CVO Monitoring Program: Keeping an Eye on Cascade Volcanoes

Video Transcript
Download Video
Right-click and save to download

Detailed Description

The good news is that volcanoes usually change behavior before they erupt, in ways that are detectable by monitoring instruments. During times of relative quiet, scientists use different sensors and instruments to help visualize and quantify the structures and processes that are occurring beneath a volcano so they can provide a better estimate of what might happen when a volcano changes behavior. Every volcano and every eruption is different, so the tools and techniques that help forecast an eruption in one situation may differ from the ones that are most useful in another time and place. In this presentation, USGS Cascades Volcano Observatory geophysicist Rebecca Kramer describes her work to plan for, install, and maintain monitoring stations on the volcanoes in Oregon and Washington, focusing on three main monitoring techniques—seismology, gas geochemistry, and geodesy. This talk was presented as part of the Sno-Isle Libraries’ 2021 Whidbey Reads program.

Details

Date Taken:

Length: 00:45:35

Location Taken: Vancouver, WA, US

Transcript

Thank you so much to the Whidbey Reads program for inviting me to speak today.

My name's Rebecca Kramer.

I'm an operational geophysicist at CVO,

the Cascades Volcano Observatory.

Some of you may have joined for talks in previous weeks from my colleagues.

They discussed geology, hazard mapping,

and some of the important lessons that we learned

from the 1980 eruption of Mount St. Helens.

All the things that they talked about really inform what I do,

which is to plan, install,

and maintain monitoring stations on the volcanoes in Oregon and Washington.

One of the most important factors when monitoring

a volcano is having enough equipment in lots

of places on and around all of the areas where an eruption is likely to occur.

We tend to think of volcanoes as erupting from their summit craters,

but that's not always what happens,

and there are things happening underground that are important,

but they can occur kilometers away from where the eruption actually occurs.

When I speak of a station throughout this talk,

I'm referring to scientific equipment that's at a location with infrastructure and

power and often real-time data transmission or telemetry.

The map you're looking at right now focuses on

deformation monitoring at Mount St. Helens.

One thing it doesn't show you is that there are actually

many more seismic stations around the volcano.

They're actually operated by our partners at

the Pacific Northwest Seismic Network at the UW.

Some of the deformation monitoring stations on here

are actually maintained by the consortium UNAVCO.

Some of them are only surveyed periodically.

Another thing on this map I want to point out,

the star in the center is a gas monitoring station

in the crater, which we'll visit later.

Let's visit one of our stations in the Mount Helen's crater.

Our stations are powered by high-capacity batteries that are charged with solar panels.

That's still the best bang for our buck in terms of power availability and reliability.

This does present challenges for us in

the Pacific Northwest with our cloudy and wintry weather.

So we always have to factor that into engineering our stations.

This is my favorite photo to emphasize how brutal our weather can

be. In addition to the rain and the snow and the clouds,

we even have to deal with fun things like rime ice at some of our stations.

Some stations are certainly more reliable than others during the winter,

but our networks are dense enough that we can

continue to monitor volcanoes with a few stations offline.

This is a similar map of Mount Rainier.

The green triangles are

the deformation survey sites I was talking about that we only visit periodically.

It's very challenging to operate stations year round at Mount Rainier.

Again, I want to emphasize,

there are a number of seismic stations not shown on this map

operated by the Pacific Northwest Seismic Network that we actually help maintain.

Here's an example of some infrastructure at

Observation Rock and one of my colleagues from the PNSN,

and a photo of a GPS monitoring antenna at Camp Muir.

If you've ever climbed one of those primary routes,

you may have walked right past our antenna.

The bottom line with volcano monitoring is that we go from a volcano at

rest to a condition of unrest and possibly an eruption,

and we want to be able to forecast how we go from point A to point B.

We're constantly asking and trying to answer a series of

seemingly simple questions about the volcanoes in the Cascades.

Of course, getting those answers is actually pretty challenging and the idea

is to form a conceptual model which is

something that Heather discussed in her talk a few weeks ago.

The important idea here is that we're not trying to

get a perfect photograph of what's happening underground,

but we're trying to visualize and describe the structures and the processes that are

occurring so that we can provide a better estimate of

what might happen when a volcano changes behavior.

Getting the answers to the questions that we're asking

about what's going on beneath the surface is not so straightforward.

The good news is that volcanoes usually change

behavior in ways that are detectable before they erupt.

But every volcano and every eruption behaves differently.

The tools and techniques that help forecast an eruption in

one situation may differ from the ones that are most useful in another time and place.

I'm going to mostly focus on

the three main monitoring techniques we use which are seismology,

gas geochemistry, and geodesy.

I'm going to make sure that my audio should be

shared. So that you don't have to listen to me the whole time,

I'm going to rely on some of my colleagues to

introduce these monitoring techniques beginning with volcano seismology.

All of the videos and animations in this presentation are publicly available on the web.

FEMALE_1:    Most earthquakes associated with young volcanoes are related to the movement of magma

FEMALE_1:    deep beneath the volcano and may indicate that a quiet volcano is becoming active.

FEMALE_1:    Although large eruptions are often preceded by

FEMALE_1:    several significant earthquakes and many small rock-breaking quakes

FEMALE_1:    there's also the continuous release of

FEMALE_1:    seismic energy associated with the underground movement of magma.

FEMALE_1:    Volcanic tremor is a seismic vibration caused by the pulsing of

FEMALE_1:    pressurized magma and gas and can last for minutes or days.

FEMALE_1:    The seismogram is of longer duration and more

FEMALE_1:    continuous than rock-breaking earthquakes of the same amplitude.

FEMALE_1:    Earthquake swarms recorded by seismometers and the ground deformation monitored by

FEMALE_1:    tiltmeters helps scientists determine

FEMALE_1:    the location and depth of moving magma beneath the volcano,

FEMALE_1:    which in turn gives scientists information to issue hazard warnings.

Volcano seismology is extremely important because there is almost always a change in

the number or location or types of earthquakes in the time leading up to an eruption.

The top image here is showing how the combined size of all the earthquakes happening

around Mount St. Helens increased dramatically before an eruption in the 1980s.

The bottom image is showing

the same information for a few weeks in September and October in 2004.

The first explosion of that eruption was on October 1st.

You can see how the seismicity increased,

and then the explosion occurred,

and then the seismicity dropped dramatically before jumping back up again.

A really important fact that Heather mentioned that's worth restating is

that most of the time when activity at a volcano increases,

it means there are changes happening underground,

but most of the time these changes don't lead to an eruption.

A combination of the geologic knowledge and widespread monitoring networks is what

helps us improve our forecasts of what may happen if there's unrest.

The most important vocabulary word in volcano seismology is wave.

Sometimes it's helpful to get back to basics.

Seismic waves are energy waves that travel underground.

When a seismic wave shakes a seismometer,

which is the instrument we use to detect the ground shaking,

we can identify lots of interesting parameters.

The parameters look really different when they get to our instruments than they did at

the source because of all the stuff they have to travel through.

These waves contain information not just about the original earthquake,

but about the layers of the earth that they traveled through.

A couple of things that we look at.

The amplitude of a seismic wave can tell us something about

how big the earthquake was or maybe how close it was.

The wavelength is literally how long the wave is.

Another way to think about this is frequency.

A typical seismic signal looks much messier than this example.

Let's go to our local slideshow volcano,

which is not exactly to-scale or representative of real life

but let's think about how ripples reverberate when you drop a rock in a pond.

If an earthquake occurs,

the seismic waves travel outward in every direction.

As they travel, they change speed and they

change frequency as they pass through the different layers of the earth.

So the amplitude and the frequency of those waves gets affected.

Some types of seismic waves namely the pressure waves or P waves,

travel faster than the surface waves or the S waves.

The path from an earthquake to a seismic station is really complex.

For example, only P waves can travel through liquids,

and we can use the velocity difference between the P waves and

the S waves to do some really cool creative things that I'll talk about momentarily.

Anything that causes the ground to shake creates seismic waves.

This can include explosions,

faults moving, tides, even people walking.

Our seismic instruments are really sensitive so

they can pick up any of these disturbances if they're close enough.

I have a couple of these examples from Mount St. Helens.

Seismologists are able to pick out these different signals

and identify which ones are relevant and which ones aren't.

In addition to earthquakes,

we can detect processes that are happening above the surface,

like the rockfall that you just saw.

We can also detect earthquakes that happen on the other side of the earth.

For example, an earthquake happening in Chile takes

a little bit of time to reach our stations at Mount St. Helens,

but we can detect that.

Seismometers, which are the tools that we use to detect earthquakes,

have changed a lot over the decades.

On the top left is an old smoke drum seismograph where basically you have a pen hanging

like a pendulum and it scratches a piece of paper covered in soot.

As of a few decades ago,

we had the slightly more advanced pen and paper system.

But in recent decades,

seismometers have all gone digital.

In the very basic sense,

a free-moving magnet inside of a coil gets

shaken around and creates an electric signal that we can digitize.

These advances are really important because we can start to see signals from

much smaller earthquakes and much larger earthquakes than we used to be able to.

We can also use software to more efficiently and completely

identify and characterize all the earthquakes happening under volcanoes.

One thing that hasn't changed is that seismometers work best when they're underground.

They're sensitive even to daily temperature changes.

The deeper we can place them the better.

In my job, we spend a lot of time digging holes.

Some are not very small.

We're also trying to tell the difference between lots of different kinds of earthquakes.

You can see visually that each of these earthquakes is different.

Those differences tell us something about the cause of each event.

Some of them like the one on the top,

result from normal tectonic stress or basically rocks breaking.

Some of the ones at the bottom are caused by

resonance as magma travels through cracks underground.

There are some really cool things we can do with volcanic earthquakes.

Using multiple stations, we can figure out where the earthquakes originated.

It's important to have seismic stations

surrounding earthquake locations to get good precision.

It's also important to have seismometers high on the edifice

at a volcano so that you can locate shallow earthquakes.

We can also determine, for example,

how a fault moves in an earthquake,

which tells us something about the stress state under a volcano.

We can do some relatively simple statistics to see if anything has changed over time.

For example, volcano seismologists

calculate what's called a b-value and you don't need to know what that stands for.

Bottom line is that over time,

there are a lot more small earthquakes happening than large earthquakes.

If we see a change in the relative number of different sizes of earthquakes,

then that may get our attention.

If our resident seismologist could put seismic stations everywhere, he would.

It's not really practical,

but sometimes we do have opportunities to do just that over the short term.

The project iMUSH or Imaging Magma Under St. Helens,

was a really cool collaborative project.

It involved multiple universities, the USGS,

the Forest Service, and the National Science Foundation.

In addition to having 70 temporary seismic stations,

the program created its own earthquakes in the form of small controlled explosions.

They knew the exact location and size and

frequency of the ground shaking that they caused.

By recording the seismic waves at all those seismic stations,

they were basically able to take an MRI

of the subsurface under and around Mount St. Helens.

We look at these images in terms of seismic wave velocity.

The changing speeds that the seismic waves travel at

underground are related to the subsurface geology.

We can get an idea,

for example, of where magma is stored.

We had way more sensitive instruments before the 2004 eruption than we did in 1980

but there wasn't an uptick in seismic activity in

2004 until just a few days before the first explosion.

Part of the reason was that this was a very different type of eruption from 1980;

it was contained in the crater,

it tapped a very shallow magma source,

and there were fewer volcanic gases.

There was less strain happening underground,

fewer earthquakes, and the eruption was less explosive.

This is showing the earthquakes and seismicity and

the different types of volcanic earthquakes from September 23rd through 26th.

Again, just about a week before the first explosion on October 1st.

Studying the seismic signals prior to and

during the eruption is a key component of what we do.

We need to know what the background earthquakes look like,

to know if there's a change in the behavior.

There's plenty of background activity at Cascade volcanoes,

and mostly they're related to tectonic forces.

We see recurring patterns of earthquakes beneath Mount St. Helens,

Mount Rainier, Mount Hood,

and even Newberry, and Crater Lake occasionally in southern Oregon.

There are multiple hazards associated with eruptions.

Joe's talk covered a number of them when he described

hazard maps. And our sensitive seismometers can pick up more than just earthquakes.

I showed the example of rockfall,

but we can also detect what's called a lahar.

An important project we're working at CVO is

expanding our lahar detection ability at Mount Rainier.

The word comes from the Javanese language,

and it describes mudflows and debris flows that originate on volcanoes.

Cascade volcanoes are steep,

they have lots of unstable material and lots of groundwater and

surface water that's available to trigger those mudflows.

Lahars usually occur when there is an eruption which displaces

lots of material and rapidly melts the snow and the glaciers.

The mudflows typically travel down existing channels or

drainages as Joe described when he discussed hazard maps.

Lahars are extremely destructive.

They're one of the most deadly and far-reaching volcanic hazards.

In recent years we've been adding

seismic monitoring stations along the Puyallup and Carbon River drainages

on the west side of Mount Rainier as part of an upgrade

to an old system for detecting lahars.

Downstream communities, such as Orting are actually built on

old deposits from lahars that originated on Mount Rainier.

In this figure, you're seeing seismic waves and their energy at different frequencies at

seven different monitoring stations

in Mount Rainier National Park and along the Puyallup River.

There was a small debris flow in

the Mowich drainage and it was clearly picked up by our sensors.

These little events will give us practice with signals that are

similar to what we may see for a larger, more dangerous event.

This image is a couple of years old.

There are actually a few more sites now where we have seismic monitoring stations.

You can see we have stations in the park,

as well as a series of stations tracking along the Puyallup River,

we've added a few more there and at the Carbon in recent years,

and we're working on adding more monitoring on additional drainages,

including the Kautz, the Nisqually, and White River.

If seismicity is a volcano's heartbeat,

then gas emissions are like it's breathing or other bodily functions, if you like.

Cindy Werner:    When the magma is very deep beneath the volcano,

Cindy Werner:    it has a lot of gas dissolved in it,

Cindy Werner:    but when that magma starts to rise towards the surface,

Cindy Werner:    the gases will come out of the magma.

Cindy Werner:    If this happens very quickly,

Cindy Werner:    then the volcano might erupt explosively.

Cindy Werner:    At volcanoes like Mount St. Helens,

Cindy Werner:    it's important to monitor the gases on a routine basis because,

Cindy Werner:    at any time, there could be another injection of gas-rich magma from below,

Cindy Werner:    which might cause the volcano to enter another eruptive phase.

Volcanic gases are really interesting to us.

They're the driving force behind eruptions.

It's ultimately these gases that cause the volcanic tremor and

the surface deformation that we see using other monitoring techniques.

The gas composition is also interesting because it points straight to

the geology and geochemistry of the magma system.

Two important things that I will focus on are

the types of gases and how quickly they come out of the ground.

The bottom line is that molten material is

generated in the mantle under high pressures and temperatures.

As this material moves toward the surface,

the pressure and temperature changes.

As this happens, gases that were dissolved in

the magma start to come out of the liquid and into their vapor phase.

This is just like the release of CO2 gas when you

suddenly opened a soda bottle and lower the pressure inside of it.

Different gases come out at different depths and temperatures.

For example, CO2 comes out much deeper than SO2.

Water starts to vaporize out of magma at

about a kilometer below the surface, so very shallow.

Water is important, it's

the most important volcanic gas because it's what will really drive an eruption.

This is a photo of one of my colleagues with a group

from the Indonesian equivalent of the USGS,

standing in front of a gas monitoring station that they installed at Timbang Crater.

The area is a critical agricultural center for the island of Java.

At Timbang Crater, we don't see a lot of eruptions there,

we see steam explosions.

But more importantly, there are

toxic carbon dioxide plumes basically that come out of the crater.

Because it's denser than air,

the carbon dioxide flows downhill.

This region has seen fatalities from this phenomenon in the past.

Basically, the CO2 comes out of the ground and then pools in these low-lying areas.

You can see on the right the percentage of

CO2 detected over a few months at the station.

At four percent CO2,

you are considered an immediate danger by OSHA,

thirty minutes at five percent will lead to feeling intoxicated.

A few minutes at 7-10 percent CO2 leads to unconsciousness,

and at around 14 percent,

you're pretty much at instantaneous lights out.

What gases does a volcano produce?

We've gone over CO2 a little bit,

I mentioned a couple others.

Carbon dioxide separates from magma at great depths.

Sulfur dioxide exsolves or comes out of the magma at shallower depths,

it might tell us that magma is closer to the surface.

Hydrogen sulfide comes from magmatic gases interacting with groundwater.

It smells like rotten eggs.

You may have smelled hydrogen sulfide

if you've visited hot springs or been to Yellowstone National Park.

Hydrogen chloride and hydrogen fluoride are halides,

and they also indicate magma near the surface,

but they dissolve really easily in water.

They can actually lead to acid rain and

other negative health effects if people are exposed to them.

Finally, water. Water is really tricky to measure.

It likes to condense,

it gets emitted in places that are really tough to access,

but water and carbon dioxide are

really important barometers for how eruptible a magma is.

They're just really hard to measure.

CO2 becomes a vapor at great depths.

It's colorless and odorless,

and as we talked about,

it can be hazardous and it kills trees and vegetation.

Sulfur dioxide or SO2 comes out at shallower depths.

It's colorless, but it has this pungent,

acidic odor and it really irritates your mucus membranes,

your eyes, and your nose and throat.

Hydrogen sulfide, or H2S,

again, is caused by sulfur interacting with water.

It's colorless and it's flammable,

and it smells like rotten eggs.

Actually your nose is much better at

detecting H2S than any gas monitoring instrument at low levels

but at toxic levels,

it actually becomes odorless.

Gas geochemists try to do creative things.

They try to relate how much gas is coming out

of a volcano to how much material has erupted.

The approach works okay in places like Hawai'i because sulfur

is easy to detect and it dissolves easily in basalt.

In the 1980s in Alaska,

there seemed to be a correlation between

gas emission rates and the likelihood of an eruption from some volcanoes,

but this was purely empirical.

Every volcanic system is different,

and even a dramatic change in gas emissions

does not necessarily mean an eruption is imminent.

How do we sniff gas?

The misspelling is intentional and you'll see why as soon.

We either directly sample the gas or we detect it remotely.

Direct sampling tells us what kinds of gases are present,

and remote sensing tells us how fast the gas is coming out.

One is like checking to see if a stove burner is hot by touching it,

and the other one is holding up a thermal camera to check.

Direct gas sampling has been around for over 100 years.

If you see this font on a USGS publication then you know it's more than a few years old.

The gas is collected in glass bottles.

They have to be sealed,

they have to be shipped to a lab and analyzed.

It's tricky to collect the samples,

it's tricky to transport them,

and it takes awhile to get results.

USGS scientists developed a gas sampling lab that can operate as its own station.

The equipment is called MultiGAS,

and it includes optical and electrochemical sensors.

It was designed and built in-house by

the Volcano Disaster Assistance Program and the USGS,

and it solves the problem of needing to hike around with gas flasks everywhere,

and it runs every day of the year.

Our other method of detecting volcanic gases is to use

remote sensing to see what the flux is.

In other words, the rate of gas emissions.

We have a remote sensing instrument called a DOAS scanner.

It has a camera that captures light from the sun coming through the atmosphere.

Volcanic gases cause different wavelengths of light to reflect

and refract in predictable ways that we can measure.

If you think about how the atmosphere filters out blue light at low angles

and causes the sunset to be red and orange, it's like that.

The DOAS station is actually pretty complex,

but above the surface,

it just looks like a little camera.

We put our gas monitoring instruments at

different places on the volcano depending on their purpose.

MultiGAS is meant to tell us the types of gases coming out at the source,

so it needs to be right in the gas plume.

DOAS or spectrometers need to be outside the plume so

that they can scan a wide area and see how fast the gas is coming out.

Using these two methods in combination helps us see if there are changes in a system.

Again, changes don't mean an eruption is eminent.

Even external factors like seasonal groundwater changes can affect the gases we detect.

The USGS has done campaign style MultiGAS measurements at multiple Cascade volcanoes.

The bottom line is that there isn't much gas.

There are signals consistent with magma cooling at Mount Hood and Lassen and Mount Baker.

Most of the gases seen are typical for hydrothermal areas.

In some isolated instances,

the gases do create their own hazard.

Mammoth Mountain has seen low lying areas with CO2 build-up.

That's in northern California.

You sometimes hear stories of climbers needing to be

rescued from caves or holes high on Mount Hood,

and they report experiencing headaches, shortness of breath,

and nausea, which is consistent with exposure to hydrogen sulfide.

We have one permanent MultiGAS station in the Cascades on the new dome at

Mount St. Helens, and that station is named SNIF, S-N-I-F.

Here's a selfie of me helping service the station a few years ago,

wearing my safety gear.

There's too little gas to be detectable, much less hazardous.

This is what the inside of a gas station looks like,

mostly batteries and gas canisters.

I'm going to let my colleague, Peter,

tell you a little bit more about monitoring at SNIF.

Peter:    Here we are at Mount St. Helens, in the crater.

Peter:    This is the Mount St. Helens SNIF site.

Peter:    This is a gas monitoring site that was put in at the end of August 2014,

Peter:    and its purpose is to monitor gas plumes within Mount St. Helens.

Peter:    This plume here contains mostly water vapor,

Peter:    but also contains carbon dioxide,

Peter:    sulfur dioxide, and a really trace amount of hydrogen sulfide.

Peter:    Its composition is consistent with a very shallow, hot,

Peter:    oxidized melt, and so it's

Peter:    a magmatic gas essentially that's being emitted to the atmosphere.

Peter:    It has a very low carbon-sulfur ratio which is consistent with low pressure degassing.

Peter:    It's mainly sulfur dioxide,

Peter:    very little hydrogen sulfide,

Peter:    which is also consistent with high temperature, low pressure degassing.

Peter:    This is the only place in the Cascades where we presently get sulfur dioxide degassing.

Peter:    Everywhere else there's hydrothermal systems that are

Peter:    in-between the point where the gas

Peter:    separates from the magma and then makes its way to the surface.

Peter:    It interacts with hydrothermal fluids and rocks and water and it's during that process,

Peter:    any sulfur dioxide that's degassed,

Peter:    gets converted into hydrogen sulfide and other species.

Peter:    It's not actually emitted to the atmosphere.

Peter:    This station, which you can see in front of you,

Peter:    is actually a station to monitor just those four gases,

Peter:    water vapor, carbon dioxide,

Peter:    sulfur dioxide, and hydrogen sulfide.

What do we learn from SNIF?

There's not a lot of gas at Mount St. Helens between eruptions.

Why? We don't know yet.

In fact, most of the gases emitted are at levels that are too low to be detected,

especially as we get further and further away from the 2004-2008 eruption.

We did have a little blip in background levels in 2017 that went quickly back to normal.

The cause isn't fully understood.

The plots on the right side of this image just show calibration information.

There are gas cylinders at SNIF

that contain known concentrations of these gases for comparison.

The cylinders have to be changed out every two years or so.

I'm going to wrap this up with my favorite suite

of monitoring tools which are used to detect ground deformation.

Here's an introduction by my friend DZ.

DZ:    One of the things moving magma tends to

DZ:    do is it tends to deform the surface of the earth.

DZ:    It tends to cause the surface to bulge upward or to spread apart.

DZ:    One of the instruments we use to measure deformation is a tiltmeter.

DZ:    A tiltmeter measures very subtle changes in the surface of

DZ:    the earth as magma accumulates beneath the station or moves upward, for example.

DZ:    Another instrument we use is the Global Positioning System or GPS.

DZ:    We put several of them out on the volcano,

DZ:    we record signals from satellites orbiting above the earth,

DZ:    and we look for movement of one of the stations with respect to the other.

DZ:    As the ground deforms,

DZ:    the volcano changes its shape,

DZ:    those stations move, and that

DZ:    tells us something about what's going on beneath the surface.

As things change below the surface of a volcano,

we may or may not see earthquakes,

we may or may not see a change in volcanic gases,

and we may or may not see the ground surface move.

Usually these movements are so

small that there are very few ways that we can detect them.

But if we do detect motion at multiple locations,

we can actually use the data to create a model for what's happening underground.

For example, if there's a magma intrusion

and it causes ground swelling and we have enough stations,

we can start to figure out the approximate volume

and the shape and the depth of that intrusion.

One instance of very dramatic ground deformation that we could see with our eyes,

was the bulge that grew on the north side of St. Helens in 1980.

It was identified in March of that year,

and this was pre-GPS.

Back then, scientists used instruments where they bounced the laser beams off of

reflectors that were on the mountain and that's how they could see ground movement.

They could determine if the reflectors were moving towards them or away from them.

The obvious over-steepening and swelling of the volcano is really what

primed it for that catastrophic landslide and

the lateral blast that destroyed so much of the area.

But because the hazard was identified in advance,

the area was at least kept closed to

the general public despite really high demand for access.

I'm going to use the term GPS to describe our modern deformation monitoring stations.

But if you want to impress your friends at dinner parties,

the correct terminology is now GNSS,

which stands for Global Navigation Satellite System.

It includes GPS, which is operated by the United States,

but it also incorporates satellite systems launched by other countries.

The funny-looking antenna and tripod are basically

super advanced versions of the GPS antenna in your cell phone.

Satellites flying above the earth at 20,000 kilometers altitude,

at a speed of about 14,000 kilometers an

hour emit signals that are picked up by our antennas.

GPS satellites have really accurate clocks,

so we know what time the signal was sent,

and then we can see how long the signal takes to reach us

and determine the distance the signal traveled.

If you record signals from multiple satellites with known orbits,

you can start to triangulate your own position.

Then if you move,

you'll calculate a new position.

Your phone does an okay job of this.

It gets you down to tens of meters of accuracy.

A little handheld GPS unit will get you a few meters.

But that's not good enough for us.

A large magma intrusion can cause the earth's surface to deform less than a centimeter,

so that's why we need such ridiculous looking antennas.

Our advanced equipment can deal with issues like effects from the atmosphere.

Our equipment can filter out phantom GPS signals that bounce off the ground surface.

The antennas do have to be secured to the ground by

an extremely stable structure that ideally won't bend too much in the wind or snow,

or even respond to big temperature changes.

Picking up these tiny signals is made even more challenging

because the ground beneath our feet is always moving even if we can't feel it.

Big earthquakes cause big ground motions.

Tectonic plates are always shifting and rotating.

On the left, you see arrows and those are pointing in the direction of movement

of hundreds of these GPS stations that are installed throughout Washington and Oregon.

Most of them are operated by UNAVCO and

the arrows may represent just a few centimeters of movement over many years.

But you can see that the whole region is shifting and rotating.

We also have really interesting very deep,

slow earthquakes that happen in the Pacific Northwest.

They're too slow to feel,

but they do cause GPS stations to move.

This is known as episodic tremor and slip.

It's not related to our volcanoes directly,

but we have to be aware of

all these ground movements so that we can subtract them from our data,

so that the only movement we see is caused by volcanic processes.

We did see ground movement during and after the 2004 Mount St. Helens eruption.

When the eruption began,

there was a GPS station at Johnston Ridge Observatory to the north.

When the eruption started,

the GPS began moving to the south and down,

so basically toward the volcano.

This was happening as the eruption drained material from the reservoir underground.

It's like deflating a balloon.

Ever since the eruption,

the domes from the 1980s and early 2000s have been cooling and contracting.

The station in the bottom figure was located on the 1980s dome.

Sadly, the station was destroyed by snow a few years ago.

But it showed a steady trend moving down to

the east as the domes cooled over the span of decades.

Another method for seeing the ground move is using tiltmeters.

They're basically really sensitive bubble levels put in boreholes in the ground,

and if the ground tilts, the bubble goes off level and we see the signal.

They're wonderfully sensitive, but they can't handle really big movements.

They're also very difficult to install.

We have to drill at least 10 feet underground.

There were a lot of interesting tiltmeters signals on Mount St.

Helens during the 2004-2008 eruption.

During a period of dome growth in the crater,

there were these pulses of motion that would last a few hours where the surface would

inflate a tiny amount very quickly and then deflate very gradually.

The tiltmeters in the crater would move away from

the eruption vent and then move back closer together.

Another method that we use for monitoring ground deformation is called InSAR

or Interferometric Synthetic Aperture Radar.

Say that 10 times fast.

Just like the reflection measurements in the 1980s,

they're satellites transmitting electromagnetic waves that

measure the signals that are reflected back to them from the ground.

They do the same thing over and over as they pass over an area,

and we can compare before and after images to see changes.

In recent years, satellite imagery is becoming more available than ever.

It can show us deformation over

really wide areas and in places where we don't have monitoring stations.

These satellite methods do have limitations.

The types of electromagnetic waves they use don't travel well through trees.

They don't reflect well off snow.

We have a lot of trees and snow in the Cascades.

The satellites only pass by about once every two weeks,

which during an eruption is a very long time to wait.

The method also isn't as sensitive to tiny movements as our GPS or tilt instruments.

Let's put this all together.

A perfect volcanic eruption at a perfectly well-behaved volcano is

preceded by earthquakes that show us that magma is

moving and we can track those earthquakes underground.

We maybe see CO2 gas coming out and

then H2S and then SO2 and water comes in a predictable way.

That's rarely the reality.

Volcanoes are really complex and every system and eruption is unique.

Because of this inherent complexity,

we have to use lots of different tools above and beyond what I've discussed here.

For example, infrasound uses

special microphones to detect pressure waves in the atmosphere.

They can be caused by explosions or by big lahars traveling down a drainage.

Gravity changes tell us if the mass of the earth below us has changed or moved.

Yes, gravity is different at every point on earth and it can change.

Monitoring stream temperatures and chemistry can be important diagnostic tools.

It's also important to track stream levels and sediment loads

to assess flooding hazards even when there's no volcanic unrest.

Thermal imagery has been very beneficial during

Kīlauea's erupted activity since the 1980s in Hawai'i.

Finally, lightning detection is playing

an increasingly important role in volcano monitoring.

Volcanic lightning is real.

It's hands down the coolest natural phenomenon.

You can try to argue that if you want.

Lightning can be detected using sensors that

are thousands of miles away spread around the globe.

This can help identify eruptions that are occurring

at remote volcanoes that don't have any monitoring equipment.

That has really important implications for air travel.

Even with all this information,

we can't predict exactly when or how a volcano will erupt.

We can track changes carefully.

We can estimate how likely different events are to occur,

and we provide the best possible information to officials and the public.

But typically, a volcano at rest tends to stay at rest.

Just like me early in the morning.

It's really important to have multiple tools to monitor volcanoes.

They all behave differently.

There are examples of eruptions where the ground deformed,

but there weren't a ton of earthquakes.

There are examples where earthquakes increased dramatically,

but there wasn't a noticeable change in gas before an eruption.

We use the information provided by geologists to help plan our monitoring networks,

including the locations of our stations and the types of equipment that we install.

We recently added seismic and GPS stations higher on the edifice at Mount Hood.

We're planning on adding MultiGAS there later.

Our lahar monitoring network at Mount Rainier is ever expanding,

and we're adding improved lahar detection capabilities at Mount St. Helens.

We'll also be installing a scanning DOAS gas instrument

at Mount St. Helens this summer,

and we're working to add stations around Glacier Peak.

What is our most important monitoring tool?

I would argue that our most important tool is our eyes.

What we see helps us identify how volcanoes behaved in

the past and what hazards they may present in the future.

Geologists see the deposits from past eruptions and from lahars.

People like the park ranger on the left here,

have witnessed small debris flows at Mount Rainier.

The eyewitness accounts and photos from the Mount St.

Helens 1980 eruption helped document

a landslide and lateral eruption that were

completely unheard of in the world of volcano science.

On that note, one technology that's really

expanded our abilities in recent years is unmanned aerial systems.

A small group of amazing people worked for many years to

add UAS capabilities to our toolbox.

The program was scaled really rapidly during

the Kīlauea 2018 eruption response and that eruption had everything;

earthquakes, ground deformation, gas emissions.

But it was also an exciting opportunity to explore and test new monitoring techniques.

None of the techniques really capture the impressive nature of an eruption,

quite like the photos and videos.

So please enjoy some of this drone footage from the 2018 response.

Thank you so much for your time and attention and

I would be happy to answer any questions.