WEBVTT
Kind: captions
Language:  en

00:00:01.920 --> 00:00:08.600
[inaudible background conversations]

00:00:09.520 --> 00:00:11.920
Good evening.

00:00:11.920 --> 00:00:13.690
Welcome to the
U.S. Geological Survey

00:00:13.690 --> 00:00:17.580
in another installment of our
evening public lecture series.

00:00:17.580 --> 00:00:20.750
My name is Leslie Gordon,
and it’s always my pleasure

00:00:20.750 --> 00:00:25.670
and honor to introduce
the speakers each month.

00:00:25.670 --> 00:00:27.420
Before I do introduce
tonight’s speaker,

00:00:27.420 --> 00:00:30.090
I want to make sure you
all come back next month.

00:00:30.090 --> 00:00:33.949
So there are flyers on the back
corner table if you didn’t pick one up.

00:00:33.949 --> 00:00:38.260
Our speaker next month is
hydrologist Marjorie Schulz.

00:00:38.260 --> 00:00:40.519
She will be speaking
about the Marine Terraces

00:00:40.519 --> 00:00:44.290
of California: Landscapes
from the Waves.

00:00:44.290 --> 00:00:46.280
And any of you that have ever
been over to the coast --

00:00:46.280 --> 00:00:48.469
San Mateo County,
Santa Cruz County --

00:00:48.469 --> 00:00:52.920
you’ve seen those significant land forms,
the marine terraces, and they have a

00:00:52.920 --> 00:00:59.399
really interesting geological story.
So please do join us next month.

00:01:02.160 --> 00:01:04.100
I know we have
many people here that have

00:01:04.110 --> 00:01:07.229
been coming for years,
literally, to our show.

00:01:07.229 --> 00:01:11.479
And if you have, many of you have
actually heard Ross Stein speak before.

00:01:11.479 --> 00:01:16.330
He’s one of our most popular repeat
speakers in this lecture series.

00:01:16.330 --> 00:01:19.310
And so it is my pleasure
again to introduce Ross Stein

00:01:19.310 --> 00:01:23.289
for tonight’s lecture.
Ross is a seismologist.

00:01:23.289 --> 00:01:27.910
He’s recently retired here
in Menlo Park for the USGS.

00:01:27.910 --> 00:01:32.229
He studies how earthquakes interact
by the transfer of stress.

00:01:32.229 --> 00:01:37.959
He’s a consulting professor of
geophysics at Stanford University,

00:01:37.959 --> 00:01:41.619
and as I mentioned,
a scientist emeritus at USGS.

00:01:41.619 --> 00:01:45.649
He is also president-elect
of the tectonophysics section

00:01:45.649 --> 00:01:49.399
of the American
Geophysical Union.

00:01:49.399 --> 00:01:52.739
Ross gave a wonderful
TEDx Talk in 2012.

00:01:52.739 --> 00:01:55.539
I’m sure it’s online
if you’re interested.

00:01:55.539 --> 00:01:57.660
It was called
Defeating Earthquakes.

00:01:57.660 --> 00:01:59.429
He was also the
keynote speaker for the

00:01:59.429 --> 00:02:04.209
Presidential Awards for Excellence in
Mathematics and Science Teaching.

00:02:04.209 --> 00:02:07.399
And in 2014, he was a
distinguished lecturer

00:02:07.399 --> 00:02:11.720
at the Stanford School
of Earth Sciences.

00:02:11.720 --> 00:02:14.970
Within the USGS,
Ross has been roundly recognized.

00:02:14.970 --> 00:02:16.900
He received the
Eugene Shoemaker

00:02:16.900 --> 00:02:21.079
Distinguished Lifetime
Achievement Award from the USGS.

00:02:21.079 --> 00:02:23.260
He received an
Excellence in Outreach Award

00:02:23.260 --> 00:02:26.239
from the Southern
California Earthquake Center.

00:02:26.239 --> 00:02:30.450
And was also recognized by
NOAA with an award called the

00:02:30.450 --> 00:02:35.430
Outstanding Contributions and
Cooperation in Geosciences.

00:02:35.430 --> 00:02:39.540
So we are very fortunate that,
although Ross retired last year,

00:02:39.540 --> 00:02:43.239
geologists and scientists never
really retire from the USGS.

00:02:43.239 --> 00:02:46.250
We just call them emeriti,
and they remain involved

00:02:46.250 --> 00:02:48.870
with USGS and
earthquake science.

00:02:48.870 --> 00:02:52.700
So it is my pleasure to introduce
Ross Stein, who will be speaking

00:02:52.700 --> 00:02:57.390
about the Gold Rush and the 1906
Earthquake: How They Combined

00:02:57.390 --> 00:03:02.920
to Create the Breakthrough Discovery
of Modern Seismic Science.

00:03:02.920 --> 00:03:05.480
[ Applause ]

00:03:05.480 --> 00:03:06.740
- Thank you, Leslie.

00:03:06.740 --> 00:03:10.560
[ Applause ]

00:03:11.599 --> 00:03:14.579
Let me begin at the ending.

00:03:14.579 --> 00:03:20.959
Nearly everything we love about the
Bay Area is brought to us by the faults.

00:03:23.310 --> 00:03:26.950
Napa Valley, Portola Valley,
and Silicon Valley.

00:03:26.950 --> 00:03:31.420
Tomales Bay, San Francisco Bay,
and Monterey Bay.

00:03:31.420 --> 00:03:34.739
Absent the San Andreas
and the Hayward Faults,

00:03:34.739 --> 00:03:39.310
there would be no San Francisco Bay,
the only deep water-protected port

00:03:39.310 --> 00:03:44.530
along the California coast, and so
the wellspring for the Gold Rush.

00:03:49.859 --> 00:03:55.939
The Hayward Fault lifted up
the Berkeley and Oakland hills,

00:03:55.939 --> 00:03:59.959
giving spectacular sunset
views of the Golden Gate.

00:03:59.960 --> 00:04:02.960
Faults make Big Sur big.

00:04:02.960 --> 00:04:07.260
A bend in the San Andreas
thrust up the Santa Cruz Mountains,

00:04:07.260 --> 00:04:11.599
the spine of the peninsula,
and the Marin Headlands.

00:04:11.599 --> 00:04:14.999
These coastal ranges
temper our climate,

00:04:14.999 --> 00:04:19.090
bathe us in fog,
and crown us in redwoods.

00:04:19.090 --> 00:04:24.720
What I want you to see is that we enjoy
the fruits of the faults every day.

00:04:24.720 --> 00:04:31.199
And so we have to learn to live with their
occasional spoils as befell the Bay Area in

00:04:31.199 --> 00:04:40.240
1906 and 1989 and the
Southland in 1857 and 1994.

00:04:40.240 --> 00:04:44.930
While we cannot predict earthquakes,
we know where the hazard is high,

00:04:44.930 --> 00:04:48.120
and we know
why it’s high.

00:04:48.120 --> 00:04:53.919
And we can erect structures to
withstand any shaking that the

00:04:53.919 --> 00:04:58.780
Bay Area faults can throw at them.
And this is what we must do.

00:04:58.780 --> 00:05:02.310
So tonight, we’ll move from
the discovery of gold to the

00:05:02.310 --> 00:05:05.960
discovery of what an earthquake is.
And from there, how earthquakes

00:05:05.960 --> 00:05:12.599
interact, which is highly
important for the Bay Area.

00:05:12.599 --> 00:05:19.699
And so I’d like to begin this story with
showing you the world -- or the home

00:05:19.699 --> 00:05:27.680
planet as it was given to us 50 years ago
in the plate tectonics revolution.

00:05:27.680 --> 00:05:32.039
Where the discovery was made that
the surface of the Earth is not fixed

00:05:32.039 --> 00:05:36.500
and rigid, but highly mobile,
with about a dozen great plates

00:05:36.500 --> 00:05:41.690
moving around from fractions
of an inch to several inches a year.

00:05:41.690 --> 00:05:45.710
Where these plates pull apart,
these yellow zones,

00:05:45.710 --> 00:05:49.500
basalt is welded on to
the trailing edges.

00:05:49.500 --> 00:05:54.349
And where the plates collide,
these magenta zones,

00:05:54.349 --> 00:05:57.500
are the great subduction
zones of the Earth.

00:05:57.500 --> 00:06:00.340
And what happens there
is that the denser of the two

00:06:00.340 --> 00:06:04.270
colliding plates gets shoved
under the lighter one.

00:06:04.270 --> 00:06:08.190
And this is where the world’s
greatest earthquakes occur.

00:06:08.190 --> 00:06:13.180
Now, profoundly important for
the discovery of plate tectonics

00:06:13.180 --> 00:06:18.289
was the 1906 earthquake that
occurred 60 years beforehand.

00:06:18.289 --> 00:06:21.560
This was the first realization
that the surface of the Earth

00:06:21.560 --> 00:06:25.060
was mobile and active.

00:06:25.060 --> 00:06:28.090
But that discovery
never would have been made

00:06:28.090 --> 00:06:31.949
had not the Gold Rush
occurred 60 years earlier.

00:06:31.949 --> 00:06:36.349
And it’s that story that I’d like
to begin the evening with.

00:06:42.139 --> 00:06:47.259
So here it is, the two-paragraph
announcement in the Californian

00:06:47.259 --> 00:06:51.990
in 1848 announcing that
gold had been discovered.

00:06:51.990 --> 00:06:54.710
And as soon -- and it’s kind of
an amazing line at the bottom.

00:06:54.710 --> 00:06:58.310
It’s like they are foretelling the tech
boom rather than the Gold Rush.

00:06:58.310 --> 00:07:02.409
[laughter]
So that’s incredible foresight.

00:07:02.409 --> 00:07:07.909
And as soon as the federal Army
confirmed and assayed the gold,

00:07:07.909 --> 00:07:13.590
60,000 people around the world
abandoned their jobs, their families,

00:07:13.590 --> 00:07:17.750
and their communities and boarded
ships to sail to San Francisco.

00:07:17.750 --> 00:07:20.050
And they were right.

00:07:20.050 --> 00:07:24.930
Because within two years, all of
the arrivals were getting the dregs.

00:07:24.930 --> 00:07:31.039
The real lucrative gold finds
occurred in those first few years.

00:07:31.039 --> 00:07:33.490
And so here’s how it
was supposed to work.

00:07:33.490 --> 00:07:37.389
There would be some miners
on there at Mile Rock,

00:07:37.389 --> 00:07:40.629
and you’d just wave the ships in
with a hanky. [laughter]

00:07:40.629 --> 00:07:44.280
But it didn’t really
turn out that way.

00:07:44.280 --> 00:07:47.509
And that’s because, if you’ve
ever been outside the Golden Gate,

00:07:47.509 --> 00:07:52.310
you’ll know that it has the
fastest current anywhere on the coast --

00:07:52.310 --> 00:07:57.750
faster, often,
than these boats could sail.

00:07:57.750 --> 00:08:02.289
It has steep waves, high winds,
and often dense fog.

00:08:02.289 --> 00:08:05.789
And so many of these ships,
after having come halfway

00:08:05.789 --> 00:08:10.610
around the world, came to
grief in the Golden Gate.

00:08:10.610 --> 00:08:14.060
And so the federal government
sent out surveyors from the

00:08:14.060 --> 00:08:18.469
Coast and Geodetic Survey to
create decent navigation charts.

00:08:18.469 --> 00:08:22.449
Because this is the chart that
was used in 1851. [laughter]

00:08:22.449 --> 00:08:25.500
It looks like nothing like
the California coast. [laughter]

00:08:25.500 --> 00:08:29.389
And the idea -- and this is part
of the chart -- was that the captain

00:08:29.389 --> 00:08:32.729
was supposed to line the ship up
so that they could see Alcatraz

00:08:32.729 --> 00:08:36.849
in the middle of the gate and then
just shoot the rapids. [laughter]

00:08:36.849 --> 00:08:40.159
And it just really
didn’t work out that way.

00:08:40.159 --> 00:08:44.450
So what’s interesting about
mapmaking in the 1850s

00:08:44.450 --> 00:08:47.840
is that you could not
measure a long distance.

00:08:47.840 --> 00:08:50.470
Once you came to the
end of a tape, you were done.

00:08:50.470 --> 00:08:52.770
And in fact, we couldn’t
measure long distances

00:08:52.770 --> 00:08:56.840
until the 1970s with the
development of the laser.

00:08:56.840 --> 00:08:58.960
What you could
measure were angles.

00:08:58.960 --> 00:09:03.690
So what was done is you’d use a
theodolite and you’d go to a promontory

00:09:03.690 --> 00:09:07.650
or a mountaintop, and you’d measure
an angle to two other mountaintops.

00:09:07.650 --> 00:09:10.970
And thousands of these
angles would be measured.

00:09:10.970 --> 00:09:14.580
And that’s how locations
were determined.

00:09:14.580 --> 00:09:18.440
Because the instruments were heavy
and inaccurate, every five or 10 years,

00:09:18.440 --> 00:09:22.560
the surveys were repeated with
better and lighter instruments.

00:09:22.560 --> 00:09:27.570
They saw changes in those angles, which
they attributed to better instrumentation.

00:09:27.570 --> 00:09:31.010
So that mapmaking continued,
but there’s something interesting

00:09:31.010 --> 00:09:35.200
about what happens when you’re
measuring angles to make a map.

00:09:35.200 --> 00:09:36.730
Let me illustrate is this way.

00:09:36.730 --> 00:09:41.280
Let’s say I put a square on a balloon,
and I inflate the balloon,

00:09:41.280 --> 00:09:42.770
and the square gets bigger.

00:09:42.770 --> 00:09:46.810
Do any of those angles change?
No.

00:09:46.810 --> 00:09:50.210
Expansion and contraction
was invisible to them.

00:09:50.210 --> 00:09:53.170
They had no way
of measuring that.

00:09:53.170 --> 00:09:56.880
Now what happens
if I shear that square?

00:09:56.880 --> 00:10:00.980
Now two angles have
gotten smaller and two larger.

00:10:00.980 --> 00:10:05.810
So as crude as this measurement was,
they were very sensitive to shear.

00:10:05.810 --> 00:10:09.610
And that accident
changed everything.

00:10:10.980 --> 00:10:15.860
Now, this is the map that they
made that was completed in 1856.

00:10:15.860 --> 00:10:20.720
And this beautiful map
is as accurate as any we have today,

00:10:20.720 --> 00:10:24.030
despite how crude
these measurements were.

00:10:24.030 --> 00:10:30.640
And what I love about this geodetic map
is it actually has sailing directions in it.

00:10:30.640 --> 00:10:34.260
And you can also see San Francisco
during the Gold Rush.

00:10:34.260 --> 00:10:36.410
Let’s zoom in on that.

00:10:36.410 --> 00:10:39.220
So here is the city in 1853.

00:10:39.220 --> 00:10:42.590
It had already burned
to the ground twice by this time.

00:10:42.590 --> 00:10:47.620
And the city had gone
from 15,000 people to 100,000 people

00:10:47.620 --> 00:10:49.320
in a year and a half.

00:10:49.320 --> 00:10:52.700
So when that city burned down,
that was quite a conflagration.

00:10:52.700 --> 00:10:56.940
Now, you see a couple of things.
You see the marina on top.

00:10:56.940 --> 00:11:00.500
You see Yerba Buena Cove --
which we’ve never heard of --

00:11:00.500 --> 00:11:03.090
and Mission Bay.

00:11:03.090 --> 00:11:06.600
These were where the Gold Rush
ships dropped anchor.

00:11:06.600 --> 00:11:10.100
And to give you a picture of what
Yerba Buena Cove looked like

00:11:10.100 --> 00:11:14.440
at that time, here’s a
contemporary photograph in 1851.

00:11:14.440 --> 00:11:19.390
By the way, look how built up
San Francisco is at that time.

00:11:19.390 --> 00:11:27.310
Now, when I look at this picture, I count
about 70 ships in 1851 in port at anchor.

00:11:27.310 --> 00:11:31.540
Now let me show it to you
two years later in 1853.

00:11:31.540 --> 00:11:37.000
It’s hard to count, but there are
about 700 ships in this picture.

00:11:37.000 --> 00:11:40.600
Now, here’s what’s really interesting.
Take a closer look.

00:11:40.600 --> 00:11:44.080
These ships are a wreck.
They’re falling apart.

00:11:44.080 --> 00:11:46.110
So here’s what happened.

00:11:46.110 --> 00:11:48.510
These ships would travel
halfway around the world.

00:11:48.510 --> 00:11:53.310
The most lucrative
merchant line anywhere.

00:11:53.310 --> 00:11:59.800
The passengers would dis-board --
disembark and jump onto steamers

00:11:59.800 --> 00:12:03.860
and head to Sacramento, and from
there to the gold fields.

00:12:03.860 --> 00:12:06.700
And so would the crews.
[laughter]

00:12:06.700 --> 00:12:09.860
Every single crew member
abandoned ship.

00:12:09.860 --> 00:12:13.300
Because they could earn 10 to
100 times more in the gold field.

00:12:13.300 --> 00:12:18.910
And not one captain was able to
raise a crew to sail the most lucrative

00:12:18.910 --> 00:12:22.140
route back and pick up
more passengers.

00:12:22.140 --> 00:12:26.190
Every one of these ships rotted
in place, completely abandoned.

00:12:26.190 --> 00:12:29.550
And eventually, San Franciscans
simply imported sand from

00:12:29.550 --> 00:12:32.800
what is today Golden Gate Park,
dumped it on the ships,

00:12:32.800 --> 00:12:37.510
and that is the financial district.
[laughter]

00:12:37.510 --> 00:12:45.150
So here is one of these ships
unearthed in 2001 between

00:12:45.150 --> 00:12:48.330
the Transamerica tower
and Portsmouth Square.

00:12:48.330 --> 00:12:49.540
And look at Telegraph Hill.

00:12:49.540 --> 00:12:52.940
Look at what the city
looked like in 1851.

00:12:52.940 --> 00:12:55.520
So let’s come
back to that map.

00:12:55.520 --> 00:12:58.700
And let’s compare it to what
San Francisco looks like today.

00:12:58.700 --> 00:13:03.380
So you can see, San Francisco
has really grown quite a bit.

00:13:03.380 --> 00:13:05.570
So what do we see?

00:13:05.570 --> 00:13:10.170
What was Yerba Buena Cove
is now part of the banking boom.

00:13:10.170 --> 00:13:12.250
It’s the financial district.

00:13:12.250 --> 00:13:16.940
What was Mission Bay and its marsh
is now SoMa, part of the tech boom.

00:13:16.940 --> 00:13:19.370
And what was the marina --
another place where ships

00:13:19.370 --> 00:13:24.060
dropped anchor -- is now the center
of the dating boom. [laughter]

00:13:24.060 --> 00:13:26.170
And that’s what we inherited.

00:13:26.170 --> 00:13:30.790
And this turns out to be
San Francisco’s Achilles heel.

00:13:30.790 --> 00:13:36.390
Because this junk land, to put it kindly,
shakes and liquefies in earthquakes.

00:13:36.390 --> 00:13:39.670
And this is where the
greatest damage was in 1906.

00:13:39.670 --> 00:13:41.810
And because we didn’t
learn our lesson, it was where

00:13:41.810 --> 00:13:47.000
the greatest damage was in 1989.
And it will be a problem again.

00:13:47.000 --> 00:13:50.160
So think about that when we
see this enormous boom

00:13:50.160 --> 00:13:54.230
going on in the SoMa area.

00:13:54.230 --> 00:13:58.990
But all this changed
on April 18th, 1906.

00:13:58.990 --> 00:14:03.370
Now, the most powerful words
ever written -- contemporary words

00:14:03.370 --> 00:14:07.250
written about this earthquake --
were penned by Jack London,

00:14:07.250 --> 00:14:12.020
who wrote for Collier’s magazine,
two weeks after the earthquake.

00:14:12.020 --> 00:14:14.690
It was published on May 5th.

00:14:14.690 --> 00:14:17.670
And what I’d like to do as
I show you some of these images

00:14:17.670 --> 00:14:23.370
is just read a passage from
Jack London’s report.

00:14:23.370 --> 00:14:27.890
San Francisco is gone.
Nothing remains of it but memories

00:14:27.890 --> 00:14:31.940
and a fringe of dwelling houses
on its outskirts.

00:14:31.940 --> 00:14:38.230
Its industrial section is wiped out.
Its business section is wiped out.

00:14:38.230 --> 00:14:42.720
Its social and residential
section is wiped out.

00:14:42.720 --> 00:14:47.720
The factories and warehouses and
great stores and newspaper buildings,

00:14:47.720 --> 00:14:52.750
the hotels and the palaces
of the nabobs are all gone.

00:14:52.750 --> 00:14:56.160
Remains only the fringe of
dwelling houses on the outskirts

00:14:56.160 --> 00:14:59.980
of what was once
San Francisco.

00:15:02.140 --> 00:15:05.750
On Wednesday morning at a quarter
past five came the earthquake.

00:15:05.750 --> 00:15:09.430
A minute later,
flames were leaping upward.

00:15:09.430 --> 00:15:12.600
In a dozen different quarters
south of Market Street,

00:15:12.600 --> 00:15:17.730
in the working-class ghetto,
and in the factories, fires started.

00:15:17.730 --> 00:15:20.180
There was no
opposing the flames.

00:15:20.180 --> 00:15:24.420
There was no organization,
no communication.

00:15:24.420 --> 00:15:27.250
All the cunning adjustments
of a 20th-century city

00:15:27.250 --> 00:15:30.890
had been smashed
by the earthquake.

00:15:30.890 --> 00:15:33.170
An enumeration of the
buildings destroyed

00:15:33.170 --> 00:15:36.500
would be a directory
of San Francisco.

00:15:36.500 --> 00:15:38.490
An enumeration of the
buildings undestroyed

00:15:38.490 --> 00:15:42.320
would be a line
and several addresses.

00:15:42.320 --> 00:15:44.450
An enumeration of the
deeds of heroism

00:15:44.450 --> 00:15:48.830
would stock a library and bankrupt
the Carnegie Medal fund.

00:15:48.830 --> 00:15:51.990
An enumeration of the
dead will never be made.

00:15:51.990 --> 00:15:57.060
All vestiges of them
were destroyed by the flames.

00:15:57.060 --> 00:16:01.890
I went inside with the owner of
a house on the steps of which I sat.

00:16:01.890 --> 00:16:04.660
He was cool and
cheerful and hospitable.

00:16:04.660 --> 00:16:10.580
Yesterday morning, he said,
I was worth $600,000.

00:16:10.580 --> 00:16:18.050
This morning, this house is all I
have left. It will go in 15 minutes.

00:16:18.050 --> 00:16:22.430
He pointed to a large cabinet.
That is my wife’s collection of china.

00:16:22.430 --> 00:16:27.820
This rug upon which we stand
is a present. It cost $1,500.

00:16:27.820 --> 00:16:33.630
Try that piano. Listen to its tone.
There are few like it.

00:16:33.630 --> 00:16:37.390
The flames will be here in 15 minutes.
There are no horses.

00:16:37.390 --> 00:16:42.300
And he was right.
Everything was destroyed.

00:16:42.300 --> 00:16:46.070
Nothing was left.

00:16:46.070 --> 00:16:49.100
After the earthquake,
which nobody understood --

00:16:49.100 --> 00:16:52.040
nobody in the United States
understood that an earthquake

00:16:52.040 --> 00:16:56.920
was an explosion, an implosion --
nobody knew.

00:16:56.920 --> 00:16:58.290
Geologists from
Stanford and Berkeley

00:16:58.290 --> 00:17:03.820
started to walk the path of destruction
north and south of San Francisco.

00:17:03.820 --> 00:17:10.120
And what they saw were things like
this -- a tear -- a rent in the landscape.

00:17:10.120 --> 00:17:15.120
By the way, this is Alice Eastwood,
who saved the botany collection

00:17:15.120 --> 00:17:20.250
of the California Academy of
Sciences during the earthquake.

00:17:20.250 --> 00:17:23.350
And even more startling,
they discovered that, wherever

00:17:23.350 --> 00:17:29.610
they found this tear in the landscape,
anything that crossed it was offset.

00:17:29.610 --> 00:17:35.419
So a road or a
fence line or a stream.

00:17:35.419 --> 00:17:40.370
So these orange points
were original side-by-side.

00:17:40.370 --> 00:17:42.779
And what they noticed was,
whichever side you’re on,

00:17:42.779 --> 00:17:46.750
the other side has moved to the right,
which they called right lateral.

00:17:46.750 --> 00:17:50.620
And they saw that the
amount of this displacement

00:17:50.620 --> 00:17:53.169
varied from
about 6 to 12 feet.

00:17:53.169 --> 00:17:59.539
But over the entire 250-mile-long
rupture, this is what they saw.

00:17:59.539 --> 00:18:03.509
And they realized for the first time
that an earthquake involves

00:18:03.509 --> 00:18:08.179
permanent displacement
of the land.

00:18:08.179 --> 00:18:10.779
Then they realized, wait a minute,
we need those triangulation

00:18:10.779 --> 00:18:12.029
measurements to be redone.

00:18:12.029 --> 00:18:15.509
Because we need to figure out,
is this what you see, and as you

00:18:15.509 --> 00:18:20.110
move away from the fault, do you
also see this 12-foot offset?

00:18:20.110 --> 00:18:24.039
So the Geodetic Survey came back,
and they surveyed points

00:18:24.039 --> 00:18:27.960
that they had last surveyed
in 1905 in 1906 and ’07.

00:18:27.960 --> 00:18:33.340
And if you put a dotted line
down perpendicular to the fault,

00:18:33.340 --> 00:18:37.250
and you slide all those displacements
that they measured from promontories

00:18:37.250 --> 00:18:40.309
into their position at the same
distance from the fault, you can see,

00:18:40.309 --> 00:18:44.750
yes, there is 12 feet of
displacement across the fault.

00:18:44.750 --> 00:18:47.509
But as you move farther away,
those diminish.

00:18:47.509 --> 00:18:52.600
So it’s like the ground had
leaped ahead along the fault.

00:18:52.600 --> 00:18:57.559
But as you got farther away,
you began to see that diminish.

00:18:57.559 --> 00:19:01.000
And then H.F. Reid from
Johns Hopkins University said,

00:19:01.000 --> 00:19:03.940
wait a minute, let’s go back
and look at those surveys that were

00:19:03.940 --> 00:19:11.200
done from 1851 all the way up to 1905
and see what we can learn from those.

00:19:11.200 --> 00:19:15.259
And he found something
truly astonishing.

00:19:15.259 --> 00:19:19.370
Which was that those points
had been moving all along.

00:19:19.370 --> 00:19:23.169
There had been no movement
along the fault during those years.

00:19:23.169 --> 00:19:26.279
But there was plenty of movement
as you moved away from it.

00:19:26.279 --> 00:19:29.710
So if you look at the motion
of the Farallons out in the west,

00:19:29.710 --> 00:19:32.799
compared to Mount Diablo,
they were moving at about

00:19:32.799 --> 00:19:36.990
an inch a year in opposite
directions from each other.

00:19:36.990 --> 00:19:41.090
So this was an indication that the
earthquake wasn’t the whole story.

00:19:41.090 --> 00:19:46.169
The land was always moving.
But it wasn’t moving along the fault.

00:19:47.880 --> 00:19:53.120
And so, if you line up those points
along this dotted line, what you see

00:19:53.129 --> 00:19:57.180
is that the entire area
was being sheared.

00:19:57.180 --> 00:20:00.360
Now, remember I told you
that triangulation measurements

00:20:00.360 --> 00:20:04.149
are uniquely and exclusively
sensitive to shear.

00:20:04.149 --> 00:20:07.419
So this was pure serendipity,
not only that the measurements

00:20:07.419 --> 00:20:11.970
were there from 1850 to 1905
so you could see this accumulation

00:20:11.970 --> 00:20:18.350
of strain, but that they could detect that
strain with such a crude instrument.

00:20:18.350 --> 00:20:22.009
And so here’s how
Reid put it together.

00:20:22.009 --> 00:20:25.529
He said -- he said, well, if it’s
moving at an inch a year,

00:20:25.529 --> 00:20:28.830
imagine you let a couple
hundred years go by, and you’d

00:20:28.830 --> 00:20:33.610
see about 12 feet net movement between
the Farallons and Mount Diablo.

00:20:33.610 --> 00:20:39.960
Then we’re going to add the earthquake
displacements on top of that.

00:20:39.960 --> 00:20:42.629
And now you see
something really simple.

00:20:42.629 --> 00:20:45.460
The fault has just been offset.

00:20:45.460 --> 00:20:48.759
So the fault doesn’t leap ahead.
It catches up.

00:20:48.759 --> 00:20:52.879
The whole thing has been shearing,
but it’s stuck along the fault.

00:20:52.879 --> 00:20:58.240
And suddenly, that frictional resistance
is overcome, and the fault catches up,

00:20:58.240 --> 00:21:02.289
and we simply have
an offset on the fault.

00:21:02.289 --> 00:21:05.639
And that’s the earthquake
machine that we have.

00:21:05.639 --> 00:21:09.570
And that’s what Reid discovered.
And that has been confirmed

00:21:09.570 --> 00:21:13.480
in everything that has been --
has happened since.

00:21:13.480 --> 00:21:17.210
So with Reid’s inheritance,
we can look at the Bay Area

00:21:17.210 --> 00:21:21.610
and say, it’s not just one fault.
But we have lots of faults,

00:21:21.610 --> 00:21:27.850
including the Hayward that
gave us the magnitude 7 in 1868.

00:21:27.850 --> 00:21:31.350
And when we look at all these faults,
they all have this right lateral motion

00:21:31.350 --> 00:21:37.059
because they’re all in
this giant shear system.

00:21:37.059 --> 00:21:41.590
And what happened in 1906
is the biggest player of those faults,

00:21:41.590 --> 00:21:45.029
the San Andreas,
had its event.

00:21:46.000 --> 00:21:48.060
Now, Reid had these crude tools.

00:21:48.070 --> 00:21:51.980
But what Reid was telling us was,
look, you don’t have to wait for

00:21:51.980 --> 00:21:56.279
the earthquake to come to know
where the hazard is going to be high.

00:21:56.279 --> 00:21:59.519
What you should be doing
instead is measure the strain.

00:21:59.520 --> 00:22:01.380
Everywhere you can on Earth.

00:22:01.389 --> 00:22:05.549
And today, with GPS,
we can measure displacements

00:22:05.549 --> 00:22:07.659
extremely accurately everywhere.

00:22:07.659 --> 00:22:11.559
We have 20,000 GPS
receivers around the world.

00:22:11.559 --> 00:22:17.370
And here is what we now see in terms of
the distribution of strain on the globe.

00:22:17.370 --> 00:22:20.879
So this is the Japanese
subduction zone.

00:22:20.879 --> 00:22:26.190
And at the bottom, you see where the
2004 Sumatra earthquake occurred.

00:22:26.190 --> 00:22:32.250
But what happened to my globe?
We have a thief. [laughter]

00:22:33.440 --> 00:22:36.779
I spent so much time
painting the boundaries on.

00:22:36.779 --> 00:22:39.830
But what I want to show you
is that the plate tectonics view

00:22:39.830 --> 00:22:43.389
of the globe is
not the whole story.

00:22:43.389 --> 00:22:45.970
I drew these lines,
but what you’re seeing here

00:22:45.970 --> 00:22:49.320
is that the deformation extends
well into the continents.

00:22:49.320 --> 00:22:51.960
In the oceans, yes,
it’s a thin line.

00:22:51.960 --> 00:22:55.759
But in the continents,
deformation extends way in,

00:22:55.759 --> 00:23:01.429
as do our faults in the western U.S.
extending all the way to the Wasatch.

00:23:01.429 --> 00:23:06.120
And so the picture with great data
that’s since collected gives us

00:23:06.120 --> 00:23:09.879
a much richer sense of
where earthquakes have occurred

00:23:09.879 --> 00:23:12.299
and where they will
occur in the future.

00:23:12.299 --> 00:23:15.200
In the center of the picture,
you’re seeing now the

00:23:15.200 --> 00:23:20.490
North Anatolian Fault --
that bright line.

00:23:20.490 --> 00:23:24.090
And that gave us the
most spectacular falling-domino

00:23:24.090 --> 00:23:30.559
sequence of earthquakes ever
recorded between 1939 and 1999.

00:23:30.559 --> 00:23:34.559
And here’s our piece of
the tectonic puzzle --

00:23:34.559 --> 00:23:39.289
the San Andreas system you can see
working its way through California

00:23:39.289 --> 00:23:44.709
with a branch that goes up the
eastern side of the Sierras as well.

00:23:45.920 --> 00:23:51.380
Okay. What I want to do now
is take that view that I just gave you

00:23:51.389 --> 00:23:53.990
and make it a
little bit more real.

00:23:53.990 --> 00:23:58.119
Because you will see that the system
that we inherited from Reid

00:23:58.119 --> 00:24:03.360
can really be simplified down to
what I’m going to show you now.

00:24:03.360 --> 00:24:08.600
So the interiors of these plates are
moving every day at a fixed rate.

00:24:08.600 --> 00:24:10.919
The plates get hung up
on their boundaries,

00:24:10.919 --> 00:24:12.659
and that’s why
we have earthquakes.

00:24:12.659 --> 00:24:17.679
So this casting reel represents the
steady motion of the plate’s interiors.

00:24:17.679 --> 00:24:20.700
This rubber band
represents the rubberiness --

00:24:20.700 --> 00:24:24.090
the elasticity
of the Earth’s crust.

00:24:24.090 --> 00:24:27.110
If the Earth’s crust weren’t elastic,
we wouldn’t have earthquakes.

00:24:27.110 --> 00:24:32.619
I’ll grant you, it’s very stiff
elastic material, but it’s elastic.

00:24:32.619 --> 00:24:35.940
And then here are some
granite slider blocks

00:24:35.940 --> 00:24:40.110
that are going to be the guys that
are affected by the earthquake.

00:24:40.110 --> 00:24:44.940
And this surface is just a non-skid.
Just industrial non-skid.

00:24:44.940 --> 00:24:46.620
Okay. So I’m going
to crank steadily.

00:24:46.629 --> 00:24:49.249
I’m not going to do anything
to create earthquakes.

00:24:49.249 --> 00:24:51.709
And I just want
you to watch.

00:24:52.400 --> 00:25:02.180
[scraping sounds]

00:25:02.860 --> 00:25:06.020
Okay. So you saw that there would be
a couple hundred years, then we’d have

00:25:06.029 --> 00:25:10.240
an earthquake, then we wait maybe
a hundred years then -- right?

00:25:10.240 --> 00:25:15.110
So in fact, there’s nothing I can do to
prevent this from producing earthquakes.

00:25:15.110 --> 00:25:18.080
I can’t just make it slide.
It’s going to produce earthquakes.

00:25:18.080 --> 00:25:20.029
Okay. Now here’s
a question for you.

00:25:20.029 --> 00:25:24.879
Were they all the same size?
You sure about that?

00:25:24.879 --> 00:25:28.149
That’s very bad news.

00:25:28.149 --> 00:25:31.149
If I can’t make regular
repeating earthquakes

00:25:31.149 --> 00:25:37.519
with Home Depot sandpaper [laughter],
a rubber band, a casting reel,

00:25:37.519 --> 00:25:42.629
and kitchen countertop samples,
we’re never going to get it in the Earth.

00:25:42.629 --> 00:25:44.679
That is the bitter pill
that we’ve swallowed

00:25:44.679 --> 00:25:48.860
from 30 years of research on
the behavior of earthquakes.

00:25:48.860 --> 00:25:53.279
We can never talk about a
fault as being 10 months’ pregnant.

00:25:53.279 --> 00:25:55.289
They are not that regular.

00:25:55.289 --> 00:25:58.269
And what’s interesting about
that observation is, when you

00:25:58.269 --> 00:26:03.259
think about it, that irregularity can’t
come from this because I was steady.

00:26:03.259 --> 00:26:05.570
It can’t come from
the rubber band because

00:26:05.570 --> 00:26:08.450
it’s always within
its elastic range.

00:26:08.450 --> 00:26:10.669
It can only come
from the friction.

00:26:10.669 --> 00:26:14.100
Which means, even though when I put
my hand over this, it feels extremely

00:26:14.100 --> 00:26:19.740
regular, tiny, little irregularities cause
faces of the fault to get hung up.

00:26:19.740 --> 00:26:23.629
And when those grains pop,
things begin to happen.

00:26:23.629 --> 00:26:29.119
And it’s remarkably unpredictable
exactly when they’re going to let go.

00:26:29.119 --> 00:26:31.669
So that lack of
earthquake predictability

00:26:31.669 --> 00:26:35.500
is a fundamental part
of the earthquake machine.

00:26:35.500 --> 00:26:37.240
But that doesn’t mean
we can’t do anything.

00:26:37.240 --> 00:26:38.639
We can do tons.

00:26:38.639 --> 00:26:42.749
So let me ask you
another question.

00:26:42.749 --> 00:26:46.539
What if we just have
a one-bricker spot.

00:26:46.539 --> 00:26:51.240
Am I going to get more earthquakes or
less if I’m cranking at the same rate?

00:26:51.240 --> 00:26:53.960
Are they more frequent or less frequent?
- [audience] More.

00:26:53.960 --> 00:26:56.159
- More.
- Are they going to be smaller or larger?

00:26:56.159 --> 00:26:59.019
- [inaudible responses]
Okay, so let’s see what we get.

00:26:59.920 --> 00:27:07.420
[scraping sounds]

00:27:07.429 --> 00:27:09.980
So they’re about
twice as frequent, right?

00:27:09.980 --> 00:27:11.679
And they’re about half the size.

00:27:11.679 --> 00:27:15.559
And that makes sense when you
think about the fact that the motion

00:27:15.559 --> 00:27:18.879
is being resisted by the weight
of those bricks times the friction.

00:27:18.879 --> 00:27:22.009
The friction isn’t changing,
so if I have the weight,

00:27:22.009 --> 00:27:25.749
then there’s less resistance,
and it can go forward.

00:27:25.749 --> 00:27:28.249
And we have a one-bricker
place on the San Andreas.

00:27:28.249 --> 00:27:29.960
It’s called Parkfield.

00:27:29.960 --> 00:27:36.960
It knocks off earthquakes every 20 to 40
or so years, and they’re 6s or 6-1/2s.

00:27:36.960 --> 00:27:42.779
So it’s interesting to think that,
same elasticity, same friction,

00:27:42.779 --> 00:27:44.409
same speed of the San Andreas --
an inch a year -

00:27:44.409 --> 00:27:48.679
but I can get places that
produce 6s rather than 8s.

00:27:48.679 --> 00:27:52.820
If I can just find some places that are
only one-brickers, and we have them.

00:27:52.820 --> 00:27:54.869
Okay, next question.

00:27:54.869 --> 00:28:00.070
We’ve sandblasted the bottom of
these samples to create the friction.

00:28:00.070 --> 00:28:03.690
And the top side
has the original polish.

00:28:03.690 --> 00:28:08.560
So what if I just turn it down -- turn it
over so the polished side is down?

00:28:08.560 --> 00:28:11.680
Now, I’ve only changed the friction
on one side of the fault.

00:28:11.690 --> 00:28:15.950
The other side of the fault’s the same.
What’s going to happen?

00:28:15.950 --> 00:28:19.130
- [inaudible responses]
- Okay, let’s see.

00:28:21.440 --> 00:28:24.710
It’s very interesting, right?
It’s almost just creeping,

00:28:24.710 --> 00:28:28.100
but it’s producing lots and lots
and lots of little earthquakes.

00:28:28.100 --> 00:28:30.830
Well, we have such
a spot on the San Andreas.

00:28:30.830 --> 00:28:32.710
We call it the creeping section.

00:28:32.710 --> 00:28:34.809
It extends from
Hollister down to Parkfield.

00:28:34.809 --> 00:28:37.350
It’s a hundred miles long.

00:28:37.350 --> 00:28:41.269
And it basically moves every day,
and it produces more earthquakes

00:28:41.269 --> 00:28:43.269
than any other spot
on the San Andreas.

00:28:43.269 --> 00:28:47.389
And it’s only got
low friction on one side.

00:28:47.389 --> 00:28:49.759
One side is granite.
One side is serpentine,

00:28:49.759 --> 00:28:54.700
which is very, very mushy,
plate-like material.

00:28:55.680 --> 00:28:58.519
Now, you might also say,
well, now, wait a minute.

00:28:58.519 --> 00:29:01.190
There are not that
many faults in isolation.

00:29:01.190 --> 00:29:07.110
The San Andreas is 700 miles long.
So what about faults interacting?

00:29:07.110 --> 00:29:10.239
So now what I’m going to do is
I’m going to hook these guys up

00:29:10.239 --> 00:29:17.859
in series and ask you to
tell me who’s going to go first.

00:29:19.900 --> 00:29:22.059
And then I’m going to ask you
who’s going to go second

00:29:22.059 --> 00:29:23.049
and who’s going
to go third.

00:29:23.049 --> 00:29:26.070
Okay, so who thinks
this guy’s going first?

00:29:26.070 --> 00:29:30.539
All right. Who thinks
this guy’s going to go first?

00:29:30.539 --> 00:29:34.830
All right. Who thinks this
guy’s going to go second?

00:29:34.830 --> 00:29:38.470
Who thinks this guy’s going to go third?
[laughter] Okay. All right.

00:29:38.470 --> 00:29:41.710
For those of you who were willing
to bet, okay, here we go.

00:29:42.540 --> 00:29:46.900
[scraping sounds]

00:29:47.620 --> 00:29:50.400
It gets interesting
in a hurry, doesn’t it?

00:29:50.409 --> 00:29:52.749
Okay, so first it was
a total trick question,

00:29:52.749 --> 00:29:54.360
and you never should
trust a speaker.

00:29:54.360 --> 00:29:59.820
This guy went three times, right, before
he was able to tension this guy up to go.

00:29:59.820 --> 00:30:03.220
But once this guy did get that
tension going, this guy went.

00:30:03.220 --> 00:30:06.289
And sometimes when that guy went,
they all went together.

00:30:06.289 --> 00:30:08.649
And sometimes
they didn’t.

00:30:08.649 --> 00:30:11.200
And that’s exactly
what faults do.

00:30:11.200 --> 00:30:17.080
In general -- we have independent
cases -- but we had, in 1992,

00:30:17.080 --> 00:30:19.989
in southern California,
the Landers earthquake,

00:30:19.989 --> 00:30:27.580
where four wannabe faults joined up
together and produced a managed 7.3.

00:30:27.580 --> 00:30:29.289
Faults we never thought
could have ruptured together

00:30:29.289 --> 00:30:32.019
produced a
very large earthquake.

00:30:32.019 --> 00:30:34.739
And so sometimes that happens.
Sometimes that process

00:30:34.739 --> 00:30:38.789
happens in 20 seconds, and
sometimes it happens in 20 years.

00:30:38.789 --> 00:30:43.580
And that variability further complicates
the business of predicting earthquakes

00:30:43.580 --> 00:30:47.610
because no little earthquake is
marked for future greatness.

00:30:47.610 --> 00:30:50.779
It doesn’t know it’s
going to be a big one.

00:30:50.779 --> 00:30:53.970
Most of them don’t make it,
and occasionally they do,

00:30:53.970 --> 00:30:56.999
and we can’t figure out
which of those ones -- which of those

00:30:56.999 --> 00:31:03.659
little earthquakes that are actually going
to cascade into something much larger.

00:31:03.659 --> 00:31:06.080
But all is not lost.

00:31:06.080 --> 00:31:10.889
Because there’s still something
very important about the condition

00:31:10.889 --> 00:31:15.220
that tells us that,
once we’ve had an earthquake,

00:31:15.220 --> 00:31:19.440
we can say something
about what might happen next.

00:31:19.440 --> 00:31:24.159
Okay, so if I tense --
if I get things almost ready to go,

00:31:24.159 --> 00:31:27.700
what’s going to happen if I lift a brick?
- [inaudible response]

00:31:27.700 --> 00:31:30.769
- Everybody feel that way?
Good. Okay.

00:31:30.769 --> 00:31:34.649
So I unclamped the fault, right?
And the San Andreas is like that.

00:31:34.649 --> 00:31:38.210
Same deal. I’ve unclamped a little bit,
so it makes it easier to go.

00:31:38.210 --> 00:31:42.470
And, by the same token,
if it’s just about ready to go,

00:31:42.470 --> 00:31:46.779
and I pull on the spring,
I can trigger an earthquake, right?

00:31:46.779 --> 00:31:50.889
But if it’s just gone,
and this thing isn’t tense --

00:31:50.889 --> 00:31:59.259
oops, that didn’t -- it’s just --
it’s suction. Okay, nothing?

00:31:59.259 --> 00:32:00.480
Nothing.

00:32:00.480 --> 00:32:04.629
So only if faults are very
close to failure do they become

00:32:04.629 --> 00:32:08.809
sensitive to the combination of
increasing the shear stress

00:32:08.809 --> 00:32:12.970
or un-clamping the fault, which
is called the Coulomb criteria.

00:32:12.970 --> 00:32:16.919
So if we can calculate that
in the field around an earthquake,

00:32:16.919 --> 00:32:19.080
we can say something about
where the aftershocks are

00:32:19.080 --> 00:32:23.470
more likely to occur and where
the next larger shocks can occur.

00:32:23.470 --> 00:32:27.389
And that is our way in
to a very hard problem.

00:32:27.389 --> 00:32:30.169
That’s the easy way in
to a very hard problem.

00:32:30.169 --> 00:32:32.249
Yes?
- Are you taking questions?

00:32:32.249 --> 00:32:34.729
- I’m taking yours.
[laughter]

00:32:34.729 --> 00:32:40.159
- What natural phenomenon is
equivalent to lifting a brick?

00:32:40.159 --> 00:32:43.460
- Okay, great question.
So imagine the San Andreas,

00:32:43.460 --> 00:32:46.129
which extends 10 miles down ...
- What was the question?

00:32:46.129 --> 00:32:50.029
- Oh, sorry. She said, well, what’s the
physical phenomena of lifting a brick?

00:32:50.029 --> 00:32:57.220
And lifting a brick is whatever process
will allow a fault to unclamp.

00:32:57.220 --> 00:33:01.440
So if some other earthquake in the
neighborhood creates some kind of

00:33:01.440 --> 00:33:06.230
tension perpendicular to the fault,
then that stress will be lowered.

00:33:06.230 --> 00:33:13.019
If it’s a subduction zone like this, maybe
you’re subtracting material above it.

00:33:13.019 --> 00:33:17.159
So anything that changes the stress
which keep these faces under

00:33:17.159 --> 00:33:23.049
enormous stresses in close contact
will tend to promote failure.

00:33:23.049 --> 00:33:26.730
Okay, so now I’m going to --
we’re going to go one step further.

00:33:26.730 --> 00:33:31.850
So this was the one-dimensional view
of life, which is my favorite dimension.

00:33:31.850 --> 00:33:34.629
And now we’re going to go to
two dimensions in a series of animations

00:33:34.629 --> 00:33:39.529
to take you a little farther of this
world of earthquake interaction.

00:33:40.800 --> 00:33:44.780
[ Silence ]

00:33:45.340 --> 00:33:47.680
And we’re going to do this
for the North Anatolian sequence

00:33:47.690 --> 00:33:51.320
because that’s that falling-domino
sequence of earthquakes.

00:33:51.320 --> 00:33:54.779
And that is a clue that
earthquakes interact,

00:33:54.779 --> 00:33:56.929
and so we want to
be able to explain that.

00:33:56.929 --> 00:34:02.480
So we’re going to go to the site of the
last earthquake in the sequence in 1999,

00:34:02.480 --> 00:34:06.509
put up the fault, and we’re going to
put a fence across the fault,

00:34:06.509 --> 00:34:08.270
just as you saw across
the San Andreas.

00:34:08.270 --> 00:34:12.280
And we’re going to watch
200 years of stress accumulate.

00:34:12.280 --> 00:34:16.220
And you can see that that bends
because the whole area’s being sheared,

00:34:16.220 --> 00:34:20.140
and then suddenly the
fault ruptures, and it catches up.

00:34:20.140 --> 00:34:23.550
Exactly what I showed you
that Reid discovered, right?

00:34:23.550 --> 00:34:25.990
The squares sheared
into parallelograms.

00:34:25.990 --> 00:34:29.020
The earthquake occurred,
and they un-sheared.

00:34:29.020 --> 00:34:33.640
Okay. And so that’s the
earthquake cycle. So far, so good.

00:34:33.640 --> 00:34:35.770
Now the question I want to ask
is a little different, though.

00:34:35.770 --> 00:34:39.110
I want to say, what about
beyond the ends of the fault?

00:34:39.110 --> 00:34:41.520
Okay. This guy’s had its earthquake.

00:34:41.520 --> 00:34:44.429
Beyond the ends of the fault,
if this fault continued, have we

00:34:44.429 --> 00:34:49.330
brought that fault closer to failure
as a result of this one or farther?

00:34:49.330 --> 00:34:51.980
And that’s a question
we have to understand.

00:34:51.980 --> 00:34:54.570
So I’m just going to
repeat this experiment.

00:34:54.570 --> 00:34:56.950
But rather than just
putting a fence across the fault,

00:34:56.950 --> 00:34:59.450
I’m going to
put a whole grid.

00:34:59.450 --> 00:35:02.100
Densify the grid,
and watch stress accumulate.

00:35:02.100 --> 00:35:06.440
And let’s focus our attention just beyond
the end of what will be a blue rupture.

00:35:06.440 --> 00:35:09.030
And does this
shearing parallelogram

00:35:09.030 --> 00:35:12.230
get more sheared or less
after the earthquake?

00:35:12.230 --> 00:35:15.850
You can see it’s more.
And if we color more sheared red,

00:35:15.850 --> 00:35:20.100
we have these kind of blowtorch zones
beyond the end of the rupture.

00:35:20.100 --> 00:35:24.960
The blue areas are places where
we removed some of that stress.

00:35:24.960 --> 00:35:27.060
Now here we’ve stretched this guy.

00:35:27.060 --> 00:35:29.850
This is where we’ve lifted
a brick and lifted a brick here.

00:35:29.850 --> 00:35:33.560
And here is where we’ve
added a brick -- here and here.

00:35:33.560 --> 00:35:38.120
And if we add those two together,
we get the Coulomb stress.

00:35:38.120 --> 00:35:43.080
So what it says is if there’s a fault
out here, this guy has --

00:35:43.080 --> 00:35:45.800
now closer to failure as a result of
the neighboring earthquake,

00:35:45.800 --> 00:35:48.760
as would be a fault
over here or over here.

00:35:48.760 --> 00:35:52.600
But a fault here is in
what we call the stress shadow --

00:35:52.600 --> 00:35:57.580
this profoundly blue zone, where we’ve
reduced the likelihood of failure.

00:35:57.580 --> 00:36:00.980
Okay, now let me just show you
one other way -- same deal.

00:36:00.990 --> 00:36:04.870
So these are these places
where we’ve pulled on the spring.

00:36:04.870 --> 00:36:10.110
These red zones are
just where we’ve done this.

00:36:10.110 --> 00:36:14.500
And then here where we lifted a brick --
here and here -- or added a brick.

00:36:14.500 --> 00:36:17.770
And we multiply this panel
by the friction coefficient,

00:36:17.770 --> 00:36:21.760
which for most rock materials,
is somewhere around 1/2.

00:36:21.760 --> 00:36:25.880
And we add them together, and we
get the Coulomb failure condition.

00:36:25.880 --> 00:36:28.290
So now let’s try this out
on the North Anatolian Fault,

00:36:28.290 --> 00:36:32.700
but we’ll first try it out on
just a simple ironed-out fault.

00:36:32.700 --> 00:36:35.040
But I’ll put each rupture on.

00:36:35.040 --> 00:36:39.770
And what I want you to see is that each
rupture has a blowtorch zone beyond the

00:36:39.770 --> 00:36:44.950
end, and that seems to be where the next
rupture nucleates and then propagates.

00:36:44.950 --> 00:36:50.560
So we can see how we could get a
falling-domino sequence of earthquakes.

00:36:50.560 --> 00:36:53.430
Now let’s try it out on the real fault.

00:36:53.430 --> 00:36:58.810
Real life is a little bit more grungy
than that, but the process is the same.

00:36:58.810 --> 00:37:05.260
So we had this zone that gets filled by
an earthquake out beyond 1943, ’44.

00:37:05.260 --> 00:37:09.660
Both of these spots will get filled
by earthquakes over here -- ’57.

00:37:09.660 --> 00:37:12.040
Then we’re going to
get to ’67 in here.

00:37:12.050 --> 00:37:16.540
This turns it red where
the August 1999 earthquake occurs.

00:37:16.540 --> 00:37:21.510
Now that turns this guy red,
which ruptured in November.

00:37:21.510 --> 00:37:27.630
Okay, so you can see how
earthquake interaction is a helpful tool

00:37:27.630 --> 00:37:30.180
to understand what
can happen next.

00:37:30.180 --> 00:37:34.780
But now I want to show you -- tell you
something very important about this.

00:37:34.780 --> 00:37:37.020
This map has
basically turned blue.

00:37:37.020 --> 00:37:40.390
In other words, if we have
any parallel faults -- and remember,

00:37:40.390 --> 00:37:42.440
the Bay Area is a
series of parallel faults --

00:37:42.440 --> 00:37:48.300
then they’re all in the stress shadow,
and they’ve all been shut down.

00:37:48.300 --> 00:37:59.520
Okay, so now let’s go back to the slides
to see how this plays out in the Bay.

00:37:59.530 --> 00:38:02.900
So this is the record of
earthquakes in the San Francisco Bay

00:38:02.900 --> 00:38:08.290
from the start of occupation
by the Spanish.

00:38:08.290 --> 00:38:11.530
And what’s astonishing
about this 75 years is that

00:38:11.530 --> 00:38:16.080
magnitude 6s and 7s are
ricocheting around the Bay.

00:38:16.080 --> 00:38:21.300
Two magnitude 7s -- one on either side
of the Bay, and lots and lots of 6s.

00:38:21.300 --> 00:38:28.020
Now, compare that to the 75 years
after the 1906 earthquake.

00:38:28.020 --> 00:38:31.030
Almost complete shutdown.

00:38:31.030 --> 00:38:36.800
When we calculate the stress transmitted
to those faults by the 1906 earthquake,

00:38:36.800 --> 00:38:42.250
you can see that the Bay fell into
this profound seismic shadow,

00:38:42.250 --> 00:38:45.840
which probably explains
the shutdown of earthquakes,

00:38:45.840 --> 00:38:51.580
which means we’ve been living in a
very unusually quiet period for the Bay.

00:38:51.580 --> 00:38:58.040
Now, the gods continue to crank
the casting reel, right? [laughter]

00:38:58.040 --> 00:39:00.670
That rubber band
is getting tighter.

00:39:00.670 --> 00:39:04.210
And we have 110 years
of cranking that’s going on.

00:39:04.210 --> 00:39:08.260
So our best realization
of what the stresses look like today

00:39:08.260 --> 00:39:10.620
are about like this.

00:39:10.620 --> 00:39:13.960
And because the stress has
profoundly dropped the most

00:39:13.960 --> 00:39:18.220
on the fault, the stresses work
their way in from the sides.

00:39:18.220 --> 00:39:22.660
And so you can see that the
Hayward Fault in the East Bay

00:39:22.660 --> 00:39:28.540
and to the North Bay, we have
some very high stresses building up.

00:39:28.540 --> 00:39:31.860
Which is why we have to
concern ourselves with

00:39:31.860 --> 00:39:34.360
being prepared for
the next earthquake.

00:39:34.360 --> 00:39:37.550
We are almost certainly moving
into a period where we will

00:39:37.550 --> 00:39:40.510
see more earthquakes
because these stresses

00:39:40.510 --> 00:39:44.740
have come back, and the
whole area is shearing.

00:39:44.740 --> 00:39:49.270
Okay. So I’m going to take you to
the new Bay Bridge, but I think

00:39:49.270 --> 00:39:55.120
maybe this is [laughter] a moment
for a little bit of gallows humor.

00:39:55.120 --> 00:40:01.580
[laughter]
See? You have choices no matter what.

00:40:01.580 --> 00:40:04.330
Okay. So here is the Bay Bridge.

00:40:04.330 --> 00:40:11.630
I mean, here is the paragon of
seismic design. And why so?

00:40:11.630 --> 00:40:18.190
Because it has a tall, slender column, or
mast, that takes the compression loads.

00:40:18.190 --> 00:40:22.480
It has stainless steel cables
that take the tensile loads.

00:40:22.480 --> 00:40:26.890
It’s light on top,
and it’s heavy on the bottom.

00:40:26.890 --> 00:40:31.840
Except I’m not talking about the bridge.
I’m talking about the boat.

00:40:31.840 --> 00:40:33.320
[laughter]
That’s right.

00:40:33.320 --> 00:40:38.570
The boat is made of
light, strong materials --

00:40:38.570 --> 00:40:43.660
fiberglass, aluminum,
stainless steel cable.

00:40:43.660 --> 00:40:46.660
It’s built with
triangles and curves,

00:40:46.660 --> 00:40:50.670
which are inherently
stronger than a cube.

00:40:50.670 --> 00:40:54.000
And it’s heavy at the bottom
and extremely light on top.

00:40:54.000 --> 00:40:56.310
And here is the
incredible thing.

00:40:56.310 --> 00:41:01.400
There is not a sailboat in the Bay
that could not withstand anything

00:41:01.400 --> 00:41:05.900
that the San Andreas or
Hayward Fault could throw at it.

00:41:05.900 --> 00:41:08.020
And you’re saying, oh, well,
that’s just because they’re

00:41:08.020 --> 00:41:10.890
bobbing around in the water.
No, you’re wrong.

00:41:10.890 --> 00:41:15.940
If you dug a hole in the ground, and you
hoisted any sailboat into the ground,

00:41:15.940 --> 00:41:20.640
that boat would do just fine with the
largest earthquake you could put on it.

00:41:20.640 --> 00:41:22.260
And it’s even better than that.

00:41:22.260 --> 00:41:24.510
It’s not just structural
damage to the boat.

00:41:24.510 --> 00:41:28.790
If we go inside one of these boats,
you’re going to see that, in the galley,

00:41:28.790 --> 00:41:33.860
which is the sailing term for kitchen,
or the saloon -- sailing term for

00:41:33.860 --> 00:41:37.010
the dining area, that there
are no sharp corners.

00:41:37.010 --> 00:41:42.180
There are handholds everywhere.
There’s no glass that could fly around.

00:41:42.180 --> 00:41:43.950
Every drawer is latched.

00:41:43.950 --> 00:41:47.390
Does this look like your kitchen?
Probably not.

00:41:47.390 --> 00:41:50.000
In fact, the stoves are gimballed.

00:41:50.000 --> 00:41:55.270
You could make soup during the Loma
Prieta earthquake on any boat in the Bay.

00:41:55.270 --> 00:42:01.030
Okay. So I ask you, how could
naval architects be so smart,

00:42:01.030 --> 00:42:04.060
and architects be so stupid?

00:42:05.100 --> 00:42:09.900
Well, it’s really that the Darwinian
process of natural selection in

00:42:09.900 --> 00:42:13.860
boat design works really fast compared
to building design. [laughter]

00:42:13.860 --> 00:42:16.310
Anybody who designs a boat
and sends it out into the Bay

00:42:16.310 --> 00:42:20.880
on a summer day, you know,
with high winds and a lot of chop,

00:42:20.880 --> 00:42:24.350
that boat, if it isn’t built for it,
is going to the bottom.

00:42:24.350 --> 00:42:26.800
And that design isn’t
going to do very well.

00:42:26.800 --> 00:42:31.030
And so quickly yacht designers
learn what is necessary to

00:42:31.030 --> 00:42:36.030
make a vessel that can handle
these very, very high accelerations.

00:42:36.030 --> 00:42:39.500
And it’s even more true
for boats that go off shore.

00:42:39.500 --> 00:42:44.750
So this is a part of the design.
It’s not built into any boating codes.

00:42:44.750 --> 00:42:46.540
It’s boating sense.

00:42:46.540 --> 00:42:49.150
Now, compare that to buildings.

00:42:49.150 --> 00:42:53.310
Very few architects have ever
experienced a large earthquake.

00:42:53.310 --> 00:42:59.180
And it’s not in anybody’s experience
or anybody’s life preparation.

00:42:59.180 --> 00:43:01.530
So they’re basically
building to codes.

00:43:01.530 --> 00:43:05.920
And the codes produce only
the minimal safety standard.

00:43:05.920 --> 00:43:11.420
So I want you to compare this kitchen
to any bar you choose in the Bay area.

00:43:11.420 --> 00:43:12.930
So here’s one that I chose.
[laughter]

00:43:12.930 --> 00:43:15.760
Because it’s local.
I’m a field scientist,

00:43:15.760 --> 00:43:20.660
so I do a lot of research, and I have
to go out and sample everything.

00:43:20.660 --> 00:43:23.390
But look, we know what’s
going to happen in an earthquake

00:43:23.390 --> 00:43:25.950
in that bar, right?
It’s going to be horrible.

00:43:25.950 --> 00:43:30.350
And is this bar in any
kind of problem location? No.

00:43:30.350 --> 00:43:34.390
Everything’s fine.
It’s a mile from the San Andreas.

00:43:34.390 --> 00:43:35.900
So here’s your
homework assignment.

00:43:35.900 --> 00:43:41.680
I challenge you to go to any bar in the
Bay area that restrains the liquor bottles.

00:43:43.340 --> 00:43:47.360
Now, I spend a lot of time in Japan,
actually in bars in Japan.

00:43:47.370 --> 00:43:53.010
I’ve never been to a bar in Japan that
does not restrain its liquor bottles.

00:43:53.010 --> 00:43:55.350
This is denial.

00:43:55.350 --> 00:44:01.030
Okay. I have to make one back step.
There is a kind of boat that

00:44:01.030 --> 00:44:04.290
doesn’t meet any of the glorious
standards that I describe.

00:44:04.290 --> 00:44:07.900
And it’s called a houseboat.
[laughter]

00:44:07.900 --> 00:44:11.810
And if you took this thing out on
the Bay on any summer afternoon

00:44:11.810 --> 00:44:16.540
and turned it broadside to the wind or the
waves, it’s basically going to keel over,

00:44:16.540 --> 00:44:19.080
all those windows are going to break,
everything’s going to fall apart,

00:44:19.080 --> 00:44:20.770
and it’s going to
sink to the bottom.

00:44:20.770 --> 00:44:24.540
So if there were justice in the world,
that would be its naval architect

00:44:24.540 --> 00:44:27.510
being pooped out the transom.
[laughter]

00:44:28.800 --> 00:44:35.880
Okay, guys, here’s the bad news.
We all live in beached houseboats.

00:44:35.880 --> 00:44:40.460
And let me tell you
what I mean by that.

00:44:43.830 --> 00:44:46.400
Here is a building that you’ll
find anywhere in the Bay area,

00:44:46.400 --> 00:44:48.320
and pretty much anywhere
around the world.

00:44:48.320 --> 00:44:53.300
We have decided to build
buildings out of stacks of cubes.

00:44:53.300 --> 00:44:56.120
Cubes have no
structural integrity.

00:44:56.120 --> 00:44:58.840
They’re only held together
by their corners.

00:44:58.850 --> 00:45:02.670
No matter -- this is bolted down to the
ground -- well, actually Velcro-ed.

00:45:02.670 --> 00:45:07.230
And no matter how strong
I make the columns and the beams,

00:45:07.230 --> 00:45:09.690
this building is only
as strong as its corners.

00:45:09.690 --> 00:45:11.850
What’s going to happen
in an earthquake?

00:45:11.850 --> 00:45:14.780
The ground
goes side-by-side.

00:45:14.780 --> 00:45:18.800
Or -- and also torsion.
See how weak this building is?

00:45:18.800 --> 00:45:25.050
If you don’t believe me when I say cubes
have no structural integrity, look at this.

00:45:25.050 --> 00:45:27.450
Ikea house.
You can put it in your station wagon

00:45:27.450 --> 00:45:31.700
and drive it home.
[laughter]

00:45:31.700 --> 00:45:35.650
Okay. This is the exact same building.
The exact same building, but I’ve

00:45:35.650 --> 00:45:43.240
snapped in -- literally snapped in these
little corner braces with sewing snaps.

00:45:45.120 --> 00:45:46.500
Watch.

00:45:49.360 --> 00:45:50.900
Nothing.

00:45:51.460 --> 00:45:55.660
Okay. How much more do you think
it takes -- how much -- what do you

00:45:55.660 --> 00:46:00.490
think the additional cost is of building
this building than this building?

00:46:00.490 --> 00:46:02.410
Any guesses?
- 5%.

00:46:02.410 --> 00:46:06.950
- Half a percent.
- 1 to 2%.

00:46:06.950 --> 00:46:13.830
And if you’re retrofitting this building to
go to this building, maybe it’s 2 to 3%.

00:46:14.680 --> 00:46:18.080
So this is what we have to do.
This is how we have to live.

00:46:18.080 --> 00:46:20.800
And this is, incidentally,
why a boat is strong.

00:46:20.800 --> 00:46:25.850
Because it’s filled with triangles and
curves, and it’s not a stack of cubes.

00:46:25.850 --> 00:46:28.740
So as long as we
want to build tall cubes,

00:46:28.740 --> 00:46:32.190
we are facing an incredibly
difficult engineering problem.

00:46:32.190 --> 00:46:36.410
It’s ironic to think that, if we lived
in geodesic domes, we would not

00:46:36.410 --> 00:46:40.000
be having this seminar because nobody
would care about earthquakes.

00:46:40.000 --> 00:46:41.640
Because they
wouldn’t kill anybody.

00:46:41.640 --> 00:46:46.520
But we have chosen
buildings that attack us.

00:46:46.520 --> 00:46:49.550
And so we have to
move away from that.

00:46:49.550 --> 00:46:54.900
So I want to leave you with the idea
and the recognition that so much of

00:46:54.900 --> 00:46:59.410
the wealth and beauty
and history of California

00:46:59.410 --> 00:47:04.270
is bound up in
its seismic setting.

00:47:04.270 --> 00:47:12.510
And so we -- in a sense -- in a very real
sense, serendipitously and ironically,

00:47:12.510 --> 00:47:17.000
the Bay Area’s loss has been
science’s gain -- how much

00:47:17.000 --> 00:47:20.000
was learned from
our seismic past.

00:47:20.000 --> 00:47:26.280
And the best way I, as a scientist,
can return this to you is to say --

00:47:26.280 --> 00:47:31.190
is to try to encourage you
to move from this to this

00:47:31.190 --> 00:47:36.470
and work to make yourself safer
for the earthquakes that lie ahead.

00:47:36.470 --> 00:47:38.250
Thank you.

00:47:38.250 --> 00:47:45.740
[ Applause ]

00:47:45.740 --> 00:47:48.420
And I just want to point out,
seismically strong buildings

00:47:48.420 --> 00:47:53.470
can be absolutely beautiful.
And I also want to acknowledge

00:47:53.470 --> 00:47:57.170
the people who made these
wonderful tools with me.

00:47:59.580 --> 00:48:01.220
- I’m sure many of you
have questions,

00:48:01.220 --> 00:48:04.320
and Ross said that he’d be
happy to answer them.

00:48:04.320 --> 00:48:08.640
We do have two microphones
in the two aisles back here.

00:48:08.640 --> 00:48:12.280
We like you to use a microphone, not
only so others in the room can hear you,

00:48:12.280 --> 00:48:18.070
but we’re web-streaming this live so
people online can hear you as well.

00:48:18.070 --> 00:48:21.290
So if you can, please line up
at one or two of the microphones,

00:48:21.290 --> 00:48:23.980
and we’ll go
back and forth.

00:48:23.980 --> 00:48:28.120
If it’s difficult for you to do that, wave
at me, and I’ll bring you a microphone.

00:48:28.120 --> 00:48:32.160
But let’s -- why don’t we start with
the gentleman in the orange shirt?

00:48:32.160 --> 00:48:33.210
Go ahead with
the first question.

00:48:33.210 --> 00:48:36.550
- It’s going to be a funny question.
Where do you stand on fracking, then?

00:48:36.550 --> 00:48:39.620
That’s supposed to be relieving
and making small earthquakes.

00:48:39.620 --> 00:48:43.930
Does that make it better? Worse?
Or who knows?

00:48:43.930 --> 00:48:48.370
- So the USGS actually is doing a lot of
work in trying to understand the

00:48:48.370 --> 00:48:53.500
role of oil field and gas field activities
in inducing earthquakes.

00:48:53.500 --> 00:48:57.010
First, you should know that
18 months ago, Oklahoma dethroned

00:48:57.010 --> 00:49:00.400
California as the most seismically
active state in the union.

00:49:00.400 --> 00:49:02.820
We’re very upset about that.
[laughter]

00:49:02.820 --> 00:49:04.890
Taking it very personally.

00:49:04.890 --> 00:49:09.150
And most of these earthquakes are
almost certainly induced, but they’re not

00:49:09.150 --> 00:49:13.270
induced by the fracking, per se, which is
happening at relative shallow depths.

00:49:13.270 --> 00:49:16.100
They’re induced by what’s
called produced water --

00:49:16.100 --> 00:49:20.040
all the water that comes out of the
ground with the gas and oil

00:49:20.040 --> 00:49:25.430
and the fracking fluids that need to be
re-injected very deep into the ground

00:49:25.430 --> 00:49:27.750
so that they don’t
contaminate the groundwater.

00:49:27.750 --> 00:49:31.430
By injecting, under high pressure,
these very toxic fluids,

00:49:31.430 --> 00:49:33.630
they’re reaching
fault zones.

00:49:33.630 --> 00:49:36.770
And as long as the rubber band is
sufficiently tight, these are not

00:49:36.770 --> 00:49:41.540
fault zones where the crank is cranking
very fast, but it is still cranking.

00:49:41.540 --> 00:49:45.020
So some of these faults are close to
failure, and basically they’re being

00:49:45.020 --> 00:49:51.380
lubricated, and large -- or moderate
to large earthquakes are occurring.

00:49:51.380 --> 00:49:52.710
Yes?

00:49:53.760 --> 00:49:55.360
- Go ahead.

00:49:56.580 --> 00:50:02.300
- Can you discuss the program that
USGS and Caltech and Berkeley

00:50:02.300 --> 00:50:07.099
have regarding early warning
detection for earthquakes?

00:50:07.099 --> 00:50:11.150
- Sure. So earthquake early warning
is a very exciting development

00:50:11.150 --> 00:50:16.850
on the horizon for all of us here
that can give us seconds to

00:50:16.850 --> 00:50:21.160
tens of seconds of warning
before earthquakes.

00:50:21.160 --> 00:50:26.720
And there’s a great deal that that can
accomplish in terms of dropping fuel

00:50:26.720 --> 00:50:33.980
rods in nuclear reactors and stopping
laser eye surgery, and the list is long.

00:50:33.980 --> 00:50:37.460
However -- and so this is important,
and many different groups

00:50:37.460 --> 00:50:39.530
are working to
bring this fruition.

00:50:39.530 --> 00:50:44.170
And many parts of the world
already have such a system.

00:50:44.170 --> 00:50:48.380
With that said, remember that
none of us want a situation

00:50:48.380 --> 00:50:53.010
in which we all got warning,
and every building fell down.

00:50:53.010 --> 00:50:55.750
We do not want to have
to rebuild the Bay Area.

00:50:55.750 --> 00:50:57.580
We want to build it strong.

00:50:57.580 --> 00:51:02.370
So the proper combination
of things is to have early warning,

00:51:02.370 --> 00:51:06.190
but then also to recognize
that we can take steps so that,

00:51:06.190 --> 00:51:09.830
when that earthquake occurs,
it’s not just a matter of preserving

00:51:09.830 --> 00:51:14.620
life safety, that we have a habitable,
recoverable community.

00:51:14.620 --> 00:51:15.780
Yes?

00:51:15.780 --> 00:51:17.700
- So two questions.

00:51:17.700 --> 00:51:26.060
The first one is, do you see any
future long-range in lubricating faults

00:51:26.060 --> 00:51:30.460
from what we’re finding out
about fracking, and therefore

00:51:30.460 --> 00:51:33.400
avoiding big quakes?
That’s question one.

00:51:33.400 --> 00:51:35.670
- Well, you know,
careful what you wish for.

00:51:35.670 --> 00:51:38.090
- Yeah.
- Is all I can say.

00:51:38.090 --> 00:51:43.050
You know, it’s certainly true that --
and it’s been known since the ’60s

00:51:43.050 --> 00:51:45.560
that lubricating faults can
trigger earthquakes.

00:51:45.560 --> 00:51:50.500
But you don’t know -- here’s the
problem. It’s a numbers game.

00:51:50.500 --> 00:51:55.330
So if -- let’s say what we’re trying
to do is get rid of the 7s.

00:51:55.330 --> 00:51:59.100
Well, a 7 is 1,000 times
larger than a 5.

00:51:59.100 --> 00:52:05.010
So you need a 5 -- you need a
thousand 5s to get rid of that 7.

00:52:05.010 --> 00:52:10.890
And so you simply can’t get enough
moderate earthquakes to remove

00:52:10.890 --> 00:52:15.190
the importance or the need
for the larger earthquakes.

00:52:15.190 --> 00:52:18.110
So in a world in which we
could lubricate all the faults,

00:52:18.110 --> 00:52:20.480
we’d have big earthquakes
every day to get rid of the

00:52:20.480 --> 00:52:22.940
likelihood of something
really large.

00:52:22.940 --> 00:52:26.380
And it’s just as possible that,
by lubricating the fault,

00:52:26.380 --> 00:52:30.460
we will actually create the large
earthquake that we would rather avoid.

00:52:30.460 --> 00:52:34.470
And there’s evidence from
filling of reservoirs that

00:52:34.470 --> 00:52:38.300
large earthquakes have
occurred in that fashion.

00:52:38.300 --> 00:52:42.960
- And you said that the nature
of earthquakes wasn’t really

00:52:42.960 --> 00:52:46.140
discovered until they
started observing the displacement

00:52:46.140 --> 00:52:51.420
from the -- from the ’06 quake.
Was Japan any further ahead on this?

00:52:51.420 --> 00:52:53.940
Did they know what was going on,
since they’ve had so much history?

00:52:53.940 --> 00:52:57.650
- There was an earthquake in Japan
in 1894 called the Nobi earthquake,

00:52:57.650 --> 00:53:01.830
where the Japanese began
to see similar things.

00:53:01.830 --> 00:53:05.690
The 1906 earthquake was
so much larger and so much

00:53:05.690 --> 00:53:09.800
better surveyed that
the message was --

00:53:09.800 --> 00:53:13.540
really kind of burst into
awareness around the world.

00:53:13.540 --> 00:53:18.560
So the Japanese, as they are often,
today, ahead of us in seismic science,

00:53:18.560 --> 00:53:20.760
began to see things
just before we did.

00:53:20.760 --> 00:53:26.960
But the picture immediately
snapped into focus after 1906.

00:53:26.960 --> 00:53:28.080
Yes?

00:53:28.080 --> 00:53:31.480
- Okay, I heard that
the One Rincon building

00:53:31.480 --> 00:53:38.750
had a ballast of water at the top to
counter the sway from an earthquake.

00:53:38.750 --> 00:53:41.150
Would that work,
do you think?

00:53:41.150 --> 00:53:44.360
So I heard that, too, but then I talked to
an engineer recently -- and of course,

00:53:44.360 --> 00:53:50.270
I’m not an engineer -- and I was told it
does not have this water baffle system.

00:53:50.270 --> 00:53:55.340
However, I was in the
Taipei 101 building,

00:53:55.340 --> 00:53:59.930
which is really -- this is the Taiwanese
definition of the word "hutzpa."

00:53:59.930 --> 00:54:04.630
They built a 101-story building at the
foot of a fault in the path of typhoons.

00:54:04.630 --> 00:54:06.060
[laughter]
- Oh my gosh.

00:54:06.060 --> 00:54:10.010
- Okay. So at the top of this incredible,
beautiful building -- it’s got this

00:54:10.010 --> 00:54:12.230
beautiful, kind of
bamboo-like look to it.

00:54:12.230 --> 00:54:15.480
The Taiwanese call it a
stack of Chinese take-out cartons.

00:54:15.480 --> 00:54:21.540
But at the top of it is a giant steel ball
as wide as this room that’s suspended

00:54:21.540 --> 00:54:24.180
and has dashpots
at the bottom.

00:54:24.180 --> 00:54:30.500
And this is designed for an earthquake or
a typhoon, that if the building tilts over,

00:54:30.500 --> 00:54:33.300
the ball then goes out of alignment,
and the ball sucks the building

00:54:33.300 --> 00:54:35.680
back into position.

00:54:35.680 --> 00:54:38.980
But what’s really great about this is
there -- you’re going to kill about this.

00:54:38.980 --> 00:54:41.150
There’s a bar around the ball.
[laughter]

00:54:41.150 --> 00:54:46.560
And you can sit there with a --
with a cocktail, and you can

00:54:46.560 --> 00:54:50.220
watch this ball displace.
And you go, oh, I’m 100 stories up,

00:54:50.220 --> 00:54:53.580
and if that ball’s moving,
it means actually I’m moving.

00:54:53.580 --> 00:54:56.670
So I think that’s one of the
best geological experiences

00:54:56.670 --> 00:55:00.180
you could ask for.
[laughter]

00:55:00.180 --> 00:55:01.640
Yes?

00:55:01.640 --> 00:55:06.460
- So I’m curious about the development
of the tectonic theory and why,

00:55:06.460 --> 00:55:11.790
after they noticed the -- you know,
the ground slippage, why did it take

00:55:11.790 --> 00:55:15.110
so long for them to, you know,
extrapolate that hey, you know,

00:55:15.110 --> 00:55:16.800
this thing actually
would go on and on and on,

00:55:16.800 --> 00:55:19.500
and then suddenly you’d
have faults all over the world?

00:55:19.500 --> 00:55:21.060
So I don’t know if you
know anything about that.

00:55:21.060 --> 00:55:22.160
- It’s a great question.

00:55:22.160 --> 00:55:25.600
You know, what the -- what was really
powerful about Reid’s discovery is,

00:55:25.600 --> 00:55:30.270
if you could move 12 feet on a fault
every 200 or 300 years,

00:55:30.270 --> 00:55:33.920
then in 10 million years,
you’d move 150 miles.

00:55:33.920 --> 00:55:36.760
And that was
inconceivable at the time.

00:55:36.770 --> 00:55:39.500
The thought was,
the only way the Earth

00:55:39.500 --> 00:55:42.840
could change if it
expanded or contracted.

00:55:42.840 --> 00:55:44.570
And this didn’t
make any sense.

00:55:44.570 --> 00:55:49.080
So, like many things in science,
some powerful clue lay there

00:55:49.080 --> 00:55:54.690
kind of waiting, germinating,
for other ideas to come along.

00:55:54.690 --> 00:55:57.470
And Wegener’s hypothesis
of continental drift

00:55:57.470 --> 00:56:00.980
had the continents plowing
through the oceanic material.

00:56:00.980 --> 00:56:04.670
Which geophysicists couldn’t stand
because they couldn’t see how

00:56:04.670 --> 00:56:09.270
lighter material could actually move
through the denser oceanic material.

00:56:09.270 --> 00:56:15.680
So it wasn’t until submarine data
began to reveal -- and oceanic data

00:56:15.680 --> 00:56:20.920
began to reveal the nature of the
sea floor and these mid-ocean rifts,

00:56:20.920 --> 00:56:25.160
they were put together
with earthquakes like 1906

00:56:25.160 --> 00:56:29.560
and the 1966 Parkfield earthquake,
and things snapped into focus.

00:56:29.560 --> 00:56:33.820
And like most discoveries,
once those pieces started to assemble,

00:56:33.820 --> 00:56:35.610
everything happened
really fast.

00:56:35.610 --> 00:56:37.760
Everything happened in
a matter of a year or two.

00:56:37.760 --> 00:56:42.640
It was -- it was all over
between 1966 and 1968.

00:56:42.640 --> 00:56:47.940
And this is just as profound as natural
selection -- this change of thinking.

00:56:47.940 --> 00:56:49.320
- Thanks.

00:56:49.820 --> 00:56:52.060
- One other question.
- I saw you before.

00:56:52.060 --> 00:56:54.300
- I know.
[laughter]

00:56:54.310 --> 00:57:00.190
I heard the Northridge earthquake
was a new kind of blind thrust fault.

00:57:00.190 --> 00:57:04.110
Was that known about before?
Or if it wasn’t, how many

00:57:04.110 --> 00:57:06.160
more don’t we know about?
[laughter]

00:57:06.160 --> 00:57:10.990
- Hm. So blind thrust faults were
something that began to be clearer in --

00:57:10.990 --> 00:57:14.910
between about 1980 to 1983, when
we had the Coalinga earthquake.

00:57:14.910 --> 00:57:17.580
That was the first large
blind thrust earthquake --

00:57:17.580 --> 00:57:25.410
a magnitude 6.7 in the oil patch in --
east of the San Andreas.

00:57:25.410 --> 00:57:30.140
And there had been a
magnitude 7.5 earthquake in Algeria

00:57:30.140 --> 00:57:31.630
on a blind thrust fault, too.

00:57:31.630 --> 00:57:34.190
And it just means it’s
blind to the geologists.

00:57:34.190 --> 00:57:37.680
The fault doesn’t come to the surface.
So you can’t map it at the surface.

00:57:37.680 --> 00:57:42.810
And what it does is it
folds the rocks at the surface.

00:57:42.810 --> 00:57:46.950
And so now they’re
detected by finding places

00:57:46.950 --> 00:57:50.490
where the strata had
been warped or folded.

00:57:50.490 --> 00:57:54.690
And so many of the important
blind faults now are part of

00:57:54.690 --> 00:57:57.880
the USGS models and
other models around the world.

00:57:57.880 --> 00:58:01.960
But they are the hardest faults to find,
and I’m sure there are more out there.

00:58:01.960 --> 00:58:04.600
And I might add, you know,
in my 30 years of being a geologist

00:58:04.600 --> 00:58:07.360
and standing in front of the news
after an earthquake, about

00:58:07.360 --> 00:58:11.010
one of every three earthquakes occurred
on a fault we didn’t know about.

00:58:11.010 --> 00:58:14.100
This is a very humbling field, okay?
[laughter]

00:58:14.100 --> 00:58:17.110
If you want to be smug,
choose another field because

00:58:17.110 --> 00:58:20.630
this happens over and over again,
even without blind thrust faults.

00:58:20.630 --> 00:58:21.940
Yes?

00:58:22.580 --> 00:58:28.420
- You mentioned in the case of loading
a reservoir or injecting fracking fluid,

00:58:28.420 --> 00:58:31.970
you seem to mention lubrication
is the primary driver for earthquakes

00:58:31.970 --> 00:58:37.340
from these -- from these
things that humans do.

00:58:37.340 --> 00:58:41.340
How would you consider
lubrication versus loading

00:58:41.340 --> 00:58:44.860
for what’s the primary
driver of those earthquakes?

00:58:44.860 --> 00:58:48.720
- It’s a great question, and they’re
probably both important.

00:58:48.720 --> 00:58:55.290
The Wenchuan earthquake in 2008
in China, which was a magnitude 7.9,

00:58:55.290 --> 00:58:59.790
and killed 60,000 people --
20,000 of them schoolchildren,

00:58:59.790 --> 00:59:02.410
the epicenter of that earthquake
was underneath a reservoir

00:59:02.410 --> 00:59:05.210
that the Chinese
had recently filled.

00:59:05.210 --> 00:59:09.840
And there is some evidence
that the filling of that reservoir,

00:59:09.840 --> 00:59:15.950
possibly by both diffusion of water
and lubrication and changing of the load

00:59:15.950 --> 00:59:19.350
might have been the factor
that triggered that earthquake.

00:59:19.350 --> 00:59:23.350
The Chinese, of course, very
vigorously deny this possibility.

00:59:23.350 --> 00:59:29.080
And the evidence is not overwhelming.
You know, you don’t see a sequence

00:59:29.080 --> 00:59:32.480
of earthquakes migrating to
the future epicentral site.

00:59:32.480 --> 00:59:36.000
So there are lots of cases of
earthquakes that are associated

00:59:36.000 --> 00:59:40.530
with dams up to magnitude 6.7,
and this possible one at 7.9,

00:59:40.530 --> 00:59:46.460
but it’s very hard to be
certain of those relationships.

00:59:46.460 --> 00:59:49.860
So undoubtedly, both processes
are important, and the point is --

00:59:49.860 --> 00:59:52.410
your point is well-taken.
- Thank you.

00:59:53.270 --> 00:59:56.550
- One of the slides you had of the
Bay Area showed the Hayward Fault

00:59:56.550 --> 00:59:59.830
in kind of an orange and then
another fault off the shore,

00:59:59.830 --> 01:00:02.570
but the San Andreas
outlined in blue.

01:00:02.570 --> 01:00:06.119
Would you explain what that means
for the future of the San Andreas?

01:00:06.119 --> 01:00:13.280
- Well, you know, if -- imagine that the
San Andreas slipped a uniform amount.

01:00:13.280 --> 01:00:18.780
Then the stress shadow would be all
about the same along the fault, and the

01:00:18.780 --> 01:00:23.890
San Andreas would be kind of the last
to go because it has gone most recently.

01:00:23.890 --> 01:00:28.390
But the fault is not a straight line.
It’s got bends and breaks.

01:00:28.390 --> 01:00:31.620
And the slip is irregular,
so there are probably concentrations

01:00:31.620 --> 01:00:35.020
of stress on the San Andreas
that we don’t know about

01:00:35.020 --> 01:00:38.840
that could give us 6s and 7s
on the San Andreas.

01:00:38.840 --> 01:00:42.890
The San Andreas did have,
on the peninsula, near Filoli,

01:00:42.890 --> 01:00:46.150
there is evidence for
a magnitude 7 in 1838.

01:00:46.150 --> 01:00:51.540
So we know the San Andreas does not
just knock off 8s -- or actually 7.8s.

01:00:51.540 --> 01:00:53.860
It also produces
moderate earthquakes.

01:00:53.860 --> 01:00:58.300
So it’s very difficult to know
what it could do next.

01:00:58.300 --> 01:01:02.180
But let me put it this way.
No seismologist would be shocked

01:01:02.180 --> 01:01:06.600
if the San Andreas -- any part of it
up here -- produced a 7.

01:01:06.600 --> 01:01:09.880
It’s just part of the game.

01:01:09.880 --> 01:01:11.480
Yes?

01:01:11.480 --> 01:01:14.000
- So you mentioned about
the early warning system.

01:01:14.000 --> 01:01:17.520
How do they -- how do they work?
Are you measuring strain,

01:01:17.520 --> 01:01:21.840
or are you measuring early movement
somewhere more closer to the epicenter,

01:01:21.840 --> 01:01:24.140
or is there something else?
- It’s a great question.

01:01:24.140 --> 01:01:28.440
Basically, it’s just -- it’s just another geo-
trick since we can’t predict earthquakes.

01:01:28.440 --> 01:01:30.930
So we get enough instruments
out there that we can detect

01:01:30.930 --> 01:01:35.530
the earthquake as it’s rupturing --
as it’s beginning to nucleate.

01:01:35.530 --> 01:01:38.400
And just so you know,
an earthquake nucleation is a process

01:01:38.400 --> 01:01:41.990
where the fault is normally
moving an inch a year,

01:01:41.990 --> 01:01:46.930
and it speeds up to 3,000 miles
an hour in about 3 seconds.

01:01:46.930 --> 01:01:50.180
This is one of the reasons why
these things are so hard to predict.

01:01:50.180 --> 01:01:53.710
If you have instruments
close enough, you will detect that.

01:01:53.710 --> 01:01:56.810
If you have three or four,
then you can be certain that it is

01:01:56.810 --> 01:01:59.970
not a disturbance or a
malfunctioning of one instrument.

01:01:59.970 --> 01:02:04.770
So those associations are made.
The determination of the earthquake.

01:02:04.770 --> 01:02:08.550
And then, as it starts to grow,
its size becomes evident.

01:02:08.550 --> 01:02:11.370
And then there’s
enough computer power

01:02:11.370 --> 01:02:15.670
to determine roughly when that
shaking will reach everybody.

01:02:15.670 --> 01:02:17.960
And then, on your cell phone,
you’ll have an app for that,

01:02:17.960 --> 01:02:21.900
and you’ll get a ring.
And it’ll tell you how strong

01:02:21.900 --> 01:02:25.920
the shaking should be for you
and when it’ll get there.

01:02:25.920 --> 01:02:33.810
Now, this is in testing now, and it’s
a far cry from failsafe or bulletproof.

01:02:33.810 --> 01:02:36.620
There are missed earthquakes.
There are false earthquakes.

01:02:36.620 --> 01:02:37.900
There are false alerts.

01:02:37.900 --> 01:02:42.110
But the system is really just a
matter of detecting the earthquake

01:02:42.110 --> 01:02:45.070
and getting that to you
before the seismic waves do

01:02:45.070 --> 01:02:48.040
because they’re moving
at 3,000 miles an hour.

01:02:48.040 --> 01:02:51.230
And they’re going to get there
more slowly than that signal.

01:02:51.230 --> 01:02:53.540
- All right. Just had a
follow-up question.

01:02:53.540 --> 01:02:56.850
I heard of some experiment with
electromagnetic disturbances

01:02:56.850 --> 01:03:02.320
that were detected at 1906 and 1989
and, I mean, there’s a theory that, okay,

01:03:02.320 --> 01:03:05.280
there might be some electromagnetic
disturbances before that we could --

01:03:05.280 --> 01:03:08.369
do you have any
thoughts on that?

01:03:08.369 --> 01:03:11.940
- Well, the USGS isn’t
pursuing that line of research.

01:03:11.940 --> 01:03:15.900
There’s some independent groups
that are looking to see if there is

01:03:15.900 --> 01:03:18.960
a consistent and reliable
indicator of that kind.

01:03:18.960 --> 01:03:25.030
But I want to say that -- that’s why
I called it a bitter pill to swallow.

01:03:25.030 --> 01:03:27.550
Just about everything
we’ve tried hasn’t worked.

01:03:27.550 --> 01:03:30.750
And the best evidence from the
laboratory and the field is that,

01:03:30.750 --> 01:03:33.790
if we were thinking about
a managed 6 or 7 earthquake,

01:03:33.790 --> 01:03:37.410
the preparation zone, if there is one,
is about the size of this room.

01:03:37.410 --> 01:03:42.010
And it’s 10 miles down.
So it’s a pin prick.

01:03:42.010 --> 01:03:46.820
And to be able to detect
that phenomena on the

01:03:46.820 --> 01:03:49.170
surface of the Earth
is extremely difficult.

01:03:49.170 --> 01:03:52.550
Parkfield, the place I mentioned --
the one-bricker spot,

01:03:52.550 --> 01:03:54.400
so we kind of know where
it was going to be because

01:03:54.400 --> 01:03:59.860
it kept happening, we had -- we put
everything we had at Parkfield.

01:03:59.860 --> 01:04:01.900
It ruptured on the other end
of the fault than we thought,

01:04:01.900 --> 01:04:06.150
so our best instruments were about,
oh, 15 miles away.

01:04:06.150 --> 01:04:10.270
We could measure
tiny changes in pressure.

01:04:10.270 --> 01:04:13.700
We could measure pressure changes
that would be equivalent of

01:04:13.700 --> 01:04:17.390
dropping a jigger of
gin in Lake Tahoe.

01:04:17.390 --> 01:04:19.360
And we saw nothing
before that earthquake.

01:04:19.360 --> 01:04:21.290
Nothing whatsoever.

01:04:21.290 --> 01:04:26.230
So I invite you [laughter]
to work on this problem.

01:04:26.230 --> 01:04:32.070
Because it is hard, and it needs young
people to make progress there.

01:04:32.070 --> 01:04:33.270
Yes?

01:04:33.270 --> 01:04:39.349
- This is a question regarding your non-
collapsible model that you have on there.

01:04:39.349 --> 01:04:41.249
- My good guy?
- Your good guy, yeah.

01:04:41.249 --> 01:04:42.460
- Yes.

01:04:42.460 --> 01:04:50.300
- Do the USGS scientists and
geo-advisers partner with

01:04:50.300 --> 01:04:57.330
construction firms to create some type
of seismic retrofitting standardization?

01:04:57.330 --> 01:05:00.270
Because it seems like the
only people doing the testing

01:05:00.270 --> 01:05:03.120
are the manufacturers
of the parts.

01:05:03.120 --> 01:05:08.920
And if you call any local contractor
to get input on how they approach

01:05:08.920 --> 01:05:11.800
retrofitting your home,
it’s all over the map,

01:05:11.800 --> 01:05:16.530
including cutting more parts away
of your foundation rather than adding,

01:05:16.530 --> 01:05:20.310
you know, in layers
on the exterior frame.

01:05:20.310 --> 01:05:22.540
- It’s a great question.

01:05:22.540 --> 01:05:25.450
The Survey -- USGS is not
involved in this process.

01:05:25.450 --> 01:05:28.740
There’s the Earthquake Engineering
Research Institute and the

01:05:28.740 --> 01:05:32.530
Applied Technology Council,
and to some extent, the California

01:05:32.530 --> 01:05:36.300
Seismic Safety Commission,
all of which address this.

01:05:36.300 --> 01:05:39.970
There’s also Earthquake Brace & Bolt,
which is administered by the

01:05:39.970 --> 01:05:42.790
California Earthquake Authority,
that gives retrofit grants and

01:05:42.790 --> 01:05:46.250
are focusing on
research on retrofit.

01:05:46.250 --> 01:05:52.240
But I completely agree with you that, 
overall, the retrofit business is a mess.

01:05:52.240 --> 01:05:56.840
There is no licensing
for retrofitting.

01:05:56.840 --> 01:05:59.470
I just want you to know,
you cannot color a woman’s hair

01:05:59.470 --> 01:06:01.570
without a license.
You can retrofit a home

01:06:01.570 --> 01:06:02.820
without a license.

01:06:02.820 --> 01:06:05.480
That makes a lot of sense.
[laughter]

01:06:05.480 --> 01:06:06.760
And it’s worse than that.

01:06:06.760 --> 01:06:11.220
I just refinanced -- or my wife
and I just refinanced our home,

01:06:11.220 --> 01:06:16.800
and the California-certified real estate
appraiser came to my house.

01:06:16.800 --> 01:06:19.630
And I said -- his name was Joe --
I said, Joe, would you like to know

01:06:19.630 --> 01:06:22.290
if I retrofitted my house?
And he says, no.

01:06:22.290 --> 01:06:24.240
I said, what do
you mean, no?

01:06:24.240 --> 01:06:27.700
He says, it has no effect
on the value of the house.

01:06:28.930 --> 01:06:32.050
So this is a broken
part of the system.

01:06:32.050 --> 01:06:33.820
And so you put your
finger on something.

01:06:33.820 --> 01:06:36.860
If we can standardize it,
if we can accredit it,

01:06:36.860 --> 01:06:41.520
if we can properly train for it,
this is a relatively inexpensive fix --

01:06:41.520 --> 01:06:46.130
for most of us, $5,000 to $10,000
that confers an enormous increase

01:06:46.130 --> 01:06:53.250
in safety and survivability
and loss of -- reduction of damage.

01:06:53.250 --> 01:06:55.060
So thank you.

01:06:55.060 --> 01:06:57.160
Okay, I think this is our last question.

01:06:57.160 --> 01:07:02.720
- I have a question about
the new reservoir in China

01:07:02.730 --> 01:07:06.880
that was blamed
for the earthquake.

01:07:06.880 --> 01:07:10.900
Was not that earthquake
inevitable anyway?

01:07:10.900 --> 01:07:18.820
And could that reservoir have
possibly lessened the damage?

01:07:18.820 --> 01:07:25.760
- Well, certainly the first part of your --
of your premise is correct,

01:07:25.760 --> 01:07:31.170
that this is a fault that is being loaded.
Its spring is being pulled.

01:07:31.170 --> 01:07:34.980
And so it would have an
earthquake sometime anyway.

01:07:34.980 --> 01:07:38.640
So at most, filling of the reservoir
advanced that earthquake

01:07:38.640 --> 01:07:41.140
by perhaps
a hundred years.

01:07:41.140 --> 01:07:45.350
But certainly, if that’s your family,
that hundred years really matters.

01:07:45.350 --> 01:07:51.120
So if you’re going to change the
natural process, then you’re going to

01:07:51.120 --> 01:07:55.870
put yourself into a situation
where you could be blamed.

01:07:55.870 --> 01:07:59.570
In Hawaii, if a lava flow
goes down the flank,

01:07:59.570 --> 01:08:03.210
and it engulfs a home,
there’s no liability issue.

01:08:03.210 --> 01:08:06.480
But if you deflect that lava flow,
then anything else that

01:08:06.480 --> 01:08:09.540
happens downstream,
you’re responsible for it.

01:08:09.540 --> 01:08:12.980
So I think that it’s an issue
of responsibility,

01:08:12.980 --> 01:08:16.380
and it will be very hard to
prove one way or another.

01:08:16.380 --> 01:08:21.489
But the larger problem is this,
and we should just end here.

01:08:21.489 --> 01:08:25.440
In countries like India
and China, to fuel their

01:08:25.440 --> 01:08:29.960
magnificent economic engine,
they need water and power.

01:08:29.960 --> 01:08:32.509
And where do they need
to build these reservoirs?

01:08:32.509 --> 01:08:36.980
In the -- in the foothills, which are
laced with faults in both countries.

01:08:36.980 --> 01:08:39.460
So they are not really
being straight with the public

01:08:39.460 --> 01:08:42.549
because they feel they have to
produce the water and power.

01:08:42.549 --> 01:08:45.880
And as a result, we are seeing
hydroelectric facilities

01:08:45.880 --> 01:08:50.219
go up in places where inevitably,
we will be in harm’s way.

01:08:50.219 --> 01:08:51.949
That’s the larger problem.

01:08:51.949 --> 01:08:54.829
Thank you for your
wonderful questions, and good night.

01:08:54.829 --> 01:09:01.929
[ Applause ]

01:09:01.929 --> 01:09:04.819
- Thank you, Ross.
And thank you all for coming.

01:09:04.819 --> 01:09:07.179
And some of you probably noticed
there was a little slip of paper

01:09:07.179 --> 01:09:09.480
on your chair
when you arrived.

01:09:09.480 --> 01:09:11.839
We don’t usually ask for your --
well, we’re always open

01:09:11.839 --> 01:09:15.359
to your opinion, but tonight
we’re explicitly asking for it.

01:09:15.359 --> 01:09:18.480
You can leave your responses
on the table in the back of the room

01:09:18.480 --> 01:09:20.170
if you care to give us
your opinion.

01:09:20.170 --> 01:09:25.150
Thank you, and I’ll see you next month.
- Thank you.