Friday, May 16, 2008

Volcano monitoring: last night's PGS meeting

Last night I attended my first meeting of the Potomac Geophysical Society (PGS). The PGS meets on Thursday nights, and I usually can't make it because I teach on Thursday nights. (I do however attend meetings of the Geological Society of Washington quite regularly, but those are on Wednesday nights.) Now that the semester is over, I was able to make it to the final PGS meeting of the spring.

The meeting was held at Fort Meyer Officer's Club. It's on a military base adjacent to Arlington National Cemetery, and before entering, my Prius had to be searched for bombs (as did all other civilian vehicles). The Officer's Club was about what you would expect, I guess -- kind of 1950's decor, elegant once. I noticed they had compact fluorescent light bulbs in all the sockets, which pleased me. PGS meetings consist of: (1) beer downstairs in the lounge, (2) dinner upstairs in the "Campaign Room," (3) business details, and (4) a talk by a guest speaker.

Last night's speaker was Bill Burton, from the USGS's volcano hazards and monitoring program. Bill's office will be launching a comprehensive new volcano website later this year, and he gave us a brief preview of its features in last night's talk. If you'd like a look for yourself, they have a beta version of the site online now.

Bill reviewed the efforts of his office to monitor a whole lot of volcanoes in U.S. territory (including the protectorate of the Mariana Islands). The map above shows how they've got their work cut out for them.

I was struck by two things about Bill's talk. First, that the volcanoes that the USGS has fully equipped with monitoring instruments (high-precision GPS, seismographs, etc.) provide a wealth of information which makes it relatively easy (relative to say, earthquakes) to predict eruptions. We've learned enough that we can use subtle signals to issue eruption watches and warnings, for both people on the ground and planes in the air.
But the second thing I noted is that they don't have every volcano in their area of responsibility fully instrumented. In fact, they don't even have every "dangerous" volcano in their areas fully instrumented. A striking example of this was a story Bill told while showing us this image:

That's Cleveland Volcano, one of the Aleutians. The photograph was taken by an astronaut, who then called the duty officer in the Alaska Volcano Observatory and told them "Cleveland's erupting." The AVO doesn't have instrumentation on Cleveland, so this phone call (from space!) was the first they knew about the eruption! That's a pretty big gaping hole in the program, it seems to me. We should know at the very least when our volcanoes are already erupting. Even better would be to fully instrument all our potentially-dangerous volcanoes to the fullest possible extent, so we can predict in advance when they will erupt.

At the end of the talk, I asked Bill about this: "How much more money would your office need to outfit the under-instrumented volcanoes to the level you think they need?" Bill estimated $80 million, both for installation and the constant upkeep these instruments require. This morning on the drive into work, I heard that the new farm bill has $450 million set aside for Chesapeake Bay environmental work, so while $80 million seems like a lot to me, I guess it's not all that much in the grand scheme of things. In a post this morning, Nicholas at The Critical Zone examined this issue of huge numbers, and all the science that could get done with that money.
In the wake of the recent tragedies in Myanmar and China, it seems like the US would be wise to invest some money in outfitting our volcanoes with the full suite of monitoring equipment. An ounce of prevention is worth a pound of cure. Who knows how many human lives an ounce of seismograph is worth? When a destructive eruption does happen, we're going to wish we had spent that $80 million when we had the chance.

Labels: , , , ,

Monday, March 3, 2008

Mineralogy of the atmosphere

On Thursday, I posted some reflections on one of the talks at the most recent meeting of the Geological Society of Washington. At the same meeting, there was another talk that got my attention, and I wanted to briefly share its findings with the geoblogosphere. The talk was entitled "Mineralogy of the Atmosphere: Assessing environmental and health impacts of airborne particulate matter." It was given by Reto Giere, of the University of Freiburg, Germany. (He's currently in DC as a Visiting Investigator at the Geophysical Laboratory of the Carnegie Institution of Washington.)

Reto's research has lately focused on particulate matter in the air. He collects it and then evaluates it using transmission electron microscopy, X-ray diffraction, and other techniques. The first point he made in Wednesday's talk is that "soot" is a matter of definition. Natural and anthropogenic sources can both be found in the sub-1-micrometer range. If you look at small particulates, Switzerland's environmentally-lauded train system actually generates three times as much "soot" as their traffic output.

So what's in that "soot?" Turns out that a lot of it is anglesite, PbSO4 and some of it is gunningite, ZnSO4'H2O. (There are also droplets of elemental selenium, Se.) The majority of these metal sulfates (and others) are coming from flue gases from power plants. And the thing is, because they're so small, all these goodies end up in our lungs. Reto has run modeling experiments to see what weight-percent of the average person's dose of inhaled metals gets extracted by the lung fluid. In one week, 80% of the zinc was absorbed by the lungs, 55% of the nickel, and 35% of copper. Yum! (I would have been interested to see the actual masses of these absorbed metals compared to the quantities present in a typical vitamin pill, but that wasn't covered.)

There's good news that stems from Reto's work too: the particular "cocktail" of minerals in a sample may be diagnostic of a specific source, which would be useful for forensic identification of polluters. Overall, I found it an interesting talk, on something I'd never really thought about before.

Labels: , ,

Thursday, February 28, 2008

Volcanoes are coffeemakers?

Last night at the meeting of the Geological Society of Washington, we were treated to a couple of really entertaining talks. The first was by John Eichelberger, of the U.S. Geological Survey in Reston (formerly of UAF). John is interested in Plinian eruptions -- the ones where volcanoes shoot massive amounts of ash and gas upwards in an eruption column. He made the point that while Plinian eruptions are widely characterized as "explosive," they are actually a steady state phenomenon with a high volume, "like a firehose."

John suggested that, contrary to convential wisdom, Plinian eruptions do not require viscous magma. "Basalt erupts this way," he said, "Water erupts this way." To underscore his new way of thinking about the topic, John did the funniest thing I've seen yet at GSW: He showed a photograph of two dimes (10-cent coins) lying on a black background. Then he showed another, similar photograph, but in the second one, the dimes had been moved to the left by an inch or so. His explanation of this apparent act of performance art? "A paradigm shift!" (pair... of... dimes...) It got big laughs; We're geeks.

John then started a detailed discussion of the inner workings of a $9 coffee maker he had bought at Wal-Mart. He introduced the idea of a "magma table" akin to a water table, and showed how the relationship between density, pressure, gravity, and the height of the water table determined how coffeemaker water 'erupts' out of a conduit (black in the image above, stolen from the "How Coffeemakers Work" page at howstuffworks.com), in spite of the conduit's greater height. It has to do with lowering the density of the material in the conduit by heating it to be partially steam. He suggested that this is akin to how a geyser erupts out of a conduit, as sufficient heating lowers the density, which lowers the pressure on the water below, which flashes to steam, which lowers its density, which lowers the pressure on the water below that, and so on. This chain reaction propagates downward, and it keeps working until the geyser's subterranean reservoir is emptied. (Note that the same principle applies to coffeemakers: they use up all the water inside, and make it into coffee -- there's nothing left sloshing around in there when it's done.) John returned to volcanoes when he invoked the same process to explain Plinian eruptions.

It was a lively, thoughtful presentation that emphasized simple physical relations and familiar analogies to explain one of the most distinctive phenomena of our planet. Thumbs up!

Labels: ,

Thursday, January 24, 2008

Yellowstone, Newberry, & lithospheric thickeness

Last night at the year's first meeting of the Geological Society of Washington, Derek Schutt of the National Science Foundation gave a talk entitled "The Yellowstone hotspot and how it got that way." Derek mainly focused on the evidence for there indeed being a mantle plume under Yellowstone, possibly caused by the destabilization of the core-mantle boundary layer when subducted Farallon lithosphere sank down to the bottom of the mantle.

But the thing that he said that really caught my attention has to do with one of the weird aspects of Yellowstone. Yes, to the southwest of Yellowstone's modern caldera is the Snake River Plain, a series of ancient calderas which overlap one another, getting older and older the further to the southwest you travel, until you get to the oldest one at 17 Ma. That part of it looks pretty much like a classic hotspot track, a la Hawaii. But there's a weird aspect to Yellowstone that doesn't fit the traditional hotspot stereotype: starting at that same 17 Ma caldera/"rift," another series of eruptions propagated away to the west/northwest, including the voluminous Columbia River flood basalts and leading to the Newberry Caldera, which Derek described as "the largest basalt dome in the United States." (See the map above, from Schutt's collaborators Gene Humphreys and John Hernlund.)

So, the question is: What's up with that? It kind of looks like two hotspots heading in different directions. Is this linked to the stretching of the western U.S. via the Basin and Range? Derek pitched another idea, which is based on the thickness of the lithosphere (crust + uppermost mantle). His idea is summarized in the diagram below, which I drew this morning based on my rough sketch of the diagram he put up on the screen at GSW last night. (My apologies to Derek if I've gotten any of the details wrong.)



The basic idea is that the North American lithosphere is thicker to the east, under Yellowstone, which Derek (admittedly loosely) defined as the Wyoming Craton. He suggested that the lithosphere was thinner to the west under Newberry and the Columbia River Plateau, since those were accreted terranes added to North America during the Mesozoic. The mantle plume came up underneath the thicker lithosphere, and punched a hole through right above it (Yellowstone), but part of the plume slid upwards and westwards towards the thinner lithosphere, where it broke through in multiple locations, producing first the Columbia River flood basalts and then later the eruptions culminating in Newberry. I like the idea, and the picture Derek showed is elegant. I can picture this happening, if the suggested lithosphere thicknesses are true. The question is, are they? I don't know enough about that region of the country (yet) to assess the validity of this model. I wanted to use this blog post to share the notion, and see what people think. If you're familiar with that area, please clue me in to the details.

An additional difference between Yellowstone and Columbia River/Newberry (CR/N) is that Yellowstone's magma is rhyolitic and CR/N's magma is basaltic. Rhyolitic magma is a lot more explosive than basalt, and indeed Yellowstone's eruptions have been among the most powerful observed in the geologic record. (The Huckleberry Ridge tuff, which erupted from Yellowstone 2.1 Ma, is deposited over something like half of the Lower 48!) CR/N, on the other hand, appear to be gentler eruptions more like Hawaii's oozing of basalt. I suppose this too can be explained by Derek's model: partial melting of the more-felsic crust under Yellowstone (as hot plume magma heats that thicker slab of continental crust), but a shallower Moho to the west, producing mafic magma a shorter vertical distance from the surface.

PS -I must also add that it was great to meet Tuff Cookie of Magma Cum Laude at the meeting. If there had been one more of us there, it could almost have been a geology blogger's conference.

Labels: ,