WEBVTT
Kind: captions
Language: en-US

00:00:00.865 --> 00:00:03.905
[inaudible background conversations]

00:00:03.905 --> 00:00:06.640
Okay.
Let’s kick things off.

00:00:06.640 --> 00:00:12.115
And I will introduce Tim Dawson
to introduce our speakers.

00:00:14.705 --> 00:00:17.970
- All right, people out in
the hallway. Get in.

00:00:20.000 --> 00:00:21.907
Put the coffee down.

00:00:23.520 --> 00:00:26.560
I don’t have many introductory
remarks except for I’m pleased to

00:00:26.560 --> 00:00:31.600
be back in Menlo Park, as are some
of you USGS folks who used to

00:00:31.600 --> 00:00:35.816
hang out around here who
have moved to Moffett Field.

00:00:35.840 --> 00:00:40.640
I’d like to kick off this morning’s
mid-morning session on the

00:00:40.640 --> 00:00:44.800
Ridgecrest earthquake sequence
with the theme of fault complexity –

00:00:44.800 --> 00:00:48.296
lessons from Ridgecrest.
And our first speaker will be

00:00:48.320 --> 00:00:52.640
Ken Hudnut, who was instrumental
in the post-earthquake response,

00:00:52.640 --> 00:00:56.240
along with a lot of people in this room.
I think almost every geologist in this

00:00:56.240 --> 00:01:00.536
room somehow participated
in the earthquake response.

00:01:00.560 --> 00:01:03.680
And the title of his talk is
Cross-Fault Interaction in the

00:01:03.680 --> 00:01:08.240
July 2019 Ridgecrest Earthquake
Sequence, Southern California –

00:01:08.240 --> 00:01:11.680
what we did see in the field
from imagery and GPS data

00:01:11.680 --> 00:01:15.859
prior to the M 7.1.
Take it away, Ken.

00:01:17.748 --> 00:01:21.885
[silence]

00:01:21.910 --> 00:01:27.976
- So this is a story of serendipity
and teamwork and being ready.

00:01:28.000 --> 00:01:31.360
I won’t read all the names here,
but I just want to say how much I

00:01:31.360 --> 00:01:35.600
appreciate all of those, especially those
who got out into the field right away.

00:01:35.600 --> 00:01:38.320
This proves to be an example
of how important that can be

00:01:38.320 --> 00:01:43.416
when it comes to making
observations of perishable data.

00:01:43.440 --> 00:01:48.160
You’ve seen this figure many times.
What I’d like to do is zoom in

00:01:48.160 --> 00:01:52.960
progressively to that yellow square –
the intersection of the 6.4 rupture,

00:01:52.960 --> 00:01:59.040
shown by the lower-right red star –
the 6.4, and the 7.1 epicenter,

00:01:59.040 --> 00:02:01.360
shown by the
upper-left red star.

00:02:01.360 --> 00:02:05.440
So the intersection of those two main
ruptures is where we will zoom into.

00:02:05.440 --> 00:02:09.840
And we’ll be looking, in many cases,
on these figures at the best available

00:02:09.840 --> 00:02:13.120
compilation as of now.
This is a ongoing effort,

00:02:13.120 --> 00:02:16.800
as Tim mentioned, of this large group
effort represented by the SCEC poster

00:02:16.800 --> 00:02:20.800
by Kendrick et al., the submitted SRL
Ponti et al. paper, and the Dawson et al.

00:02:20.800 --> 00:02:25.200
to-be-submitted paper, as well as
slip distributions from DuRoss et al.

00:02:25.200 --> 00:02:29.256
So many, many people,
as Tim mentioned, worked on this.

00:02:29.280 --> 00:02:33.200
But I will focus my discussion
here today on the observations

00:02:33.200 --> 00:02:36.376
from prior to the 7.1.

00:02:36.400 --> 00:02:39.600
What I wanted to do in sort of
setting it up is talk about cross-fault

00:02:39.600 --> 00:02:43.200
interaction in the 1987 Superstition
Hills earthquake sequence, which,

00:02:43.200 --> 00:02:47.440
of course, really got off to a start
with the 6.2 Elmore Ranch event.

00:02:47.440 --> 00:02:52.480
You see in the A panel upper left,
left-lateral on a northeast-southwest-

00:02:52.480 --> 00:02:58.080
oriented fault. Followed 11.4 hours
later by Panel D, below left, and that is

00:02:58.080 --> 00:03:04.000
a right-lateral, northwest-southeast-
striking right-lateral fault, 6.6.

00:03:04.000 --> 00:03:07.016
And so you hopefully
know this case well.

00:03:07.040 --> 00:03:12.160
Just one figure from the paper we did
in GRL in ‘89 showing our stress

00:03:12.160 --> 00:03:16.616
change model. And, in particular,
how we explained the delay

00:03:16.640 --> 00:03:20.400
between those two events.
Well, how can this cross-fault

00:03:20.400 --> 00:03:24.480
triggering hypothesis help to
understand the 2019 sequence?

00:03:24.480 --> 00:03:29.096
I’ve used green to indicate yes –
good, and yellow, maybe.

00:03:29.120 --> 00:03:33.120
So the delay between foreshock and
main shock – I think that our model

00:03:33.120 --> 00:03:38.720
is one that works also for Ridgecrest.
It’s not required. There are alternative

00:03:38.720 --> 00:03:42.400
models that could also explain
the 34.3-hour delay that was

00:03:42.400 --> 00:03:45.896
seen in Ridgecrest
between the 6.4 and 7.1.

00:03:45.920 --> 00:03:49.600
The rupture propagation, on the other
hand, I give that a maybe in yellow

00:03:49.600 --> 00:03:55.920
color here because it’s interesting that,
when we do flip Superstition to look at

00:03:55.920 --> 00:04:00.560
it as – and try to explain Ridgecrest
in a similar way, remember that

00:04:00.560 --> 00:04:07.096
Ridgecrest looked like this – a capital L
shape with the epicenter of the 6.4 here.

00:04:07.120 --> 00:04:12.456
And this is after the 6.4 but
before the 7.1, it looked like this.

00:04:12.480 --> 00:04:16.856
And so, what we’re going to look at
is surface rupture on this leg of the L.

00:04:16.880 --> 00:04:21.360
Okay, but, in general, at the time,
from seismology, we knew the epicenter

00:04:21.360 --> 00:04:24.720
was close to where they intersected.
And so we were presuming that

00:04:24.720 --> 00:04:28.160
there was right-lateral slip here,
left-lateral slip here.

00:04:28.160 --> 00:04:32.400
Well, now we flip that. And here
we have 87 Superstition Hills,

00:04:32.400 --> 00:04:35.920
where the 6.2 was like this on the
Elmore Ranch fault, and the 6.6

00:04:35.920 --> 00:04:38.480
like this on the Superstition
Hill main rupture.

00:04:38.480 --> 00:04:43.040
So that’s all I’ve done here is just flip it.
And so you see the problem now,

00:04:43.040 --> 00:04:47.656
in the lower right, with the three
question marks in the yellow oval.

00:04:47.680 --> 00:04:51.440
In the Ridgecrest 7.1,
the rupture propagated

00:04:51.440 --> 00:04:56.616
from its epicenter of the 7.1
toward the southeast.

00:04:56.640 --> 00:05:00.640
And then, somehow – I’m not quite
sure why – it kept on going beyond

00:05:00.640 --> 00:05:04.720
the intersection with the cross-fault.
So just simply from the geometry

00:05:04.720 --> 00:05:09.280
on Ridgecrest, the 6.4 left-lateral
would have unclamped this portion

00:05:09.280 --> 00:05:13.840
where the 7.1 occurred. But it would
have also clamped this portion.

00:05:13.840 --> 00:05:17.896
So the rupture kept going.
That’s a surprise.

00:05:17.920 --> 00:05:20.880
Now, in terms of the delay,
the equation in the lower right

00:05:20.880 --> 00:05:24.080
of this slide is straight
from our ‘89 paper.

00:05:24.080 --> 00:05:28.880
The delay may well be for a similar
or same reason it would work –

00:05:28.880 --> 00:05:32.157
this mechanism would
work to explain it.

00:05:33.840 --> 00:05:37.680
Okay. So, in terms of addressing
the complexity, as Sarah said,

00:05:37.680 --> 00:05:42.720
this session is about trying to get a
handle on fault rupture complexity.

00:05:42.720 --> 00:05:46.720
Steve gave some examples, too,
of these incredibly complex earthquakes

00:05:46.720 --> 00:05:49.680
that we’ve seen in recent time.
And I would argue that’s partly

00:05:49.680 --> 00:05:53.760
because of all these improvements
in our capability to observe –

00:05:53.760 --> 00:05:56.216
as Steve mentioned –
all the new techniques.

00:05:56.240 --> 00:06:01.256
Among them, seismicity relocation
methods have gotten so much better.

00:06:01.280 --> 00:06:04.480
I’ll show some examples from the
work of David Shelly, but of course,

00:06:04.480 --> 00:06:08.856
you’ve seen the Ross et al. Science
paper and Hauksson’s talk at AGU.

00:06:08.880 --> 00:06:13.336
Other examples where seismicity
relocation is really sharpening

00:06:13.360 --> 00:06:16.856
our knowledge of
rupture complexity.

00:06:16.880 --> 00:06:21.200
So, in this case, what you see in blue
is all of the seismicity associated

00:06:21.200 --> 00:06:25.200
with the 6.4 and the hours
after it prior to the 7.1.

00:06:25.200 --> 00:06:30.880
And in red is the 7.1 and after it.
And so what you’ll see here as we,

00:06:30.880 --> 00:06:38.480
again, zoom in toward that
yellow square, in the 6.4 – whoops.

00:06:38.480 --> 00:06:41.416
I got ahead of myself there.

00:06:41.440 --> 00:06:45.280
I wanted to mention that there is some
complexity and some cross-faulting

00:06:45.280 --> 00:06:48.880
between the time of the 6.4 and the 7.1
that is important to understand better.

00:06:48.880 --> 00:06:51.440
I’m not going to get into
that in detail here today.

00:06:51.440 --> 00:06:57.200
But there was a 5.4 about
16 hours before the 7.1.

00:06:57.200 --> 00:06:59.440
And that appears to
have been on a cross-fault.

00:06:59.440 --> 00:07:05.440
There was also this 5.0 that – those of
us that were in Ridgecrest for the 7.1,

00:07:05.440 --> 00:07:08.320
we also were in the earthquake
clearinghouse meeting, and we also

00:07:08.320 --> 00:07:12.160
felt this 5.0 that was just
a few minutes before the 7.1.

00:07:12.160 --> 00:07:14.960
It may have been on a cross-fault,
but we need to really zoom in

00:07:14.960 --> 00:07:18.216
and examine that
more closely.

00:07:18.240 --> 00:07:23.520
Okay, so the question of whether the
6.4 simply triggered the 7.1 in this

00:07:23.520 --> 00:07:27.120
simplified cross-fault triggering
mechanism like Superstition, it’s really

00:07:27.120 --> 00:07:30.720
more nuanced. It’s more subtle.
It’s much more interesting than that.

00:07:30.720 --> 00:07:36.056
And there’s actually a series of
cross-faults that broke going from that

00:07:36.080 --> 00:07:41.440
apex of the L toward the northwest,
migrating in a way – not cleanly and

00:07:41.440 --> 00:07:47.496
simply, but in a more complex way –
toward the eventual epicenter of the 7.1.

00:07:47.520 --> 00:07:51.120
All of this, by the way – the Southern
California Seismic Network data may

00:07:51.120 --> 00:07:55.600
be cited using this DOI number.
And I hope you’ll cite these other works

00:07:55.600 --> 00:07:59.976
as well as the SRL data mine article
by David Shelly that’s just come out.

00:08:00.000 --> 00:08:03.760
Okay, so now these dots are moving.
That’s because these are showing

00:08:03.760 --> 00:08:08.856
you three GPS stations from
the GAGE NOTA network.

00:08:08.880 --> 00:08:11.200
And nothing’s happening much,
and then the earthquake starts.

00:08:11.200 --> 00:08:16.136
And notice this left-hand panel.
This is P595 at 5 hertz.

00:08:16.160 --> 00:08:19.120
That station started
moving toward the southeast.

00:08:19.120 --> 00:08:23.176
That’s what we would expect in
this big earthquake – the 7.1.

00:08:23.200 --> 00:08:26.480
That station, which is in the southeast
quadrant, would start moving towards

00:08:26.480 --> 00:08:30.800
the southeast. But, unexpectedly then,
after about just a few seconds,

00:08:30.800 --> 00:08:34.320
it takes off towards the northeast.
And that’s not explained well by

00:08:34.320 --> 00:08:37.520
any of the models I’ve seen yet.
Maybe Julian Lozos is going to

00:08:37.520 --> 00:08:40.480
explain that with his latest model
when he gets up here next.

00:08:40.480 --> 00:08:44.536
But I would say that this is indicating
some rupture complexity in the 7.1

00:08:44.560 --> 00:08:49.576
that may involve cross-faulting
during the 7.1 as well.

00:08:49.600 --> 00:08:53.920
Okay, so let’s leave that aside for the
moment and keep blasting through.

00:08:53.920 --> 00:08:59.200
Let’s back up now to the 6.4 and look
right here – I’ll bring the cursor up.

00:08:59.200 --> 00:09:04.936
Nope. Look at the –
what’s in the blue circle there.

00:09:04.960 --> 00:09:10.400
On a northwest-southeast-oriented
structure, that’s where the seismicity

00:09:10.400 --> 00:09:13.896
began prior to the 6.4.

00:09:13.920 --> 00:09:18.720
So now, back to this idea that we
might be able to apply cross-fault

00:09:18.720 --> 00:09:24.400
triggering to the 2019 sequence, well,
that indicates that this started on a

00:09:24.400 --> 00:09:30.560
northwest-southeast right near the 6.4.
And then activity migrated, then,

00:09:30.560 --> 00:09:35.360
onto the northeast-southwest fault.
So, again, this is happening at multiple

00:09:35.360 --> 00:09:39.120
scales, including small scale,
different time intervals – this is

00:09:39.120 --> 00:09:43.760
over 30 minutes prior to the 6.4
that you had a magnitude 4.

00:09:43.760 --> 00:09:48.160
But this is a – sort of an escalation
and a migration of seismicity

00:09:48.160 --> 00:09:51.920
from one fault to another.
In this case, to start the whole thing,

00:09:51.920 --> 00:09:54.800
it actually began on a northwest-
southeast-oriented fault.

00:09:54.800 --> 00:09:59.600
Rather surprising, but now that we see
these details in retrospect, I just find it

00:09:59.600 --> 00:10:04.184
fascinating that the cross-faulting
was occurring at different scales.

00:10:06.720 --> 00:10:07.976
Okay.

00:10:08.000 --> 00:10:11.336
I’m going to swing into a quick
play-by-play mode here.

00:10:11.360 --> 00:10:16.696
On the 4th of July, Janis Hernandez and
I went straight up to the field together

00:10:16.720 --> 00:10:21.896
and got out there and started taking
photographs at about 7:00 p.m.

00:10:21.920 --> 00:10:24.560
And here’s what we came across
when we got to the field.

00:10:24.560 --> 00:10:29.760
A wide zone – about 165 meters
wide in total – and measured across

00:10:29.760 --> 00:10:33.040
all those strands and summed up,
we got about 50 centimeters of

00:10:33.040 --> 00:10:37.336
left-lateral slip on a north-
northeast-striking zone.

00:10:37.360 --> 00:10:41.440
And there were two main zones of
rupture across Highway 178 that

00:10:41.440 --> 00:10:45.840
we initially mapped. And Janis had
with her the spray paint, and she was

00:10:45.840 --> 00:10:52.616
able to lay down some straight stripes
to check for any afterslip later.

00:10:52.640 --> 00:10:55.920
And we were joined in the field
by Kelly Blake from the Navy,

00:10:55.920 --> 00:10:59.360
Ben Brooks, Todd Ericksen
from USGS, who helped us

00:10:59.360 --> 00:11:02.000
make all these observations.
And you’ll see some photographs

00:11:02.000 --> 00:11:04.880
down on the right just indicating
that these were our first observations

00:11:04.880 --> 00:11:07.760
of surface fault rupture.
We could see that it was big and

00:11:07.760 --> 00:11:10.400
interesting, and we wanted to
get back out there the next day.

00:11:10.400 --> 00:11:15.120
So we set everything up for initial
aerial reconnaissance that took place

00:11:15.120 --> 00:11:19.200
the afternoon of July 5th.
This is our flight plan on the left

00:11:19.200 --> 00:11:22.080
that we provided to the Navy,
and they approved it.

00:11:22.080 --> 00:11:25.920
We needed to take this helicopter
inside the Navy base, and this is when

00:11:25.920 --> 00:11:31.040
we found out how tight security is at
Naval Air Weapons Station China Lake.

00:11:31.040 --> 00:11:32.960
But we were helped by
them tremendously,

00:11:32.960 --> 00:11:35.816
and Kelly Blake really
opened the doors for us.

00:11:35.840 --> 00:11:38.720
Okay, so this is that L-shaped
pattern that I was talking about.

00:11:38.720 --> 00:11:42.240
So we wanted to go out
and recon both legs of that L.

00:11:42.240 --> 00:11:44.960
In the end, you can see from the
right panel, which is what we

00:11:44.960 --> 00:11:48.720
actually got to do, we really had to
cut it short on recon on the

00:11:48.720 --> 00:11:52.240
northwest-southeast segment.
We were able to go across it a couple of

00:11:52.240 --> 00:11:57.600
times, but because of the short duration
of this California Highway Patrol flight,

00:11:57.600 --> 00:12:00.776
we didn’t get to do everything
that we really intended to do.

00:12:00.800 --> 00:12:03.200
But we were able to get
a lot done in a hurry.

00:12:03.200 --> 00:12:07.680
And so, on the right, what you’re
looking at here is our ground base

00:12:07.680 --> 00:12:09.840
reconnaissance from
the morning of the 5th.

00:12:09.840 --> 00:12:14.080
So, as we were waiting for the CHP
helicopter to come in in the afternoon,

00:12:14.080 --> 00:12:19.280
we got out there to – on the ground
into what we call Skytop Valley.

00:12:19.280 --> 00:12:24.960
So, on the right-hand panel, that’s got
a dashed, heavy orange line around it.

00:12:24.960 --> 00:12:30.160
So that’s up near the epicentral area.
And, in that Skytop Valley, Janis,

00:12:30.160 --> 00:12:34.400
Kelly, and I and some Navy escorts
and unexploded ordnance experts

00:12:34.400 --> 00:12:38.080
came with us, and we identified some
places where there was minor cracking,

00:12:38.080 --> 00:12:41.360
but we really couldn’t map
surface rupture on any of those.

00:12:41.360 --> 00:12:44.320
But it was really during the aerial
reconnaissance in the afternoon,

00:12:44.320 --> 00:12:48.800
we flew out to Highway 178, and from
there, recon-ed to the northeast and

00:12:48.800 --> 00:12:52.880
southwest and got air photos, which
turned out later to be very useful.

00:12:52.880 --> 00:12:58.136
And I’ll spend some time emphasizing
that part of the air recon that we did.

00:12:58.160 --> 00:13:01.360
So the actual duration of the aerial
reconnaissance – we took off

00:13:01.360 --> 00:13:05.440
and landed at the Navy airfield
inside the base, and it really lasted

00:13:05.440 --> 00:13:11.280
only from 1:40 p.m. to 2:50 p.m.
But I want to start with the ground

00:13:11.280 --> 00:13:17.336
recon in Skytop Valley in the upper
right – that orange dashed line.

00:13:17.360 --> 00:13:22.080
So that’s where we were able to
first make some observations.

00:13:22.080 --> 00:13:26.216
And then – nothing large enough
to measure, though, like I said.

00:13:26.240 --> 00:13:33.120
We then met the Navy and took off and
began – we moved from Highway 178

00:13:33.120 --> 00:13:36.720
toward the epicenter, then we went all
the way down to the southwest end,

00:13:36.720 --> 00:13:39.840
back up, and continued to try to
chase down – figure out where

00:13:39.840 --> 00:13:42.696
the rupture was within
the yellow square there.

00:13:42.720 --> 00:13:47.840
So, along the way, I captured stereo
overlap photos of most of the fault

00:13:47.840 --> 00:13:54.080
rupture that we identified from the 6.4.
We also then took single images –

00:13:54.080 --> 00:13:56.560
as Janis and I were flying,
we were just taking pictures when

00:13:56.560 --> 00:14:00.160
we saw cracks on the ground.
And so some of those are single,

00:14:00.160 --> 00:14:03.840
not stereo overlapped, and I’ll show
how some of that is also very

00:14:03.840 --> 00:14:08.616
interesting in terms of figuring out
exactly what broke in the earthquake.

00:14:08.640 --> 00:14:11.360
So, but yeah, it was all said
and done before 3:00 p.m.

00:14:11.360 --> 00:14:15.280
And then after that is when we went
in and conducted the emergency

00:14:15.280 --> 00:14:19.907
operations center briefings, first with
the Navy and then with the city.

00:14:21.120 --> 00:14:25.920
Okay, so one of the things that we
were very interested in, and Janis and

00:14:25.920 --> 00:14:28.960
I finally had an opportunity to just
sit down and sort of go through

00:14:28.960 --> 00:14:31.120
what we have from
this initial air recon.

00:14:31.120 --> 00:14:34.856
Just the other day, we actually did this
for the first time, believe it or not.

00:14:34.880 --> 00:14:38.376
But things started
happening fast, I got to say.

00:14:38.400 --> 00:14:43.840
So these are circles of places where,
either on the ground or in the air,

00:14:43.840 --> 00:14:50.320
on July 5th, we either drove by or flew
over, and we think we would have seen

00:14:50.320 --> 00:14:56.080
surface rupture if it had been there.
And so we’re looking for absence

00:14:56.080 --> 00:14:59.680
of rupture observed.
And so, in the Skytop Valley –

00:14:59.680 --> 00:15:03.280
that’s where we’re zoomed in here,
and these are showing you places from

00:15:03.280 --> 00:15:06.480
the compilations that I mentioned –
Kendrick et al., Ponti et al.,

00:15:06.480 --> 00:15:10.480
Dawson et al., DuRoss et al. – those
compilations that are shown here in

00:15:10.480 --> 00:15:15.520
the purple and orange lines,
for example, later, on the basis

00:15:15.520 --> 00:15:21.040
of satellite imagery and ground
mapping, ruptures were mapped later.

00:15:21.040 --> 00:15:25.600
And so these are places circled where
I think, and Janis and I agree,

00:15:25.600 --> 00:15:30.240
that we would have seen rupture
on the 5th if it had been there.

00:15:30.240 --> 00:15:35.920
And so we’re trying to be quantitative
about where we did not observe rupture.

00:15:35.920 --> 00:15:40.160
So I’ll show some more circles like this
on some other upcoming figures.

00:15:40.160 --> 00:15:46.136
The orange circle there is one that
Janis visited again in the field,

00:15:46.160 --> 00:15:50.056
not only on the morning of the
5th of July, but then again later.

00:15:50.080 --> 00:15:53.520
And, at the time when she visited
again later, she was quite certain

00:15:53.520 --> 00:15:58.320
that there was more offset on
those fractures than on the 5th.

00:15:58.320 --> 00:16:00.936
And so that’s interesting.
That’s what we’re looking for

00:16:00.960 --> 00:16:04.960
is evidence that some of these
breaks occurred in the 6.4

00:16:04.960 --> 00:16:08.400
and then re-ruptured in the 7.1.
And I’ll come to it.

00:16:08.400 --> 00:16:12.800
We have some other examples where
we know that rupture occurred in both

00:16:12.800 --> 00:16:19.016
events and that the slip from the
6.4 was increased by the 7.1.

00:16:19.040 --> 00:16:24.320
Okay, so back to the initial plan
for our day – the red pattern and

00:16:24.320 --> 00:16:26.640
then the yellow intersection
area on the left there.

00:16:26.640 --> 00:16:30.936
And then we’re zooming into that
yellow square at the intersection.

00:16:30.960 --> 00:16:34.880
Again, that same overlay from all of
the compilation work that’s been done

00:16:34.880 --> 00:16:38.880
to include both the 6.4 rupture
and 7.1 rupture, but trying to

00:16:38.880 --> 00:16:41.816
separate out what
happened in the 6.4.

00:16:41.840 --> 00:16:47.200
So here you see, zoomed in on it, these
are the flight tracks as well as the GPS –

00:16:47.200 --> 00:16:50.696
the GPS tracks from the
ground recon in the morning.

00:16:50.720 --> 00:16:54.480
In the upper center of that
right-hand panel is part of

00:16:54.480 --> 00:16:57.120
what I’m going to refer to
as the Bunker Complex.

00:16:57.120 --> 00:17:01.896
This is out in the clip area
at the NAWS China Lake.

00:17:01.920 --> 00:17:04.800
And so – and then there’s the
red pond, which also serves as

00:17:04.800 --> 00:17:11.360
reference in the lower left of that.
But the salty crust on the ground

00:17:11.360 --> 00:17:14.080
surface there, that’s
called Salt Wells Valley.

00:17:14.080 --> 00:17:17.736
And so the northeast-southwest-
oriented main surface rupture

00:17:17.760 --> 00:17:22.400
and the intersection with what
would become the 7.1 main rupture

00:17:22.400 --> 00:17:26.000
is what we’re looking at here.
And you can see how we really

00:17:26.000 --> 00:17:29.120
flew around that intersection area
quite a bit – Janis and I did

00:17:29.120 --> 00:17:33.200
on the afternoon of the 5th.
And so this is where, as we were

00:17:33.200 --> 00:17:36.400
coming toward the northeast,
we could see the rupture

00:17:36.400 --> 00:17:39.120
very prominently on the
ground and take photos of it.

00:17:39.120 --> 00:17:43.416
And then it just stopped.
And so what’s interesting is how,

00:17:43.440 --> 00:17:49.920
in the 6.4, that eventual intersection
with the 7.1 was very important,

00:17:49.920 --> 00:17:52.240
apparently, in just
terminating the rupture.

00:17:52.240 --> 00:17:55.200
And yet also there’s all this
complexity right there.

00:17:55.200 --> 00:17:59.096
So what we’re going to take a look
at further as we zoom in is that,

00:17:59.120 --> 00:18:03.280
in that same intersection area, there
was rupture on the northwest-southeast-

00:18:03.280 --> 00:18:09.816
oriented fault, but in the 6.4,
as well as re-rupture in the 7.1.

00:18:09.840 --> 00:18:13.040
And this may be also the best point to
mention, there are a couple of other

00:18:13.040 --> 00:18:16.880
single photos up in the Bunker
Complex and also northwest of the

00:18:16.880 --> 00:18:23.040
main strand of the cross-fault that are
single shots and triplets over there.

00:18:23.040 --> 00:18:26.080
There is rupture in those,
but that’s kind of a sidelight issue.

00:18:26.080 --> 00:18:27.600
I’m not even going to
dwell on it anymore.

00:18:27.600 --> 00:18:32.160
But, in places where, in the main
compilation, we still don’t have faults

00:18:32.160 --> 00:18:36.640
mapped, there were definitely ruptures
in there as well that we need to get

00:18:36.640 --> 00:18:39.760
onto the main compilation.
But this is all part of the sort of

00:18:39.760 --> 00:18:44.400
work-in-progress. One further thing
to mention is that we did fly over –

00:18:44.400 --> 00:18:48.160
as we were returning to the base,
we flew over what became later

00:18:48.160 --> 00:18:52.400
in the 7.1 the main strand ruptures
farther to the northwest.

00:18:52.400 --> 00:18:56.720
And we were higher and moving faster
at this time because the CHP needed

00:18:56.720 --> 00:19:00.160
to return, and we ended up not being
able to refuel and get back out there.

00:19:00.160 --> 00:19:03.840
But this shows you those two red
circles in the right panel that we did

00:19:03.840 --> 00:19:08.400
fly over areas that later then had large
surface rupture, and we did not observe

00:19:08.400 --> 00:19:12.056
surface rupture at those places
on the afternoon of the 5th.

00:19:12.080 --> 00:19:17.416
And this would have been about 2:45
on the afternoon of the 5th of July.

00:19:17.440 --> 00:19:20.080
Now, I’m going to go through
several slides quickly here.

00:19:20.080 --> 00:19:24.080
This is back at the intersection,
and you see here those features

00:19:24.080 --> 00:19:26.320
we’ve described – the Bunker
Complex in the upper right,

00:19:26.320 --> 00:19:30.296
Salt Wells Valley in the lower left,
and the red pond for reference.

00:19:30.320 --> 00:19:34.056
So this is the full compilation
of the 6.4 and the 7.1.

00:19:34.080 --> 00:19:37.200
And now we’re going to look
at the observations that are in

00:19:37.200 --> 00:19:40.160
the latest version of the
compilation from Ponti et al.

00:19:40.160 --> 00:19:43.200
And this is places where people
have been to on the ground since

00:19:43.200 --> 00:19:46.320
then to make observations.
So you can see that, even though

00:19:46.320 --> 00:19:50.480
a lot of observations have been made,
many of the key areas, there has

00:19:50.480 --> 00:19:53.680
not been a – we haven’t yet been
able to get back in there and

00:19:53.680 --> 00:19:57.016
do a concerted effort to really
do detailed mapping to look for,

00:19:57.040 --> 00:20:01.736
for example, cross-cutting relationships
of the surface rupture in here.

00:20:01.760 --> 00:20:06.560
Again, here are the photos that we got.
And the yellow square zoomed in.

00:20:06.560 --> 00:20:10.800
Just to get you that reference again.
And so this is an area of great interest,

00:20:10.800 --> 00:20:13.736
of course, for understanding
fault interaction.

00:20:13.760 --> 00:20:16.160
So here, again, and with
a COSI-Corr overlay,

00:20:16.160 --> 00:20:20.720
this is including the 6.4 and 7.1
ruptures and the photo centers from the

00:20:20.720 --> 00:20:27.336
7.5 air recon and the flight tracks from
that air recon at the intersection area.

00:20:27.360 --> 00:20:30.720
You see the stereo overlap
photos along the main rupture.

00:20:30.720 --> 00:20:34.560
You see where the main rupture stopped
at the eventual 7.1 main rupture.

00:20:34.560 --> 00:20:38.800
And then those single photos that
I mentioned – 0073, practically in the

00:20:38.800 --> 00:20:41.360
center of the photo, is one
that I’ll show in more detail,

00:20:41.360 --> 00:20:45.040
and we’ll zoom in on that.
The other ones on the left-hand side

00:20:45.040 --> 00:20:48.240
that I’m not going to show today that
do have rupture in them, we haven’t

00:20:48.240 --> 00:20:51.680
made sense out of that yet, but it’s
probably some additional rupture on

00:20:51.680 --> 00:20:55.040
northwest-southeast and northeast-
southwest-oriented faults.

00:20:55.040 --> 00:20:58.640
But over off the main ruptures –
these were not captured in any

00:20:58.640 --> 00:21:01.416
of the imagery by Milliner
or Donnellan or anybody.

00:21:01.440 --> 00:21:04.456
All right. So back again
and zooming in.

00:21:04.480 --> 00:21:08.080
Oops. I was going backwards instead
of forwards there. Forgive me.

00:21:08.080 --> 00:21:10.560
So this is now with
some interpreted line work

00:21:10.560 --> 00:21:13.176
from Alex Morelan
from CGS.

00:21:13.200 --> 00:21:17.440
And now that same view but without
the COSI-Corr just to show you the

00:21:17.440 --> 00:21:21.680
background imagery [phone ringing] –
sorry. Time is up already.

00:21:24.800 --> 00:21:28.160
Okay. I’ll keep moving.
So here we go. [chuckles]

00:21:28.160 --> 00:21:34.240
Yeah. Sorry. 0073 that I mentioned
shows very interestingly that we had

00:21:34.240 --> 00:21:38.880
northwest-southeast-oriented rupture
associated with the 6.4, very clearly.

00:21:38.880 --> 00:21:41.600
And here we zoom in
onto that frame of imagery.

00:21:41.600 --> 00:21:45.656
And the little red arrows in this
left panel show you that. Okay?

00:21:45.680 --> 00:21:49.600
So, as we zoom in, we see that, yep,
northwest-southeast-oriented,

00:21:49.600 --> 00:21:55.096
right-lateral rupture – because we see
that stepping en echelon left-lateral.

00:21:55.120 --> 00:22:00.720
And then here it is after the 7.1.
This is from a shot that Gordon Seitz

00:22:00.720 --> 00:22:04.640
took looking back toward the
southeast along the main 7.1 rupture

00:22:04.640 --> 00:22:08.320
but after the 7.1.
So this is the same place shown

00:22:08.320 --> 00:22:11.760
with the yellow bracket here where
we clearly have rupture after the

00:22:11.760 --> 00:22:17.435
6.4 northwest of the
intersection of the cross-fault.

00:22:18.720 --> 00:22:21.840
So that’s a pretty amazing observation.
And here are more photos of that same

00:22:21.840 --> 00:22:25.360
intersection, just differently oriented.
And this is new material.

00:22:25.360 --> 00:22:29.120
A lot of this since AGU, so even
those of you that saw that talk.

00:22:29.120 --> 00:22:33.256
Again, also this is new from Alex
Morelan who took photos that I took

00:22:33.280 --> 00:22:37.360
on the 5th and that Gordon and I took
on the 12th of the intersection area.

00:22:37.360 --> 00:22:42.080
We used COSI-Corr to show that the –
not only did the northwest-southeast-

00:22:42.080 --> 00:22:47.360
oriented fault re-rupture in the 7.1, so
did the northeast-southwest cross-fault.

00:22:47.360 --> 00:22:50.320
It also re-ruptured in
the 7.1 at this location.

00:22:50.320 --> 00:22:53.120
Now, Ben Brooks has already
shown from his data that,

00:22:53.120 --> 00:22:57.120
down at Highway 178, first of all,
they re-measured those afterslip lines,

00:22:57.120 --> 00:23:01.016
and those did not show slip.
And the GPS didn’t either.

00:23:01.040 --> 00:23:05.520
But up near the intersection, in the 7.1,
there was additional left-lateral slip.

00:23:05.520 --> 00:23:09.040
Again, this a hot new result.
And here, just as we’re trying to

00:23:09.040 --> 00:23:12.400
summarize where we did not see slip,
there are a bunch of places where

00:23:12.400 --> 00:23:16.640
Janis and I believe we would have seen
slip on the 5th if it had been there.

00:23:16.640 --> 00:23:19.520
So, again, we’re trying to
be quantitative about the

00:23:19.520 --> 00:23:22.400
absence of observations.
And those blue circles show you

00:23:22.400 --> 00:23:25.919
the off-main ruptures
that we talked about earlier.

00:23:28.080 --> 00:23:32.080
Sorry to have taken too much time,
but I will just go briefly through the –

00:23:32.080 --> 00:23:34.880
in terms of cross-fault interaction,
it’s reminiscent of ‘87

00:23:34.880 --> 00:23:38.560
Superstition Hills. That fluid
diffusion is a good explanation,

00:23:38.560 --> 00:23:41.840
we think, also, for Ridgecrest.
There are issues with the fault

00:23:41.840 --> 00:23:46.000
interaction in general and rupture
in the 7.1 of Ridgecrest going through

00:23:46.000 --> 00:23:49.120
the intersection area
and continuing southeast.

00:23:49.120 --> 00:23:52.536
Not well understood.
Very interesting.

00:23:52.560 --> 00:23:56.800
And ‘87 appears simpler.
Maybe that’s because the data now

00:23:56.800 --> 00:24:01.520
are so much better and also because
the events were bigger, so more to study.

00:24:01.520 --> 00:24:06.923
And so also we had no surface
rupture observations in ‘87 after the –

00:24:06.923 --> 00:24:10.376
after the Elmore Ranch but before
the Superstition main shock.

00:24:10.400 --> 00:24:14.056
So here we have these great
observations. That’s really helpful.

00:24:14.080 --> 00:24:17.360
And we’re just working our way
through continuing to study this

00:24:17.360 --> 00:24:21.440
example of fault interaction.
We find it very significant

00:24:21.440 --> 00:24:27.096
that we have this definitive
evidence now of re-rupture

00:24:27.120 --> 00:24:29.816
on both the cross-fault
and main rupture.

00:24:29.840 --> 00:24:33.336
Both of them broke in
the 6.4 and again in the 7.1.

00:24:33.360 --> 00:24:36.880
And it really underscores the
importance of doing rapid-response

00:24:36.880 --> 00:24:40.240
aerial reconnaissance and
getting out there quickly.

00:24:42.480 --> 00:24:45.736
And I know I don’t have
any time to get into last point.

00:24:45.760 --> 00:24:47.940
So I’ll do it very fast.

00:24:47.965 --> 00:24:50.480
[laughter]

00:24:50.480 --> 00:24:53.680
Sorry about this, so yeah, this is –
I didn’t have a slide when

00:24:53.680 --> 00:24:57.416
I talked about this at AGU,
so I created a slide just for you.

00:24:57.440 --> 00:25:01.680
We looked at the Gulf of California,
Salton Trough, Eastern California

00:25:01.680 --> 00:25:04.560
Shear Zone, Walker Lane,
and the possible future of that.

00:25:04.560 --> 00:25:07.920
It’s been written about
in Wired and so forth.

00:25:07.920 --> 00:25:11.280
But you know the story.
A lot of this has come from GPS.

00:25:11.280 --> 00:25:15.600
But really, basically, the upper-right
panel shows how, if you orient the

00:25:15.600 --> 00:25:19.280
whole system to show the plate
boundary structure, it’s really kind of

00:25:19.280 --> 00:25:22.480
weird that you’ve got the big bend,
and it really would be simpler to just

00:25:22.480 --> 00:25:26.320
take a straight shot all the way through,
not only to Cape Blanco, but maybe

00:25:26.320 --> 00:25:30.456
even eventually to the
Haida Gwaii Fairweather System.

00:25:30.480 --> 00:25:34.800
Furthermore, where we see cross-fault
interaction in the northwest end of

00:25:34.800 --> 00:25:38.800
El Mayor-Cucapah and Superstition and
Landers-Big Bear and in Ridgecrest –

00:25:38.800 --> 00:25:42.000
also Chalfant Valley and 9-Mile
Ranch examples, it’s really along

00:25:42.000 --> 00:25:44.960
this alignment where we’re between
the right-lateral shear of the

00:25:44.960 --> 00:25:47.760
San Andreas and the extension
of the Basin and Range.

00:25:47.760 --> 00:25:52.240
So these red circles in the center upper
panel are where I think we may have a

00:25:52.240 --> 00:25:55.520
tendency to have cross-fault interaction,
and that may well be where we have

00:25:55.520 --> 00:25:58.936
detachments and metamorphic
core complex formation.

00:25:58.960 --> 00:26:01.200
Sorry to take too much time.
Thank you.

00:26:01.200 --> 00:26:04.800
[Applause]

00:26:05.699 --> 00:26:07.920
- We have some time built in
for discussion at the end,

00:26:07.920 --> 00:26:11.896
so I’m going to hold off the
questions and introduce Julian,

00:26:11.920 --> 00:26:17.280
who will be – oh, yeah.
Please pull yours up – who will

00:26:17.280 --> 00:26:21.520
be talking about dynamic rupture
simulations of the July 2019 Ridgecrest

00:26:21.520 --> 00:26:25.718
earthquakes and implications
for other cross-faults.

00:26:28.885 --> 00:26:36.800
[Silence]

00:26:36.800 --> 00:26:41.040
All right. So thank you all for inviting
me. Thank you all for coming.

00:26:41.040 --> 00:26:44.240
Hopefully, because this started
as an AGU talk that I added

00:26:44.240 --> 00:26:47.496
a few more slides to, I will
make up for some of this time.

00:26:47.520 --> 00:26:52.080
But anyway, so I’m here from SoCal to
talk about some SoCal earthquakes, but

00:26:52.080 --> 00:26:56.160
I think they do have, as we’ve already
heard, larger-scale implementations.

00:26:56.160 --> 00:26:59.840
And so the title I was given was
Dynamic Rupture Simulations of

00:26:59.840 --> 00:27:03.600
the 2019 Ridgecrest Earthquakes and
Implications for Other Cross-Faults.

00:27:03.600 --> 00:27:07.200
And this is work I’m doing with
Ruth here – Ruth Harris, and this

00:27:07.200 --> 00:27:09.680
photo really has very little
to do with the actual talk.

00:27:09.680 --> 00:27:12.885
It’s just, as a dynamic rupture modeler,
I almost never get to put a cool

00:27:12.885 --> 00:27:18.296
field photo in a talk.
So here’s a right-laterally offset bush.

00:27:18.320 --> 00:27:21.600
But anyway [laughs] – which was –
I thought was pretty cool.

00:27:21.600 --> 00:27:26.960
But anyway, so I guess I don’t need to
give a lot of introduction to Ridgecrest

00:27:26.960 --> 00:27:30.480
at this point, both from the last talk
and because I know that there’s been

00:27:30.480 --> 00:27:33.840
a lot of energy focused across
many organizations, especially

00:27:33.840 --> 00:27:37.920
the USGS, on these earthquakes.
I also know this is an older figure

00:27:37.920 --> 00:27:41.336
of the geometry by now,
so please be patient with me.

00:27:41.360 --> 00:27:45.840
But the super-duper basics.
So it started with a 6.4 on a –

00:27:45.840 --> 00:27:50.560
predominantly on a northeast-striking
left-lateral fault at 10:33 a.m.

00:27:50.560 --> 00:27:53.760
on July 4th. Many of us dynamic
rupture modelers were flying back

00:27:53.760 --> 00:27:57.120
from a dynamic rupture meeting
in Europe at the time and landed

00:27:57.120 --> 00:28:01.520
and were, like, oh, no.
But anyway, predominantly on –

00:28:01.520 --> 00:28:04.800
pointer went away.
Okay. Here it is.

00:28:04.800 --> 00:28:09.200
Predominantly on this
northeast-southwest-striking

00:28:09.200 --> 00:28:13.360
left-lateral feature.
And then, about 34 hours later

00:28:13.360 --> 00:28:18.480
on the following night, a 7.1
predominantly on this mostly

00:28:18.480 --> 00:28:23.576
orthogonal – like, this orthogonal
mostly right-lateral strike-slip structure.

00:28:23.600 --> 00:28:26.640
A long, complex rupture.
As we’ve just heard, this is a complex

00:28:26.640 --> 00:28:30.880
surface-rupturing sequence with
many cross-faults at different scales.

00:28:30.880 --> 00:28:33.360
And I’m going to apologize right
now that, you know, in dynamic

00:28:33.360 --> 00:28:36.560
rupture land, meshing all the
cross-faults can be complicated.

00:28:36.560 --> 00:28:40.000
But we also started working on this,
like, less than a week after

00:28:40.000 --> 00:28:43.280
the earthquake. So, for my purposes
today, I’m going to be focusing

00:28:43.280 --> 00:28:47.960
on two primary strands.
So the primary northeast- and

00:28:47.960 --> 00:28:51.200
southwest-striking left-lateral strand,
the primary northwest-southeast-

00:28:51.200 --> 00:28:54.240
striking right-lateral strand and
finding primary where the

00:28:54.240 --> 00:28:57.360
most consolidated surface rupture
with the largest amount of slip.

00:28:57.360 --> 00:29:00.160
So I’m not going to be able to
get to Ken’s question about

00:29:00.160 --> 00:29:02.160
some of the smaller
interactions yet.

00:29:02.160 --> 00:29:04.480
That is something we’re
hoping to do in the future.

00:29:04.480 --> 00:29:08.456
But, right now this
is a two-fault model.

00:29:08.480 --> 00:29:11.440
But it still, I think, is useful for
addressing some questions about what

00:29:11.440 --> 00:29:15.120
happened in Ridgecrest and also
some just general questions about

00:29:15.120 --> 00:29:19.416
cross-fault interactions.
So we had a couple of big questions.

00:29:19.440 --> 00:29:22.400
We had four – well, five,
but I’m not going to talk about

00:29:22.400 --> 00:29:25.360
the Garlock Fault today.
But our four questions about these

00:29:25.360 --> 00:29:30.000
cross-fault interactions in particular is,
first of all, what prevented the first

00:29:30.000 --> 00:29:32.400
earthquake from just rupturing
through all of this at once?

00:29:32.400 --> 00:29:36.160
Why didn’t we just get one larger
earthquake on two faults?

00:29:36.160 --> 00:29:41.256
Or on, I guess, however many faults
you want to quantify it to in our model?

00:29:41.280 --> 00:29:46.800
And, in that case, then, because they
didn’t rupture together, did the 6.4

00:29:46.800 --> 00:29:51.200
actually create the conditions necessary
for the 7.1 to even happen on that fault?

00:29:51.200 --> 00:29:55.040
Like, was the right-lateral fault, like,
not too close to being ready,

00:29:55.040 --> 00:29:59.440
and then the 6.4 really made it ready?
Or was it the kind of thing where the

00:29:59.440 --> 00:30:02.960
right-lateral fault was already
pretty close, and the 6.4 just

00:30:02.960 --> 00:30:06.320
accelerated the clock on that event
and made something that would have

00:30:06.320 --> 00:30:11.096
probably happened fairly soon
anyway happen even sooner?

00:30:11.120 --> 00:30:13.440
And so those are
sort of related questions.

00:30:13.440 --> 00:30:17.360
And then our other related questions
are, was the right-lateral fault

00:30:17.360 --> 00:30:20.000
significantly involved
in the 6.4 rupture?

00:30:20.000 --> 00:30:23.680
So I’m going to – I’m going to
verbally add the word “significantly”

00:30:23.680 --> 00:30:28.640
because, as we’ve seen with time and
further understanding of this rupture

00:30:28.640 --> 00:30:31.680
trace, everything is super complex,
and there’s a lot of structures and

00:30:31.680 --> 00:30:35.200
different orientations involved.
So, in this case, we’re really going to

00:30:35.200 --> 00:30:38.720
talk about the two main strands.
And then, as an extension of that,

00:30:38.720 --> 00:30:42.000
if the right-lateral fault was
substantially involved,

00:30:42.000 --> 00:30:46.560
or even a little bit involved,
what allowed that to be re-ruptured?

00:30:46.560 --> 00:30:49.840
Why was this able to rupture twice?
And I guess, now we can get at,

00:30:49.840 --> 00:30:54.136
why was the left-lateral fault able to
rupture or at least slip twice?

00:30:54.160 --> 00:30:56.400
And I guess, before I go on,
I’m going to distinguish now,

00:30:56.400 --> 00:31:01.760
when I say “rupture,” I really mean,
like, a propagating rupture front causing

00:31:01.760 --> 00:31:06.160
substantial stress drop and
actually really slipping at depth

00:31:06.160 --> 00:31:07.920
and at the surface.
Whereas, when I’m going to

00:31:07.920 --> 00:31:12.640
say “triggered slip,” I mean much
more of just a surface opening of

00:31:12.640 --> 00:31:17.440
small response – usually just centimeter
scale – in response to the stress changes

00:31:17.440 --> 00:31:20.800
and movement on nearby faults.
Because, as we’ll – as I’ll show you

00:31:20.800 --> 00:31:24.800
soon, we see both of those things in
our models, and I think that’s a point to

00:31:24.800 --> 00:31:30.216
look at in terms of interactions in cross-
faults in general, not just Ridgecrest.

00:31:30.240 --> 00:31:34.880
So anyway, sort of the biggest –
I guess the biggest debate on this was,

00:31:34.880 --> 00:31:38.320
was the 6.4 a conjugate or
orthogonal or two faults,

00:31:38.320 --> 00:31:41.016
or whatever you
want to call it, or not?

00:31:41.040 --> 00:31:45.416
And so, again, Ken’s talk was a really
great introduction to this question.

00:31:45.440 --> 00:31:49.440
So the aftershocks suggest
an L-shaped rupture.

00:31:49.440 --> 00:31:54.560
And that’ – so, on this left panel here
is the – it’s from the Southern

00:31:54.560 --> 00:31:57.920
California Seismic Network.
It’s all the aftershocks between –

00:31:57.920 --> 00:32:01.440
it’s from one minute after the 6.4
to one minute before the 7.1.

00:32:01.440 --> 00:32:04.616
This is very definitely an L.

00:32:04.640 --> 00:32:08.560
But, as Ken was talking about,
there’s not – I mean, there’s a little bit

00:32:08.560 --> 00:32:12.480
of right-lateral displacement,
but there’s not really evidence

00:32:12.480 --> 00:32:17.256
for major right-lateral surface
structures slipping in a – in a large way.

00:32:17.280 --> 00:32:21.440
And similarly, the satellite imagery –
I should have made this larger.

00:32:21.440 --> 00:32:24.240
So this is from Milliner and
Donnellan’s SRL paper

00:32:24.240 --> 00:32:28.960
that just came out.
And so these – the blue lines –

00:32:28.960 --> 00:32:32.880
actually, I think the green are the
ones where they’re really definitely –

00:32:32.880 --> 00:32:34.240
I’m forgetting which is
green and which is blue.

00:32:34.240 --> 00:32:37.520
There’s no scale on this figure.
Basically, the stuff that is marked on

00:32:37.520 --> 00:32:41.440
this figure are the things that they see –
so there’s one satellite image from

00:32:41.440 --> 00:32:45.680
between the 6.4 and the 7.1, and so
this is differenced between that

00:32:45.680 --> 00:32:50.800
and prior to the whole sequence.
And so, in the 6.4, they see a whole

00:32:50.800 --> 00:32:55.680
bunch of these northeast-southwest-
striking structures – maybe, like,

00:32:55.680 --> 00:33:00.240
a teeny-weeny bit in a couple
of places northwest-southeast.

00:33:00.240 --> 00:33:03.600
But, for the most part, this is actually
a pretty complex network of

00:33:03.600 --> 00:33:07.816
mostly parallel faults that are
not parallel to the 7.1 main shock.

00:33:07.840 --> 00:33:13.040
So, on the one hand, the seismicity said,
yeah, this was a conjugate rupture.

00:33:13.040 --> 00:33:16.800
Surface rupture pattern
says it probably wasn’t.

00:33:16.800 --> 00:33:19.040
And there are definitely a bunch
of different inversions that

00:33:19.040 --> 00:33:23.874
see a bunch of different things.
So, for one thing, you have the

00:33:23.874 --> 00:33:28.400
Ross et al. inversion that says that
this was absolutely an L-shaped

00:33:28.400 --> 00:33:32.880
rupture and that this sort of branch of
the right-lateral fault was involved.

00:33:32.880 --> 00:33:36.960
On the other hand, Chen Ji’s group
has a seismic inversion that says

00:33:36.960 --> 00:33:40.000
that it wasn’t involved.
There were geodetic inversions.

00:33:40.000 --> 00:33:44.696
I looked the longest at Gareth Funning’s
poster because he did some stuff where,

00:33:44.720 --> 00:33:49.896
when you told the right-lateral
fault to rupture, it did.

00:33:49.920 --> 00:33:52.000
Or it showed – it worked
with the geodetic inversion.

00:33:52.000 --> 00:33:54.560
But, when you didn’t
require it to, it didn’t.

00:33:54.560 --> 00:34:00.000
So correct me if that’s changed, Gareth,
but – so this is a – this is a question.

00:34:00.000 --> 00:34:03.360
Like, was this involved?
And so that’s one of the things

00:34:03.360 --> 00:34:06.160
we really want to get at
with dynamic rupture modeling.

00:34:06.160 --> 00:34:08.640
Because we don’t have to
assume it was or wasn’t.

00:34:08.640 --> 00:34:11.896
We just put the fault there
and see what happens.

00:34:11.920 --> 00:34:17.280
And so, in my case, what I’m trying
to match when I’m trying to see

00:34:17.280 --> 00:34:19.520
what happens is, you can
make a model do anything.

00:34:19.520 --> 00:34:22.720
I want to make it look
like what we actually saw.

00:34:22.720 --> 00:34:26.720
So, in this case, I’m trying to make
two earthquakes – one that is a 6.4

00:34:26.720 --> 00:34:31.200
and one that is a 7.1 – that match –
at this point, when we started

00:34:31.200 --> 00:34:34.240
working on this, pretty much
what we had was surface slip data.

00:34:34.240 --> 00:34:38.800
And so trying to match the surface –
the overall surface slip distribution

00:34:38.800 --> 00:34:41.200
and the magnitude.
So I’m not necessarily trying to do,

00:34:41.200 --> 00:34:44.560
like, every single site.
Because, so first of all, the resolution

00:34:44.560 --> 00:34:47.600
of our model might not
capture the exact sites.

00:34:47.600 --> 00:34:50.160
And also because each
site is so specific.

00:34:50.160 --> 00:34:52.960
So we’re trying to match the
overall shape of where there’s

00:34:52.960 --> 00:34:56.160
more or less slip and the overall
magnitude of slip in that area

00:34:56.160 --> 00:34:58.696
as well as the magnitude
of the earthquake.

00:34:58.720 --> 00:35:02.400
So – and, again, we have two faults.
So, if you see, in the figure with

00:35:02.400 --> 00:35:05.280
the aftershocks, the things that are lined
up in white – although I guess they’re

00:35:05.280 --> 00:35:09.576
hard to see under the aftershocks,
those are our primary fault geometries.

00:35:09.600 --> 00:35:15.120
And so, to do this, it’s a
fully dynamic 3D rupture model.

00:35:15.120 --> 00:35:18.080
I’m using FaultMod –
Michael Barall’s software.

00:35:18.080 --> 00:35:22.080
And I generated the mesh for this using
Trellis, which is a commercial meshing

00:35:22.080 --> 00:35:27.016
software that allows you to mesh pretty
much anything if you are patient.

00:35:27.040 --> 00:35:30.480
For now, I’m using slip-weakening
friction, which, in past studies,

00:35:30.480 --> 00:35:35.222
has shown to do well for full coseismic
rupture speeds just as comparable to

00:35:35.222 --> 00:35:40.056
friction laws with more complex
formulations. So we went with that for now.

00:35:40.080 --> 00:35:42.960
And, in this case, I’m starting
the earthquakes by basically

00:35:42.960 --> 00:35:46.080
kicking the fault really hard.
Basically, I’m raising the shear

00:35:46.080 --> 00:35:49.680
stress higher than the yield stress
over an area large enough to

00:35:49.680 --> 00:35:52.216
make sure that
the earthquake goes.

00:35:52.240 --> 00:35:56.400
And so I did that both – so this type
of model doesn’t allow nucleation

00:35:56.400 --> 00:35:59.840
to grow organically, so I have
to kickstart the 7.1 as well.

00:35:59.840 --> 00:36:07.256
But, in both of these cases, I kickstart
them at the – at the USGS hypocenter.

00:36:07.280 --> 00:36:10.536
So I have this big old
slide of parameters.

00:36:10.560 --> 00:36:13.840
So I’m using a regional stress
orientation from seismicity –

00:36:13.840 --> 00:36:16.136
the average for
southern California.

00:36:16.160 --> 00:36:18.560
Using the SCEC
community velocity model.

00:36:18.560 --> 00:36:21.280
Pretty standard slip-weakening
friction coefficients.

00:36:21.280 --> 00:36:24.160
And I do want to call some
attention to the principal stresses.

00:36:24.160 --> 00:36:27.520
So this might look like
a very, very large difference.

00:36:27.520 --> 00:36:32.240
But I did need to use different stresses
on – both on the right-lateral fault

00:36:32.240 --> 00:36:36.216
and the left-lateral fault to get
behaviors that match what we saw.

00:36:36.240 --> 00:36:42.080
And so – and a way to think about that,
though, is – so the overall stress on

00:36:42.080 --> 00:36:45.280
the right-lateral fault had to be
higher than on the left-lateral fault.

00:36:45.280 --> 00:36:48.400
But, if you think about it, we are
in a right-lateral plate boundary,

00:36:48.400 --> 00:36:51.600
even though the setting in the
Ridgecrest area is very complex.

00:36:51.600 --> 00:36:54.960
You still are in a predominantly
right-lateral plate boundary, and this is

00:36:54.960 --> 00:36:59.760
a right-lateral fault, so that you might
expect, for a given right-lateral stress

00:36:59.760 --> 00:37:03.920
field, to be accumulating more stress
on a right-lateral structure and perhaps

00:37:03.920 --> 00:37:07.440
less on a left-lateral structure.
And so, in the sense of sort of

00:37:07.440 --> 00:37:10.000
how stress builds on faults
in different orientations,

00:37:10.000 --> 00:37:12.320
we felt comfortable
about setting it up this way.

00:37:12.320 --> 00:37:16.480
And, as it turned out, when you
assigned the exact same principal

00:37:16.480 --> 00:37:20.216
stresses on both faults in that
orientation, it just did not

00:37:20.240 --> 00:37:21.840
match the earthquake.
You either got something

00:37:21.840 --> 00:37:23.920
that was much too large
or much too small.

00:37:23.920 --> 00:37:27.496
Or didn’t follow the pattern
that we actually saw.

00:37:27.520 --> 00:37:31.520
So our procedure – sorry for two
all-text slides in a row – so first,

00:37:31.520 --> 00:37:34.560
we just implemented a regional
stress field on both faults.

00:37:34.560 --> 00:37:40.400
North-7-east just resolved on
the complex fault geometry and said –

00:37:40.400 --> 00:37:43.920
and then we started a 6.4.
Well, we started a earthquake

00:37:43.920 --> 00:37:47.360
at the 6.4 nucleation point.
And basically, we ran a bunch of

00:37:47.360 --> 00:37:50.880
those until we got a 6.4 that
had an appropriate amount

00:37:50.880 --> 00:37:53.096
of slip for what
was observed.

00:37:53.120 --> 00:37:58.000
And then, once we had that, we then
used the final stresses from the 6.4 model

00:37:58.000 --> 00:38:02.400
as the initial stresses for the 7.1 model.
So this is close – 34 hours in time is

00:38:02.400 --> 00:38:06.480
not going to have a substantial
stress change from tectonics,

00:38:06.480 --> 00:38:10.800
so we figured that was safe to do.
And then we nucleated at the

00:38:10.800 --> 00:38:14.456
7.1 hypocenter to see if we got
something to match the 7.1.

00:38:14.480 --> 00:38:16.080
And so then it was
kind of a give-and-take.

00:38:16.080 --> 00:38:19.120
Like, maybe we got something that
matched the 6.4 but not the 7.1.

00:38:19.120 --> 00:38:22.696
So we had to go back and do
another 6.4 and try another 7.1.

00:38:22.720 --> 00:38:25.600
So there was a lot of back-and-forth.
We ran over 100 models.

00:38:25.600 --> 00:38:29.760
And what I’m going to show here are
the one or two that actually, we think,

00:38:29.760 --> 00:38:33.840
match best. To illustrate a little more
of what I was talking about with

00:38:33.840 --> 00:38:36.800
the initial stresses – I just got louder.
Have you been able to

00:38:36.800 --> 00:38:40.880
hear me this entire time?
Okay. [laughs] Cool.

00:38:40.880 --> 00:38:45.040
So here are two panels of a lot of things.
I will mention, the Garlock Fault is

00:38:45.040 --> 00:38:48.240
in here because we were looking at
stress changes on it, but I’m not talking

00:38:48.240 --> 00:38:51.496
about that today. You’re welcome
to ask me about that later.

00:38:51.520 --> 00:38:55.440
But basically, on the left
is just the regional stresses.

00:38:55.440 --> 00:38:58.240
And so – by the way,
these faults go down 12 kilometers,

00:38:58.240 --> 00:39:01.576
which is the base
of the seismicity.

00:39:01.600 --> 00:39:07.280
And, in this case, both normal
and shear – so normal stress and

00:39:07.280 --> 00:39:09.760
shear stress we show on
a similar color scale, but note that

00:39:09.760 --> 00:39:14.160
the scales have different top.
So the reason it’s kind of stripey is,

00:39:14.160 --> 00:39:17.680
again, because the fault is bent,
and you’re resolving a homogeneous

00:39:17.680 --> 00:39:21.280
stress direction on a bent structure.
So you get different accumulation of

00:39:21.280 --> 00:39:24.880
stress or different resolution of stress.
And the reason that it’s tapered is that

00:39:24.880 --> 00:39:28.400
we tapered it because stress gets
less as you get towards the

00:39:28.400 --> 00:39:32.080
surface of the Earth.
And so, for our homogeneous one,

00:39:32.080 --> 00:39:34.880
you can see it’s – you know,
you see a lot of complexity

00:39:34.880 --> 00:39:37.840
just due to the fault geometry.
These are rotated to the show

00:39:37.840 --> 00:39:40.216
the left-lateral fault and
the right-lateral fault.

00:39:40.240 --> 00:39:45.736
But then, on the right, we have
the stress field post-6.4, pre-7.1.

00:39:45.760 --> 00:39:51.440
So, in this case – so the 7.1 hypocenter
is around here, and there’s very little –

00:39:51.440 --> 00:39:59.840
there’s no obvious difference between
the two cases at the nucleation point.

00:39:59.840 --> 00:40:03.496
But, when you look closer to the
intersection, it’s especially noticeable

00:40:03.520 --> 00:40:07.680
on the left-lateral fault. You see
where the shear stress has dropped.

00:40:07.680 --> 00:40:12.296
You see some places where
normal has increased at bends.

00:40:12.320 --> 00:40:16.000
And then also I realize that the
color scale on this figure is not great,

00:40:16.000 --> 00:40:18.616
and I’m revising it
for the paper.

00:40:18.640 --> 00:40:21.458
But – oh, I just
lost my pointer.

00:40:21.458 --> 00:40:25.360
Makes it – oh, here we go.
So around here – and actually, I guess

00:40:25.360 --> 00:40:28.560
I should have put in – I have a different
figure that shows this better. My bad.

00:40:28.560 --> 00:40:37.976
But there is some increased normal
stress around the junction point.

00:40:38.000 --> 00:40:42.160
So just think about the sense
of slip for – so you would –

00:40:42.160 --> 00:40:47.520
there’s increased normal stress
on the side of the – on the right-lateral

00:40:47.520 --> 00:40:50.320
fault because – just thinking about the
sense of slip on the left-lateral fault,

00:40:50.320 --> 00:40:53.600
it’s moving, and it’s pushing it.
It’s clamping it.

00:40:53.600 --> 00:40:57.120
And similarly, there’s a reduction
in normal stress on this side.

00:40:57.120 --> 00:40:59.760
This is – I definitely put, actually,
the wrong figure in here.

00:40:59.760 --> 00:41:03.016
I do have a better figure of
this I can show you later.

00:41:03.040 --> 00:41:08.000
But also there’s a reduction of shear
stress around this area just because the

00:41:08.000 --> 00:41:12.240
slip on the left-lateral fault actually
produces something of a stress shadow.

00:41:12.240 --> 00:41:16.776
So, again, I apologize for not
putting in the best figure of this.

00:41:16.800 --> 00:41:23.690
So this is our initial stress from the –
or the – before the 6.4 initial stress –

00:41:23.690 --> 00:41:26.055
after the 6.4,
before the 7.1.

00:41:26.080 --> 00:41:28.720
And so I’m going to jump
straight to our preferred model.

00:41:28.720 --> 00:41:32.640
So our preferred model of the 6.4.
So here, this is the same earthquake.

00:41:32.640 --> 00:41:36.800
I just rotated it to show the right-lateral
fault and the left-lateral fault.

00:41:36.800 --> 00:41:38.880
Notice that these are
different color scales.

00:41:38.880 --> 00:41:43.920
So I only go up to 10 centimeters
on the right-lateral fault and 1.5 meters

00:41:43.920 --> 00:41:46.400
on the right-lateral fault to show
what we’re – excuse me –

00:41:46.400 --> 00:41:49.600
on the left-lateral fault,
and this is the right-lateral fault.

00:41:49.600 --> 00:41:56.720
So, in this case, we see, you know,
3/4 of a meter to just over a meter

00:41:56.720 --> 00:42:00.640
of slip along the surface of the
documented left-lateral rupture.

00:42:00.640 --> 00:42:04.856
Which is consistent with the
observations we saw at the time.

00:42:04.880 --> 00:42:09.280
Or, that we got at the time from people.
I did not measure any of this. [laughs]

00:42:09.280 --> 00:42:14.320
And then, interestingly, looking at the
right-lateral fault, there actually is

00:42:14.320 --> 00:42:18.800
a little bit of – we’re going to call
this triggered slip on the surface of the

00:42:18.800 --> 00:42:24.135
right-lateral fault near the intersection.
This is a place where there was,

00:42:24.160 --> 00:42:27.840
you know, just some unclamping
that allowed this to slip a little.

00:42:27.840 --> 00:42:31.680
And so, in this case, we’re talking on
the scale of, like, 5 to 7 centimeters.

00:42:31.680 --> 00:42:34.160
It’s not a lot, but it’s the
kind of thing that, because it is

00:42:34.160 --> 00:42:38.136
on the surface, you might –
it might show up – might see.

00:42:38.160 --> 00:42:43.176
So – and that doesn’t extend further
north into the whole aftershock zone.

00:42:43.200 --> 00:42:46.696
So anyway, there’s our 6.4.
Then going on to the 7.1,

00:42:46.720 --> 00:42:51.040
when we nucleated the 7.1 hypocenter
under those initial stresses from the 6.4,

00:42:51.040 --> 00:42:55.520
we get a 7.1 that ruptures through
the whole right-lateral fault.

00:42:55.520 --> 00:42:58.800
And, again, this colored bar is
another thing I’m fixing for the paper.

00:42:58.800 --> 00:43:03.040
But south – so, north of here, we do,
in fact, get the highest slip around

00:43:03.040 --> 00:43:06.080
the hypocentral area,
which is what was observed.

00:43:06.080 --> 00:43:11.920
Here we get slip – we were trying to
match the magnitudes of 4-ish meters

00:43:11.920 --> 00:43:14.400
of slip, which is what
we do have in this area.

00:43:14.400 --> 00:43:16.640
I will be adding some
contours in the paper.

00:43:16.640 --> 00:43:19.120
And then, in the real earthquake,
there was less slip south of

00:43:19.120 --> 00:43:23.818
the junction than there was north
of it, and we do see that here.

00:43:24.320 --> 00:43:27.440
And there’s still – there’s still more
than in the real thing, but the real

00:43:27.440 --> 00:43:31.200
earthquake also splayed into three
right-lateral strands down there.

00:43:31.200 --> 00:43:34.640
So that’s something that,
when you sum across those strands,

00:43:34.640 --> 00:43:37.496
it’s consistent with
what we have.

00:43:37.520 --> 00:43:40.800
And also notice that here – again,
this scale bar goes up to 3 meters.

00:43:40.800 --> 00:43:43.440
For the left-lateral fault,
we go up to 50 centimeters, and do,

00:43:43.440 --> 00:43:47.360
in fact, see some – I hesitate to
call this re-rupture because

00:43:47.360 --> 00:43:50.320
it’s just at the surface.
But I would say this is a, you know,

00:43:50.320 --> 00:43:54.880
triggered slip of, you know, 20 to 30
centimeters on the left-lateral fault.

00:43:54.880 --> 00:43:58.536
So that’s the kind of thing that could
account for those observations of

00:43:58.560 --> 00:44:02.080
re-rupture or more consolidated
fractures along the left-lateral fault,

00:44:02.080 --> 00:44:06.216
even if this isn’t necessarily
a propagating rupture front.

00:44:06.240 --> 00:44:09.816
Okay, here’s the other slide that
I was talking about that’s better.

00:44:09.840 --> 00:44:12.640
So dynamic clamping and stress
shadowing, we think are what

00:44:12.640 --> 00:44:16.400
are prevented the rupture front –
both faults from going at once.

00:44:16.400 --> 00:44:19.760
So now this is just normal stress
and shear – normal stress and

00:44:19.760 --> 00:44:22.640
shear stress change
rather than amounts.

00:44:22.640 --> 00:44:26.376
So this is ignoring pre-stress.
It’s just difference.

00:44:26.400 --> 00:44:29.760
So, again, in the – after the 6.4 –
so this is between the two earthquakes –

00:44:29.760 --> 00:44:33.840
after the 6.4, you have a reduction of
normal stress here because, you know,

00:44:33.840 --> 00:44:36.160
this side – this left-lateral
fault is going that way.

00:44:36.160 --> 00:44:39.440
It’s pulling away and
reducing normal stress here.

00:44:39.440 --> 00:44:43.040
Where, meanwhile, this side is
going this way and is increasing

00:44:43.040 --> 00:44:48.480
normal stress here. And so,
in the case of the 6.4, this clamping

00:44:48.480 --> 00:44:51.496
would have very much just said,
no, rupture, don’t go south.

00:44:51.520 --> 00:44:54.880
But, in this case, the fact that there’s
a little bit of reduction to the north,

00:44:54.880 --> 00:44:58.056
this is exactly where we see
the triggered slip in our model.

00:44:58.080 --> 00:45:03.280
Meanwhile, with the shear stress,
you see this blue means a reduction

00:45:03.280 --> 00:45:08.560
on both sides of the junction point.
And so, in this case, increased normal

00:45:08.560 --> 00:45:12.776
stress and decreased shear stress is a
super, like, rupture, don’t go this way.

00:45:12.800 --> 00:45:15.920
Which could have both stopped the
rupture from going south in the 6.4

00:45:15.920 --> 00:45:19.840
and also actually led to some of
that decrease in slip – that lower slip

00:45:19.840 --> 00:45:24.240
south of the junction in the 7.1.
Meanwhile, even though the decreased

00:45:24.240 --> 00:45:28.400
normal stress kind of started letting the
rupture go north a little, this decreased

00:45:28.400 --> 00:45:33.656
shear stress might not have given it
enough fuel to actually take off.

00:45:33.680 --> 00:45:36.376
So, we think this is
what’s going on here.

00:45:36.400 --> 00:45:40.080
But, as for the second question,
was the right-lateral fault, you know,

00:45:40.080 --> 00:45:45.680
actually brought to failure by the 6.4?
Well, again, looking here, you know,

00:45:45.680 --> 00:45:49.840
maybe – it’s a little pink – it’s maybe
a slight increase in both normal stress

00:45:49.840 --> 00:45:53.840
and shear stress, but it’s a very
insubstantial change at this nucleation

00:45:53.840 --> 00:45:57.760
point compared to at the junction.
And so, as it turns out, when we ran

00:45:57.760 --> 00:46:03.976
a model with just the regional
stress field – no effects of the 6.4 –

00:46:04.000 --> 00:46:08.400
and started at the 7.1 nucleation point,
we actually still got the 7.1.

00:46:08.400 --> 00:46:11.200
And, again, I’ll be putting contours
on this in the paper.

00:46:11.200 --> 00:46:15.360
There is – the slip north of the junction
is about the same as in the model

00:46:15.360 --> 00:46:19.520
with the effects of the 6.4.
South of the junction, this is redder.

00:46:19.520 --> 00:46:22.936
It’s a lot most saturated.
The slip down here is higher.

00:46:22.960 --> 00:46:27.280
But we still actually, either way,
see some of that triggered slip

00:46:27.280 --> 00:46:31.120
on the left-lateral fault, but it’s less,
perhaps because the fault hasn’t

00:46:31.120 --> 00:46:35.176
already weakened and isn’t
just sort of open and ready to go.

00:46:35.200 --> 00:46:37.520
So what this says is, yeah,
the right-lateral fault was

00:46:37.520 --> 00:46:39.600
probably already
really, really close.

00:46:39.600 --> 00:46:44.000
And so it might be just that the
6.4 advanced the nucleation clock.

00:46:44.000 --> 00:46:48.240
Oh, I lied about time.
Anyway, I’ll try to finish.

00:46:48.240 --> 00:46:52.000
so we think the main right-lateral
fault probably wasn’t involved –

00:46:52.000 --> 00:46:54.560
not in a substantial way.
It might have triggered slip.

00:46:54.560 --> 00:46:58.080
There’s definitely aftershocks showing
that there’s a stress change there.

00:46:58.080 --> 00:47:00.800
And, as we showed, there was
definitely a stress change there.

00:47:00.800 --> 00:47:04.240
But we were able to reproduce the
magnitude and observe slip distribution

00:47:04.240 --> 00:47:08.536
of the 6.4 without involving
a major right-lateral strand.

00:47:08.560 --> 00:47:11.040
And I should change it to not,
no observed surface rupture,

00:47:11.040 --> 00:47:15.120
but minor observed surface rupture.
And so we wonder if these aftershocks

00:47:15.120 --> 00:47:20.856
are actually the early stages of the
nucleation process of that part of the

00:47:20.880 --> 00:47:25.040
right-lateral fault growing towards
instability and actually – rather than

00:47:25.040 --> 00:47:28.960
being aftershocks of the 6.4, being
foreshocks of the 7.1, which I know

00:47:28.960 --> 00:47:32.560
is a little semantic, but I think
interesting from a physics standpoint.

00:47:32.560 --> 00:47:37.176
And, again, I will note we haven’t
modeled the right-lateral splays yet.

00:47:37.200 --> 00:47:39.600
Our next step is, we’re going to
add more fault strands.

00:47:39.600 --> 00:47:41.280
We’re going to see
what made this delay.

00:47:41.280 --> 00:47:43.520
We’re going to use rate-state
friction because that’s useful

00:47:43.520 --> 00:47:47.760
for doing the delay.
But that’s coming later.

00:47:47.760 --> 00:47:50.936
But what is – but, you know,
this is Ridgecrest.

00:47:50.960 --> 00:47:54.000
Not all cross-fault ruptures
behave like Ridgecrest.

00:47:54.000 --> 00:47:57.360
So we just heard a lot from Ken about
Elmore Ranch and Superstition Hills,

00:47:57.360 --> 00:48:03.120
which were pretty similar to Ridgecrest.
You had one fault go and then, 11 hours

00:48:03.120 --> 00:48:07.760
later, a cross-fault, separated in time.
But, on the other hand, you have

00:48:07.760 --> 00:48:12.720
something like Off Sumatra in 2012,
which was an 8.5 strike-slip rupture

00:48:12.720 --> 00:48:17.400
that went through this network –
the actual title of Meng et al.’s paper

00:48:17.400 --> 00:48:19.600
is Earthquake in a Maze.
Went through this really

00:48:19.600 --> 00:48:23.360
complex network of
cross-faults in one event.

00:48:23.360 --> 00:48:27.256
So that’s kind of the opposite
end is this really big sucker.

00:48:27.280 --> 00:48:29.920
And then there’s other examples.
Thing like – I just learned about

00:48:29.920 --> 00:48:34.376
this one – the Kita Tango earthquake
of 1927 in Japan, where you had

00:48:34.400 --> 00:48:38.160
the first rupture on this left-lateral
strike-slip fault, which kind of stops,

00:48:38.160 --> 00:48:43.120
and then jumped a maybe 5- or
6-kilometer step-over onto a cross-fault.

00:48:43.120 --> 00:48:46.880
And so that’s the kind of thing that is –
that’s the distance the Garlock is

00:48:46.880 --> 00:48:49.920
from Ridgecrest, but also this is
the kind of thing – if that fault is

00:48:49.920 --> 00:48:54.456
really ready, it might go, even
at more of a distance like that.

00:48:54.480 --> 00:48:56.400
And then, on the right,
another thing you can have.

00:48:56.400 --> 00:49:00.000
So the Kumamoto earthquakes
in Japan – again, predominantly

00:49:00.000 --> 00:49:03.360
on these right-lateral structures.
But there were a lot of cross-faults

00:49:03.360 --> 00:49:07.040
that were activated with triggered slip.
They show no indication of coseismic

00:49:07.040 --> 00:49:11.120
rupture, but still, a triggered slip fault in
your house is still a fault in your house.

00:49:11.120 --> 00:49:14.240
This is something you still need to
think about from a hazard standpoint.

00:49:14.240 --> 00:49:18.776
So there are a lot of things cross-faults
can do. Ridgecrest is just one thing.

00:49:18.800 --> 00:49:22.400
And so I think some things that boil
down to is, it’s not just geometry.

00:49:22.400 --> 00:49:24.776
Pre-stress is really key.

00:49:24.800 --> 00:49:28.800
Cross-faults need to be close
to failure or – in order to go.

00:49:28.800 --> 00:49:31.760
Dynamic clamping,
in and of itself, doesn’t do it.

00:49:31.760 --> 00:49:35.200
And sometimes the stress orientation
leads to things like triggered slip

00:49:35.200 --> 00:49:38.720
even if there isn’t major rupture.
So this example here is one of

00:49:38.720 --> 00:49:41.920
our models that didn’t work.
It’s a 6.8 where I started at the

00:49:41.920 --> 00:49:45.760
6.4 nucleation point, and it actually
did rupture through the junction,

00:49:45.760 --> 00:49:49.200
just because the stresses weren’t
right for Ridgecrest, but, you know,

00:49:49.200 --> 00:49:52.160
the stresses for Ridgecrest
aren’t the stresses everywhere.

00:49:52.160 --> 00:49:57.176
You can still get a situation like this.
So, last slide, for reals.

00:49:57.200 --> 00:50:00.560
Ideas for moving forward.
It’s multi-disciplinary, for sure.

00:50:00.560 --> 00:50:02.560
So I think we need to
identify these things.

00:50:02.560 --> 00:50:06.640
I know that Tim has said it was, what,
70-some percent of the Ridgecrest

00:50:06.640 --> 00:50:10.536
surface rupture traces were something
you could have identified in the past.

00:50:10.560 --> 00:50:13.600
So find them. And maybe if you
can find a way to assess that the

00:50:13.600 --> 00:50:17.920
cross-faults and the main faults
go at the same time, that’s useful.

00:50:17.920 --> 00:50:20.720
Definitely get a sense of the
stress field and see how it –

00:50:20.720 --> 00:50:24.240
how stress over time accumulates
on faults of different orientations.

00:50:24.240 --> 00:50:27.200
And definitely I think we need
some dynamic models of just basic

00:50:27.200 --> 00:50:30.240
interactions of cross-faults.
And so hopefully my SCEC

00:50:30.240 --> 00:50:33.600
proposal for that will be funded.
But this is an inter-disciplinary

00:50:33.600 --> 00:50:36.720
question. And so I think these are
some things that can be not just for

00:50:36.720 --> 00:50:41.016
Ridgecrest, but – not just for
northern California, but for anywhere.

00:50:41.040 --> 00:50:42.800
So, with that, I’d like to
thank a lot of people.

00:50:42.800 --> 00:50:46.560
And sorry [laughs]
for going so long.

00:50:46.560 --> 00:50:51.867
[Applause]

00:50:53.432 --> 00:50:56.080
- Thank you, Julian. Again, I
think we can take questions

00:50:56.080 --> 00:51:02.960
at the end and I’ll introduce
David Schwartz back from Japan

00:51:02.960 --> 00:51:08.880
to talk about conjugate faulting –
potential occurrence and implications

00:51:08.880 --> 00:51:12.357
for the Bay Area and
northern California.

00:51:15.042 --> 00:51:18.300
- All right. Let’s see if
we can get me up here.

00:51:22.960 --> 00:51:25.016
All right.

00:51:25.040 --> 00:51:28.800
Great. Well, thank you, Tim.
This is an instance where my wife says

00:51:28.800 --> 00:51:34.865
to me, David, why are you doing all of
this work without pay? You’re retired.

00:51:36.213 --> 00:51:38.133
[laughs]

00:51:38.640 --> 00:51:43.360
You’ve heard two really,
really detailed presentations

00:51:43.360 --> 00:51:49.040
on what went on in Ridgecrest.
And I’d like to say something else

00:51:49.040 --> 00:51:53.360
about cross-faults and the
potential implications for the

00:51:53.360 --> 00:51:55.816
Bay Area and
northern California.

00:51:55.840 --> 00:51:59.336
And, to put this – and why am
I giving this talk to begin with?

00:51:59.360 --> 00:52:04.160
When the organizing committee
was sitting around, and Ken’s talk

00:52:04.160 --> 00:52:08.400
and Julian’s talk were sort of put on
the agenda, somebody said, well,

00:52:08.400 --> 00:52:10.320
that’s great, but that’s
southern California.

00:52:10.320 --> 00:52:13.600
What are the implications
for the Bay Area?

00:52:13.600 --> 00:52:16.960
And I think I said, yeah, what are
the implications for the Bay Area?

00:52:16.960 --> 00:52:23.016
So, Sarah – Sarah Minson looked at me.
And then I looked for the door,

00:52:23.040 --> 00:52:26.080
but the door was closed.
And she said, well, David,

00:52:26.080 --> 00:52:28.640
you’re going to tell us
what the implications are.

00:52:28.640 --> 00:52:31.230
So here I am today.

00:52:35.280 --> 00:52:39.200
Conjugate faults, cross-faults,
are nothing new.

00:52:39.200 --> 00:52:41.680
They’ve been mapped
ever since geologists have

00:52:41.680 --> 00:52:44.456
been doing
geologic mapping.

00:52:44.480 --> 00:52:47.840
And really, though, it’s been in the
last several decades where there’s

00:52:47.840 --> 00:52:54.480
been an appreciation for coseismic
cross-faulting – two different faults,

00:52:54.480 --> 00:53:00.376
or a series of faults, rupturing
together or very close in time.

00:53:00.400 --> 00:53:09.440
And we saw some examples of that.
Cross-faults and conjugate faults come

00:53:09.440 --> 00:53:14.400
in a range of magnitudes, lengths,
and faulting types, and I’ll run through

00:53:14.400 --> 00:53:19.200
a number of just quick examples.
You have the Elmore Ranch/

00:53:19.200 --> 00:53:23.040
Superstition Hills,
which we’ve heard about.

00:53:23.040 --> 00:53:27.040
In the Brawley Seismic Zone,
you have cross-faults sort of

00:53:27.040 --> 00:53:30.560
connecting the Imperial Fault
with the San Andreas Fault.

00:53:30.560 --> 00:53:33.760
And these are really
well-defined in the seismicity.

00:53:33.760 --> 00:53:38.480
You can go to a place like Lake Tahoe,
and just to the north, there’s the

00:53:38.480 --> 00:53:42.000
Polaris Fault and the
cross-cutting Dog Valley Fault.

00:53:42.000 --> 00:53:47.096
These are Holocene and Quaternary
faults that haven’t ruptured yet.

00:53:47.120 --> 00:53:50.000
They will at some time in the future.
You could ask the question,

00:53:50.000 --> 00:53:54.696
will they go together,
or will they be independent?

00:53:54.720 --> 00:53:58.960
In a really interesting paper by
Fukuyama, he brought up a number

00:53:58.960 --> 00:54:03.520
of points that Julian mentioned,
and he kind of summarized a range

00:54:03.520 --> 00:54:10.880
of conjugate and cross-faults,
and these are all earthquakes in Japan.

00:54:10.880 --> 00:54:16.720
And, in the top, you’re looking at
strike-slip faults, the lower part of

00:54:16.720 --> 00:54:20.880
the diagram, thrust faults, and he
grouped these into what he called

00:54:20.880 --> 00:54:26.000
conjugate rupture which occurred
simultaneously with the main rupture.

00:54:26.000 --> 00:54:32.397
So this is a set of faults.
They all went at the same time.

00:54:33.600 --> 00:54:39.656
And then he had a conjugate rupture
that occurred after the main rupture.

00:54:39.680 --> 00:54:45.200
Superstition Hills we’ve already seen,
and then two large earthquakes in the

00:54:45.200 --> 00:54:51.176
ocean – the Wharton Basin 2000 –
a 7.8 followed by another 7.8.

00:54:51.200 --> 00:54:56.560
And Julian showed a nice slide
of the 2012 Sumatra earthquake –

00:54:56.560 --> 00:55:01.440
an 8.6 where these all went
together, followed by an 8.2.

00:55:01.440 --> 00:55:07.496
So this is just some of the variability
in the way these conjugate faults work.

00:55:07.520 --> 00:55:12.160
And, coming closer to home,
we could look at some smaller events.

00:55:12.160 --> 00:55:17.176
The 1984 Round Valley,
California, earthquake.

00:55:17.200 --> 00:55:24.160
And that started on a northeast-trending
left-lateral strike-slip fault,

00:55:24.160 --> 00:55:32.160
which is here. And then, an hour later,
there was a 5.2, and within 24 hours,

00:55:32.160 --> 00:55:38.160
seismicity had propagated onto
a northwest-trending right-lateral

00:55:38.160 --> 00:55:43.440
fault with a – with a different dip.
So this is a small rupture sequence,

00:55:43.440 --> 00:55:45.896
but just an example
of cross-faulting.

00:55:45.920 --> 00:55:49.976
And the 1994 Double Spring Flat,
Nevada, earthquake.

00:55:50.000 --> 00:55:56.456
It started on a vertical northeast-
trending left-lateral fault.

00:55:56.480 --> 00:56:00.880
And then, within eight days,
seismicity had migrated down along

00:56:00.880 --> 00:56:07.520
a northwest-trending right-lateral fault.
This went on, actually, for a significant

00:56:07.520 --> 00:56:12.376
period of time, and the
seismicity extended further.

00:56:12.400 --> 00:56:18.696
One more small event –
a recent 5-1/2 in Osaka, Japan.

00:56:18.720 --> 00:56:23.816
The earthquake nucleated on a
northwest-trending thrust fault

00:56:23.840 --> 00:56:28.240
and then seismicity developed
along the northeast-trending

00:56:28.240 --> 00:56:31.520
vertical strike-slip fault.
And then, with time,

00:56:31.520 --> 00:56:38.080
additional strike-slip faults lit up,
extending out of the vertical.

00:56:38.080 --> 00:56:42.880
So these are – these are examples
of kind of small-ish events

00:56:42.880 --> 00:56:48.000
that occur everywhere.
And the last example I want to

00:56:48.000 --> 00:56:54.880
give is a favorite of mine.
This is the 1976 magnitude 7-1/2

00:56:54.880 --> 00:57:03.576
Motagua Fault rupture. The fault
ruptured for 230, or 240, kilometers.

00:57:03.600 --> 00:57:07.496
My Ph.D. dissertation
area was right here.

00:57:07.520 --> 00:57:10.080
So I have a lot of good
feelings about the fault.

00:57:10.080 --> 00:57:12.856
Not about the earthquake
and its results.

00:57:12.880 --> 00:57:19.256
And then, south of the Motagua
are a series of extensional

00:57:19.280 --> 00:57:28.400
north-south-northeast-trending graben.
And, six hours after the main Motagua,

00:57:28.400 --> 00:57:34.800
the Mixco Fault, just west of
Guatemala City, had a 5.9 or a 6,

00:57:34.800 --> 00:57:42.000
and it produced 16 kilometers of surface
rupture, spread over about 8 kilometers.

00:57:42.000 --> 00:57:45.440
So this was sort of a
surprise to everybody.

00:57:45.440 --> 00:57:51.016
And it was 20 kilometers
from the main Motagua.

00:57:51.040 --> 00:57:58.320
And I think this is a nice example of
triggering of ruptures at faults that

00:57:58.320 --> 00:58:03.520
are distant from the main fault
and do not have direct rupture

00:58:03.520 --> 00:58:07.200
propagation involved.
All right?

00:58:07.200 --> 00:58:13.040
So, with that sort of little feeling
for the – a range of some of the types

00:58:13.040 --> 00:58:18.696
of conjugate ruptures you can get,
let’s look at northern California.

00:58:18.720 --> 00:58:21.329
And the Bay Area.

00:58:22.240 --> 00:58:25.600
I put this figure in.
This is an old seismicity map.

00:58:25.600 --> 00:58:32.776
You’ll see the seismicity is 1984
to 2002, but I think it really shows

00:58:32.800 --> 00:58:37.600
some things that I want to point out.
This is really focused on the East Bay.

00:58:37.600 --> 00:58:42.480
The red dots are historical
events of 5 and larger.

00:58:42.480 --> 00:58:48.480
There aren’t many of those out there.
And Livermore, Pleasanton,

00:58:48.480 --> 00:58:55.680
Dublin, Alamo, Walnut Creek –
I live here in Danville –

00:58:55.680 --> 00:58:59.096
Hayward, Fremont,
to sort of orient you.

00:58:59.120 --> 00:59:03.416
And I want to talk about
three locations. One is this,

00:59:03.440 --> 00:59:10.376
which is the magnitude 5.7
Mount Lewis rupture in 1986.

00:59:10.400 --> 00:59:17.096
I want to talk about this, which is the
Greenville-Las Positas rupture in 1980.

00:59:17.120 --> 00:59:20.136
And then we’re going to
talk about the San Ramon Valley

00:59:20.160 --> 00:59:23.336
and earthquake swarms.

00:59:23.360 --> 00:59:29.920
So, the 1986 Mount Lewis earthquake
is really, really fascinating.

00:59:29.920 --> 00:59:34.240
I went back and I read the paper
by Kilb and Rubin, and I would

00:59:34.240 --> 00:59:39.976
recommend everybody go and do that.
It really was – it was a small event.

00:59:40.000 --> 00:59:43.736
It was a vertical strike-slip
fault right in here.

00:59:43.760 --> 00:59:47.120
It was about – estimated to
be about – it’s a right-lateral,

00:59:47.120 --> 00:59:52.960
about 2, 2-1/2 kilometers long.
And then, with time, this very, very

00:59:52.960 --> 00:59:58.880
complex aftershock sequence
grew to the north, so it extended

00:59:58.880 --> 01:00:04.216
out to about 20 kilometers.
It grew east to west.

01:00:04.240 --> 01:00:08.776
And this is in an area that really
had very little seismicity.

01:00:08.800 --> 01:00:15.416
The fault itself is at a very high angle to
all of the major plate boundary faults.

01:00:15.440 --> 01:00:19.656
And, if you look closely,
you see this whole series of

01:00:19.680 --> 01:00:24.320
east-west-trending
seismicity distributions.

01:00:24.320 --> 01:00:30.080
These are all left-lateral,
vertical, strike-slip faults.

01:00:30.080 --> 01:00:36.240
And they called these – they called
these sliced bread faults because it

01:00:36.240 --> 01:00:42.080
was like having – taking a
loaf of bread and slicing it up.

01:00:42.080 --> 01:00:47.120
And so this is – this is a young fault.
There was no surface rupture.

01:00:47.120 --> 01:00:52.640
There’s no expression, as far as I know,
in the geology itself of this area.

01:00:52.640 --> 01:00:59.600
And I think this is part of a zone
of similar-oriented seismicity

01:00:59.600 --> 01:01:05.040
that is helping transfer slip
from the Calaveras Fault eventually

01:01:05.040 --> 01:01:12.136
over to the Greenville Fault.
So this is one type of cross-faulting

01:01:12.160 --> 01:01:18.640
in this part of the East Bay.
In 1980, there was a 5.8 followed

01:01:18.640 --> 01:01:22.880
by a 5.5 a couple days later
on the Greenville Fault –

01:01:22.880 --> 01:01:25.336
one of the biggies
in the East Bay.

01:01:25.360 --> 01:01:30.720
And then it was noted that the Las
Positas Fault – so this is right-lateral.

01:01:30.720 --> 01:01:35.040
The Las Positas is left-lateral.
And there was rupture.

01:01:35.040 --> 01:01:39.840
There was rupture along the
Greenville anywhere between

01:01:39.840 --> 01:01:45.016
4-1/2 and 6-1/2 kilometers.
And then rupture was found

01:01:45.040 --> 01:01:50.362
on the Las Positas Fault.
And the rupture was small.

01:01:51.410 --> 01:01:55.920
Another view of this conjugate,
but this is – this is one of the few

01:01:55.920 --> 01:01:59.496
conjugates in the Bay Area where
something actually happened.

01:01:59.520 --> 01:02:04.056
And Lawrence Livermore Lab
is right here.

01:02:04.080 --> 01:02:07.120
So the amount –
the amount of slip was small.

01:02:07.120 --> 01:02:11.840
Only up to about 3 centimeters
on the Greenville and about a

01:02:11.840 --> 01:02:17.280
centimeter and a half on the Las Positas.
And a lot of it was afterslip.

01:02:17.280 --> 01:02:24.616
But this is a really nice conjugate
relationship in this part of the region.

01:02:24.640 --> 01:02:29.736
And then, here’s the
Greenville-Las Positas.

01:02:29.760 --> 01:02:33.840
This is Mount Diablo with
a blind thrust underneath.

01:02:33.840 --> 01:02:37.520
This is the San Roman Valley.
The Calaveras Fault.

01:02:37.520 --> 01:02:44.400
So the Calaveras is right-lateral,
and these arrows represent the

01:02:44.400 --> 01:02:49.120
direction of compression associated
with the Mount Diablo blind thrust.

01:02:49.120 --> 01:02:53.680
So this area in the San Ramon Valley
where I live is really being –

01:02:53.680 --> 01:02:58.320
it’s being torn apart. It’s being
squeezed. It’s being compressed.

01:02:58.320 --> 01:03:02.578
It’s breaking in all sorts
of different complex ways.

01:03:03.840 --> 01:03:08.560
And this is the area where the
San Ramon swarms occurred.

01:03:08.560 --> 01:03:11.416
And, if you want to
just look at it in 3D,

01:03:11.440 --> 01:03:15.840
this is from the
northern California ...

01:03:21.088 --> 01:03:24.696
... fold and fault model.

01:03:24.720 --> 01:03:28.320
This is the Las Positas Fault.
We’re looking north.

01:03:28.320 --> 01:03:33.256
This is the Mount Diablo blind thrust.
This is the Calaveras Fault.

01:03:33.280 --> 01:03:40.480
And, actually, I think the Calaveras
extends – it has branches that –

01:03:40.480 --> 01:03:45.200
or, splays that extend into the Napa
Valley, driving the system there.

01:03:45.200 --> 01:03:49.520
But this area is just really
being compressed and broken

01:03:49.520 --> 01:03:51.976
and pulled apart.

01:03:52.000 --> 01:03:59.176
And there have been a series of
earthquake swarms in this region

01:03:59.200 --> 01:04:03.736
between 1970 and the
most recent one was 2018.

01:04:03.760 --> 01:04:06.960
The most recent one was
actually about a kilometer

01:04:06.960 --> 01:04:10.216
from my house, so I took
a lot of interest in that.

01:04:10.240 --> 01:04:16.080
And what I just want to show is,
this is a map trace of the Calaveras.

01:04:16.080 --> 01:04:23.200
And it’s shown as a – as an active
structure ending here just

01:04:23.200 --> 01:04:28.056
a little bit north of Danville.
And you can see this.

01:04:28.080 --> 01:04:34.456
This is the swarm.
This is the swarm from 1990.

01:04:34.480 --> 01:04:40.000
The largest swarm in this area.
There were 351 earthquakes

01:04:40.000 --> 01:04:45.976
over 42 days. There were four
earthquakes larger than magnitude 4.

01:04:46.000 --> 01:04:48.960
And one – and it’s left-lateral.
The slip is left-lateral.

01:04:48.960 --> 01:04:53.920
Right-lateral on the Calaveras.
And there was concern at the time

01:04:53.920 --> 01:05:00.640
that slip along this Alamo swarm
sequence might unclamp the

01:05:00.640 --> 01:05:05.920
Calaveras and allow it to rupture.
And maybe when the Calaveras –

01:05:05.920 --> 01:05:09.040
the northern Calaveras
does rupture, this will be

01:05:09.040 --> 01:05:13.256
the mechanism
that promotes it.

01:05:13.280 --> 01:05:16.400
These other swarms
have different orientations.

01:05:16.400 --> 01:05:19.200
Some of them are normal
to the Calaveras.

01:05:19.200 --> 01:05:24.240
Some are at high angle and are
made up of multiple fault patches.

01:05:24.240 --> 01:05:30.720
But, again, this is an area
of observable conjugate,

01:05:30.720 --> 01:05:34.376
and cross-faulting
in the region.

01:05:34.400 --> 01:05:36.320
All right.

01:05:36.320 --> 01:05:39.416
As I’m losing my voice.

01:05:39.440 --> 01:05:43.440
When you step back and take a look
at the fault map of the Bay Area –

01:05:43.440 --> 01:05:49.016
and this is from Graymer et al.
It was put together for the

01:05:49.040 --> 01:05:54.536
2006 100th anniversary
of the 1906 earthquake.

01:05:54.560 --> 01:05:59.576
And you look very closely,
and the colors are hard to see.

01:05:59.600 --> 01:06:04.640
You’re hard-pressed, except for
what’s going on out in the East Bay

01:06:04.640 --> 01:06:08.560
where I just showed you,
to see any faults that have been

01:06:08.560 --> 01:06:16.536
mapped conjugate to the main
northwest-trending structures.

01:06:16.560 --> 01:06:18.936
So I challenge
you to go do it.

01:06:18.960 --> 01:06:23.716
And, when we go –
here’s San Pablo Bay.

01:06:24.640 --> 01:06:28.960
When we go north of San Pablo Bay –
and San Pablo Bay is right here,

01:06:28.960 --> 01:06:34.136
this is the Rodgers Creek Fault,
the Maacama Fault, The Geysers.

01:06:34.160 --> 01:06:38.137
Santa Rosa is located here.
The San Andreas.

01:06:38.800 --> 01:06:43.440
There’s one – in this entire area,
there’s one – well, there are

01:06:43.440 --> 01:06:47.656
actually two, but there’s one fault
called the Wight Way Fault.

01:06:47.680 --> 01:06:53.280
I was completely unaware of that,
and I tried to find some information

01:06:53.280 --> 01:06:54.720
on it, and I couldn’t.

01:06:54.720 --> 01:06:59.680
But it’s conjugate with the Maacama,
and then there’s a little unnamed fault

01:06:59.680 --> 01:07:05.840
out here that also has this sort of
a conjugate geometric orientation.

01:07:05.840 --> 01:07:13.120
But the Bay Area as a whole is pretty
much without mapped cross-faults.

01:07:13.120 --> 01:07:19.816
And this is a figure that we had
in a fact sheet that we released

01:07:19.840 --> 01:07:23.416
with the probabilities from
UCERF3 for the region.

01:07:23.440 --> 01:07:30.320
And this is a list of 33 additional
faults which, if you look closely, are

01:07:30.320 --> 01:07:36.400
shown in yellow – very hard to see –
that were incorporated into the analysis.

01:07:36.400 --> 01:07:41.096
And there’s one cross-fault
out of all of these minor faults.

01:07:41.120 --> 01:07:44.400
And that’s up here.
That’s the Wight Way Fault.

01:07:44.400 --> 01:07:51.257
So I think I’m going to end it here.
Am I on time? Good.

01:07:52.640 --> 01:07:58.960
I think ruptures such as Ridgecrest
provide critical insights into fault

01:07:58.960 --> 01:08:02.640
rupture processes and their controls.
This is invaluable.

01:08:02.640 --> 01:08:07.840
You can take that understanding and
build models of rupture in other places.

01:08:07.840 --> 01:08:12.616
And so I love what’s
being done at Ridgecrest.

01:08:12.640 --> 01:08:15.280
I may have gone down there,
but our granddaughter was

01:08:15.280 --> 01:08:19.416
born that day, and
that was – that was it.

01:08:19.440 --> 01:08:24.640
For the hazard perspective,
for the Bay Area …

01:08:27.280 --> 01:08:32.080
Or, just in general, complex and
unexpected surface faulting associated

01:08:32.080 --> 01:08:37.120
with conjugate ruptures could locally
increase damage from surface ruptures.

01:08:37.120 --> 01:08:41.360
So we want to – if we want to think
about the hazards in general associated

01:08:41.360 --> 01:08:46.320
with these faults, I think the potential
for surface rupture and what it

01:08:46.320 --> 01:08:50.800
might do is number one.
But, in the Bay Area and northern

01:08:50.800 --> 01:08:54.960
California, the potential for
Ridgecrest-type conjugate

01:08:54.960 --> 01:09:00.936
faulting to occur is low.
If these events do occasionally occur,

01:09:00.960 --> 01:09:07.336
there’s little basis to change established
shaking hazard estimates for the region.

01:09:07.360 --> 01:09:12.640
And I think an important area of
research dealing with stress changes

01:09:12.640 --> 01:09:20.720
and complexity is to think about
large triggered events associated

01:09:20.720 --> 01:09:23.520
with ruptures on
some of these large faults.

01:09:23.520 --> 01:09:27.840
These are events that are on faults
that are not involved in

01:09:27.840 --> 01:09:32.720
direct propagation of rupture.
And certainly, in other parts of the

01:09:32.720 --> 01:09:38.400
world, in intraplate settings,
we see tremendously large faults

01:09:38.400 --> 01:09:41.840
triggered by earthquakes
on other large faults.

01:09:41.840 --> 01:09:48.696
So this is just another area of concern.
So, in the end, I would say that,

01:09:48.720 --> 01:09:52.400
let’s keep this great work going
on Ridgecrest, but when you’re

01:09:52.400 --> 01:09:58.400
driving home tonight, don’t think you
have to worry about conjugate faulting,

01:09:58.400 --> 01:10:02.790
at least tonight, in the Bay Area.
Thank you very much.

01:10:02.790 --> 01:10:08.093
[Applause]

01:10:09.332 --> 01:10:12.616
- So we’ve got some
time for some questions.

01:10:12.640 --> 01:10:17.773
If you have a question for David –
I’ll keep him up at the microphone.

01:10:18.480 --> 01:10:21.674
If you’re going to ask a question,
use the microphone.

01:10:21.674 --> 01:10:25.039
- I left my hearing aids home
today, so you better talk loud.

01:10:25.039 --> 01:10:26.900
- [inaudible]

01:10:28.986 --> 01:10:33.200
- David, this is more
of a comment to add to

01:10:33.200 --> 01:10:36.669
your Bay Area
conjugate fault database.

01:10:37.680 --> 01:10:42.080
In addition to the Wight Way Fault,
if you look on the west – to the west

01:10:42.080 --> 01:10:47.360
of the Maacama Fault, the south side
of Ukiah Valley, there is a zone of

01:10:47.360 --> 01:10:51.176
faulting that goes through Anderson
Valley in the Boonville area.

01:10:51.200 --> 01:10:56.216
You might want to take a look at
my poster [laughs] that’s hanging up

01:10:56.240 --> 01:11:00.640
that that fault is shown.
And the other fault is in the Clear Lake

01:11:00.640 --> 01:11:06.056
area, and it would be conjugate
to the Bartlett Springs, I believe.

01:11:06.080 --> 01:11:10.240
And that would be the
Cross Springs Fault, which bounds

01:11:10.240 --> 01:11:16.456
the Cache Formation by
a Pleistocene basin there.

01:11:16.480 --> 01:11:20.224
Those are two to add to
your database. [laughs]

01:11:20.224 --> 01:11:22.560
- Great. I’ll definitely
come to your poster.

01:11:22.560 --> 01:11:25.635
And everybody else should also.

01:11:27.338 --> 01:11:32.313
- Any other questions?
Questions for Ken or Julian?

01:11:35.250 --> 01:11:40.080
- So this is a question for Julian.
Thanks for the – thanks for showing

01:11:40.080 --> 01:11:44.640
the modeling, and that actually I
guess those conjugate ruptures are

01:11:44.640 --> 01:11:49.440
sort of not as intimately related
as they might appear, right?

01:11:49.440 --> 01:11:53.440
The stress changes are
more subtle and nuanced.

01:11:53.440 --> 01:11:57.680
The question is, you talk about the
different stresses resolved on these

01:11:57.680 --> 01:12:01.336
faults, right? You required these
different shear and normal stresses.

01:12:01.360 --> 01:12:05.040
The first part is, are those different
stress states on the faults compatible

01:12:05.040 --> 01:12:10.136
with a single – or, relative
uniform regional stress field?

01:12:10.160 --> 01:12:12.240
And then the next is more
speculative or hypothetical.

01:12:12.240 --> 01:12:15.200
What if the faults were not
at sort of 90-degree angles

01:12:15.200 --> 01:12:21.360
but at a more conventional
60- or 30-degree – you know,

01:12:21.360 --> 01:12:24.335
at a more
conventional conjugating?

01:12:25.303 --> 01:12:28.880
- So, for the first question –
so we didn’t change the orientation

01:12:28.880 --> 01:12:30.960
of the stress field.
So, even though we had different

01:12:30.960 --> 01:12:35.256
amounts of stress, it was still
at that north-7-east.

01:12:35.280 --> 01:12:39.920
So it’s – I mean, it makes more sense
to assume just different stress

01:12:39.920 --> 01:12:44.088
on different faults than the same
amount of stress everywhere anyway.

01:12:44.088 --> 01:12:47.496
We kept the orientation the same.
It was just the amplitude.

01:12:47.520 --> 01:12:50.240
And that’s what we needed to match.
But, as for the second question –

01:12:50.240 --> 01:12:55.280
so that’s – I haven’t tested that yet
because it was specifically

01:12:55.280 --> 01:12:58.480
a Ridgecrest model, but it’s very
much something I’m interested in.

01:12:58.480 --> 01:13:01.760
because I anticipate you would have
some of the same general behaviors

01:13:01.760 --> 01:13:05.600
because you still have faults
with a different sense of motion

01:13:05.600 --> 01:13:08.560
intersecting each other.
You would still have that clamping

01:13:08.560 --> 01:13:12.000
and unclamping, and you would still
have the stress shadowing associated

01:13:12.000 --> 01:13:16.560
with one fault dropping its stress.
But, in terms of how those affects

01:13:16.560 --> 01:13:19.840
would matter based on angle,
that’s – I can’t tell you yet.

01:13:19.840 --> 01:13:21.734
I want to model it.

01:13:22.560 --> 01:13:25.080
Sorry that’s a non-answer.

01:13:26.724 --> 01:13:29.203
- Any other questions?

01:13:31.840 --> 01:13:38.240
- Maybe I just haven’t heard anything,
but do your models give you any idea

01:13:38.240 --> 01:13:41.600
of what these earthquakes may mean
for a potential earthquake

01:13:41.600 --> 01:13:44.560
on the Garlock Fault?
- So that is something

01:13:44.560 --> 01:13:48.400
we did look at in our paper. I figured
I didn’t want to talk about that

01:13:48.400 --> 01:13:52.720
here today because I was asked to
talk about cross-fault interactions.

01:13:52.720 --> 01:13:56.560
But we did do, actually, some models
where we took the stress changes from

01:13:56.560 --> 01:14:01.040
our dynamic models of the Ridgecrest
sequence and resolved them under

01:14:01.040 --> 01:14:05.520
the Garlock Fault. And so you see
the similar pattern of, where you would

01:14:05.520 --> 01:14:09.360
see some stress increase at the end 
of the right-lateral, but also you see

01:14:09.360 --> 01:14:14.000
some decrease from the sense of slip.
And so basically, long story short,

01:14:14.000 --> 01:14:17.600
and I can show anyone figures later
if they want, we saw that, basically,

01:14:17.600 --> 01:14:23.200
in order to get an increased chance
of rupture, you have to have both

01:14:23.200 --> 01:14:25.680
increased shear stress
and decreased normal stress.

01:14:25.680 --> 01:14:28.720
And the area of the Garlock Fault
that had that in our models

01:14:28.720 --> 01:14:31.440
was actually pretty small.
And it lines up with where the

01:14:31.440 --> 01:14:34.880
triggered slip and aftershocks were.
So that matches nicely.

01:14:34.880 --> 01:14:39.280
But, for the most part, we hesitate to
say anything because we don’t know

01:14:39.280 --> 01:14:42.160
what the pre-stress was on the Garlock.
And the nice thing about doing this

01:14:42.160 --> 01:14:45.120
with a dynamic model is we didn’t
have to assume the pre-stress.

01:14:45.120 --> 01:14:47.896
We didn’t have to assume
the coefficient of friction.

01:14:47.920 --> 01:14:50.640
So there’s a couple of spots that
look like they might be closer

01:14:50.640 --> 01:14:52.720
just from that – a couple
that look like they’re less close.

01:14:52.720 --> 01:14:55.120
I think ultimately, it’s going to
be one of those, how ready

01:14:55.120 --> 01:14:57.416
was the Garlock already?

01:14:57.440 --> 01:15:00.800
I think definitely a lot of people
have been talking about things like

01:15:00.800 --> 01:15:03.760
the Blackwater Fault to the south,
which is a right-lateral structure,

01:15:03.760 --> 01:15:08.560
being more likely. And it just had
a 4.5 last week, but who knows.

01:15:08.560 --> 01:15:12.280
And I’m not going to say anything
about that, either. [chuckles]

01:15:13.192 --> 01:15:16.033
- Any other questions?
Ah.

01:15:20.720 --> 01:15:30.320
- Julian, you spoke about having to
kick off the 6.4 with a certain stress pop

01:15:30.320 --> 01:15:36.560
that you imposed upon one place.
And that gave you the 6.4, and then

01:15:36.560 --> 01:15:40.376
you go through the
dynamic rupture calculations.

01:15:40.400 --> 01:15:46.856
And then you get to the 7.1,
and you have to kick that off too.

01:15:46.880 --> 01:15:51.920
To what extent – well, what are those
numbers, and to what extent did the

01:15:51.920 --> 01:15:59.176
dynamic rupture help or
hinder the 7.1, in terms of

01:15:59.200 --> 01:16:03.920
what you needed to do there?
- So we started the rupture the

01:16:03.920 --> 01:16:07.760
same way in both cases, where
we’re just raising the shear stress

01:16:07.760 --> 01:16:13.520
to 10% about the yield stress
over a patch that was – we did a –

01:16:13.520 --> 01:16:16.000
like, a 2.5-kilometer
radius for the patch.

01:16:16.000 --> 01:16:21.680
So it equals about a – like,
a 5.6 or 5.4 – I forget exactly.

01:16:21.680 --> 01:16:24.320
It’s a mid-5s earthquake
nucleation zone.

01:16:24.320 --> 01:16:29.520
So that’s kind of – with this method,
it is a little bit brute force, but because

01:16:29.520 --> 01:16:32.880
this doesn’t run interseismic or
nucleation, it’s kind of what

01:16:32.880 --> 01:16:37.280
you’d have to do to get it started.
Certainly, our patch for the 6.4 was

01:16:37.280 --> 01:16:40.720
on the end of the fault and didn’t –
you know, once the rupture gets started,

01:16:40.720 --> 01:16:46.480
it is running on its own, and we –
and similarly, with the 7.1, actually,

01:16:46.480 --> 01:16:51.040
the documented hypocenter is
pretty far from the junction point.

01:16:51.040 --> 01:16:54.400
So, by that point, the rupture
was propagating on its own.

01:16:54.400 --> 01:16:57.440
So nucleation is always
an issue in dynamic models.

01:16:57.440 --> 01:17:01.360
And there’s different
approaches to it, but in this case,

01:17:01.360 --> 01:17:04.856
a rupture was going for a while
before it hit the intersection.

01:17:04.880 --> 01:17:07.532
Hope that answers
your question.

01:17:10.119 --> 01:17:15.680
- Okay. Any other burning questions?
Otherwise, I’d like to thank all of our

01:17:15.680 --> 01:17:19.200
speakers again for an awesome session.
[Applause]

01:17:19.200 --> 01:17:22.697
And I believe it’s time for lunch.

01:17:24.626 --> 01:17:26.856
- Okie dokie. A couple of
quick announcements.

01:17:26.880 --> 01:17:29.680
There is a fire engine
demonstration downstairs.

01:17:29.680 --> 01:17:33.600
And let’s call lunch, which is morphed
on the agenda as lunch plus

01:17:33.600 --> 01:17:37.040
poster session as, instead,
lunch plus fire engine session.

01:17:37.040 --> 01:17:40.320
And, if you have a chance,
go down and explore that.

01:17:40.320 --> 01:17:43.120
If you have paid for lunch,
lunch is ready for you outside.

01:17:43.120 --> 01:17:46.080
If you have not paid for lunch,
please don’t eat the lunch. [chuckles]

01:17:46.080 --> 01:17:53.033
And come on back, and we will pick up
again at 1:15 with lightning talks.

01:17:55.135 --> 01:18:01.769
[inaudible background conversations]