PubTalk 10/2016 — Rockfalls in California's Sierra Nevada

Video Transcript
Download Video
Right-click and save to download

Detailed Description

Rock falls in California’s Sierra Nevada - Pursuing explanations for exfoliation and seemingly spontaneous fracture of rock


Date Taken:

Length: 01:09:49

Location Taken: Menlo Park, CA, US


[ Silence ]

Well, good evening, and welcome
to our public lecture for this month.

My name
is Mark Reid.

I’m a landslide researcher
here in Menlo Park.

And before tonight’s presentation,
I’d like to just mention

next month’s presentation.
There’s a flier on it in the back.

It’ll be given by Keith Miles.
He’s the director of the

Western Ecological Research
Center in Sacramento.

The title of his talk is Ecological
Stressors – It’s a Lot of WERC.

W-E-R-C – it’s a pun.

Science pun, I guess.

Anyway, tonight’s presentation
is by Brian Collins.

I’ve had the good fortune to
work with Brian over the years.

But he’s a research civil engineer
with the landslide hazards program

here in Menlo Park.
He’s worked on a lot of different hazards

projects for the past 20 years,
including 10 years with the USGS.

He currently leads a variety of
programs and projects ranging from

predicting shallow landslides and
debris flows here in the Bay Area

to understanding the mobility
of the deep-seated 2014 Oso slide.

I’ve worked with him on that
up in Washington.

And he’s looked at studying
rockfall hazards in Yosemite.

Now, Brian has a degree
from Purdue University,

a master’s from CU – University of
Colorado in Boulder, and a Ph.D.

in geotechnical engineering from the
University of California-Berkeley.

And he’s a registered
professional engineer in California.

But I think importantly,
Brian is a rock climber.

He’s climbed many of
Yosemite’s iconic faces.

And so that background helps
him pursue his research here.

So I’ll – with that, I’ll let
Brian try to inform you on that.

[ Applause ]

- Thank you.

[ Applause ]

Thanks, Mark.
And are we good on this?

Good. Okay. Thanks a lot.
And thank you all

for coming out tonight.
I know it’s rainy, but hopefully

I’ll keep you entertained with
stories of rockfalls in the Sierra.

This is one of my favorite
subjects to study.

As Mark pointed out,
I am a – I do research

on a number of different
landslide hazard projects.

This one’s been a favorite of mine,
partially because I get to work in

Yosemite, which is a beautiful place
to be, but also because it’s

quite fascinating to take all the
different aspects of science

and apply them to what ends up
being a fairly complex problem,

and that is looking at
rockfall triggering.

So today we’re going
to be looking at

rockfalls that are specifically caused
from exfoliation-type processes.

And as I mentioned,
the triggering processes –

what leads to the
detachment of the rockfalls.

- Can't hear you.

- This is, I think, the voice that I need
to speak on for the – back there, so …

Now, our studies in the
Sierra Nevada are –

obviously Yosemite
gives us that aspect.

We’re also going to be looking at –
and I’m going to share some

research from
Twain Harte.

Twain Harte is located on the way
to Sonora Pass, east of Sonora.

And so we’ll be looking at
some of those as other examples

of exfoliation processes.

So we’re going to start with
a blast here, and this is Twain Harte.

As I mentioned, Twain Harte
is on Highway 108.

And this granitic dome that you see
in the foreground forms the

left abutment of a 30-foot-tall concrete
arch dam that holds a small reservoir.

And in 2014, this dome
spontaneously fractured.

And some of you may
have seen this on YouTube.

I’m going to share
a video here of that.

One of our colleagues and a consultant
on the job to investigate this failure,

Scott Lewis,
took this video.

And it’s quite spectacular, so I’m
going to play it a couple times.

[inaudible speaking from video]

[laughter from audience]

So let me start again, and I’ll
pause it here so you can ..

pay attention to this area.
That’s the area that fractured.

And this is not – there’s no
dynamite involved in this.

These are people out there who
heard popping, loud cracking noises,

ran out there with cameras,
and happened to capture this.

So this is not a anthropogenic-type thing.
This is just, wow, what happened.

[laughter from video]

And you see
one more time.

Now, it turns out that
that was some of the

smaller exfoliation that
happened at this location.

This was the third event
out of a sequence of five.

And the first two really damaged
another part of the dome.

And so that’s why these people were
keen on being out there, so they knew

that it had been cracking days prior.
And so when they heard it again,

they said, let’s get out there and see how
it works, and they were pretty close to it.

These popping noises had been
happening for a couple minutes,

and so that’s why they were
chasing around with cameras,

like, where’s it coming from,
where’s it coming from.

So I’ll start off by defining exfoliation
as a process that’s forming these

rock joints in the Sierra Nevada
and other places in the world.

So exfoliation is common in
granitic rock specifically,

but it can form in basically all rock
types, and we see evidence for that in

things like the Navajo sandstone of
Utah and metamorphic rocks as well.

So this process is how
surface-parallel joints form.

So surface-parallel fractures –
it’s that formation.

And you see that here.
This is a side of Half Dome.

The cables route is just
over here off of the picture.

And then this is the northwest face –
the steep side of Half Dome.

So the exfoliation joints are typically
curved. You see that here in this aspect.

And they’re thicker and
more widely spaced at depth.

So what you see here is the surface.
This is the shallow part of it.

And they’re very thin,
and they’re thinly spaced.

People have found
exfoliation-type features

down to about 100 meters, I think.
Then they typically do die out.

Geologically, they’re young.
In the Sierra, they’re post-glacial,

so that means they’re
less than 10,000 years old,

which in geological time,
is very, very young.

So we’ll look at that.

And as I mentioned, we’re going to
look at exfoliation in the Sierra Nevada.

But I should say that the
processes that we’re talking about are –

occur worldwide.
And I have a couple of examples here.

One exfoliation feature
is down low here.

Some nice exposures
of exfoliation-type sheets.

And then this is Cannon Cliff in
New Hampshire on the East Coast.

And all this area is great
exfoliation examples.

So we see those examples here in
the U.S. and also around the world.

So why do rocks
exfoliate in the first place?

Why are exfoliation joints there?
That is a very complicated issue.

And I have a fairly wordy, complicated
slide to just start talking about that.

But we’re not actually going to –
I’m not going to be talking too much

about the initiation and the
formation of exfoliation joints.

Instead, I’m going to be focused on
what happens once they’re formed

and how do they finally detach
into rockfall processes?

That said, I think it’s important to
at least point out some background

on the complex history and the
complex research that’s gone into

trying to explain why we have
dome structures in Tuolumne Meadows

and big, huge slab failures
like this in the Yosemite Valley.

The first observations – first written
accounts, at least, come from Germany

in the early 1800s where they were
just trying to explain why these

mountains have these domal-type shapes
and have these layering-type structures.

And in California,
closer to home,

Whitney, in the mid-1800s,
tried to explain that.

And a multitude of different processes
has been suggested for these.

Whitney suggested that the exfoliation
joints were there when the pluton –

when the Sierra Nevada actually –
and the batholith cooled,

when the
granite cooled.

And then the cooling process
caused these fractures to form.

That’s pretty much been dismissed
because the cooling was at depth,

and if it cooled, that should
happen for a lot larger depth.

And these granites have now been
exhumed all the way to the surface.

And we know that these –
for the most part, these joint sets

are younger than the age of the pluton
being hundreds of millions of years,

where here we know that
they’re less than 10,000 years.

So some of these theories
have been proposed

and dismissed by
various researchers.

A early researcher in the 1800s
suggested that temperature

might have something
to do with it.

Those were observations made by Shaler
of the East Cost of the United States

in New England who just wrote
a paper that basically suggested that,

because the temperatures on the
coast of Maine are very, very cold

and it gets very, very hot
in the summertime,

that that might be suggestive that
the temperature is playing a role.

We have a multitude of
other reasons that have

more to do with the stresses in the
rock and the stresses in the surface.

Gilbert suggested that the removal
of overburden and the removal

of glaciers from the valleys in the
Sierra might be leading the rocks to

flex outwards to go into tension,
and that could be the cause of that.

Similarly, Jahns on the East Coast
looking at quarries – rock quarries –

granite quarries in New England
suggested that, when you remove

the rock on the surface,
that was related to the opening up

of these
exfoliation joints.

We have examples from
Australia as well that suggest

that the rock is in some form of
compression and that the compressive

forces are actually what’s forcing
these slabs up off the ground.

More recently, a researcher that I work
with in Hawaii – I haven’t been there,

but he’s based at the University of
Hawaii – Steve Martel suggests that,

if you have a curved surface in
compression, you can actually get

a tensile force that’s strong enough
to open up an exfoliation joint.

And so this is sort of rewriting
a lot of the – or revisiting a lot of

the research that’s gone
into this over the past 200 years.

So the debate still continues.
People have their own theories,

and people have dismissed and
projected and suggest other theories.

But it’s a
really active field.

And I think some of the research that’s
coming out, particularly by Steve Martel,

is going to help us explain
exfoliation even more.

And that’s the formation story.
So I think that there’s a lot of things

that could be causing exfoliation
joints to form in the first place.

It’s probably
a complex story.

And fortunately, I’m going to
just side-step that whole issue.

And we’ll move on instead to, not what
makes these beautiful landscapes in the

first place, but – oops – but what causes
the fracturing once they’re formed.

So here is an example.

This is Charlotte Dome in the southern
Sierra in Kings Canyon National Park.

And we have these
beautiful exfoliation sheets.

And the question is, at what point is
this sheet going to fall down?

Or this sheet, or this sheet, or this sheet,
or this sheet, and so on and so forth?

So the joints are already there. The
exfoliation sheets are already formed.

But they’re somehow
still attached to the rock.

What’s it going to
take to cause a rockfall?

And why is that important?
Well, when we have lots of visitors

to Yosemite National Park and
there’s rockfalls at the same time,

then that poses a hazard and
considerable risk to the public.

And so this has been work
that I’ve gotten involved with,

working with Yosemite National Park
over the past seven years.

And working in particular with the first
national park geologist, Greg Stock, who

has spearheaded many, many different
types of studies on rockfall in the park.

So this confluence of
the rockfalls and the public

gives us the
motivation for studies.

And particularly for trying to do
something to make headway

to look at the hazard and risk
for situations like this.

This is a cabin – hard-sided cabin that
got hit by a very large boulder in 2008.

The people that were in this cabin
had woken up early for breakfast

that morning, and nobody was in there.
But you can imagine that, if their

schedule had changed just a little bit,
this would have been a very serious case.

So as part of these studies,
we’ve done a bunch of different things.

One of the most important,
I think, is performing

a hazard and risk assessment
for Yosemite Valley for rockfall.

And we had the good fortune that
the administration was very receptive

to actually moving on
the suggestions of these reports.

And so, because of that, they –
Yosemite moved and closed

different tent cabins and
repurposed different situations.

And that’s actually reduced
the rockfall risk considerably.

It was slightly unpopular
in some people’s minds

because moving tent cabins
and hard-sided cabins

that had been there for a long
time had a historical basis.

You know, some people would have
liked to keep them that way,

but the scientific story is that
those cabins are actually built

on the depositional
environments of rockfalls.

They were built on talus slopes.
And we generally don’t like

to put anything on a talus slope
when we know that this is

where the rocks are going to land.
So that’s been a big, big first

and important step for
addressing rockfall risk.

Another thing that we’ve done in
Yosemite is compile a database

of all the rockfalls that have been
recorded over time starting in 1857.

And this database was released in 2011.
It’s publicly available.

It has narratives and descriptions
from people that were observing

the rockfalls at the time.
And this was a study that was initially

begun by Gerry Wieczorek who worked
at the USGS for a long time and

really was one of the pioneers
of the USGS for working in Yosemite.

He did a bunch of different studies, and
he published, along with Jim Snyder,

who was the park historian,
the first database.

And then Greg and I,
a few years ago, updated that.

This database contains 925 rockfall
descriptions and provides an ability

to start looking at the
statistics of rockfalls there.

So on average,
there’s about –

currently we record a rockfall
about once a week, on average.

Now, that’s not to say that there
aren’t more than that that aren’t –

that done go recorded as well.
But it’s basically proof that Yosemite,

being a wild place,
is geologically active in this way.

So this is a plot
that Greg has put together –

Greg Stock has imaged –
the park geologist.

This looks at the number of rockfalls
here on the Y axis here – over time.

1857 on the left.
2011 on the right.

So the number of recorded rockfalls,
you can see that not much

was recorded
in the 1800s.

And then there was some reporting
here in the early to mid-1900s.

And then, all the sudden,
in the 2000s, there’s all sorts

of rockfalls being reported.

And I keep rockfalls
reported because,

coincidentally, in 2006,
or I should say not coincidentally,

Greg Stock started working there
and started talking to all the rangers

and saying, hey, if you see a rockfall,
tell me about it. I’m the park geologist.

And so that reporting rate, you know,
at first glance, might suggest that

the number of rockfalls in Yosemite
is increasing, but really it’s the

reporting rate, that people are
observing these things, and everybody

is walking around with a camera
taking pictures of these things

and posting it online
or telling people about it.

So this database is
really increasing in size.

The red plot on the same graph is
the number of visitors to Yosemite.

And basically, what that shows is,
currently about 4 million people

per year, that really gives us
that impetus for looking at

this hazard-and-risk-type

Now, the bottom plot is another
sort of proof that we use to show

that this reporting rate is the
reason for these seemingly increasing

number of rockfalls
over time.

When we look at just the largest
rockfall during any time period,

that pretty much stays the same at
about 10,000 cubic meters each event.

And that – the fact that that
hasn’t changed over time

gives us an indication.
We know that, over time,

basically somebody should have
been reporting the biggest ones.

These are the ones that,
you know, are being felt

far and wide
throughout the valley.

And so those haven’t really changed,
and so we presume that the rockfall rate

is at least not increasing –
not from the data that we have.

So as I mentioned, there’s people
walking around Yosemite all the time

with cameras taking pictures
of the amazing scenery.

And one of those was caught
spectacularly on camera by a visitor

from New York, Robert Atkinson,
who took this video.

Look right up here.

- [from video] Look at that piece!

[crashing sounds]

- So that’s 78 cubic meters.

About the size of
24 mid-size cars.

[crashing sounds]

Coming off that cliff.

Okay, one more time.

- [from video] Look at that piece!

[crashing sounds]

- Now, he captured what would be the
third event of the – of that sequence.

There was two other pieces
that popped off here earlier,

I think a few
minutes prior.

And so that was part of the reason
why he captured such good video

was something’s going on over there,
and turns his camera and got that.

That sort of video being provided
to the park allows us to do

so many things with it, from looking at,
you know, just the magnitude,

the size of these, to understanding
where the source area of the rockfall is.

That’s often very difficult.
We know – you know, in some cases,

there’s a rockfall deposit, but where
on the cliff did it come from?

So these sorts of observations are really
spectacular and very useful for us.

So looking at that database in a little bit
more detail – the where, the when,

and the why, I have shown here a
geological map of Yosemite Valley.

Let’s get the mouse back up.
Here’s Half Dome.

Here’s El Capitan, Yosemite Falls,
and Glacier Point for reference.

And these are all
the different rock types.

And I’ve shown here,
in this key – hopefully you can –

yeah, you can read
this text here at the bottom.

So the Sentinel granodiorite
outcrops here in eastern Yosemite,

it has the greatest number
of events per cliff area per year.

So this is normalized over
the amount of cliff and over time.

So that’s sort of a
standard unit to compare

each rock type,
one to the other.

So these two units here between
the Kuna Crest granodiorite

and the Sentinel have the
largest number of events per year.

We think that could be partially
because there’s a lot more reporting

that goes on in eastern Yosemite Valley.
That’s where all the visitor facilities

are – Curry Village, Glacier Point,
the Ahwahnee, so on and so forth.

And so that could be
part of the reason.

The other reason is
that these rocks are –

actually have some fabric to them –
some alignment of the minerals.

And they tend to be – they form
more highly jointed rocks.

And so that could be providing
weaknesses that then different triggering

mechanisms can take advantage of.
That could explain that.

We see very large volumes of
rockfalls here from El Capitan.

And so that’s on
the other Y axis here,

the volume of the
rockfalls per cliff area per year.

And we think that’s because this –
the El Capitan granite is so massive

that it doesn’t have
a lot of joints in it.

And so when it does break,
it breaks in very large pieces.

When we look at the timing of rockfalls,
this tells not much of a story.


It basically tells us that rockfalls
happen all year long,

and there actually
is not a rockfall season.

And we think that that has to do
with the fact that there’s

a multitude of different
triggering mechanisms.

So in the wintertime, freeze/thaw-type
processes could be affecting it.

In the springtime, precipitation,
rainfall could be causing rockfalls.

So we can’t really tease out too much
of a pattern from this sort of data

other than taking home that you should
be aware of rockfalls all the time.

Now, the why
gets to this triggering.

And this will lead into the majority
of what I want to talk about today.

Looking at all the potential triggering
mechanisms for rockfalls here on the

right graph, precipitation-related
rockfalls account for 62% of the data.

Freeze/thaw-caused rockfalls is only 7%,
and that was pretty interesting because

I think a lot of people think that
in the high Sierra, that freeze/thaw

is the predominant reason
for rockfalls happening.

It’s actually just a small fraction compared
to these other mechanisms.

- How do you know that?
How do you make

that determination that …
- Yeah, that’s – right.

So that’s a good question, and the reason
for that is basically temporal coincidence.

So if there is an earthquake,
and a rock falls off of a cliff,

then we usually assign
an earthquake-type trigger.

If it’s raining, and a rock falls off
the cliff, similarly, precipitation.

So we – it’s a temporal coincidence,
I would say, that we’re trying

to link those processes
to what’s going on that day.

Not to say that, you know, some other
thing could have been happening.

But it’s – I think that’s the
best we can do with the data –

the observations that
we have out there.

When we group all of the
other triggering mechanisms,

put those here, 18%.
Precipitation accounts for 29%.

We still have,
out of the entire database,

50% – over 50% are
unexplained in some way.

So this was a subset of the
database that we could actually

positively assign a trigger on.
So we had enough information

from the observations, either
meteorologically or seismically,

to say, yeah, this is – this is pretty
much the reason for that rockfall.

That’s 361 events
out of 925.

This subset is 770 out of 925.
And the reason why it’s not 925

out of 925 is because some of them,
we just have literally zero information on.

There’s a report from a superintendent’s
note that says there was a rockfall in,

you know, 1890 or
something like that.

So we really can’t even
attempt to do anything with it.

The other two categories –
unrecognized and unknown –

are a little bit
more specific.

Unknown means that
there’s enough information to know

that the rockfall, let’s say, came from
a particular place, but we can’t make any

determination because it’s
not enough information.

The other cases, we literally can’t
even assign any information to it.

Here we have a little bit known,
but not enough to make anything.

The unrecognized, on the other hand, are
rockfalls where they were fully captured,

so we know everything about them.
They were observed.

We have pictures or video
or firsthand accounts.

We know what the weather
conditions were at the time.

We know whether there
was an earthquake or not.

And we can’t explain it.

These are rockfalls similar to the video
that we showed – that I just showed,

saying that this is very well –
we know exactly what was happening,

and we still
don’t know why.

So it wasn’t raining that day,
and there hadn’t been an earthquake,

so why that rockfall then.
So that’s the unrecognized ones.

And those are the ones that were
particularly interesting to us

when we started wanting to
investigate rockfall triggering more.

So why can’t we
explain 26% of the data?

Something has to
be causing it.

So this led us to start looking a little bit
more in-depth at triggering mechanisms.

And, as I mentioned, some of these are –
there’s obvious triggers,

and there’s not-so-common triggers –
wind, lightning, root wedging.

In fact, if you look at the database,
the lightning entries are quite interesting.

This is when somebody
unfortunately was out on a cliff

and saw
lightning strike.

They probably shouldn’t have
been out on the cliff at the time,

but they saw a rockfall immediately
after the lightning strike.

The rocks fell from where
the lightning struck the rock.

So that’s a lightning-type triggering
mechanism that gets assigned.

But many are still unexplained,
as I mentioned.

And in particular, since Greg has
been in the park for a long time,

he started noticing that
a lot of these were during

the summertime during these
Sierra beautiful blue sky days.

And he couldn’t assign a trigger to
them because there was nothing obvious.

So those are the ones we said, we should
think about this a little bit more.

And that led us to this question.

Why do rockfalls sometimes occur in
the absence of any identified trigger?

And here’s two photos
of those instances.

Look at the color of the sky there.

There’s just – it’s a beautiful day.

And we don’t know why
these rockfalls are occurring.

I also want to point out
the time and the day on those.

The left one is in July and
early evening, we should say.

And the right one is in
August in mid-afternoon.

That’s important.
Keep those in mind.

Because then we thought, okay,
hot summer days, hot afternoons –

maybe has something to do
with the temperature.

Let’s go revisit what Shaler
first proposed in New England.

And there’s lots of examples of this,
including those from India

where people have documented
the lighting of fires to quarry rock.

So they can build fires along
a rock surface, get that rock to heat up,

and it actually separates from the rock,
and then they got a nice slab

of granite they can pull
away and build with it.

So this was some of the inspiration for
taking on a study on thermal triggering.

Is heat enough to
cause rockfalls?

And there’s been – you know,
this isn’t a theory that we’re proposing.

As I mentioned, it’s been
proposed several times.

But we wanted to see if we could
quantify this in much more detail.

And so when we look at the record –
when we look at that database,

what we saw was that,
between July and September –

the hot months in Yosemite –
between 12:00 p.m. and 6:00 p.m. –

the hot afternoons, there were
2-1/2 times more rockfalls happening

compared to if that was spread
over a random distribution.

So the number of rockfalls,
if it was randomly distributed

throughout those months and
throughout those times,

the data set shows that 2-1/2 more
times is actually occurring.

So that gave us even more inspiration,
I guess, to take on this study.

And so we designed
an experiment.

And the experiment was,
let’s go find an exfoliation flake

that’s partially attached –
partially detached from the rock.

Let’s measure it.
Let’s see if it moves when it heats up.

And let’s measure the temperature,
both on the inside and the outside.

And we’ll make
measurements for a while.

And a while, I think, in our mind,
was, let’s do this for a few months.

And actually, upon inspiration from
Mark Reid, who introduced me,

we decided to do it a little bit longer, and
a little bit longer, and a little bit longer.

And we’re really glad we did, so thanks
to Mark for pushing us in that direction.

Because we captured what I’ll show,
I think, is a pretty spectacular data set.

So our experiment is here.

This is the Royal Arches.

This is the
Rhombus Wall.

And there’s a hotel down here
that some may know by this name,

some may know by that name.

And this is the rock study
area here in the red box.

And this is
the exfoliation sheet.

It’s laid back about 70 degrees.
It’s fairly steep, but not dead vertical.

We installed custom-designed
crack meters, is what we call these.

This is a sensor that measures how much
that opening opens and closes over time.

And we installed temperature
sensors on the outside of the flake

and temperature sensors
on the inside of the flake.

And the geometry is such that
the entire flake is detached.

It’s completely open.
It’s hollow in the back.

If you slam on it,
you hear that hollow sound.

It is attached up at the top,
and it’s attached down at the bottom.

It was about 19 meters tall,
4 meters wide, and you can see here,

about 10 centimeters thick.
So it’s a very long, thin feature.

This is actually very typical of lots
of exfoliation sheets in Yosemite.

This wasn’t an anomalous sheet at all.
This was very typical.

Now we had these crack meters
designed to be very, very precise.

We weren’t sure what we were
going to measure out there.

We figured it’d be a fraction
of a millimeter, so in the conversations

I had with the manufacturer, I said, you
know, please give us your best design –

you know, the best resolution,
the best precision, the best accuracy

in this instrument.
And they did.

They gave us something that would
read down a 100th of a millimeter.

And I had to apologize to them
shortly thereafter because it turned out

we didn’t need
that kind of accuracy.

These sheets were actually moving
by almost a centimeter a day.

And this is the
data from that.

So this is three days
in July of 2010.

And this is the relative deformation, so
starting here, this is – they were zeroed.

And this shows that signal that,
over three days, moving inwards

during the early hours of the day,
moving back outwards, moving inwards,

moving outwards, and so on and
so forth. But look at the magnitude.

It’s going 6 millimeters in the
positive direction, 4 in the negative.

So a span of almost
a centimeter per day.

We had a control crack meter.
That’s one that we put in there

that was not attached to the flake
just to make sure that the instruments

themselves weren’t picking up
some thermal signal of their own –

that the instruments weren’t
heating up over time.

That is the thin blue line.
And you can see that it does heat up.

It has a slight signature to it,
but it’s much, much smaller

than the signal that we
were measuring from the sheet.

And these are the values of
three different crack meters –

upper, middle, and lower.
So they all measure that same pattern.

Now with that in mind, we said,
wow, that’s a whole lot.

Why do we need these crack meters?
Why don’t we just go out there

with a tape measure?

So we did. We had to prove to ourselves
that we were actually making sense.

That we were actually
getting this right.

So we climbed back up there and
measured it with a tape measure.

It’s there. It’s true.
You can measure it in the daytime,

and measure it in the nighttime, and
you get the – virtually the same answer.

But we also do lots of sophisticated
monitoring for other projects,

and we have access to these tools,
which one of those is terrestrial Lidar,

which is something I do
a lot of in other projects.

And so we applied that tool
to this to see if we could

actually measure the
deformation of the entire sheet.

So these were just point measurements
we were getting at the three instruments.

Terrestrial Lidar is a surveying
technique that collects thousands

of points per second.
It’s like a very high-powered radar.

It can measure the
exact trajectory and

the exact orientation,
the exact distance to a point.

So we set this up at the
base of the cliff and shot it from –

over the course of several hours,
including over a single night.

And when you compare the data from,
let’s say, late afternoon until the

very next morning, when the sheet
should be moving in the most,

what we got was this pattern here –
a contractive pattern, shown in orange,

of the sheet moving back
into the flake – back into the cliff.

About that same magnitude –
here in the middle,

6 millimeters for this particular day,
and 9 millimeters.

So we were, again, convinced we
were capturing the right magnitude.

This data isn’t from the same day
as I just showed, but it did coincide

exactly with the values from the
crack meters for that particular time.

So the flake’s moving in
and out quite a bit.

What are the
temperatures doing?

Temperatures, as we would expect, rise
during the day and fall during the night.

What was very interesting about that
was that the inside of the flake

really doesn’t – well, it also
increases in temperature,

but there’s a big temperature gradient 
between the outside and the inside.

And that’s important
when we do analyses.

Now we have two reasons for the
sheet of rock potentially deforming.

One is that the temperature
of the whole thing is going up.

The second is that we
have a thermal gradient across.

And a thermal gradient will
actually cause rocks to behave

in a certain manner, and we’ll –
I’ll talk about that in a little bit.

But those two components
are important, that the whole rock

is heating up and there’s
a gradient across that sheet.

When we look at the
temperature signal and the

deformation signal in the same plot,
that’s what this looks like.

So we have temperature
here on the Y axis,

and the crack aperture,
so that’s the gap behind the sheet.

Here we have four days plotted,
and I’ll step you through it

starting here at
midnight on June 12th.

And what we see is that,
between midnight and

6:00 in the morning,
the temperatures are still dropping.

So the temperature started about
14 degrees, and now they’re

10 degrees at 6:00 in the –
6:00 in the morning.

This is – temperature is
in degrees Celsius.

And the crack was starting to –
or was continuing to close.

At 6:00 in the morning,
things started to heat up,

and it just absorbed heat for three hours.
It didn’t move at all.

It was just out there absorbing and
getting hotter and hotter and hotter.

At 9:00, all the sudden it turns a corner,
and the sheet starts to move outwards.

It’s also continuing to get hot during
the day, but it’s really moving outwards.

And here, on this particular day,
about 7 millimeters.

And then, in the mid-afternoon,
starts to cool off.

It gives off heat back to the environment,
starts to cool down, doesn’t have any

movement for a few hours,
and then starts to come back again.

So that’s day one.

- That temperature is the
[inaudible], temperature, right?

- That is a near-surface
rock temperature.

- Okay.
- Yep.

So then the
second day is here.

And the temperatures were
slightly hotter on this day.

But we see the same pattern again
of it opening back up again,

cooling, and closing
back down again.

Here’s the third day, June 14th, and it
absorbs heat similarly, comes out, comes

back down again, and then the fourth
day of what I’ve shown here on the plot.

So not only is it going around in a cycle,
but over the course of four days,

it’s actually building, and it’s moving
further and further outwards over time.

And that’s pretty interesting because
now we understand that these aren’t –

the rock’s just not moving
back and forth according to,

you know,
fixed positions.

It’s actually – can sequentially
move outwards over time.

That cycle has analogies to it.

And when I showed this cycle to a
mechanical engineering friend of mine,

he instantly pointed out that
that must be a Carnot cycle.

And for those not familiar with
Carnot cycle, this was an invention –

discovery by Sadi Carnot,
who worked on this in the early 1800s.

He was looking at the
efficiency of steam engines.

At the time, they only had
a few percent efficiency,

and he knew that there were
ways to improve upon that.

And as part of those studies,
he actually defined what would be

a theoretically perfectly
efficient heat engine.

And that is the definition
of a Carnot cycle.

This is temperature here
on the Y axis and entropy –

the measure of randomness
in the system on the Y axis.

It’s not too important
for understanding,

but I’ll step you through what works –
how a heat engine works.

You heat – you give the heat engine
some increase in temperature.

It absorbs energy and then
does work on the system.

So if it’s a – if it’s a stem piston, then
that piston is going to move up, let’s say.

And then we lose heat
to the environment,

and the piston moves down again.
And in a perfect cycle, you wouldn’t

have any heat losses, and you wouldn’t
have any frictional losses, and you’d get

back to exactly where you started again.
That would be the perfect engine.

And the reason why this was a
big discovery at the time because

it showed that they shouldn’t be
just dealing with a few percent,

that they can do a lot better than that,
and that this is the best they can do.

So this was the goal, and that led to a lot
of advances in steam engine efficiency.

Now, the reason why I talk about
Carnot cycles is because this

friend of mine who pointed out this
analogy, I think, was quite correct.

And that, although here
we’re measuring crack aperture,

and on the – in the Carnot cycle,
we’re measuring entropy –

clearly two totally different
measurements, but the same

analogy could be made that here,
the flake absorbs heat just like

a heat engine would.
And then does work on the system.

It moves outwards,
just like a piston does.

And then it loses heat back to the
environment, and it moves back again.

And so the cycle of the flake
moving in and out and in and out –

it’s doing work on its environment.
And that working – that work is

both the movement and
potential deformation of the rock.

So those are the daily cycles, and the
short-term cycles, over several days.

When we look at seasonal
cycles and yearly cycles,

we see something
also very interesting.

One of the things that’s potentially
obvious would be that the crack –

that the sheet moves in the
most during the wintertime.

So here’s December ’10 – 2010,
and December 2011.

So it’s closest to the
cliff when it’s coldest.

Closest to the cliff
when it’s coldest.

And it’s farthest away
from the cliff when it’s hottest –

during the hottest months.
But over the three-year time span,

we saw that the opening
was actually increasing.

And what I show here, these dots, are
the maximum monthly crack aperture.

So out of every month
that we measured, and this is –

we measured these crack
meters at five-minute intervals.

So we take the maximum of every
month, and we plot that on this graph.

And the maximums actually increased
over time over those three years.

Now, the most obvious thing to check
on that was, well, did the temperatures

increase also over those –
course of the three years?

They did not, it turned out.

So we know that this is presumably
capturing permanent deformation

of the rock face, that over time,
it’s slowly moving away from the cliff.

That makes perfect sense when you
consider that rockfalls are happening –

the rocks are not becoming
more attached to the cliff.

Eventually, they’re falling off.
And I think that this actually is some

clear proof that, over time, these sheets
can open up and eventually detach.

And so that opening
is captured here.

Now, the opening is about
1/2 to 1 millimeter per year

for this particular flake
that we measured.

We don’t think that’s a linear value
that you can project back in time.

If that’s the case, then this sheet was
only there for about a hundred years,

which we think is
a little short.

But the point is that it could
be a nonlinear relationship,

where maybe a long time ago,
it was only moving outwards by,

let’s say,
0.01 millimeters per year.

And then, as it moved farther
and farther out, it was able to absorb

more heat because now the back
and the front can become hotter

during the day and cooler at the night,
and so on and so forth.

So this is a metric that I think is
important but can’t be applied

ubiquitously back in time –
probably a nonlinear relationship.

So one question I get asked a lot about
this is, how did we collect our data.

And, as Mark pointed out,
I am a rock climber, and so is Greg.

And that came in
very handy for this study.

So I have here a picture of the
two of us collecting our data.

So here’s the data logger.
Here’s Greg.

That’s what it looks
like on the cliff.

We’re hanging from
ropes that we set up.

And most of the time, we had this great
weather to do it, but it wasn’t always fun.

We resorted to raincoats and umbrellas

when we were trying to get that data.
Because this was a data logger

that’s recording that data, and we
had to go get that data periodically.

So that’s how we do this
kind of experiment up there.

Pretty neat.
Pretty fun thing to do.

- Did rock climbers interfere with your –
with your things that were –

were these on spots that people
were climbing [inaudible]?

- Rock climbers did not disturb this.
We had done quite a bit of

public outreach with the rock-climbing
community, and we had notes on this

equipment saying, please don’t pull me.

Or clip gear to me
or fall on me.

And it turned out this route
isn’t that popular of a route.

So it was –
we mitigated that effect.

So turning back to the effects of
this movement – this deformation

in and out all day, is it really
enough to fracture rock, though?

We know that it’s moving,
but this rock sheet did not detach,

and we were
thankful of that.

Because I think we would have
been blamed for sticking a carjack

in behind it and prying it loose.
That we did not do.

But it did not fall down.
And so, you know, we don’t have

that proof positive that it
can actually fracture rock.

But we can examine
rock fracture in other ways.

We can look at it and find these beautiful
exposures at all different scales.

So here’s 4 centimeters of a little
tiny piece of rock that’s actually

peeling away from the rock surface –
probably thermally induced.

Here’s a much larger feature.
This is 10 meters long,

similar to the one that we monitored –
a little bit smaller in scale.

And I want to use three analogies.
And this comes from my

civil engineering background of bridges,
buildings, and boats to explain

how rock fracture might be
occurring and how the heat

contributes to the deformation
of these exfoliation sheets.

So we’ll use principles of
thermodynamics, structural engineering,

and fracture mechanics
to get there.

And, as I mentioned earlier,
this is an exciting topic for me,

but one of the reasons behind that
is being able to grasp all of

these different areas of science
and engineering and bring them

together to try to explain
what we see out there.

And I think all of these
pieces are actually necessary

to get to the point
that we’re at.

So rocks are like bridges,
and here’s how.

When you heat up a bridge – for
example, and unfortunately, in a fire –

this is Liberty Bridge in Pittsburgh that
caught fire in March of this past year.

And this was, I believe,
a welding job that caught

some of the scaffolding,
or the tarps caught fire.

This led to some deformation of the
steel beams that you can see here.

And so how are
rocks like bridges?

Well, the steel beams heated up,
and they wanted to deform.

And this is the same sort of analogy as
we have expansion joints in bridges that,

when the bridge heats up,
you have to accommodate

the thermal expansion
of the material.

And if you don’t,
then you get deformations.

And so that’s the same thing as what we
see in rocks, is that when you heat the

rock face up, it does want to expand.
But it can’t because it’s actually pinned.

This isn’t free to expand
as much as it wants on both sides.

It’s actually attached to a much larger
cliff face that’s providing resistance to it.

So when it heats up like that,
stresses are built up,

forces are built up,
at the end members.

And Newton’s Law tells us that there’s
a equal and opposite reaction into it.

So the exfoliation sheet wants to expand,
but instead, it’s being held there

between two end members,
and so there’s some forces and

stresses that develop at the cliff –
at the ends of the exfoliation sheet.

This is a image taken by colleagues of
ours that come over periodically from

Switzerland to do research in Yosemite.
They obviously have lots of rockfall

issues in Switzerland, and so we
cooperate very closely with them.

They applied a thermal camera
to take pictures sequentially over time

and caught that same signal,
but we were able to see the

entire signal over the entire rock
with that thermal camera image.

Basically what this tells us is,
yes indeed, the rock is heating up

everywhere like this,
and we can do some

pretty neat experiments
with that data.

So what happens when
the rocks heat up?

And what happens when buildings
heat up and bridges heat up?

Well, they end up deforming.

And the way that they end up deforming,
unfortunately in fires, is that they buckle.

And you can see here – this is the
Windsor Tower in Madrid, Spain,

in 2005 that caught fire. The building
was about a 30-story building.

It was built in 1979, and it
was undergoing renovation.

Unfortunately for them at the time,
they were actually putting in

sprinkler systems into the building
and doing some fireproofing.

Nobody was in the building at the time.
Nobody died from this building fire.

But it did give an example of some
quite dramatic deformations in the –

in the steel members.

And you can see some of that
deformation here – the buckling.

This is what it
looked like afterwards.

And you can see here that, when those
steel beams heated up, they weren’t

allowed to expand because they’re
attached to the rest of the structure.

And so instead, they buckled –
they deformed outwards.

So again, this analogy of the rock
heating up, and instead of expanding –

it can’t, and so instead,
it’s going to bow.

And it’s the same sort of thing
if you take a yardstick between

your hands and push on it,
it’s going to push out.

It’s going to
bow and deform.

And so that’s the analogy
we can make with buildings,

that rocks are like buildings,
and that, if you heat them up,

and you don’t allow them to move,
they’re going to bow outwards.

They’re going to buckle under
those stresses at the ends.

And a final analogy is, what happens
when that does happen, that it does

bow outwards, and then it cools down,
and it bows back inwards again?

And let’s say you do that
many, many, many times.

Well, there’s been some
very interesting research

that comes to us out of
the boat design industry.

And I have an example here
of some of the Liberty ships

that were built
during World War II.

This one, the U.S.S. Ponaganset,
was built here in Sausalito in 1943 and

served in the Pacific arena for several
years until the end of World War II.

It was a water tanker that was supplying
potable water to troops in the Pacific.

Following World War, it –
well, I should say it spent

all of its time in the
warm South Pacific Ocean.

That’s important because
after World War II, it was decided

to be repurposed and salvaged,
and it was sent over to the East Coast.

And when it spent some time
in the cold, icy waters off the coast

of Boston, this happened.
The boat split right in two.

And upon investigation,
it was determined that some of

the welds that had been put
on this boat, those welds had been

tested under the temperature
regime that it thought the boat was

going to experience, which was the
warm waters of the Pacific Ocean.

It hadn’t been tested at the colder 
temperatures of the Atlantic.

And so the microflaws in the
welding process formed into

larger and larger cracks, and that
cyclical nature that a boat undergoes

during the course of sailing over
thousands of millions of ocean waves

eventually caused those cracks
to become bigger, bigger, bigger,

and then it hit a critical
stress propagation,

and the cracks just riddled
right through that ship.

And so this wasn’t just a – one case.
They actually had a lot of examples

of these ships breaking in two like this.
So that led to lots of discoveries in terms

of fracture mechanics and how you make
sure that welds do not break over time.

It’s the reason why we have
sort of oblique, circular windows

on airplanes too.
They like to avoid sharp corners

because those are stress concentration
points that then fracture eventually.

And those, you know,
airplane loads are subjected to

many, many cycles over time.
So lots of – this was unfortunate

failure of this boat,
but lots of big discoveries

in terms of material science
came out of that.

So what’s the analogy back
to the rock sheets of Yosemite?

Well, eventually, if we move
that sheet back and forth over time,

and it’s bowing, and it’s coming back,
and it’s bowing, and it’s coming back,

it’s going through
that same type of cycles.

And we think that that’s
enough for it to fracture.

And we can see examples of that
throughout the Sierra Nevada.

This is an example of an event that
happened in Twain Harte – back to that

video at the beginning that I showed
of the rock buckling, basically.

Just snapping in two like that
and moving upwards.

Maybe it had been subjected to
enough cycles of stress over time.

We can test this in
the laboratory, and we are.

We work with colleagues, again, in a
different laboratory in Switzerland.

We send them cores of rocks from
Yosemite, and they put it in this device.

This is a cyclic loader. And this is
a fracture toughness-type test.

So what we’re doing
right now is we’re loading –

we have a cylinder of rock –
this is Half Dome granodiorite.

We put a small crack in the middle,
and then we load it with a small load –

not enough to fracture it
right away under one cycle.

Instead, just a fraction
of that maximum load.

And then we do that a lot,
and we measure the number

of cycles that we –
that we get.

And the purpose of this
is to see if we can determine

how many cycles does it take
to eventually crack this rock.

And are the thermal stresses
generated by what I just showed

in all those analogies with buildings,
boats, and bridges, are those thermal

stresses that are created at
the ends enough to eventually

cause this to crack
with a number of cycles?

So that’s the research –
the current research

that I’m involved in is following
through on that part of the study.

We know that failure must be cyclic,
in a way, because if the rocks

were stressed at some point, just once,
because of temperature variations,

then it wouldn’t really explain
why all these flakes are still here.

They should have fallen off
if it was the thermal trigger.

So instead, we know that the thermal
trigger must be cyclic in nature.

It must take a long
time to develop,

and it’s got to be many, many cycles
eventually lead to the fracture.

So can the fracture lead to rockfalls?
Well, I think, you know,

we do have a lot of case studies
now that might suggest that.

This is another rockfall
off of the Rhombus Wall –

the video that I showed.
And I’ll play that again

because now you can look at it
maybe with some different eyes.

So these are all exfoliation sheets.
- [from video] Look at that piece!

- And this is a piece of that exfoliation
sheet that’s eventually falling away.

[crashing sounds]
So maybe thermal cycles are acting on

all these sheets over time, and eventually
some of them are going to fail.

And maybe at some other point in time,
another one’s going to fail.

It’s not to say that the whole cliffs are
going to fail all at once because of this.

This is just one potential trigger.
We know that there’s lots of other

triggers, as it discussed before.
But it’s one thing that can cause it.

So let’s look back at
those unrecognized triggers

that we were
talking about earlier.

I showed you this plot. This is all of
the triggers discretized by month.

And now I’m going to
show you the

unrecognized triggers
discretized by month.

And you can see that a lot of those,
as I pointed out, 2-1/2 times more often,

are during the summertime and
during summer afternoons.

What we also did was we
looked at the maximum temperatures

for those
unrecognized triggers.

And what we see is that, sure enough,
a lot of those rockfalls are happening

when the temperature is pretty hot out –
over 30 degrees Celsius.

That’s a hot day.
That’s a 90-degree-plus day.

So we think that we’re onto something
here in explaining the database

a little bit more, which was basically
the place that we started at.

We wanted to know if we could
say something about those

unrecognized rockfalls – unrecognized
triggering mechanisms for rockfalls.

And we wanted to see if we could
explain why rockfalls happen

on beautiful summer days.

So I think we’re onto something.

We think we can revisit that database
and try to explain some of those.

And certainly moving forward,
do a better job.


So this study received a lot of press
when we published a paper

this past year, including this one
in the Daily Mail that said,

is the sun destroying
Yosemite’s iconic cliffs?

That wasn’t our headline.
That was theirs.

But the point is that, yes, I guess,
in a way. [laughter]

I wouldn’t use the
word “destroying.”

But, you know, sun and the
thermal cycles probably do

have something to do with the
landscapes that we see in Yosemite.

So moving back to fracturing,
I want to move back and close

with discussing the current research
we’re doing over at Twin Harte.

And so this – I’m going to show that
video again, but the video was taken

on the other side of this bench
on the other side of the dome.

And I said that that was just a
minor event. This was the big one.

This is all new
rock surface.

They stripped off all the rock
from those first couple events.

And you can see the thickness
of the exfoliation sheets here.

This is, you know, about 40 centimeters,
something like that.

So all that rock actually popped off,
including – well, this very large area.

So let’s look at some of
the video one more time.

[voices in video]

So, yeah, you can hear somebody
laughing in the background

because this [laughs] –
pretty exciting for them.

So this event was in 2014 –
August 2014.

There was five
events in 30 days.

Each event, as I mentioned,
lasted several minutes.

That’s what allowed people to
get out there and start filming.

It had buckled slabs of
up to 40 centimeters.

This is one of those.

This is a lifeguard tower that you
see on top of the slab that was uplifted.

The lifeguard tower is there
because that reservoir

is used as a recreational
swimming facility.

And this is a slab that’s about
15 centimeters thick and popped up,

you know, 40 centimeters,
so from one of those early events.

This one wasn’t captured on film,
unfortunately, but it would have been

pretty spectacular to see something
like that happen in real time.

Cracking over about
a 1,200-square-meter area.

And no obvious cause.

No earthquake,
no rain, no snow.

This was during
the summertime.

We got involved in this study because
of its direct analogy with what

we think might be going on in
Yosemite in terms of thermal stresses.

So that’s why USGS
is involved in a study here.

And so those events in 2014 –
what I’m showing here is a date

on the bottom starting from August 2014
through December here, 2016.

These are the events –
the five events in 2014.

We started measuring
temperature at that site here.

And temperatures on
the Y axis on the right.

And the blue line
shows extensometer data.

Extensometer is an instrument
that is put in the ground.

We drill through the slab,
and we attach something

above the slab
and below the slab.

And that measures how much the
slab is going to move up and down.

So it’s a measure of that deformation.
Similar – just the same to the

crack meters that we installed.
Only this is installed a little bit

differently because the slab is beneath
our feet instead of on the side of a cliff.

And so we’ve been taking continuous
measurements for quite some time.

What you can see over the
course of a year is about

4-1/2 centimeters of
opening of that slab.

Now, we were looking at,
you know, a centimeter per day

on the slab in Yosemite.
This is 4 centimeters per year,

which is a lot more deformation
than we saw on the slab per year.

And it also has the same diurnal cycle –
the daily opening and closing.

You don’t see that here because I’m
showing such a long time period.

What happened in 2014 –
that was this big slab opening up.

In 2015, the extensometer data
showed that the sheet was moving

rapidly outwards, so moving away
from the bottom, away from the Earth,

if you – because now we’re
on a horizontal surface.

And it coincided with
some pretty hot temperatures –

the maximum temperature
was during that time period.

And at that time, there wasn’t
a big exfoliation event,

but there was cracking
throughout the dam.

So they had already poured
new concrete for some of the repairs

to the dam and the facilities,
and we were seeing cracks

opening up in that
during that time period.

And then, in 2016, just this past year,
in June and July, there were two, again,

energetic exfoliation events,
smaller in scale than the 2015 ones,

but spot on with the
maximum temperatures of 2016.

And so I don’t think it’s any coincidence
there that, in these three time periods

over the course of the past two years,
that the cracking is only happening

at the maximum temperatures.
So heat’s got to be playing a role here.

And so that’s – you’ve seen
this picture before,

but that’s what happened
this past year in 2016.

So can it really fracture rock?
I think yes.

I mean, we’re seeing it
there in Twain Harte,

and again, that’s the reason
why we’re studying this.

And we’ve actually
started putting together –

we had a lot of instruments put on the
ground to measure all sorts of things –

the stress, the strain, the deformation
of the rock slab in Twain Harte –

to see if we can learn what it would
take to cause rockfalls in Yosemite

in places that we’re trying to
minimize the hazard and risk at.

So with that, I’m going to end with
this slide. Can this really fracture rock?

I think the answer is yes.
And we see it here in Twain Harte,

and I think we’re seeing it
in a lot of places in Yosemite.

Thanks a lot for your attention,
and these are all the collaborators

I’ve worked with that I’d
like to acknowledge.

[ Applause ]

- Well, thank you, Brian.

I’m sure Brian would be happy
to answer questions from the audience.

There’s a couple of microphones in the
aisles there if people have questions.

We’d like to record the questions
so others can hear as well.

- Well, now you have a theory,
and then you have assay methods.

How are you going to use this –
how are you using this work

to mitigate – to assess risk
and mitigate risk?

- Yeah. That’s a good question.
I think, in Yosemite, what this means

is that, when there are rockfalls
happening on really hot days,

we’re not going to scratch our heads
too much more, and we’re going to

have a lot more confidence
in assigning that trigger.

Will we ever get to
the point where we say,

watch out for rockfalls
because it’s hot out?

I think that’s probably
raising the red flag a little too soon.

That would be about the
same thing as saying,

it’s raining out,
watch out for rockfalls.

It’s the same level of caution,
I think, that you’d have to take.

But I think what this does is actually –
if we can get this out to people’s minds

and visitors and they can understand
that rockfalls are an active agent

in the landscape, and there’s all sorts
of different things that can cause them.

- Just curious – oop,
mic is on. Sorry.

Do you have any anecdotal evidence
from the climbing community

about seeing this
level of deformation?

Because I know people
do put anchors in slabs.

Hopefully not in ones that are
about to come off, but if you’re

seeing movement of 4 centimeters
or more during the course of a day,

have you ever heard of anyone
saying, yeah, I put an anchor there,

but it moved
because it got hot?

- Absolutely. Yep. For sure.
And that was actually one thing

I failed to mention, but in the early
days of designing this experiment,

that was actually some of the
impetus and inspiration

for even instrumenting
one of these slabs.

Greg and myself, just talking with
lots of climbers, we’ve all put in gear

in a rock face as we’re climbing
and then, for whatever reason,

didn’t take it out within a certain
amount of time, and it got stuck there.

And I think a lot of climbers think that,
oh, that’s just bad climbing.

You know, you shouldn’t have put your
gear in that place, and you got it stuck.

That’s your fault.

And, you know, it’s easy to blame
yourself, and when you talk to a lot

of people with the same experience, all
of the sudden, it’s, like, oh, really?

Like, you’re a professional climber.
You’re really good at this.

You know better than that.
How did you get your gear stuck?

And so it started sort of cranking
the wheels and saying, well,

maybe these rocks are
closing in on the gear.

And so that was actually
one of the drivers for that.

And so that actually led us to instrument
that flake and say, well, that’s a place.

So great question, yeah.

- So your thermal, you know,
flexing datas are interesting,

and kind of that all makes sense,
and I can visualize that.

But a big blob of the pie chart
was precipitation as a cause.

And I can’t really get my head
around what could be causing that.

Do you have any thoughts
on what’s going on there?

- Yeah. So precipitation-related failures
are, as I showed, one of the –

well, it is the responsibility
of a large number of rockfalls.

The mechanism behind that is
generally that the – when it rains,

the water will flow into
fractures behind the rock cliff.

And those fractures aren’t always
perfectly drained at the bottom.

So the water can actually pond up
in back of the sheets, let’s say.

And the water pressures will
actually be enough to dislodge that.

It could also – that’s the most
commonly invoked mechanism,

and there’s lots of studies
that show that can happen.

The other way is that it can actually
just erode particles off the base

of these partially attached blocks.
Maybe something is sitting

on a ledge and then erodes
the base and that can slide off.

- So you were speaking about how the
whole slab heating up causes it to bow.

But do you have any data about
differential heating from the outside

to the inside of the slab and whether
that may have also played a role?

- And it did. Yeah. In our analyses –
so we have the temperature

from the outside. We also have
temperature data from the inside.

And that’s the thermal
gradient that I was referencing.

So the whole sheet heats up,
and that has some part of the

bowing outwards when the stresses
are enough to make it buckle.

But the thermal gradient itself
can actually make a bowing failure

as well, and so it’s a separate analysis,
but the same – I shouldn’t say

it’s the same mechanism.
It’s a different mechanism

but the same result in that
it’s going to bow it outwards.

So that thermal
gradient is part of it.

What I haven’t done is see which part
is more significant in our analysis.

We have a numerical analysis where
we plug all the numbers in equations.

We include those both – those –
both components –

the expansion and
the gradient effect.

- In your thermal studies, have you
taken into account for climate change?

And what has happened over the
years as climate has changed from –

we started it in 1858, I think,
and we’re still going on.

So we’ve had some ups and
downs in our planet’s climate.

- Yep. Yeah, thanks.
It is something that we’ve considered.

We’ve thought about it.
I think on the time scale of

this experiment,
we’re not seeing that.

Although, when we sort of carry this
into the future, we could say that –

I think the most important thing is
going to be the temperature cycle.

So will the temperatures be different
more – the low versus the high?

If the high is also – you know,
if the highs and the lows both move out,

then we have that same difference,
even – just at a hotter scale.

And so it’s something that could be,
I guess, carried out in a future analysis.

But on this time scale, it doesn’t
seem to affect our analysis.

- So sort of related to that,
it seems like the – especially the

daily thermal gradient
is largely driven by solar.

So I was just wondering if you have
mapped out where the rockfalls occur

relative to – are they on
south-facing slopes, north-facing slopes,

and their exposure
to the sun.

- Right. Well, I can – I can give
two pieces of information for that.

One is related to the solar.

We actually measured
light intensity on this.

What we found was that,
when the flake – when the sheet

was in the sun, it mattered, but it
didn’t matter as much as temperature.

The daily – just the temperature
increases are enough.

So even on a cloudy day,
it was still increasing enough

if the temperature
was going.

So the ambient temperature increase was
more important than the solar effect.

But looking at south-facing and
north-facing cliffs, we had thought

about – once this experiment
was going, we thought about –

this is a south-facing cliff,
by the way.

And that was purposeful because
we thought, if we’re going to measure

an effect, it’s going to be on a south-
facing cliff. So good point to make there.

We had thought about doing one
tandem on the north-facing cliffs

on the south side of the valley.
Just decided to sort of pursue the study.

But I think it’s something that we
could look at in the future, for sure.

- So you’ve been saying that it wouldn’t
make sense to do warnings for

high-heat days for risk, but what about
maybe geo-engineering for structurally

significant areas, like where a dam is
buttressed into one of these domes?

Perhaps painting the rock white
to reduce how much it heats up.

- Yeah. Well, so I think that –
it enters a – I shouldn’t say

politically sensitive, but it enters
a new realm of thinking.

So in Yosemite,
most of it’s wilderness.

And so very little is going
to be done to the cliffs.

If you started sort of saying,
that’s an unstable cliff and peeling off

all the loose ones, you’d have to
do it to the whole valley,

or you’d have to,

you know, lay it back to 45 degrees
and change the landscape.

So we don’t do that
in the national parks.

But there are examples of people
doing that on built infrastructure.

And in fact, that dam that
I show at Twain Harte,

what they did was they
separated the dam itself

from the rock by drilling a
series of holes through that sheet.

And so if the sheet does expand
or crack or anything like that,

it doesn’t affect the dam because now
there’s, like, this sort of buffering effect.

So there are ways to mitigate that
stress by just decoupling the part

that’s stressing from whatever you’re
trying to protect. Yeah, good point.

- Hi. So you’re watching the cycling
of the movement of the flakes.

Is there any possibility,
in high-importance areas,

that you could use something like
Lidar or Landsat or some kind of

satellite to just
monitor it constantly?

And then, if it’s bowing constantly,
a lot, maybe you know, oh,

that one’s getting near flaking off,
or something like that?

- Right.
- Is that feasible?

- Yeah. Well, we’ve pursued
several different options for

looking at the applications
of those kind of technologies.

One problem with that is that they’re –
it’s such a small amount of movement

for a remote sensing-type operation
that they wouldn’t pick up

millimeter-type movements.
Centimeter, though, could be picked up.

And we did an experiment
with a terrestrial radar –

so this is a ground-based radar –
that actually is more accurate

than even Lidar.

You can get down to millimeter
and sub-millimeter deformations.

One of the problems with that ended up
being that the pixel size – the size that

you’re monitoring is actually larger than
the exfoliation sheets that are out there.

And so that then means that you’re
actually measuring something that

could be moving and something
that’s probably not moving.

And it blended the signal.

But I think there’s a lot of potential for
those types of remote sensing methods.

I think, you know,
this is just the beginning

of some of these techniques,
especially radar.

I think the pixel size will get
smaller and smaller

with advancements
in technology.

- Okay, well, thank you, Brian.
Thank you all for coming tonight.

- Thank you.

[ Applause ]

[inaudible background conversations]

[ Silence ]