Tuesday, December 22, 2015

Explosive dangers at Kilauea volcano

- Janine

My first AGU Fall Meeting was so full of wonderful science - emergency management exchanges with Colombia to address hazards of lahars (Nevado del Ruiz), volcanic lightning, active monitoring of volcanoes, community preparedness, and all aspects of volcanic activity above, on, and below the surface. I had great conversations with people excited by their work and eager to communicate their work with the AGU masses.

One of the many posters that caught my attention was "Don't forget Kilauea: Explosive Hazards at an Ocean Island Basaltic Volcano" by USGS volcanologist Don Swanson. When I talk to people about volcanology the first thing to come up is usually how cool it must be to study lava flows at Hawaii. Well, I don't study effusive lava flows, I am on the explosive end of the spectrum with dome collapse block and ash flows, and I have not yet visited Hawaii. One thing is obvious, many people I talk to think of the relatively safe (with exceptions) pahoehoe lava flows when they think of volcanic activity, especially at Hawaii.

Kilauea Overlook, photo by NPS.
What many people are not aware of is the very dangerous explosive personality of Kilauea. Around 2.6 million people visit Kilauea volcano ever year according to the National Park Service, you may be one of them (I hope to be, one day). You may have even been lucky enough to see the beautiful lava flows like this one:

But do you know about Pele's more dangerous and explosive side?

According to Don Swanson (and references listed below) Kilauea's explosive nature includes: subplinian ash plumes up to heights of 10 km; phreatomagmatic (magma + water) and phreatic (steam) explosions; pyroclastic density currents (especially pyroclastic surge); volcanic ashfall; and ballistic ejecta (flying rocks) around the entire summit area with total eruption volumes up to 0.02 km3 (VEI 3) - based on studies of explosive events that occurred during the past 2,500 years.

These are relatively small events, but for any of those 2.6 million people visiting, or 2575 people living (2010 population) near the summit they can be deadly. Once these explosive episodes get going they can even last for centuries (Swanson and Houghton, 2015).

One of these phreatic (steam) eruptions occurred on November of 1790 produced a pyroclastic surge that killed a few hundred warriors and their families (Swanson et al., 2015). This same area is currently visited by around 5000 people per day.

Extent of the November 1790 pyroclastic surge with numbers indicating possible locations of groups impacted, Swanson and Houghton, 2015.
The below video is of a pyroclastic density current at Soufriere Hills Volcano on Montserrat. You can see the hot expansion of the ash and gas as it flows down the hill, similar processes to a pyroclastic surge that may be seen at Kilauea (this video is of a denser flow from a dome eruption). The 1790 event was likely composed mostly of fast moving superheated steam that took the victims by surprise (Swanson and Christiansen, 1973).

Eruptions that produce ballistic blocks are not uncommon at Kilauea. These are blocks that are thrown out of the crater during an explosion and deposit around the crater, as seen in the video below:
This small eruption at Tavurvur volcano shows ballistic blocks/bombs flying out of the eruption plume that then deposit around the crater.

The map below shows the locations of ballistic blocks from eruptions in 1790, 1924, and 2008 (Courtesy of Don Swanson):
Ballistic block locations from Swanson and Houghton (2015).
These blocks can get quite large - with measured blocks up to 195 cm in diameter, if you have visit the summit you would have walked right past them. See this HVO article for more information on ballistic blocks at Kilauea.

Block at the Kilauea Overlook, courtesy of Don Swanson, Swanson and Houghton, 2015.
The ash plume is estimated to have reached 30,000 feet high based on visibility reports. Depending on the wind direction at the time of eruption, ashfall could impact Ka‘ū, Puna, and Hilo, and very fine ash could reach Waikiki (Swanson and Houghton, 2015). An ash plume also produces hazards for the many trans-Pacific flights that travel near the island.

Why is this explosive potential not more widely known? There is little written history about the earlier eruptions for us to learn from. The work done by Swanson and others in uncovering the narrative and geologic evidence from the 1790 event give us a much different view of the seemingly effusive and "safe" Kilauea volcano. If you get the chance to visit Pele, keep her more explosive side in mind while you enjoy this beautiful effusive phase.

All images and information from Swanson and Houghton (2015) with the permission of Don Swanson, USGS Hawaii Volcano Observatory. Link to poster below.

References for more information:
Helz, R.T., et al., 2014. Microprobe analyses of glass from Kīlauea tephra: USGS Open-file Report OF2014-1090.

Mastin, L.G., 1997. Evidence for water influx in 1790: JGR, v. 102, p. 20,093–20,109.
Schiffman, P., Zierenberg, R., et al., 2006, Acid-fog deposition at Kilauea: Geology, v. 34, p. 921–924.

Swanson, D.A., Christiansen, R.L., 1793. Tragic base surge in 1790 at Kilauea Volcano. Geology, 1: 83-86.

Swanson, D.A., Rose, T.R., et al., 2014. Cycles of explosive and effusive eruptions at Kīlauea: Geology, v. 42, p. 631–634

Swanson, D.A., Weaver, S.J., Houghton, B.F., 2015. Reconstructing 1790 lethal eruption: GSA Bull., v.127, p. 503-515.

Swanson, D.A., Houghton, 2015. Don't Forget Kīlauea: Explosive Hazards at an Ocean Island Basaltic Volcano. AGU Fall Meeting, San Francisco, PA43C-2202.

Wolfe, E.W., Morris, J., 1996. Geologic map of the Island of Hawaii: USGS MI Map I-2524.

Friday, December 11, 2015

In the Company of Volcanoes at AGU

-Alison and Janine

The largest geoscience conference happens every year in San Francisco. The American Geophysical Union (AGU) meeting draws more than 20,000 participants from around the world every December. We gather to share our new results, catch up with friends and colleagues, and drink beer. Janine and Alison will both be attending this year. Although we have been to conferences in three other countries together, this will be Janine's first AGU and the first conference in the US where we both be in attendance. Both of us have posters to present on our research, and then will spend the rest of the visit stuffing our brains full of new information and names of new colleagues. If you are going to AGU we'd love you to stop by and chat!

Janine and Alison in Japan for the IAVCEI conference in 2013.
So what sort of topics get covered in a 20,000 person conference? Way more than we could fit in a blog post, but you can get the idea just from our examples.

Alison will be presenting in a session called "Eruptive Processes and Watery Hazards of "wet" Volcanoes on Land, in the Sea, or under Ice".

Her poster is Wednesday afternoon in Moscone South, or the poster hall that is a sea of scientists, figures, and concrete. You can read her abstract on the AGU website V33B-3104: Distribution of ejecta in analog tephra rings from discrete single and multiple subsurface explosions. A shorter title for this would be "where and how stuff gets flung out of explosion craters" (Also, Alison knows there haven't been any new explosion videos in a while. This will be rectified in the new year).

Sediment transport in progress, also known as an explosion from the 2014 experiment session.

If you want to learn more about "What dominates a crater's size, the largest single explosion of the formation process or the cumulative energy of many? Results of multiblast crater evolution experiments" Alison's colleague Ingo Sonder will be presenting a poster in the same session (V33B-3105).

For those who want to see pictures of actual volcanoes, but still think about how these experiments improve our understanding of how they form, Alison's advisor Greg Valentine will be giving a talk called (V32A-03) "Tephra ring interpretation in light of evolving maar-diatreme concepts: Stracciacappa maar (central Italy)" on Wednesday Morning from 10:50 to 11:05 in Moscone South 310.
Real deposits: Alison's favorite outcrop of maar ejecta in Frijoles Canyon, New Mexico, USA.

If you're into huge, dangerous, hot, and fast volcanic avalanches, Janine will be presenting "The 2005 and 2010 dome collapse driven block and ash flows on Shiveluch volcano, Kamchatka: Morphological analysis using satellite- and field-based data" (V23A-3066) on Tuesday afternoon in Moscone South poster hall. This is about how complex these (very) large block and ash flows are, from some of the largest dome collapse events around the world.

The February 2005 Shiveluch block and ash flow destroyed 10 square kilometers of forest and threw trees and rocks around like match sticks.

For those who want variety there are lots of posters discussing non-volcano things including a whole bunch of new stuff Pluto and its Kupier Belt friends, heliophysics (or all about our Sun), ground water, magnetospheres (planetary magnetic fields), geodesy (the shape of the Earth), geoinfromatics (where computers meet maps and geology), and paleobiology (things that aren't alive anymore). There are also a whole bunch of sessions devoted to geoscience education. For example Alison's husband, Topher Hughes, will be presenting in the GeoEd poster session on Wednesday morning, "'What's a geoscientist do?' A student recruitment and education tool" (ED31B-0891).

Geoscientists tend to love what we do. We also don't all do the same thing. When students hear early on that geoscientists aren't all old white dudes in lab coats they can better imagine themselves fitting into the field. Hopefully they can find something they love as much as Alison does these rocks from Askja Volcano in Iceland.

Janine and Alison will both try to tweet from the conference at #AGU15. Though from past experience, it is a full on conference and it is a lot of work just to keep up, so tweeting and blogging take a back seat.

Saturday, November 21, 2015

Not all holes in the ground are the same.


Since I spend a lot of time thinking about holes in the ground, from the ones I make with dynamite, to volcanic craters, I have to spend some time thinking about other mechanisms, both human and natural, that make similar looking holes in the ground. If I want to say anything about volcanic holes in the ground, such as maar volcanoes, I need to know what makes them unique. If I want to recognize just one type of hole in the ground remotely on Earth, or other planets, I need to know more about holes in the ground in general.

Google Earth Image of Hole in the Ground Maar (left) and its neighbor Big Hole (right), these maars are located in Western Oregon.
Let's start with the largest holes in the solar system, impact craters. One of the most important processes for changing the surface of a planet, or any planetary body (moons, asteroids, etc.) is meteorite impact. There is a lot of junk flying around in space. This junk (rock, dust, ice) runs into other junk and in most cases form impact craters on the bigger piece of rock/ice. These impact craters are really common on planets with no atmosphere, biosphere, or plate tectonics like Mercury or our Moon. They are fairly circular holes in the ground with ejecta spread around them radially. In that simplified description they sound a lot like my explosion craters. In fact, I just went to the Geological Society of America Meeting in Baltimore in early November to talk about the ways my explosion experiments might be of interest to planetary geologists. I hung out a lot in talks about meteorite impacts. The session where I gave my presentation combined volcanic flows and impact processes. It had never been combined quite that way before, and people from both groups quickly learned how much we have in common, and what we don’t.
This enhanced color mosaic shows (from left to right) Munch (61 km/38 mi.), Sander (52 km/32 mi.), and Poe (81 km/50 mi.) craters, which lie in the northwest portion of the Caloris basin. That means these three impact craters are inside an even BIGGER impact crater. Image and links courtesy of NASA Photo Journal.
Meteorite impacts are found on Earth, but they tend to get modified pretty quickly by water, plants, and plate tectonics. One of the best examples of a well preserved meteorite impact is Barringer Crater in Arizona, also known as Meteor Crater. It was formed 50,000 years ago, and is fairly young as geologic processes go. The crater is about a kilometer wide and more than 200 m deep (that is more than one hundred 6 ft tall people standing on each other’s heads). I almost went there last week, but we only had about 20 minutes and other places to be, so it had to wait for another trip when I can spend more time to truly appreciate it. Meteorite impacts can be tiny impacts from micrometeorites (less than a millimeter), to 1,000 km across (Caloris Basin on Mercury). Earth has experienced some large impacts in the past, some are credited with impressive things like the formation of our moon and the extinction of the dinosaurs. If you want to know where the nearest impact crater is to your home you can check out this interactive map. And while news stories occasionally surface about asteroids heading for Earth, like when HM10 did a flyby this summer, it is highly unlikely that a large asteroid is heading for Earth any time soon.
Barringer Crater/Meteor Crater, Arizona (Image Wikimedia Commons).
Other holes in the ground can be formed by the release of methane (see these holes in Siberia, which are still being studied). The current leading theory is that these are formed from the sudden release of gas hydrates, or methane gas trapped in the pore spaces between sand grains. When the gas escapes, the ground above the release is disrupted (not unlike my explosions) and the sediment collapses downwards because the spaces between grains is now empty. What is left is a steep sided hole. These are, for now, much smaller - reaching 50-100 m in diameter.

A cenote, collapse of a cave roof, common in Mexico and important source of fresh water (image Wikimedia commons).
There are also less explosive ways of making holes in the ground. Karst topography, formed by the dissolution or erosion of rock (frequently limestone), is full of caves or other complicated shapes. There is quite a diversity of karst type topography on Earth, but most people will be familiar with caves and sink holes. I am interested in them because they can form small lakes or holes in flat ground and look a lot like maar craters. In areas of permafrost, where the ground is frozen all year, thermokarst (or the melting of that frozen soils) results in lots of roundish lakes in places like Alaska and Kamchatka. Some of these places are volcanically active, and since maars are formed by the interaction of rising magma and water it is common to have maars and thermokarst lakes side by side! Karst lakes and sink holes come in a wide range of sizes from a few meters across to a few kilometers diameter.  
Permafrost lakes in Alaska (Image Wikimedia commons).
Maar craters and permafrost lakes side by side on Seward Peninsula Alaska (image from Google Earth). The Devil Mountain lakes are two of the five maars in this image.
Kettle lakes, also a glacial feature, are formed by the melting of large chunks of ice that are trapped in the deposits left behind by a glacier. These features tend to be smaller, usually a few hundred meters or less in diameter.
Kettle lake from Isunngua, Greenland (image Wikimedia Commons).
There are also other volcanic features that have round craters, like scoria cones, rootless cones, and calderas. All these different features have similarities, but also differences. If the differences are not well preserved, or not visible from a satellite image it can make identifying what processes caused that particular hole in the ground very difficult.  One of the first traits to tell these features apart is size: maar craters vary from 100 – 5000 m across, with most around 700 m across. This means they are easy to tell apart from large impacts and large calderas, or tiny little kettle lakes and smaller sink holes, but that leave a whole bunch of overlap in the middle.  The best way to confirm how a hole in the ground formed is to look at the deposits and since I love field work, I don't mind this. Since I don't have enough time or money to check them all myself, I am grateful to the work done by previous geologists to describe these many processes ,and the resulting hole, in detail.  

Crater of Tambora Volcano, Indonesia is 6 km wide and over 1 km deep (Image Wikipedia Commons).

This list isn’t all inclusive because humans and animals have also found ways to produce round holes, but as they aren’t too common on other planets, I only have to worry about them when looking at Earth. Open pit mines make for very impressive holes in the ground. Although some of them are in kimberlite pipes, which have some similarities to maar volcanoes, the largest mines are for copper.
Lavender Open Pit Mine in Arizona. This is a common approach to extracting copper from the Earth's Surface. (Image Wikimedia Commons).
Kiberley Pipe in South Africa is a famous diamond mine and where Kimberlite gets its name (Image Wikimedia Commons).
Now as you look through images on Google Earth or fly over new terrain you can play the 'guess how that hole formed' game. I bet you didn’t know there were so many ways to make round holes in the ground!

Friday, October 23, 2015

Do you live near a maar volcano?


I am always going on about maar volcanoes. So where are the maar volcanoes? These volcanoes are formed by explosions underground occur because magma interacts with water and form unassuming craters. The craters are cut below the ground surface so their outer slopes are fairly shallow, and they are frequently filled with a lake. When there isn’t a large majestic volcano to climb and take fancy sunset photos in front of, it can be hard to get excited about what looks like a little hole in the ground.
A picturesque maar crater in Michoacán Mexico.
But remember, they are explosive, it takes a lot of energy to carve a big hole in the ground. I use dynamite to make craters that are only 2 meters in diameter (a tall friend lying on their side). We would need about 10,000 sticks of the dynamite we use to get close to the size of a maar volcano.
Our experimental maar volcanoes would need to be a LOT bigger to match nature.
I have been recently comparing the shapes and sizes of maar volcanoes around the planet. While I knew they were all over the place, I am now better informed on exactly where and how big they are. An average maar crater is as wide as 17 Olympic pools end to end, or 850 m. So if you get bored doing laps in a pool you could find a small maar lake and cross it to get a good work out. The largest maars that have been found on are up to 5000 m, or 100 of those Olympic swimming pools in a row. These exceptionally large maars all occur in a cluster on the Seward Peninsula in Alaska. Most maars are less than a kilometer across with the smallest maars closer to 100 m, or just one trip across a pool and back. You can find these smaller maars in the Eifel Volcanic Field in Germany or in Iceland at Askja volcano.
Google Earth image of the exceptionally large Epsenberg maars on the Seward Peninsula, Alaska, USA. The yellow pin is in the largest maar crater called the Devil Mountain Lake. Maar craters can be confused for other lakes. There are five maars in this image and the rest of the lakes are formed in the permafrost through more subtle processes.
I have swum around in this maar at Askja Volcano in Iceland, the lake is about 100 m across and a warm(ish) 60 degrees F. If you look at the upper rim of the crater you can see the layered deposits from the explosions that formed the crater. The lower more solid layers are material that was there before the maar formed.

These craters are found all over the planet in a wide variety of environments. The giant Epsenberg maars are in the remote subartic of Alaska. There are maars in the deserts of Turkey and Mexico. You can also find a lot of maars in the tropics, like in the Philippines and Indonesia. While some of them occur in remote locations, like Kamchatka Peninsula Russia or Ukinrek in Alaska that erupted in 1977, many of them are found in the middle of cities. In fact, the city of Auckland in New Zealand is built on top of 53 volcanoes, including ~11 maars. If you live in San Luis Potosi or Morelia in Mexico, Frankfurt in Germany, Rome in Italy or Kagoshima, Japan you are only a short drive away from a maar volcano.

View of Auckland from Google Earth showing some of the maars. You can also spot some scoria cones and shield volcanoes. They are called Lake Pupuke, Onepotu, Orakei Basin, Panmure Basin, Hopua, Mangere Lagoon and Pukaki from upper left to lower right.   

Eruption of Ukinrek Alaska 1971, USGS.
Some maar volcanoes occur in isolation, where they are the only volcano in the area, like Ubehebe crater in Death Valley California. While others occur in volcanic fields mixed with other types of volcanoes, like scoria cones and stratovolcanoes. Some maar volcanoes are covered by more recent volcanic activity. Sometimes scoria cones and lava flows occur at the end of the same eruption that formed a maar. Other maars occur inside, or on the flanks of a bigger volcano like Aniakchak caldera in Alaska.
Google Earth image of Aniakchak volcano, Alaska, USA. Two maars have formed inside the caldera, noted by the yellow pins They are cleverly named Southwest and Northeast maar.
Do we know where all the maars are on Earth? Nope. There are lots of other circular lakes that can, at first glance, look like a maar: permafrost lakes, sink holes, those crazy holes in Siberia and impact craters. Also, as maars get older they are subject to weathering and erosion. Since they are holes in the ground, instead of big fancy cones, they fill up with sediment and get shallower with time. Lots of maars have lakes in them which only helps them fill with sediment. Other maars get eroded along the edges (Hunt's Hole) and stop looking like nice perfect craters. In wetter environments, the shallow slopes of volcanic ash make for fertile soils, which means that you can find a lot of circular lakes in the middle of farmland that are actually volcanoes. In some locations the ash and rocky deposits are mined for abrasives or road materials. This makes it harder to count maars and know how big they can be. My goal is to look at the maars that we have found and see what we can learn about the type of eruptions that form them by comparing shape, size and environment. The more maars I study, the better. Thankfully satellite imagery means that I can study these volcanoes without having to pay for plane tickets to get to all of these locations. But maybe I'll make it a personal goal to see as many as I can. I think I've visited at least ten already, but that there are many more maars to go.
Kilburn (top) and Hunt's Hole (bottom) maars in New Mexico.
If you live in one of the cities I mentioned you can take a hike to your nearest maar volcano and appreciate it for more than just a hole in the ground for me. Keep an eye out for a circular or elliptical crater that may have a lake in it, then check for local tourist or hiking guides to find out if there are any maar volcanoes near you.  

There are 22 maars in this image, plus a stratovolcano and several scoria cones. How many can you spot? Lamongan Volcanic Field on the Island of in Indonesia.

Wednesday, October 14, 2015

Flowing rock frozen in time at Inyo Domes, California

- Janine

What happens when you get really viscous rhyolite (high silica content which makes it very sticky) magma rising to the surface? Well, it either stops, produces a really big bang, or oozes. When it stops below the surface it forms granite, which we see a lot of nearby in Yosemite. A build up of gasses that produces very high pressures can result in an explosive eruption, like certain eruptions that have occurred in the past at Yellowstone and Long Valley calderas. When the conditions aren't right for an explosive eruption, a more quiet 'oozing' of lava occurs at the surface that creates some really fantastic looking rocks! If you want to see a great example of rocks where you can see how they moved, head over to the Inyo domes volcanic chain near Mammoth mountain in California. The Inyo domes are near the edge of the Long Valley caldera, Yellowstone's less infamous cousin, west of the Mono domes chain.

The Inyo chain is a group of rhyolitic domes and flows that was erupted about 550 years ago (see Miller, 1985 for more information) with a series of phreatic (steam) explosions and thick lava forming domes and thick flows. Two different batches of magma mixed together to give an array of intermingling textures, resulting in an incredible place worthy of a geological or touristic trip - with beautiful views!

The only recent obsidian flows that have been observed actually moving can be seen here with work done by volcanologist Hugh Tuffen at Cordon Caulle volcano, Chile, in January 2012.

Here are a few of the textures I saw at the Inyo domes, you can see different bands of black, obsidian and lighter grey pumice containing feldspar crystals (phenocrysts). When you look at the textures and patterns you can imagine the lava flowing up the volcanic conduit and on to the surface where it sits today.

These next few images are from Obsidian Dome/Flow:
Vesicles (bubbles) within the grey pumice, with the black glassy obsidian.

Banded obsidian.
These next images are from Deadman Creek Dome:

Some neat banding patterns.
Vesicular and crystal-rich pumice band.
You can see the mixing of two batches of magma here with the lighter and darker greys.
Spot the evidence left by previous geologists studying these rocks!
Banding around a lighter grey zone (brown spots are biological, not rock).
Thicker obsidian bands with the light grey pumice.
A lens of light grey pumice.
Concentric patterns in the rhyolite.
If you take a closer look around when you're near a volcano you can sometimes find incredible features formed by rock 'flowing' as it moved up the volcanic conduit and onto the surface. The more viscous rhyolite is often thought of as very explosive and dangerous, but that is not always the case as you can see here. The Long Valley Caldera is an incredible region for a huge range of volcanic features, from basaltic lavas (the more famous Devil's Postpile below) to the large ignimbrite formed during the Bishop Tuff eruption, where the Rhyolite did make a big bang.

The columnar basalt of the Devils Postpile.
We can see how complex a volcanic area can be, and how important it is to study the rocks in order to know what an area is capable of producing, or to just test how good your volcanic landform spotting really is!

Miller, C.D., 1985. Holocene eruptions at the Inyo Volcanic chain, California: Implications for possible eruptions in the Long Valley caldera, Geology, 13, 14-17, 1985.

Friday, October 9, 2015

The eruption is how big? Deposit volume story


It seems obvious to say volcanoes are big, but as with anything in geology size is not immediately obvious. Volcanoes can loom over a landscape, spread ash over large chunks of the planet and even influence our climate-that all sounds big. On the scale of an individual human any eruption is big. They are frequently faster than, slower than (yes both), hotter than and physically larger than a human being. One of the exercises I use when teaching volcanology is focused on understanding just how big, "big" is. We go about this by finding things that we already understand the size and compare them to volcanoes. Like many natural processes (and humans), volcanoes and volcanic eruptions come in all shapes and sizes. If volcanoes only had one size and style of eruption our job studying them and anticipating future eruptions would be much easier.

The only human-sized volcano I have ever seen. It was a volcano costume in Kagoshima Japan in 2013.
We need to be able to put a size to volcanoes because it is one way to wrap our heads around the hazards they pose. One of the first steps to coming up with a hazard plan at any volcano is to study what it did in the past, reconstructing its eruption history. This helps geologist answer questions like how many eruptions has this volcano produced? What type of eruptions? How big were the eruptions? Were they all the same or did the eruptions change with time? We can now measure the changes of a volcano in real time, but the life span of a volcano is far longer than we have been able to measure, and we need context for these modern changes.  Frequently, we want a worst-case scenario, so we want to figure out what was the largest eruption this volcano was, or what a volcano like it, has produced.

Prehistoric deposits from an eruption from Cotopaxi, Ecuador 2013.

So what parameters can we use to define "big" for an eruption? We could measure the height of the eruption cloud, how far it threw rocks and ash into the atmosphere.  We could measure how much material was thrown out during the eruption. We could measure how much energy was released by the eruption. But not all of these measurements are possible for eruptions that happened in the past.  

For past eruptions we might be lucky and have historical observations of the eruption plume, but estimating the height of an eruption plume is a challenging task by eye and is generally imprecise from past records. Energy is even trickier and cannot really be measured from visual or auditory observations without instruments, so we are at a loss for historical or older eruptions. It is much easier to start with eruption volume. For explosive eruptions this means figuring out how far away ash traveled form the volcano and then measuring the thickness of that deposit from that furthest point back to the volcano. The more measurements we make the better the final volume estimate, but this is a tedious and frequently challenging job. To get deposit thickness you need to find both the bottom and top of the deposit. This means digging lots of pits, hunting for road cuts, and making friends with land owners to see if you can dig holes in their property or if they are willing to share any pits they may have recently dug themselves. For modern eruptions citizen scientists and observatory volcanologists set up lots of stations to just collect the ash as it falls.
Geologists digging to find the bottom of an ash layer from a pre-historic eruption. Colorado, USA 2015.

People closer to eruptions generally don’t want to collect ash, they want to remove it. In the northern hemisphere people are more likely to be familiar with how heavy winter snow is, and what a pain it is to shovel. Volcanic ash is pieces of pulverized rock, which is much heavier than snow, and also doesn’t melt in the spring. This makes volcanic ash is a serious hazard for roof tops, crops, and anything that breathes. These are just a few reasons why understanding the potential distribution of ash from a volcanic eruption is important. Even volcanoes that do not produce enough ash to threaten rooftops present constant challenges to those who live near it. I remember visiting Sakurajima in Japan and being told that they could never hang laundry out to dry because it would come back in gritty, ash it gets into everything, and clogs air filters really quickly.
Sakurajima in Kagoshima Bay, Japan produces multiple small ash plumes every day. These images are from a visit in 2013.

Once we have the volume of a deposit calculated we have a number, but what does it mean.  Usually we talk about eruption volumes in cubic kilometers (km3). Mt St Helens 1980 eruption produced 1 km3 of ash during its climatic phase on May 18. So what does that mean? Is that big?
May 18, 1980 Mt St Helens, USGS image.
Mt Pinatubo in the Philippines erupted in 1991 and was larger than Mt St Helens, producing 10 km3 of material. Is that as big as eruptions get?  
Pinatubo erupted in 1991 producing this impressive ash plume, USGS image.
Yellowstone caldera has in the past produced some of the largest eruption deposits ever measured, which is why they can be called super eruptions (See these links: adding super to volcanoes and eruptions and why not to lose sleepover them). These eruption deposits were up to 1,000 km3. THAT is big. But it is still hard really wrap your head around.
Most people do not think in cubic kilometers, and most Americans do not think in any variety of kilometers. The metric system was designed to be able to talk about things at all sorts of different sizes and compare them. It is mostly a matter of keeping track of the number of zeros to tell big from small. So metric is ideal for science as it helps us communicate size more efficiently. I think of the imperial system (inches, miles, pounds) as a more ad hoc system where we only want to compare similar sized things. Miles were for big distances, and feet was fine for around home. Eventually someone wanted to know how many feet were in a mile so they counted it out,  but it was clearly an afterthought.  

Anyway, the best way to get used to thinking about what volume means is to think about objects that we are more familiar with, figure out what their volume is, and then and compare them to eruption deposits. XKCD had an excellent comic that found familiar shapes to compare to metric, including humans. In that spirit, I find it easy to think about 2 meters as a fairly tall person 6 ft 6 inches. You can start to measure things terms of your friend Ted who is over 6 ft tall (Ted is hypothetical, I just made him up, but you can find your own Ted. Most field geologists have a Ted).  

Now that you are ready to measure some things in Teds let us calculate the volume of an office building, to compare volumes instead of just lengths. We’ll start with the building I work in. It has 12 stories and has two rectangular towers. The floors are pretty tall so we’ll say they are 10 ft each, which is about 3 meters or 1.5 Teds. That means the building is roughly 36 meters tall, or 18 Teds. The sides of the rectangle are maybe 15 meters each (that is 7.5 Teds lying down). Each tower of my building then has a volume of 8100 m3, so the whole building is 16200 m3. That is pretty big, right? Well that is only 0.0000162 km3. Well, what about a bigger building? The Empire State Building is still only 0.001047 km3 .

Looking up at the Empire State Building, Wikimedia commons.

Ok, so tall buildings still aren’t big enough to match Mt St Helens, let alone Yellowstone. The largest building on earth is an airplane factory that reaches 0.013 km3. Sticking with manmade constructions, the Hoover Dam blocks the Colorado River to form Lake Mead. The estimated capacity, if we could fill it all the way (the lake is currently at one of its lowest levels), is about 32 km3, but in 2010 it was only at half that volume 16 km3.  So, we jumped up to Pinatubo volumes with a lake. I’ll be honest, I had trouble finding something familiar at 1 km3, so we are going to skip it for now. We have more than enough orders of magnitude to jump. Imagine, Lake Mead filled with ash and pumice and you get an idea of how much material was thrown up (and then had to fall back down) in 1991 by Pinatubo.

Lake Mead, Wikimedia commons. Imagine filling this basin with pulverized rock. Now you see why no one wants to shovel all that.

Let’s go bigger. Lake Erie, the smallest of the great lakes, is large enough that you cannot see the opposite shore and is approximately 500 km3. Now we are talking! So, Lake Erie has a volume on the same order of magnitude as the Tambora 1815 eruption. This eruption produced the "year without a summer". The Eruptions Blog did a great job of discussing the size of this eruption.

So what about those large Yellowstone eruptions? The Great Slave Lake and Lake Ontario are both just over 1,000 km3. Those are some pretty serious lakes. Was that the largest eruption on Earth? Nope! There is an eruption deposit, the Fish Canyon Tuff, that is estimated to be 5,000 km3 which would fill the Grand Canyon (4,000 km3), Lake Michigan (5,000 km3) or even Phobos (one of the moons of Mars: 5700 km3). This deposit covers a good portion of the western United States, and likely came from the Jemez Mountains in Southern Colorado.

Close up view of the Fish Canyon Tuff with a nice biotite crystal from 2014.
Everyone wants to talk about big eruptions, but what about small eruptions? They produce 0.01-10 m3. Stromboli produces small eruptions like this several times a day. These can be watched safely from a distance. The University at Buffalo explosion experiments produce 0.15-0.99 m3 of ejected material, which makes then comparable to natural eruptions! I was pretty excited about that. A big refrigerator has the volume of about 1 m3 for comparison (less than a Ted tall, about half a Ted wide, and maybe a third of a Ted deep). So more flying rock than a human wants to be near, but much smaller than one of the great lakes.

These volumes are of ash and pumice, so they are much bigger than the volume of magma involved. The bubbles that make pumice light take up a LOT of space.  So what about lava flows? How do we compare them? We can still use volume, but if we want to compare lava to ash we'll have to estimate how much of that air space is in the deposit to calculate the Dense Rock Equivalent. It is also important to remember that magma densities are variable as well, depending on composition. The magma that produces pumice, called rhyolite, is 25% less dense than basalt melt that typically produces runny lava flows. I'll save that math for another post.

In 2014 a fissure in Iceland opened up and produced almost 1.5 km3of runny lava.  So that is comparable to the material erupted from Mt St Helens. To get a good idea of the size of this lava flow you can look at satellite images. But is it a big eruption for lava flows? Not even close. In Iceland alone, there were eruptions that produced 14.7-18 km3 of lava. Flood basalts tend to have lava flows on the order of 500 km3, and are composed of hundreds of flows. Because of the differences between an explosive eruption and a lava flow producing eruption, you can begin to see the challenge of describing the size of an eruption.

The Holuhraun fissure eruption in Iceland just before it ended in Feb 2015.

The volume of a deposit also represent how much material came out in total during an eruption, not all at once. We would need to calculate the discharge rate of the eruption, or the mass per time, which is a different challenge.  But these volumes hopefully give you a better idea of how diverse volcanic eruptions are, and part of the reason the jobs of volcanologists are both challenging and interesting.

BUT I do not want these very large eruptions to make you worry that an eruption on that size is a common occurrence or something to lose sleep over! The bigger an eruption is the less frequently they occur. Smaller eruptions occur all the time. In fact, on any give day there are twenty volcanoes actively erupting, but those eruptions involve small lava flows, little explosions (like the ones mentioned above at Stromboli), steam explosions and other burps and bumps that do not even leave a deposit. The Volcano Explosivity Index (VEI) is partially defined on eruption volume and ranges from 1 to 8.  Recently this scale was expanded to include more small eruptions because they are what we observe most of the time from volcanoes (academic paper). Many processes on earth seem impossibly large, that is what makes them fascinating, and in many cases beautiful. Just because we measure something doesn't mean it is less awe inspiring, in fact, for me quantifying geologic processes helps me understand just how impressive our planet is.