that was docked

at the warehouse in the harbor
of Port Valdez located here.

And by chance somebody was
taking a home movie pointed straight

at the docks during the time of
the earthquake and the tsunami.

The landslide occurred
right at the docks, right at the –

it’s called the
delta front here.

So you can go to YouTube,
look for 1964 earthquake at Valdez,

and you see this movie. It’s actually
pretty frightening to see this tsunami –

the submarine landslide go on.

But the PI for this project was Tom
Parsons, and from those home movies,

he was actually to find the
position of the S.S. Chena,

and we can use that to constrain our
numerical models of the tsunami.

And colors, again,
is the tsunami model.

The blue represents the big negative
amplitude over the excavation region,

and the red is the effects
from the deposition region.

The positive wave comes back towards
shore, runs up into Old Valdez.

By the way, Old Valdez has
since been moved over to here

where it's New Valdez.
The Chena ran aground,

ran through some of the docks there,

and then got pushed to the southeast,
and then eventually broke free.

But this was helpful in actually
helping to constrain and verify our

motions of the landslide movement.
Otherwise, we had no idea.

You can model the landslide itself,
but that’s a whole nother

matter altogether.

Okay, so I’ll probably hear from this
from many fellow scientists hearing,

but as far as I can tell, we only have
one instrumental record that’s clearly

a submarine landslide from a tide
gauge station in the whole world,

and it happens to be in our own
backyard in Monterey Bay.

So, as I mentioned, most of the
submarine landslide tsunamis are

triggered first by the earthquakes,
and the earthquakes are often offshore.

So you have an earthquake tsunami
and a landslide tsunami going on

at the same time, and it’s really
hard to separate the two signals

because they're about
the same wavelengths.

But here – of course, this is the
Loma Prieta earthquake in 1989.

It occurred on land, of course,
along the San Andreas Fault.

Triggered some ground –
strong ground shaking, and a

tsunami was recorded at the Monterey
tide gauge station in Monterey harbor.

And produced about a 1-meter
tsunami run-up in Moss Landing.

So if you recall back at the
beginning of the talk,

what information we can
gather from tide gauge records.

First, the arrival time tells us where
the tsunami source was located.

Because we have no idea where the
tsunami source really was from this.

You could tell – you could try –
from this arrival time, you can try

and draw an arc to
possible source locations.

And not surprising, it has to be
somewhere in Monterey Canyon.

You didn't really need
the tide gauge for that.

But look at this polarity.
This is the first upswing of the tsunami.

So that means, if you recall, the
deposition region – the tsunami –

the landslide had to be moving towards
the Monterey tide gauge station.

That is, it’s seeing the wave
from the deposition region

from that landslide first.
So that means the landslide had to occur

on the north wall – somewhere along
the north wall of Monterey Canyon.

Now, this part is a little
less obvious and hopefully

it might be apparent
in this animation.

The landslide could have also occurred
along the axis of the canyon.

The waves from the deposition
region are going to refract,

or bend towards, Monterey.
And actually, this is the most probable

source for that particular landslide.
But this is an important record.

This gives us an instrumental record –
maybe the only one –

of a submarine
landslide tsunami.

Okay, now to atmospheric tsunamis.
This is something new that I've

been interested in, and it’s kind of
a joy to work with, because unlike

submarine landslide tsunamis,
we have a huge amount of data.

It’s a complicated process, but we
have a lot of data to study these.

And they don't tend to be that –
they're not mega tsunamis.

They're smaller tsunamis. They do
do some damage in small boat harbors.

This is in the Balearic Islands off Spain.
I’m going to call these meteotsunamis,

but you can see what they're called in all
these different languages and cultures.

Get an idea of where they usually occur.
They usually occur in the Mediterranean,

off Japan – in the U.S., they occur
off mostly the U.S. Atlantic coast.

The Gulf Coast – Florida has seen some
recently. And of course, the Great Lakes.

So these, surprisingly, occur,
in terms of the physics, very similar

to submarine landslides. This is
another example of a coupled tsunami.

So what happens is,
you have a pressure jump.

This pressure jump results in the
deflection of the sea surface.

It’s termed the
inverted barometer effect.

Inverted because a positive pressure
jump is going to result in a

negative deflection of the
sea surface, and vice versa.

So 1 millibar of pressure change
is going to generate 1 centimeter

of deflection in
the ocean’s surface.

Not big. I mean, that’s not even hardly
measurable in tide gauge stations.

But if this is moving
at the right velocity –

again, square root of g-h –
there’s that term again –

these waves are going to start
to build and become significant.

Then you can also have other
sources of amplification –

that shoaling amplification,
if you remember.

They call it shelf amplification here –
and some other sources of amplification.

So when conditions are right,
these can be significant waves.

The key study we’re going to look at
is from 2013 – not that long ago.

And I don't know if you remember,
there was a significant weather

system called a derecho.
A derecho is basically just an

organized squall line that’s fairly
extensive and moving quite fast.

There’s very strong winds associated
with a derecho moving in the

straight direction, as opposed
to a vortex like a tornado.

So that’s where derecho gets its name –
“derecho” meaning straight in Spanish.

So here’s – so it’s moving around
50 miles per hour across the

northeastern United States.
The National Weather Service stops

these radar pictures right at the coastline,
but of course, what we want is a radar

depiction offshore, and we have one,
fortunately, showing this derecho

moving offshore along the –
this is the Atlantic continental

shelf here, and this is the
deep part of the Atlantic here.

So there were 13 tide gauge
stations in the Atlantic

that recorded
this meteotsunami.

I’m going to show a simulation.
It’s kind of an odd perspective.

This is in the southeast direction.
All right, here we go.

You'll probably recognize
some of these features.

This is obviously Long Island,
Delaware Bay, New Jersey.

Cape Cod over here.
This is the Hudson River.

And, as I’ll show you,
that becomes important.

It goes along the continental shelf here,
and right at the continental shelf edge,

there’s Hudson Canyon.
Not as impressive as our Monterey Bay

Canyon, but still, I guess by Atlantic
standards, it’s okay. [laughter]

So here’s our simulation.

So what I’m showing – I’m not
showing anything in the atmosphere.

I’m showing what’s coupled
with the water column.

Here is our derecho coupled with
the ocean moving at a really –

really the right speed for
tsunami waves wanting to move.

So, as you can see, it starts to get
larger and larger as it goes on there.

At this point, the derecho hits
the continental shelf edge.

Now it’s not moving at the
right speed because these waves

are moving a lot faster.
This is deep water now.

So these are de-coupled waves
moving out into the Atlantic.

And interestingly, these waves
are actually recorded in Bermuda

and Puerto Rico from
this particular event.

Continue this along. What we now
see is another de-coupled wave.

So here’s the coupled wave.

The derecho is still
moving in this direction.

This wave is from the continental shelf
edge moving towards Long Island.

But what we were actually really
concerned with in terms of hazards,

some people were knocked off
the sea wall in New Jersey here.

And this reflection, I believe,
is coming from Hudson Canyon.

That’s at least my interpretation of
where this de-coupled – this wave is a

reflection from Hudson Canyon where
it intersects continental shelf edge.

Let’s see. What more to say about that?
Oh, so [chuckles] interesting story.

When we were looking at this event,
we saw all these tide gauge records

go off, and you can triangulate.
You have three tide gauge stations,

so you can triangulate where
these sources were occurring.

And we looked at it and said, okay,
these sources must be occurring

on the continental
slope or shelf edge.

These must be submarine
landslide tsunamis.

That’s your first thought
as a geologist, you know?

So I think people actually went
out there looking for the landslide.

But, no, it was just an atmospheric –
just – I mean, it was a meteotsunami.

It did some damage, but they were
coming from the continental shelf edge,

but it has reflections – de-coupled
waves, not from the coupled wave itself.

I’ll continue that along a little bit.
Try to speed things up here.

It kind of wraps around Cape Cod
a little bit and then there's

a little bit of a summary
of what’s going on there.

So another source of meteotsunamis
are atmospheric gravity waves.

These are in the
very high atmosphere.

This comes from air flow over –
for example, a mountain range

that disturbs the stable density
layers of the upper atmosphere.

Buoyancy forces try to
restore the equilibrium.

They overshoot, and oscillation
is setting up these waves

that can move at fairly
significant velocities.

The case study here, which I’m
not going to show, is the 2008,

in Boothbay Harbor, Maine,
that had a 12-foot-high meteotsunami

that was fairly damaging
inside the harbor there.

So atmospheric gravity waves –
these are just the waves

in the atmosphere. We're not
talking about the ocean yet.

These are caused by air flow over
mountains, over super cells –

thunderstorm systems.
The air can flow over the super cells.

And one other source of
atmospheric gravity waves

is directly from
volcanic eruptions.

So you can't see – well, it’s hard to
see these waves from the ground,

but satellite’s really good
at envisioning these –

they're termed
lenticular clouds here.

In this case, we have a large
volcanic eruption in Chile in 2015.

And satellite was able to capture
these concentric rings of

atmospheric gravity waves
coming off from the eruption.

So basically, the volcano is sending up
this big column of ash way up into the

stratosphere and mesosphere and setting
off these atmospheric gravity waves.

Now you can probably guess
we’re getting to the answer to

problem number 2
for Krakatoa.

It turns out that the atmospheric
gravity waves move a little bit faster.

But importantly, they don't
have to go over the ocean.

They can go over land.

They can go right over
the Indonesian archipelago.

They can go over Eurasia
to the English Channel.

So it turns out what we're seeing
on our San Francisco, slash, Sausalito

tide gauge record are gravity –
atmospheric gravity waves that are

going at the right speed to couple with
the water column and cause this

tsunami-like waves in San Francisco
as well as Kodiak, Alaska, and Hawaii.

This particular – we call it a mode –
is going at the water velocity,

I think, of the continental
shelf water depth.

So that’s – go to the summary now.
Those are our four groups of sources,

and I’m going to confuse
everybody even further by

scrambling these sources on this slide.
But there's some method behind this

madness because, rather than grouping
them by geologic feature or atmospheric

process, we’re now going to group
them by how they cause tsunamis.

So first we have
our explosions shown here.

We have our impact sources,
which obviously includes asteroids,

but maybe not obviously,
includes landslides perched on

the sides of fjords
and pyroclastic flows.

For these, I kind of envision that
efficiency of tsunami generation is

inversely proportional to the splash
that’s generated at the source zone.

Then we have these kind of odd
sources I term coupled sources.

So these have to – they can only
generate tsunamis under the

right conditions moving basically at
a velocity – again, square root of g-h –

and over a certain distance –
sufficient distance or duration.

Think of it as fetch, if you’re familiar
with the term “fetch” in terms of

generating regular ocean waves.
So those are the coupled sources.

And then we have
our earthquakes.

We can lump in volcanic
earthquakes with regular earthquakes.

It’s the same sort of
tsunami-generating process.

These are our most efficient
tsunami-generating sources.

Most often observed – 70 to 80%
of the tsunamis, again.

Then we have caldera collapses.
I have no idea what to do

with caldera collapses. [laughter]
They're kind of a beast of their own.

I tried at one point
with earthquakes,

but they're not moving
fast enough as earthquakes.

Anyway, that’s the talk.
A lot of people helped out on this talk.

Bounced a lot of ideas
with one of our brilliant postdocs,

Kenny Ryan, here at the USGS.
Anne Rosenthal.

And Christy Ryan, who is part of
the organization of this

wonderful evening
lecture series.

A lot of people dedicate a lot of time
to making this series possible.

And I really presented the research
results from a lot of different studies,

and a lot of people –
a lot of people were PIs

on these studies,
such as Gerry Wieczorek.

So thank you for your attention,
and thanks for coming out tonight.

Appreciate it.

[ Applause ]

- Thank you, Eric.
That was a great talk.

And now, if anyone has any questions,
you can either line up behind one of

the mics, or I can also
bring you a mic as well.

And we’ll go ahead
and open it up.

- I’ll wait for these
guys to walk out.

Has there ever been a
tsunami in the San Francisco Bay?

- Yes. Absolutely.
Probably the biggest was 1964

from that Alaska –
the magnitude 9.2 Alaska.

It did, in 1964 dollars, a million
dollars of damage inside the bay.

That’s not outside the bay.
That’s inside the bay.

It did some damage to
Berkeley Harbor and

some other locations
I’m forgetting right now.

But even the 2011 Japan tsunami
was observed in the bay.

There’s some interesting
video on YouTube of the

tsunami going
towards Emeryville.

So certainly they're not as big.
The bay is very shallow.

The waves move very slowly,
eat up a lot of energy.

But tsunamis do make
their way into the bay.

And one thing I – maybe it was obvious.
I didn't expressly mention it, but we

can have mega tsunamis, but tsunamis
don't depend on how high they are.

We can have micro tsunamis,
which we study too, you know,

but that just doesn't make
the news. [chuckles]

- That was wonderful.
The meteotsunamis,

what’s the relative frequency
of these things at a –

at an amplitude that’s damaging?
- That’s damaging, yeah.

- I mean, what we often refer to on
the West Coast as rogue waves.

How many of those are
meteotsunamis, and …

- Well, on the West Coast,
not many at all.

I was looking in the catalog, and there’s
a few up around Puget Sound region.

And there was one in 1910 in
San Francisco that might have been

a meteotsunami, so not –
the conditions just aren't quite right.

On the East Coast, that was
actually the focus of our study

is to get at the frequency of
meteotsunamis in general.

Damage involves risk,
which is another calculation.

- Sure.
- But they're actually pretty frequent,

but they just don't
get very large.

So I don't want to get into probability,
but if you look at a distribution tail,

it falls off very quickly
at higher amplitudes.

So in terms of –
when you said damaging waves,

those are much less frequent.
But if you look at smaller

meteotsunami waves,
they're surprisingly more frequent

than regular tsunamis, you know.
- What’s the advice, then, to people

who are in coastal areas or
shallow harbors or things like this?

What – any idea of risk?
- Yeah, in terms – if you have

a small boat tied up in a harbor –
that’s a good one.

I mean, for regular tsunamis,
the advice is, you usually get

hours of warning, and you can take
your boat not very far out and

avoid a lot of the damaging
effects inside the harbor.

But these, it’s hard to tell which
weather system – weather is turbulent,

and it’s hard to tell which one
is going to generate a meteotsunami

and which one isn't.
So it’s a good question to keep in mind.

I mean, yeah. I don't have a good answer,
unfortunately, for it, yeah.

- Thank you.

- So thank you for the talk.
To your left.

So I actually had never heard
about atmospheric tsunamis before,

so that was interesting. But to the
asteroid that hit the Yucatan …

- Yeah.
- That was about 10 kilometers wide.

So are you saying that the depth
of the water there must have been

more than three times that?
Or did the asteroid effects not have the

same effects as the earthquake effects, 
that you said if they're three times

the distance or the depth.
- Yeah. Yeah, so the water depth

it hit in was very shallow.
In fact, half of it was on – well,

I don't know what the coastline was
back in the Cretaceous there.

But it was in shallow water, for sure.
It was on the shelf.

So it would have generated long waves,
but like that simulation,

a lot of it goes into splash.
A lot of the energy – in fact,

that simulation showed about –
only about 20% goes into the ocean

of that energy. And only 1% goes into
generating surface gravity waves.

So, you know, it’s hard to speculate.
There’s these tsunami deposits

that we’ve seen that are –
you know, I’m not a geologist,

but they seem
pretty convincing.

Some say they might be storm deposit,
but the timing seems too coincidental.

I think we just don't know if they
were caused by triggered landslides

from the impact or
from the impact itself.

I think, at least in my mind,
we just can't tell at this point, you know.

- So, but normally, it would have
the same impact – the, you know,

3-to-1 ratio as an earthquake
under normal circumstances?

- Any tsunami source, to generate
long waves, has to be, ideally,

three times greater than the water depth.
Everything in hydrodynamics is relative.

So irrespective of what water depth
you’re at, it has to be three times

the water depth, but …
- But I guess because it was shallow,

and it just splashed,
the mechanics were different.

- Yeah, but something’s
generated in very shallow water,

the reverse of shoaling
amplification happens.

You get de-amplification
as it goes into deep water.

But then, as it goes back into shallow
water, it’ll go back to the same height.

So there’s virtually going to be
no amplifications for a

shallow water tsunami source.
- Okay, thank you very much.

- This is neat as an overview.
I’m David Morrison from NASA Ames.

- Oh, okay.
- And I’m the chief scientist in the

group that’s been doing the asteroid.
- Oh, great, great.

- Which is, I’m sure Darrel Robertson
will be very glad you used his image.

But you might be
interested in the result.

We had a workshop,
and Darrel’s work was part of it.

NASA, NOAA, people from FEMA
and DHS and the various

Department of Energy labs put together
to try to figure out if the tsunamis

from relatively small impacts
in the ocean were a significant risk.

We know what happens when it’s –
when an asteroid hits the ground.

- Yeah.
- And the conclusion was

they are not significant.
- Okay.

- They're down an order of magnitude.
In spite of all the

wonderful Hollywood movies …
- Right, right.

- … that show a gigantic asteroid
produce a tsunami.

It still would be a mess
if it hit near the shore, of course.

- Yeah.

- It’d be a big splash like a –
like a landslide.

But we aren't worrying about something
going in the middle of the Pacific and

coming and hitting San Francisco.
- Right, right, right, right.

Well, I don't know.

Maybe you can, you know,
help answer the Chicxulub question.

- None of the people that I was
working with are looking at that.

Fortunately, we know there are no
asteroids as large as Chicxulub out there.

- Yeah, yeah.
- So we’ve been dealing with things

that are orders of magnitude smaller.
- Right.

- Like a 100-meter.
- Right, right, right.

Thank you for that, by the way.
That’s great. Glad to hear.

- I didn't get the scale
on the tide gauges. Is …

- Okay. Let’s go to the
2004 tide gauge record.

That’s probably our best one because
it shows the scale and the time.

A lot of times, the raw records,
it’s really hard to make out

the time markings on there.
So this is hours after the earthquake.

We put T equals zero here
at the time of the earthquake.

Usually it’s just the local
time of day marked on here.

And then amplitude here.

Zero, presumably is –
usually means sea level.

- Okay.

- And then amplitude
above that in meters.

- So is this stuff all computerized?
Can you …

- It is now.
And thank you for that.

Because you can go around,
let's see, and search for

NOAA tsunami tide gauges.
And what I do – and what anybody

in the audience can do if you hear of
a tsunami warning somewhere, is to,

in real time, sit at your computer and
look at these stations and show the

tsunami coming on the tide gauges.
It shows you the ones with the tide

and the ones with the tide removed,
called the residual.

So you can watch – it’s only in the U.S.,
and there’s other sites with the

worldwide tide gauges.
You can watch the tsunamis

as they come ashore at these
tide gauge stations, so yeah.

- So has anybody gone and looked at all
of the old data to see if there’s anything

interesting that they didn't notice before?
- Yeah. I mean, the – certainly there’s

a whole group led by this person by
the name of Professor Kenji Satake –

Satake-sensei now – in Japan,
who is a fairly famous tsunamian.

He was known for inverting
all these waveforms to get at

what the source was like.
That’s how he’s famous.

So he’s been doing
that for decades now.

And it’s especially important for
areas where we don't have

earthquake records and
for non-seismic tsunamis.

So, yeah, this is – this is an
important field of research.

- Do any of the offshore
buoys help in this?

- Yeah. So the offshore buoys,
I didn't mention.

Since especially the 2004 Indian Ocean
tsunami, there's deep ocean pressure

buoys put out by NOAA
specifically for tsunami warning.

They're recording the tide.
They're very precise and

accurate instruments to record
the very clear signal of the tsunami.

They really help in
real-time tsunami warning.

But you could – and people –
especially for the 2011 Japan tsunami,

they were using those buoys records as
well as the tide gauge records to invert

for the source process of that massive
earthquake off of Japan in 2011.

- Thanks.
- Yeah.

- For your meteotsunami …
- Yeah.

- … you said it was
1 centimeter per millbar.

- That’s – yeah, that’s …

- Is that – is that measuring from
standard sea level pressure?

And is it going …
- Oh. [chuckles]

- … deflecting up or down?

- Well, it’s inverted, you know.
The increase in pressure is

going to result in a deflection
of the ocean’s surface.

I’m sure – I think that’s a
rough, rough conversion.

Probably depends on pressure,
temperature, and a bunch of other things.

I just threw that to get an order of
magnitude scaling of the pressure effect,

but I’m sure there’s a lot of
other variables in there.

- Okay.
- Yeah.

- One other – there’s – again,
it probably was a Discovery

Channel show from years ago.
- Yeah.

- Some island in the Atlantic …
- Yep.

- … that has – like, a third of
the island is slowly slipping off.

Has there been any further study
on such a large landslide that

could be generated by that?
- Yes. There’s been a lot of studies. [laughs]

That’s from the Canary Island,
Cumbre Vieja islands in particular.

The hypothesis was there
was going to be a huge landslide

fall into the ocean and create this
large mega tsunami that will

travel across the Atlantic and
potentially affect the U.S. East Coast.

It was an interesting
thought-provoking study, for sure.

Since then, volcanologists
have looked at the volcano more.

I can't really report on that.
In terms of hydrodynamics,

people are using more sophisticated,
I would say, hydrodynamic models

than in that initial study showing that
the amplitude, especially on the

U.S. East Coast, is smaller than
what was originally predicted.

But still pretty significant.
I mean, even a – well, for a comparison,

a half-a-meter tsunami
is going to knock you down.

Think of crossing a half-meter river
up in the Sierra or something like that.

So it’s – I think, in round estimates,
maybe 1 to 2 meters from a

hypothesized event if those
source parameters are correct.

But that’s a whole other
area of controversy.

But that’s still significant on the
U.S. East Coast, especially because

the coastal plain is pretty flat there.
You know, so it’s down a little bit,

but it’s still of interesting
concern, I would say.

- Thank you.

- So analogous, in many ways,
to the asteroids.

You produce an impact in the middle of
the ocean, and it doesn't propagate …

- Yeah. It’s – yeah, so land –
that is correct.

Landslide tsunamis are similar
to asteroids in that they are

generating shorter wavelength waves.
And if they're tall enough,

they're going to maybe break
at the continental shelf edge.

It’s actually given a name
called the Van Dorn Effect.

Thought of years ago and
now we’re just revisiting it.

So, yes, you’re correct.

They might break at
the continental shelf edge as well.

Have you looked at any of the
Hawaiian cliff failures,

and is there usable data there?
I think when I was down in the island,

I remember seeing
run-ups of 100, 150 meters,

so it would seem like
those were pretty big events.

- Yeah. So those were – I actually
haven't modeled those events.

And actually, I think they were
discovered by a USGS scientist.

George Moore –
is that right, Helen? Or …

- [inaudible]
- You don't remember? Yeah.

But, yeah, those were significant events.
And supposedly, they've seen some

tsunami deposits from the Hawaiian
islands from those events.

I had – I have a whole talk worth
of other case studies I was going to

talk about, but one of them was a
Hawaiian example from –

but this was a volcanic
earthquake tsunami from 1975 –

the Kalapana tsunami,
which I did model.

And that’s a complex event involving
slip on a fault but also kind of

collapse of the area around
Kilauea Volcano as well.

So it was a very complex event,
but a volcanic tsunami

that involved an
earthquake in that case.

The tsunami run-ups from the
1975 Kalapana event weren't nearly

as large as what you were talking
about here, but still significant.

It was measured on
the West Coast here, yeah.

- So I was kind of interested in
the derecho that was coming …

- Closer to the mic.
- Okay. [laughs]

I was kind of interested in the derecho
that came across the country and

ended up at Moss Beach.
Is that what you said?

- The derecho started in the Midwest and
moved in the northeast United States.

Kind of ended up going off the
continental shelf edge off –

into New England
up into Canada.

- Oh, I thought I – I thought it ended up
at Moss Beach. Did you mention …

- So the – yeah, the – I went through
things pretty quickly there. [chuckles]

- Okay.
- So there was a submarine landslide

tsunami in Monterey Bay from the
Loma Prieta earthquake in 1989

that did generate a small tsunami
run-up of 1 meter at Moss Landing.

You know, kind of halfway
around Monterey Bay there.

They had a lot of liquefaction,
which was more the problem from

the Loma Prieta earthquake.
But there was a little tsunami there.

- Okay. Because I was –
kind of interesting because

I actually know two people that –
one that just bought property

and one that also did right around there.
It’s quite funny because, in this –

the one who just bought something
in Pacifica, which kept talking –

she kept talking about liquefaction
or whatever it was and that it

wasn't going to be a problem
because of where she bought it.

And I’m thinking, wow, so what
actually happened right there?

- Yeah.
- Is this a good idea, you know,

to buy beachfront property …
- Yeah, yeah, yeah.

Well, that’s one of the questions,
unfortunately, we can't tell specific,

you know, property
questions of [inaudible] …

- It’s funny because it’s a small area.
- Yeah.

- I mean, it was so pinpointed,
and I wondered how – you know.

- Yeah. I don't know if you saw in
that simulation, but that tsunami

was really resonating. It was going
back and forth in Monterey Bay.

One of the curious things I've always
wondered about, and this happened

in the 2011 tsunami, is we saw
the damage in Santa Cruz.

Monterey doesn't seem to have the
big effects as Santa Cruz does

for some reason.
There’s something about the

wave resonance in Monterey
that Santa Cruz gets the brunt of it,

and Monterey doesn't.
I don't understand.

But I did want to mention,
for hazard purposes,

the State of California has
tsunami inundation maps online.

So if you’re interested in any part
of the coastline of – most any

populated part of the coastline of
California, you can go to the

California Geological Survey website
and pull up those inundation maps.

And you can see, that’s kind of
the worst-case scenario

if you have any
questions about that.

- Okay, we’ll go ahead
and take one more question.

- I get the last question.
- Yep.

- Do tsunami waves
dissipate on their own?

- Yes. They – yeah, once they're
generated, they dissipate

very quickly in the open ocean.
They got a lot of energy behind them.

They start dissipating
once they hit land.

This is a whole part – well,
here we go, the tide gauge record.

So here’s hours
from the earthquake.

You can see, after 12 hours, the tsunami
wave activity is still going on.

This is characterized by an exponential –
I’m getting a little technical here – but

exponential envelope with a decay time
of about 22 – in the Pacific, 22 hours.

So from the 2004 Indian Ocean tsunami,
we saw tide gauge records in India

going on for over a day – the tsunami
waves were keeping going.

But they gradually, exponentially,
start to decay.

But it’s important to realize,
first wave is not the biggest.

And it looks like in Maldives,
it was the biggest here.

In Phuket, Thailand – for some reason,
it’s often the third wave.

And people have been
wondering about that.

So this isn't the first wave
coming directly from the source.

There's some kind of coastal
interaction going on to make the –

and that’s tragically what happened
in Crescent City in 1964.

People came back too early and were
caught in the tsunami from 1964.

Because these are long-period –
as you can tell – going on for 45 minutes

or an hour between waves,
you know, so it’s easy to get caught in

and go back too soon.
So it’s always best to stay away.

- Okay. Again, I want to thank
everybody for coming tonight.

And if we can get another
round of applause for Eric.

- Thank you.

[ Applause ]

- That was a fascinating talk.
- Oh, thank you for coming. Yes.

- Quick question.
- Sure. Yeah.

- Is it standard to use the term …

[ Silence ]