Slashdot Mirror
Is the Yellowstone Supervolcano About To Blow?
An anonymous reader writes "Apparently, Yellowstone National Park has been having a very unusual number of earthquakes. Many of the most recent tremors have been deeper underground, an ominous sign. Combine that with a rapid rise in elevation over the past three years, and the possibility that earthquake activity from surrounding areas could trigger such an eruption on its own, and you've got the possible warning signs of a supervolcano eruption that would wipe out half to 2/3 of the continental US, plunge global temperatures, and wipe out a very significant chunk of world food sources. Here's a little more info to make your New Year brighter!"
Map showing recent earthquakes is over here http://earthquake.usgs.gov/eqcenter/recenteqsus/Maps/US2/43.45.-112.-110.php
Seriously Slashdot, you need to work on your reaction time. This was news two days ago.
These earthquake swarms happen frequently in Yellowstone, and this one has already ended. Yellowstone has dropped back to its ordinary low rumble.
But then again, I could be wrong.
One
I wouldn't actually be "hundreds of thousands" as you say. Go back to school and learn your VEI scale.
(emphasis mine).
Second cite:
Not just a dusting of ash, by any means. To extrapolate from a single event (Mt St Helens) which may or may not even be in the same geologic region (I don't know) is pointless when the Snake River Plain has erupted several times over - the entire landscape their bears the scars of it.
Um no, dude, you don't really get it. If Yellowstone blows, there is no volcano eruption in human history that even remotely comes close. Mt. St. Helens would look like a fart standing next to Chernobyl. Areas 400 miles away would get covered in a foot of ash. There is just nothing like it.
Here is a nice, graphical link for you to look at:
link
The number of deaths could be staggering. That foot of ash, even 400 miles away in Denver, would collapse most roofs, and any with people in them would get severely injured or die. It would be the end of the U.S. as a global superpower, and there would be wars. You are naive.
Japanese scientist: Technically, sir, tomatoes are fags. Military scientist: He means fruits.
Fortunately, the Yellowstone volcanic system shows no signs that it is headed toward such an eruption in the near future. In fact, the probability of any such event occurring at Yellowstone within the next few thousand years is exceedingly low.
...
Lava flows and small volcanic eruptions occur only rarely--none in the past 70,000 years. Massive caldera-forming eruptions, though the most potentially devastating of Yellowstone's hazards, are extremely rare--only three have occurred in the past several million years. U.S. Geological Survey, University of Utah, and National Park Service scientists with the Yellowstone Volcano Observatory (YVO) see no evidence that another such cataclysmic eruption will occur at Yellowstone in the foreseeable future.
(emphasis mine)
As for that "several million years" figure for a devastating explosion of the kind TFA is describing, consider that the United States as a nation is still less than 250 years old. I'm not saying it can't happen, but the idea that "it hasn't happened in a long time so it must be ready to happen now" is just a popular Las Vegas delusion.
Breakfast served all day!
Yellowstone's largest eruption was 2,500 times more powerful than St. Helens.
It's eruptions cover hundreds of square kilometers, not tens of thousands.
Most of the United States by area would see a few meters of ash, not a football field's worth (which would be plenty devestating enough).
Yay for mods blindly modding up posts that contain numbers as "informative."
nah, it won't quite be that bad. most predictions expect the immediate danger zone to have a radius of 1000-1600km, with pumice & ash deposit probably covering all of California and most of the Midwest. but rather than being burned, most deaths/injuries will likely be caused by ash inhalation.
luckily, modern humans have the benefit of science and technology.given enough warning, most people within range of the volcanic explosion and subsequent lava/pyroclastic flow (70,000 to 100,000+ individuals by some estimates) can be evacuated beforehand. everyone else will simply have to stay in doors for a couple of days before they too can be evacuated outside of the ash cover area.
the USGS seems pretty confident that the YVO monitoring program will detect any premonitory indicators (such as emissions of magmatic gases) of any such impending disaster. and studies indicate that, if there is a volcanic eruption, it is not likely to be a caldera-forming supervolcanic eruption due to insufficient rhyolitic magma-storage to sustain such an event.
in the event that a caldera-forming eruption takes place, then yes the ash will probably circle the entire globe and lower the temperature in the lower atmosphere for a few years, and that can have a severe impact on the ecology of the planet. but it's certainly survivable. and the chances of such an event actually occurring is still statistically insignificant--contrary to what is often reported, are are not "overdue" for a supervolcanic eruption. (the mean interval between such eruptions is 710,000 years, not 600,000 years.)
if others are interested, you can read the USGS's report on the Preliminary Assessment of Volcanic and Hydrothermal Hazards in Yellowstone National Park and Vicinity (the actual report is in PDF format).
Volcanic "ash" is not burning wood "ash". Volcanic ash is actually pulverized, powdered rock that only superficially resembles wood ash as it falls and collects on the ground. It's not the result of any burning process.
The Yellowstone one 2.1 million years ago was about same
as the Toba.
The Fish Canyon Tuft one in Colorado was bigger than both COMBINED.
That is scary as hell.
Glad it was a one off like 50 million years ago.
google "32 trillion offshore needs IRS attention"
Quality of information is the fundamental problem, coupled with the variability of real rocks.
[/self : puts on my formal "geologist" hat ; it is my job, though not this particular aspect of geology]
The area around Jellystone (and all volcanos) has a long and complex history ; particular rock units vary on a scale of centimetres to metres and larger, through bedding and faulting, to say nothing of the more subtle variations resulting from hydrothermal alteration. These variations in constitution and physical organisation of the materials lead to considerable (several orders of magnitude) variations in rock strength on quite small scales - metres, if not finer.
So, to accurately characterise the rock volume where you're intending to set off a small, controlled eruption, you need that scale of knowledge of the rock units in order to work out where you can safely set off that "small, controlled eruption", and indeed, how to set off that "small, controlled eruption".
Which is well and good - it gives us a goal of the approximate level of information that we need to plan and execute the "small, controlled eruption" plan. We'd need to characterise most of the immediate vicinity of the volcano that we're planning to "defuse" - for the Jellystone hotspot hmmm, on the order of 100km of land area to a depth of several km, say 300km^3 of rock with data at (say) 10cm spacing, and with density, triaxial strength and stiffness data, temperature, pressure, stress field (triaxial again), and a few other bytes of data. Lets say 20 bytes of data per station and around 300 x 10^9(km^3->m^3) x 1000 (data points per m^3) = 3 * 10^14 stations. So we're looking at on the order of 10^16 bytes of data for the core area, and I'd guess the same for surrounding areas at progressively decreasing data density to control for "edge effects" (I'm getting a bit hand-wavy here ; it shows!). Say 10^17 bytes of raw data and working / intermediate results. That's around 100 petabytes, or approximately 10 years worth of LHC data.
That's a serious chunk of computing power, but not incredible. It also allows us to put some sort of cost on the project - the LHC is costing on the order of 5 billion USD, so we're looking into the same sort of region of cost for working out what to do and how to do it. GIVEN that we've got the data to analyse. And that's where the problem lies.
To get the data that's necessary to do this modelling, we're going to need to measure those 20-odd bytes of data for those points, at something approaching that data density. Which we don't have techniques for. We can get some data points - for example I can measure the porosity, permeability and fluid pressure at centimetric scale in a borehole. The tools used are the MDT (if I'm working with Schlumberger equipment) or RCI (from Baker Atlas), but there are others. For measuring rock strengths ... well, I could conceive of relevant tools, and I could conceive of using them at the same time as doing the pressure measurements. Getting the triaxial stress field is a deal more involved (since drilling the borehole induces a change in the stress field, by drilling out the rock), but I can envisage doing it. So let's say that we can get our data using currently conceivable direct measurements for essentially the cost of drilling the borehole.
A 3km hole in hard rock. That would be in the region of a million USD, if you're doing it wholesale. To characterise the whole rock volume, you're going to need to drill in the order of one every 10 metre to make even a faint approach at getting the areal data density (your surface borehole is going to be nearly a metre across, so you can't go to any better data density than 1/
Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"