WEBVTT
Kind: captions
Language: en-US

00:00:00.765 --> 00:00:02.880
[silence]

00:00:02.880 --> 00:00:05.120
Hi, everyone.
My name is Matt Tarling.

00:00:05.120 --> 00:00:07.736
I’m a postdoc at
McGill University.

00:00:07.760 --> 00:00:13.440
And today we’ll be going a little further
afield than northern California, looking

00:00:13.440 --> 00:00:18.720
at earthquake rupture signatures in a
serpentinite shear zone in New Zealand.

00:00:18.720 --> 00:00:22.616
So fossil earthquakes
in serpentinite.

00:00:22.640 --> 00:00:26.560
I’ll be presenting what we believe
to be the first reported evidence

00:00:26.560 --> 00:00:30.616
for an earthquake signature in
a natural serpentinite shear zone

00:00:30.640 --> 00:00:33.920
and then briefly looking at some
numerical modeling which confirms the

00:00:33.920 --> 00:00:38.857
hypothesis and sets some bounds on
the earthquake magnitude responsible.

00:00:39.680 --> 00:00:44.400
So serpentinites are ultramafic rocks
that form from the hydration

00:00:44.400 --> 00:00:46.800
of mantle material,
such as peridotite.

00:00:46.800 --> 00:00:50.000
So this occurs wherever there’s
the opportunity for water to interact

00:00:50.000 --> 00:00:54.320
with anhydrous ultramafics.
This occurs in subduction zones,

00:00:54.320 --> 00:01:00.136
in transform faults, both oceanic and,
in some instances, continental,

00:01:00.160 --> 00:01:05.656
as well as mid-ocean ridges
and ophiolites, among others.

00:01:05.680 --> 00:01:09.840
There, however, remain some
pretty big questions with regard to

00:01:09.840 --> 00:01:13.520
serpentinite’s slip behavior,
which we need to answer if we

00:01:13.520 --> 00:01:20.296
want to really properly understand
the rheology of these tectonic settings.

00:01:20.320 --> 00:01:26.240
One important observation that’s
being made is this association between

00:01:26.240 --> 00:01:31.200
the presence of frictionally weak
serpentinite and serpentinite-associated

00:01:31.200 --> 00:01:37.576
minerals and active
creeping fault segments.

00:01:37.600 --> 00:01:43.920
Which raises an interesting question.
Do earthquake ruptures nucleate or

00:01:43.920 --> 00:01:49.440
propagate through serpentinite-bearing
fault segments in shear zones?

00:01:49.440 --> 00:01:52.400
And then, from the perspective
of the field geologist, if we look at

00:01:52.400 --> 00:01:56.376
an exhumed example of
a serpentinite shear zone,

00:01:56.400 --> 00:01:58.160
what would we find
in the rock record?

00:01:58.160 --> 00:02:03.896
What would be the signature of
a fossil earthquake in serpentinite?

00:02:03.920 --> 00:02:11.920
And, more broadly speaking,
in most rocks, the signature of

00:02:11.920 --> 00:02:16.880
an earthquake is either related to
a thermal process due to coseismic

00:02:16.880 --> 00:02:22.536
frictional heating or a mechanical
process, such as pulverization

00:02:22.560 --> 00:02:26.960
or cataclasis. But really the golden
standard when it comes to earthquakes

00:02:26.960 --> 00:02:31.360
in the rock record are frictional
melts known as pseudotachylytes.

00:02:31.360 --> 00:02:35.920
We have an example here where
this black line is the fault surface,

00:02:35.920 --> 00:02:40.400
and the black rock is pseudotachylyte.
So a frictional melt that occurred

00:02:40.400 --> 00:02:45.297
due to heating during the earthquake.
And here we have an injection vein.

00:02:46.160 --> 00:02:50.936
However, pseudotachylyte has
never been found in serpentinite.

00:02:50.960 --> 00:02:55.840
But we can get some clues
from experimental studies.

00:02:55.840 --> 00:03:00.856
So there are many high-velocity shear
experiments which have been done

00:03:00.880 --> 00:03:08.320
at seismic slip rates which can give us
some hints as to what a fossil

00:03:08.320 --> 00:03:11.520
earthquake might look like,
at least from a lab perspective.

00:03:11.520 --> 00:03:15.040
And what they show when we look at
the products of these experiments

00:03:15.040 --> 00:03:20.240
is that the serpentinite dehydrates,
as it might be expected,

00:03:20.240 --> 00:03:25.496
converting back into those primary
ultramafic minerals, such as olivine,

00:03:25.520 --> 00:03:31.096
as well as amorphizing and,
in some instances, possibly melting.

00:03:31.120 --> 00:03:37.336
So, equipped with that information
we’ll turn to the field and look at

00:03:37.360 --> 00:03:40.960
a natural example of an exhumed
serpentinite shear zone in New Zealand.

00:03:40.960 --> 00:03:46.720
So this is the Livingstone Fault.
The Livingstone Fault is a trans-crustal

00:03:46.720 --> 00:03:53.280
plate boundary that extends over
500 kilometers from the South Island,

00:03:53.280 --> 00:03:56.320
cut by the Alpine Fault, and then
through to the North Island.

00:03:56.320 --> 00:04:02.480
And it juxtaposes the continental
quartzofeldspathic schists here

00:04:02.480 --> 00:04:09.040
on the right against the ultramafic
Dun Mountain ophiolite,

00:04:09.040 --> 00:04:11.440
so peridotite and
serpentinized peridotite.

00:04:11.440 --> 00:04:15.200
That’s these brown-reddish
rocks on the left.

00:04:15.200 --> 00:04:18.480
And then, between the two,
we have the Livingstone Fault.

00:04:18.480 --> 00:04:23.760
So the Livingstone Fault is
characterized by a scaly serpentinite

00:04:23.760 --> 00:04:30.800
mélange that extends up to about
400 meters wide and is really

00:04:30.800 --> 00:04:36.080
characterized by a pervasive
anastomosing foliation.

00:04:36.080 --> 00:04:41.280
So this is a foliation where we have
slip surfaces at the millimetric

00:04:41.280 --> 00:04:48.720
up to about the meter scale defining
lenticular domains of serpentinite.

00:04:48.720 --> 00:04:51.920
And it’s pretty consistent
and pervasive all the way through

00:04:51.920 --> 00:04:53.816
this package of rock.

00:04:53.840 --> 00:04:58.960
Embedded within that –
within that anastomosing matrix,

00:04:58.960 --> 00:05:02.800
we have larger lithons, which
you can see here along the ridge.

00:05:02.800 --> 00:05:07.656
We’re looking down here
at a mountain pass exposure.

00:05:07.680 --> 00:05:10.720
So we have these lithons that are
forming spines along the ridge,

00:05:10.720 --> 00:05:15.120
and they consist of massive
serpentinite lenses as well as rodingite,

00:05:15.120 --> 00:05:21.096
a type of metagabbro
that forms in serpentinite.

00:05:21.120 --> 00:05:24.400
Here’s one of these rodingite pods.
So they tend to be self-similar

00:05:24.400 --> 00:05:30.960
with the fabric, so lenticular-shaped.
And they’re embedded within this

00:05:30.960 --> 00:05:35.520
serpentinite shear zone and typically
tend to be coated with serpentinite or

00:05:35.520 --> 00:05:42.776
even characterized by slip
surfaces that go around them.

00:05:42.800 --> 00:05:46.677
And these slip surfaces are
something pretty interesting.

00:05:47.520 --> 00:05:52.160
About the Livingstone Fault and
probably most serpentinite shear zones

00:05:52.160 --> 00:05:58.080
is that – is that we often find that,
contrasting with this broad distribution

00:05:58.080 --> 00:06:01.760
of strain in this scaly serpentinite,
there are a great number of

00:06:01.760 --> 00:06:04.376
sharp, discrete
fault surfaces.

00:06:04.400 --> 00:06:08.000
Some are which – some of which
are along the margins of pods,

00:06:08.000 --> 00:06:13.016
or in some cases, they cut through
that scaly anastomosing fabric.

00:06:13.040 --> 00:06:15.760
So we have an example here
on the bottom right of

00:06:15.760 --> 00:06:18.400
a slip surface
on a rodingite.

00:06:18.400 --> 00:06:26.616
And, on the left, this is a polished slip
surface of the edge of a chromitite pod.

00:06:26.640 --> 00:06:32.080
One of these sharp slip surfaces that
we find both on the margins of pods

00:06:32.080 --> 00:06:36.640
and cutting this anastomosing
fabric that are particularly interesting

00:06:36.640 --> 00:06:42.136
are magnetite-coated,
or magnetite-bearing slip surfaces.

00:06:42.160 --> 00:06:46.640
We have an example here where
we have patchy magnetite on the

00:06:46.640 --> 00:06:52.456
edge of a massive serpentinite pod.
And if we take a closer look,

00:06:52.480 --> 00:06:58.160
we find that the – we have here
at the top in this photomicrograph,

00:06:58.160 --> 00:07:03.200
we have that magnetite slip surface,
that polished surface being the top.

00:07:03.200 --> 00:07:05.760
And, below it, we have that
serpentinite where we have that

00:07:05.760 --> 00:07:11.840
characteristic anastomosing fabric.
So these magnetite veins – in this case,

00:07:11.840 --> 00:07:16.320
these little veinlets, delineate the
lenticular domains of serpentinite.

00:07:16.320 --> 00:07:22.456
And, if we take a much closer
look again at the magnetite layer,

00:07:22.480 --> 00:07:25.360
so on that
principal slip surface.

00:07:25.360 --> 00:07:29.120
This is that polished surface that we
were looking at in the previous photos.

00:07:29.120 --> 00:07:36.160
We find that we have a magnetite layer
of about – of a few hundred microns,

00:07:36.160 --> 00:07:39.040
maybe up to a millimeter in
some places, which has

00:07:39.040 --> 00:07:44.776
a great number of inclusions
within that magnetite surface.

00:07:44.800 --> 00:07:48.984
And it’s these inclusions that we’re
going to take a closer look at.

00:07:49.760 --> 00:07:53.360
And the reason being is that it
appears that these inclusions,

00:07:53.360 --> 00:07:57.576
within the magnetite layers,
preserve evidence for progressive

00:07:57.600 --> 00:08:03.200
amorphization and dehydration,
which we interpret as being –

00:08:03.200 --> 00:08:07.896
as having occurred due to the passage
of a coseismic thermal pulse.

00:08:07.920 --> 00:08:11.600
So we’ll start – in this little diagram,
we’re looking at the top of the slip

00:08:11.600 --> 00:08:15.440
surface here, and then going further
away from the slip surface into

00:08:15.440 --> 00:08:17.120
the surrounding
serpentinite.

00:08:17.120 --> 00:08:22.776
So we’ll start in the surrounding
serpentinite below the slip surface.

00:08:22.800 --> 00:08:27.680
And what we find here is –
well, it’s serpentine that is pretty

00:08:27.680 --> 00:08:32.136
characteristic of most of the fault.
It’s crystalline serpentine.

00:08:32.160 --> 00:08:35.256
So we’re looking at
lizardite and chrysotile.

00:08:35.280 --> 00:08:39.360
And it’s pretty typical of the ambient
deformation temperature of the

00:08:39.360 --> 00:08:42.936
Livingstone Fault, so somewhere
around 300 degrees.

00:08:42.960 --> 00:08:49.200
Now, stepping into that magnetite layer,
we find something a little bit different.

00:08:49.200 --> 00:08:54.640
The same lizardite and chrysotile
are present, but there appear to be

00:08:54.640 --> 00:08:58.296
poorly crystalline or
even amorphous, in some cases.

00:08:58.320 --> 00:09:02.080
And, if this was due to a thermal
process, this probably would indicate

00:09:02.080 --> 00:09:05.680
a temperature of about 570 degrees.
So these temperatures that

00:09:05.680 --> 00:09:09.680
I’m giving you here are
from a compilation of temperatures

00:09:09.680 --> 00:09:14.740
reported in the literature
that I’ve averaged out.

00:09:15.440 --> 00:09:20.056
Taking one step further towards that
principal slip surface, so further up

00:09:20.080 --> 00:09:25.360
into that magnetite layer, we find
the appearance of interlayered talc

00:09:25.360 --> 00:09:29.600
and serpentine, along with amorphous
serpentine and poorly crystalline

00:09:29.600 --> 00:09:33.496
serpentine, indicating that serpentine
is likely breaking down.

00:09:33.520 --> 00:09:39.096
This is indicative of a temperature
of probably around 620 degrees.

00:09:39.120 --> 00:09:41.840
Another step further towards
that principal slip surface,

00:09:41.840 --> 00:09:46.856
we see the appearance of poorly
crystalline nanogranular olivine,

00:09:46.880 --> 00:09:50.880
which co-exists with
amorphous serpentine

00:09:50.880 --> 00:09:54.616
as well as that talc
that we saw previously.

00:09:54.640 --> 00:09:56.320
If this is due to
a thermal process,

00:09:56.320 --> 00:10:01.496
this indicates a temperature
of about 700 degrees.

00:10:01.520 --> 00:10:04.480
Yet another step towards
that principal slip surface,

00:10:04.480 --> 00:10:09.496
and the talc that we saw
previously is no longer present.

00:10:09.520 --> 00:10:13.440
And the assemblage largely
consists of moderately crystalline

00:10:13.440 --> 00:10:17.016
olivine and
amorphous material.

00:10:17.040 --> 00:10:20.640
This assemblage, taken with the
disappearance of talc, presumably due

00:10:20.640 --> 00:10:28.056
to thermal dehydration, would indicate
a temperature of about 800 degrees.

00:10:28.080 --> 00:10:32.000
If we go yet another step towards
that principal slip surface,

00:10:32.000 --> 00:10:38.000
we now find that these inclusions
contain almost exclusively

00:10:38.000 --> 00:10:42.616
fairly well-crystalline
nanogranular olivine.

00:10:42.640 --> 00:10:49.200
So we’re looking at nanograins
of olivine in the 20- to 200-nanometer

00:10:49.200 --> 00:10:55.896
range. And this is indicative of a
temperature of about 833 degrees.

00:10:55.920 --> 00:11:00.320
And finally, within microns of
that top slip surface, we find

00:11:00.320 --> 00:11:04.320
the appearance of enstatite.
So now the assemblage consists

00:11:04.320 --> 00:11:10.000
of fairly well-crystalline nanogranular
olivine along with enstatite, which,

00:11:10.000 --> 00:11:13.760
if occurred due to a thermal process
from the dehydration of serpentine,

00:11:13.760 --> 00:11:19.159
would indicate a temperature
of about 925 degrees.

00:11:20.080 --> 00:11:23.920
These observations appear to
be consistent with the

00:11:23.920 --> 00:11:27.360
progressive amorphization
and dehydration of serpentine.

00:11:27.360 --> 00:11:32.240
So, at lower temperatures, serpentine
dehydrates to talc and olivine, releasing

00:11:32.240 --> 00:11:39.172
water, while at higher temperatures,
it dehydrates to forsterite and enstatite.

00:11:40.080 --> 00:11:44.640
If we compare these observations
made on this natural example,

00:11:44.640 --> 00:11:48.480
we find that they’re strikingly similar,
at least in appearance, to those that

00:11:48.480 --> 00:11:50.640
we saw from those
high-temperature experiments.

00:11:50.640 --> 00:11:54.880
So this nanogranular olivine we
have in the natural rock compared to

00:11:54.880 --> 00:11:59.120
the nanogranular olivine in the
high-velocity shear experiments

00:11:59.120 --> 00:12:04.529
as well as that enstatite
which co-exists with olivine.

00:12:05.360 --> 00:12:10.080
And so here we have a cartoon
representation of this magnetite layer.

00:12:10.080 --> 00:12:14.640
So the central part is that magnetite.
On either side, we have serpentinite.

00:12:14.640 --> 00:12:17.440
And then we have that principal
slip surface right here.

00:12:17.440 --> 00:12:22.640
And I’ve placed each of these inclusion
assemblages as a function of their

00:12:22.640 --> 00:12:27.840
distance from the slip surface as well as
the coseismic temperature that they

00:12:27.840 --> 00:12:31.360
could potentially represent.
And we see that they form what

00:12:31.360 --> 00:12:34.960
looks to be a nice curve going
away from that slip surface.

00:12:34.960 --> 00:12:39.120
And, in order to get a sense of
whether that might be consistent with

00:12:39.120 --> 00:12:43.280
an earthquake rupture, I performed
some numerical modeling.

00:12:43.280 --> 00:12:46.856
Now, I’m not going to go into
intimate detail into the modeling.

00:12:46.880 --> 00:12:50.960
It suffices to say that it was important
to take into account all of the physical

00:12:50.960 --> 00:12:55.440
processes that occur during frictional
heating and phase transformation

00:12:55.440 --> 00:12:59.760
of serpentinite. So we have the
frictional heating itself, the dehydration

00:12:59.760 --> 00:13:04.000
of serpentinite, there’s an energy cost
to that phase change, the production

00:13:04.000 --> 00:13:09.440
of porosity, the release of water,
and the pressurization –

00:13:09.440 --> 00:13:12.080
the thermal pressurization
of that water, which in turn

00:13:12.080 --> 00:13:15.496
reduces the effectiveness
of frictional heating.

00:13:15.520 --> 00:13:24.000
Now, running that model for a range
of slip distances, we find that this

00:13:24.000 --> 00:13:28.320
curve that the inclusion assemblages
represents going away from the

00:13:28.320 --> 00:13:34.640
principal slip surface is pretty well
constrained by the thermal maxima

00:13:34.640 --> 00:13:38.640
that might occur due to
coseismic frictional heating.

00:13:38.640 --> 00:13:40.800
So that’s this red line here.

00:13:40.800 --> 00:13:46.536
And this sets a nice lower bound,
so magnitude 2.7.

00:13:46.560 --> 00:13:52.056
So that’s a pretty reasonable lower
bound to the assemblages observed.

00:13:52.080 --> 00:13:58.056
And the upper bound is pretty well
constrained by a magnitude 4.

00:13:58.080 --> 00:14:02.800
That is the thermal – the thermal
maximum due to a magnitude 4

00:14:02.800 --> 00:14:05.920
earthquake. So I think it’s fairly
convincing that what we’re looking at

00:14:05.920 --> 00:14:13.040
here truly is an earthquake fossil –
a fossil earthquake due to

00:14:13.040 --> 00:14:17.496
an earthquake in a
serpentinite shear zone.

00:14:17.520 --> 00:14:24.080
And so, to conclude, I do believe that
we have fairly convincing evidence

00:14:24.080 --> 00:14:29.760
for earthquake rupture and coseismic
dehydration of serpentine in a natural –

00:14:29.760 --> 00:14:36.240
in a natural serpentinite fault.
And, on this basis, I would suggest that

00:14:36.240 --> 00:14:40.160
you may not be able to discount the
possibility that an earthquake

00:14:40.160 --> 00:14:45.096
could propagate through a
serpentinite-bearing shear zone

00:14:45.120 --> 00:14:49.576
based on the mere presence of
serpentinite itself. Thank you.

00:14:49.600 --> 00:14:56.120
If you’d like to find out more,
please do go read our paper.

00:14:58.142 --> 00:15:02.800
[silence]