Guest post by Steven Goddard
Scientific American has reported that global warming may cause an increase in volcanic eruptions, due to increased magma formation at lower pressures as glaciers melt.
This caught my attention because I used to work as a volcano researcher and igneous petrologist.
That report said that about 10 percent of Iceland’s biggest ice cap, Vatnajokull, has melted since 1890 and the land nearby was rising about 25 millimetres (0.98 inch) a year, bringing shifts in geological stresses.
They estimated that the thaw had led to the formation of 1.4 cubic km (0.3 cubic mile) of magma deep below ground over the past century.
At high pressures such as under an ice cap, they reckon that rocks cannot expand to turn into liquid magma even if they are hot enough. “As the ice melts the rock can melt because the pressure decreases,” she said. Sigmundsson said that monitoring of the Vatnajokull volcano since 2008 suggested that the 2008 estimate for magma generation was “probably a minimum estimate. It can be somewhat larger.”
Interesting theory, but does it work quantitatively? Magmas, as with most solids, do show a direct relationship between the melting point and pressure. As the pressure increases, so does the melting point. (Ice is a noticeable exception to this, and shows an inverse relationship. The reason that people can ice skate is because the pressure under the blade creates a thin later of melted ice which lubricates the surface.
Below is a phase diagram of a basaltic magma similar to that found in Iceland, showing the relationship between temperature and pressure. The melting temperature does decrease at lower pressures. From 100 km depth to 0 km the melting point drops about 300°C. That is about 3°C / km. Ice is about one third as dense as basaltic magma, so the loss of 1 km of ice lowers the melting point by about 1C, or less than 0.1%.
More precisely, this study from the Carnegie Geophysical Institute did an empirical measurement of the relationship for one basaltic mineral – diposide. They found the relationship to be
Tm = 1391.5 + 0.01297 * P
Where Tm is the melting point in degrees C and P is the pressure in atmospheres. One atmosphere pressure is equal to about 10 metres of ice, so one additional metre of ice increases the melting point by about 0.0013°C. The loss of 100 metres of ice would therefore lower the melting point by about one tenth of a degree. The thickest ice in Iceland is only 500 meters thick, so complete loss of all ice would only alter the melting point by about 0.5°C, or less than 0.05%.
The geothermal gradient of the earth is typically about 40°C per km, so a 0.5°C change in temperature is equivalent to a depth change of about 20 metres. Near mid-ocean ridges this gradient is steeper, so the equivalent depth change in Iceland would be less than 20 metres. Is it credible that a 0.5°C decrease in the melting point could stimulate excess volcanic activity? Short answer – no. Volcanic activity is caused by magma rising to the surface, not glaciers melting. However, the loss of the glaciers would reduce the amount of steam and ash generated. Ash is formed when magma is cooled and fractured by steam. So the loss of the glaciers would reduce the size of the steam/ash cloud and make the Iceland volcanoes behave more like Hawaii volcanoes.
In short, the loss of all ice in Iceland would make the volcanoes less destructive.
BTW – On Al Gore’s planet, the geothermal gradient is much higher, with core temperatures averaging millions of degrees.
http://www.youtube.com/watch?v=ag2AWst3Qv4
Sponsored IT training links:
Latest PK0-003 dumps provide the best chance to improve your score in MB2-632 exam. Get CISM certified in days using free resources.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
The latest Meteosat ‘0’ Degree Ash Iceland Real Time Imagery –
http://oiswww.eumetsat.org/IPPS/html/MSG/RGB/ASH/ICELAND/
Has anyone calculated yet how much global COOLING will be generated by the ash clouds????
It would appear that the some of the Icelandic eruptions have a pattern of extended and variable activity once activated. Previous Eyjafjallajokull volcano eruptions happened in 920, 1612 and 1821-1823. The last eruption lasted from December 1821 to spring 1823 and seemed to have several stages of eruption. An initial eruption, followed by some 6 months of more subdued activity, followed again by several months of very active eruption, and followed again by some 5-6 months of more subdued activity. Since both of the last two eruptions took place during already very cold global temperatures, namely the Dalton Minimum and the Little Ice Age, it is not clear how the ash may have affected the climate of that day.
We know that the Pinatubo eruption of June 1991, affected global climate with cooler weather through 1993. [18 months?]
Looks to me that the predicted record warm year for 2010 and a few years there after may again be at risk if this eruption turns out to be similar to the previous ones.
@Steve Goddard (23:28:49)
Have you ever tried walking on ice in rubber soled boots? I have. It’s still very slippery and you can’t claim that the sole of the boot has high pressure point contacts as you can with the skate blade because the rubber deforms easily. Yet on any other slippery-smooth surface the boot will grip like mad.
A very thorough exposition and disposition of the pressure melting dogma through the good old fashioned experimental method can be found in a 2005 paper by Robert Rosenberg published in Physics Today.
Espen (03:21:24) :
“The most important measure would be to stock more cereals again.”
We should also come up with a plan and resources to help our farming sector to rapidly adjust to a sudden change in climate. Most farms will need to change their planting & harvest schedule, many will have alter their crop and seed selections and a number may need to be moved to operate in more hospitable environments until temperatures recover.
Jordan (02:05:09) :
I understand all the potential problems. The question though is whether the actual conditions warrant the actions being taken.
MrCPhysics (07:48:45) :
Hockey skates are sharpened and concave, and the part contacting the surface of the ice is much thinner than piano wire. I’ve cut my fingers on ice skates many times.
It is much easer to have an intelligent discussion when people avoid the use of the terms “myth” “consensus” or “established science.”
Ian L. McQueen (07:40:18) :
The explosivity of volcanoes is primarily determined by steam pressure in the magma chamber.
This Los Alamos paper describes an experiment which demonstrates the concept.
http://www.ees1.lanl.gov/Wohletz/Volcanic%2520dustlike%2520particles.pdf
Bad link – this one hopefully is correct.
http://www.ees1.lanl.gov/Wohletz/Volcanic%20dustlike%20particles.pdf
A similar slant by a certain group was tried, with cyclones creating lower pressure and causing an increase in earthquakes
re: Gerry (08:06:01) :
Has anyone calculated yet how much global COOLING will be generated by the ash clouds????
————
When Pinatubo erupted in 1991, almost everyone had been predicting a warm 1992. Only one or two took note. It took until June for the rest to figure out why there was cooling for a year or so (I do not have the figures handy)
Well, i guess nobody noticed the link I posted earlier, so I’ll post it again:
http://homepages.see.leeds.ac.uk/~earcpa/2008GL033510.pdf
This is the original paper on the effect of ice loss on volcanic activity that is referenced in the Sci Am article.
In short, hot rock melts when the pressure on it drops. The pressure drop can happen in two ways: 1) as the hot rock rises through the crust (the pressure decreases as depth decreases); or 2) by removing mass at the ground surface, which lowers the pressure at depth.
In iceland, the hot rock is rising through the crust at roughly 5 cm/yr (plus or minus a few cm). The amount of mass loss from the glacier discussed in the paper is the equivalent of removing 5 cm of rock from the surface per year.
So (in the area beneath the glacier) the amount of rock melting caused by the loss of the ice mass is approximately the same as the melt generated just by the hot rock rising through the crust.
The net effect – the ice loss has (very) roughly doubled the rate of magma production in the area beneath the glacier.
Although the calculations Steve G. has done are correct, his analysis and understanding is all wrong.
When I first read the report on the link between melting glaciers and eruptions I was a bit skeptical. However isostatic rebound is known and measurable. I was a bit dismayed that everyone was so quick to dismiss it. Steve’s post and some of the articles dealing with the subject are models attempting to describe a complex situation. As such they are similar to climate models– the math and physics may be correct, but the underlying assumptions may or may not be correct.
IMO the place to start is with the rock record. A paper by Slater and others in 1998 in Earth and Planetary Science Letters reports that Iceland eruption rates at the end of the Pleistocene (when deglaciation was much greater than today) were 20-30 times greater than today. I don’t have access to the full article so I can’t comment on the stregth of their data/methods. I believe that a look at eruption rates through time in places like Iceland, Alaska etc. would be a better way to go rather than relying on theoretical models.
Doug in Seattle (08:47:55) : “I understand all the potential problems. The question though is whether the actual conditions warrant the actions being taken.”
Its hard to guess other peoples’ background across the blog. Sorry about that.
If it helps, the video in the following link confirms how measurements are being taken, so the decisions should be evidence based. That should help any concerns about over-cautious reaction.
http://news.bbc.co.uk/1/hi/uk/8626625.stm
People with expertise are offering views (like the guy in the video)m although I would expect the aitline operators will be consulting engine OEMs.
I’m conscious that these points are OT for the thread.
“Surely it is the outgassing pressure in the magma that is critical here (whatever name this critical pressure has).
The magma behaves like a soda bottle, and with sufficient pressure on the soda it is stable. Release that pressure, and suddenly it effervesces and “boils” up all over the place.”
Ah Ralph, but the soda boils up when shaken, gas pressure builds up and the cap is removed quicly. If the cap is removed slowly, like a slowly melting glacier, then nothing much happens because pressure is releasely gradually.
stevengoddard (08:55:47) :
Ian L. McQueen (07:40:18) :
The explosivity of volcanoes is primarily determined by steam pressure in the magma chamber.
Steven-
Does this mean that magma is the same composition around the world?
IanM
Ulric Lyons (04:46:35) :
The eruptions are primarily due to temerature differentials, as with all eruptions. A very cool month or so followed by a strong uplift in temperature is what triggers volcanic activity.
I find it more likely that there is a common cause behind the two. Change of Earth’s rotation for instance.
Far to many ’60s hippies spent far too much time staring at their lava lamps and drawing incorrrect, imbibed conclusions.
Dave Springer (07:42:52) :
“I’ve subscribed to Scientific American for several decades and read every issue cover to cover. Overall it’s a great source for keeping up with a broad range of the sciences. ”
I’m a fellow SA fan – it’s a nicely done attractive magazine and generally fun and informative. I had no real awareness of AGW until Climategate and then I started paying attention and from what I have found AGW is a weak conjecture, physical measurements contradict it, historical climate records contradict it, there are no ill effects much less catastrophes anyone can point to as evidence for it that cannot be easily debunked by objective arguments and simple observation.
Yet SA devotes almost the entire issue of a recent edition to what we have to do to save the planet from AGW and rounds it off with a paddy-cake, paddy-cake interview with Bill McKibben author or “Earth”. Preposterous.
Thanks Steven for debunking the latest nonsense from SA; because even if it is a “nicely done attractive magazine and generally fun and informative” they occasionally print rubbish.
The Katla volcano which is only about 25 km from the Eyjafjallajokull volcano has erupted just before the Eyjafjallajokull on three of the previous occasions. So there appears to be some correlation. The last eruption of Katla was in 1918, thus it has a current repose time of 92 years, the longest since about 1500 and Katla seems long over due since it typically erupts twice in a century. It has erupted in a major way about 9 times since 1580. Typical eruption periods were 13 days to about 120 days with last three being about 3-4 weeks.
So it seems to me with all the world focus on manmade greenhouse gases and carbon dioxide only as piloted by IPCC , we have wrongly taken our eye off other natural planetary cycles and potential risks which are far more consequential and immediate. Lets hope that there will now be more sanity about this global warming issue and we start dealing with real world problems . Manmade greenhouse gases is the least of our worries at this time .
Steven Goddard
Good post: puts the “un-” to Scientific American
Minor nit: did you mean diopside? “…relationship for one basaltic mineral – diposide.“
Thanks, Great Read
[para 6, last line, the word later s/b layer]
========
Melting ice caps may trigger more volcanic eruptions
10:38 03 April 2008 by Catherine Brahic source
http://www.newscientist.com/article/dn13583-melting-ice-caps-may-trigger-more-volcanic-eruptions.html
Vatnajökull is the largest ice cap in Iceland, and is disappearing at a rate of 5 cubic kilometres per year.
Carolina Pagli of the University of Leeds, UK, and Freysteinn Sigmundsson of the University of Iceland have calculated the effects of the melting on the crust and magma underneath.
They say that, as the ice disappears, it relieves the pressure exerted on the rocks deep under the ice sheet, increasing the rate at which it melts into magma. An average of 1.4 cubic kilometres has been produced every century since 1890, a 10% increase on the background rate.
…
But the thinning ice has another effect on volcanoes which will be more widespread.
As the amount of weight on the crust changes, geological stresses inside the crust will also change, increasing the likelihood of eruptions. “Under the ice’s weight, the crust bends and as you melt the ice the crust will bounce up again,” explains Bill McGuire of University College London in the UK, who was not involved in the study.
=========
“In short, the loss of all ice in Iceland would make the volcanoes less destructive.”
Did you mean destructive or less active?
Note: wouldn’t a gradual change in the ice cap introduce increased seismic reads if the weight is having a meaningful impact on geologic stress?
So let me get this straight – are you suggesting the report was alarmist?
All together now: WHODA THUNK IT!
Ian L. McQueen (10:39:47) :
There are several different types of magma, but Hawaiian and Icelandic volcanoes are both basalts. Hawaiian volcanoes do not normally generate a lot of ash.
The most explosive volcanoes (like Mt. St. Helens) are andesite volcanoes with magma that is more viscous than basalt.
Too much is being made of the decrease in lithostatic pressure that can occur due ti ice melting ant’s effect on temperature/pressure relationships of magma. I think Steven pretty well debunks this in his article.
The coincidence of a series of eruptions with ice loss in Iceland in the early Holocene is just that in the absence of a coherent mechanism to tie the two together. This is so far lacking.
Volcanism in Iceland is driven by oceanic plate spreading, not isostatic response to ice unloading (or fracturing in response to it). The magnitude of the extensional stresses associated with plate spreading, even on a local scale, such would be in pay in Iceland, simply dwarf those associated with the small amount of unloading that could be associated with ice loss in the past century.
I can see the possibility of local structural effects modifying the direction of magma flow, or explosive reactions of magma to water/ice rich host material during an eruptive event, but beyond that ice accumulation or loss would not affect volcanism.