Study of ancient eruptions shows modern ice sheets could be vulnerable
From THE EARTH INSTITUTE AT COLUMBIA UNIVERSITY and the “dirty ice” melts faster department
Volcanic eruptions have been known to cool the global climate, but they can also exacerbate the melting of ice sheets, according to a paper published today in Nature Communications.
Researchers who analyzed ice cores and meltwater deposits found that ancient eruptions caused immediate and significant melting of the ice sheet that covered much of northern Europe at the end of the last ice age, some 12,000 to 13,000 years ago.
“Over a time span of 1,000 years, we found that volcanic eruptions generally correspond with enhanced ice sheet melting within a year or so,” says lead author Francesco Muschitiello, who completed the research as a postdoctoral fellow at Columbia University’s Lamont-Doherty Earth Observatory.
These weren’t volcanoes erupting on or near the ice sheet, but located a thousand miles away in some cases. The eruptions heaved huge clouds of ash into the sky, and when the ash fell on the ice sheet, its darker color made the ice absorb more solar heat than usual.
“We know that if you have darker ice, you decrease the reflectance and it melts more quickly. It’s basic science,” says Muschitiello. “But no one so far has been able to demonstrate this direct link between volcanism and ice melting when it comes to ancient climates.”
The discovery comes from the cross-sections of deposits, called glacial varves, most of which had been collected in the 1980s and 1990s. Varves are the layered sediments that form when meltwater below an ice sheet routes large amounts of debris into lakes near the sheet’s edge. Like the rings of a tree, the layers of a glacial varve tell the story of each year’s conditions; a thicker layer indicates more melting, since there would have been a higher volume of water to carry the sediment.

The team also compared the varves to cores from the Greenland ice sheet, whose layers contain a record of ancient atmospheric conditions. Testing of those layers for sulfates revealed which years experienced explosive volcanic eruptions, which tend to release large amounts of ash. Matching up the ice layers with varve layers from the same time periods, the team found that years with explosive volcanic activity corresponded to thicker varve layers, indicating more melting of the northern European ice sheet.
Muschitiello and his colleagues studied a period ranging from 13,200 to 12,000 years ago, when the last ice age was transitioning into today’s warm climate. They focused specifically on volcanic eruptions in the northern high latitudes–events similar to the 2010 eruptions of Iceland’s Eyjafjallajökull volcano. Although that eruption was relatively minor, its large ash cloud shut down air traffic across most of Europe for about a week.
How much melting could an eruption like that cause? “It’s difficult to put an exact number to it,” says glaciologist and coauthor James Lea from the University of Liverpool. “It depends on many factors.” Running thousands of model simulations, the team found that the amount of melting depends on the individual eruption, which season it occurs in, the snowpack conditions at the time, and the elevation of the ice sheet. “Change any one of these and you would get different amounts of melt,” says Lea. In the worst scenarios, the model predicted that ash deposition would remove between 20 centimeters and almost one meter of ice from the surface of the highest parts of the ice sheet.
The model results should be taken with a pinch of salt, Muschitiello cautions, due to uncertainties about past conditions. However, because the team simulated a very broad range of potential conditions, he’s confident that the ice sheet’s real response lies somewhere within their range.
Michael Sigl, a paleoclimatologist from the Paul Scherrer Institute in Switzerland who wasn’t involved in the new study, says the hypothesis that ash particles might counteract the cooling effects of volcanic eruptions is intriguing. But, he said, “coincidences in the timing of rapid ice-sheet melting events and eruption dates do not automatically imply causation, and there may be other scenarios that could be consistent with the presented data.” Sigl’s own work has found a link between eruption-induced ozone depletion and deglaciation in the Southern Hemisphere. Nevertheless, he says, the new study shows that more work is needed to understand the effects of aerosol emissions from volcanic eruptions.
The preliminary results suggest that “present day ice sheets are potentially very vulnerable to volcanic eruptions,” says Muschitiello. They also point to a possible hole in the climate models that scientists use to make predictions about the future: Models currently don’t simulate the ice sheets’ response to changes in particulate deposition from the atmosphere in an interactive way.
Another intriguing implication is that previous research has suggested that melting ice sheets and glaciers could increase the frequency of volcanic eruptions in glaciated areas by lightening loads on earth’s crust, allowing underlying magma to rise. If the link between volcanism and ice sheet melting is confirmed, it could indicate the presence of a so-called “positive feedback loop” in which eruptions exacerbate melting, and more melting causes more eruptions, and so on.
Muschitiello says the study “can give us hints about the mechanisms at play when you’re expecting rapid climate change.”
###
The paper, “Enhanced ice sheet melting driven by volcanic eruptions during the last deglaciation,” is available from the authors or press@nature.com. After it is published, it will appear online at http://dx.doi.org/10.1038/s41467-017-01273-1.
There isn’t any “may melt rapidly”. There is no doubt that a soot covered ice or snow pack will absorb the heat from the Sun and melt whatever is below (snow/ ice). Every winter I put my ashes from my pellet stoves on the snow covering my south walkway and the Sun does the rest- melts it so I don’t have to shovel it.
Joe, your ashes go directly to your south walkway, without making a trip into the atmosphere. Your ashes do not have a chance to reduce sunlight the way a volcanic eruption would.
SR
True but the soot on the snow stays long after the soot in the atmosphere is gone. But yes if the Sun doesn’t reach the dirty snow then it will not effect it.
“Keep in mind this latest October 12 event is not associated with obvious earthquake swarms and proven volcanic eruptions as was the case during the 1999 – 2007 event. This earlier event was powerful but not obvious to those who did not understand its true nature. Even though it was associated with an extensive low intensity earthquake swarm, a huge methane release, and a significant series of volcanic eruptions along the Gakkel Ridge it was, and still is dismissed as insignificant by most climate scientists advocating the theory of man-made global warming”
http://www.plateclimatology.com/heat-from-deep-ocean-fault-punches-hole-in-arctic-ice-sheet/
Another research study based on predispositions regarding theories. i.e. “Confirmation Bias”
• And the ash fall on ice sheet study results are?
• Not studied or inconvenient results.
Indeed?
The researchers Ignore actual ash falls and prefer to run computer models until the results desired are produce.
• Zero ability to replicate.
• Verification or validation is near impossible.
• Model runs that depend on a host of assumptions; happily provided as input by the authors.
When dumping sand or ash, or even kitty litter on sidewalks to prevent slipping and hopefully melt some of the snow; one quickly discovers negative results.
• A) Dark grit requires sunlight to work. Cloudy weather, no melt.
• B) Thick layers of grist and especially ash insulate snow/ice from melting. Ash is a terrible conductor and layers of ash prevent convection.
• C) Ice piles covered with thick layers of grit and ash last long into very warm spring weather.
• D) Any additional snow/ice accumulation covers the grit/ash negating all benefits.
Only the outer ash layer changes the albedo. Without conductive properties, covered ice/snow will not melt. Since the far North and South have greater periods of diurnal cold; freezing and refreezing ice dominates melting.
One does wonder where the alleged researchers included ash thickness insulating properties in their “models”.
Observational evidence:
Here in Colorado* we’d get say 18″ of snow. Then.
Snow removal crews create 9-12′ snow mounds just COVERED with asphalt, dirt, oil stains, trash.
Then.
After a month or more of 70°F+ days, still there. Dying away. But still there.
*Colorado snow doesn’t stick around long. The high altitude and just about the most cloudless / sunshine days over any other state just eliminates snow.
The snow in those parking lots disappear much faster (~week) when not plowed.
I don’t really accept ‘soot’ melts glaciers. And it does appear intuitive to make that claim. But, yeah. Until observations.
Compare advance and retreat of the polar ice coverage and the glaciers with the list of volcanic eruption V4, V5 & V6 since 1900.
V4 (0.1 km3 to 1km3 DRE)*
Nabro 2011
Puyehue-Cordón Caulle 2011
Grimsvötn 2011
Merapi 2010
Eyjafjallajökull 2010
Sarychev Peak 2010
Kasatochi 2008
Chaiten 2008
Reventator 2002
Ulawun 2000
Lascar 1993
Mt. Spurr 1992
Kelud 1990
Kiluchevkoi 1987
Chikurachki 1986
Mount Augustine 1986
Colo 1983
Galunggung 1982
Pagan 1981
Alaid 1981
Mount Augustine 1976
Tolbachik 1975
Volcan de Fuego 1974
Tiatia 1973
Fernandina 1968
Mount Awu 1966
Kelud 1966
Taal 1965
Shiveluch 1964
Carran-Los Venados 1955
Mount Spurr 1953
Bagana 1952
Kelud 1951
Mount Lamington 1951
Ambrym 1950
Hekla 1947
Sarychev Paek 1946
Avachinsky 1945
Paricutin 1943–1952
Suoh 1933
Volcan De Fuego 1932
Mont Aniakchak 1931
Kliucheskoi 1931
Komagatake 1931
Komagatake 1929
Avachinsky 1926
Raikoko 1924
Manam 1919
Kelud 1919
Agrhan 1917
Tungurahua 1916
Sakurajima 1914
Mount Lolobau 1911
Grimsvötn 1903
Monut Pelee 1902
V5 (1 km3 to 10 km3 DRE)*
Mt. Hudson 1991
El Chichon 1982
Mt. St. Helens 1980
Agung 1963
Bezymianny 1956
Kharimkotan 1933
Cerro Azul 1932
Katla 1918
Colima 1913
Ksudach 1907
V6 (10 km3 to 100 km3 DRE)*
Pinatubo 1991
Katmai/Novarupta 1912
St. Maria 1902
* Volume of tephra as ‘Dense Rock Equivalent’
There is a sparse correlation with the global temperature anomaly
http://www.vukcevic.talktalk.net/GV-T.gif
Interesting…
Vuk – please see above – it takes a V5 of V6 volcano to have much cooling effect, about 0.5C.
https://wattsupwiththat.com/2017/10/25/study-ice-sheets-may-melt-rapidly-in-response-to-soot-from-distant-volcanoes/comment-page-1/#comment-2645637
What temperature data are you using?
Large eruptions drive El Nino episodes, and El Nino episodes drive strong warm pulses to the AMO, with about an 8 month lag.
https://www.esrl.noaa.gov/psd/data/correlation/amon.us.data
Yogi – you’re joking, right?
Wrong.
https://phys.org/news/2017-10-large-volcanic-eruptions-tropics-trigger.html
https://www.nature.com/articles/s41467-017-00755-6
http://www.meteor.iastate.edu/gccourse/forcing/responses.html
Yogi, with respect, I don’t have much time this morning and I have not studied this hypothesis in detail – specifically, the hypo you cited that “Large eruptions drive El Nino episodes”.
So I may be wrong – I was wrong once before, when I thought I had made a mistake and later determined that I had not. 🙂
Let’s do a full-Earth-scale test to eliminate scale-up uncertainties:
Century-scale volcanoes like El Chichon in 1982 and Pinatubo in 1991 caused global cooling of about 0.5C, which then naturally dissipated over several years.
A counter-argument to your above-cited hypo is:
1. There were naturally-occurring ocean warming spikes in the Nino3.4 area in 1983 and 1992, one year after the eruption of major volcanoes El Chichon (1982) and Pinatubo (1991+). These ocean warming spikes would normally cause a medium or strong El Nino warming spike in atmospheric temperatures.
2. The atmospheric temperature warming spikes that would normally characterize a full El Nino event were suppressed by these volcanic eruptions and the resultant global cooling of ~0.5C.
4. The global warming that followed these major volcanoes was a natural recovery of temperature as the cooling aerosols and particulates settled out of the atmosphere over several years.
5. Major and lesser El Nino events are fairly common and occur about every 4 years on average.
6. There was more than one significant volcano (V4 and above) every two years since 1900, such volcanoes are commonplace, and the alleged correlation of El Nino events after volcanoes could be spurious.
https://wattsupwiththat.com/2017/10/25/study-ice-sheets-may-melt-rapidly-in-response-to-soot-from-distant-volcanoes/comment-page-1/#comment-2645920
7. The evidence to support the above points in included in my (not Bill’s) first plot below.
Regards, Allan
https://wattsupwiththat.com/2017/09/20/from-the-the-stupid-it-burns-department-science-denial-not-limited-to-political-right/comment-page-1/#comment-2616345
NOT A WHOLE LOTTA GLOBAL WARMING GOIN’ ON!
Unlike the deeply flawed computer climate models cited by the IPCC, Bill Illis has created a temperature model that actually works in the short-term (multi-decades). It shows global temperatures correlate primarily with NIno3.4 area temperatures – an area of the Pacific Ocean that is about 1% of global surface area. There are only four input parameters, with Nino3.4 being the most influential. CO2 has almost no influence. So what drives the Nino3.4 temperatures? Short term, the ENSO. Longer term, probably the integral of solar activity – see Dan Pangburn’s work.
Bill’s post is here.
https://wattsupwiththat.com/2016/09/23/lewandowsky-and-cook-deniers-cannot-provide-a-coherent-alternate-worldview/comment-page-1/#comment-2306066
Bill’s equation is:
Tropics Troposphere Temp = 0.288 * Nino 3.4 Index (of 3 months previous) + 0.499 * AMO Index + -3.22 * Aerosol Optical Depth volcano Index + 0.07 Constant + 0.4395*Ln(CO2) – 2.59 CO2 constant
Bill’s graph is here – since 1958, not a whole lotta global warming goin’ on!
My simpler equation using only the Nino3.4 Index Anomaly is:
UAHLTcalc Global (Anom. in degC, ~four months later) = 0.20*Nino3.4IndexAnom + 0.15
Data: Nino3.4IndexAnom is at: http://www.cpc.ncep.noaa.gov/data/indices/sstoi.indices
It shows that much or all of the apparent warming since ~1982 is a natural recovery from the cooling impact of two major volcanoes – El Chichon and Pinatubo.
Here is the plot of my equation:
https://www.facebook.com/photo.php?fbid=1106756229401938&set=a.1012901982120697.1073741826.100002027142240&type=3&theater
I added the Sato Global Mean Optical Depth Index (h/t Bill Illis) to compensate for the cooling impact of major volcanoes, so the equation changes to:
UAHLTcalc Global (Anom. in degC) = 0.20*Nino3.4IndexAnom (four months earlier) + 0.15 – 8*SatoGlobalMeanOpticalDepthIndex
The “Sato Index” is factored by about -8 and here is the result – the Orange calculated global temperature line follows the Red actual UAH global LT temperature line reasonably well, with one brief deviation at the time of the Pinatubo eruption.
Here is the plot of my new equation, with the “Sato” index:
https://www.facebook.com/photo.php?fbid=1443923555685202&set=a.1012901982120697.1073741826.100002027142240&type=3&theater
I agree with Bill’s conclusion that
THE IMPACT OF INCREASING ATMOSPHERIC CO2 ON GLOBAL TEMPERATURE IS SO CLOSE TO ZERO AS TO BE MATERIALLY INSIGNIFICANT.
Regards, Allan
“A counter-argument to your above-cited hypo is:
1. There were naturally-occurring ocean warming spikes in the Nino3.4 area in 1983 and 1992, one year after the eruption of major volcanoes El Chichon (1982) and Pinatubo (1991+)”
For all you know there may have been La Nina conditions then without the eruptions. The El Nino episodes developed during 1982 and 1991, not a year after the eruptions.
The misdirection in the article is in the photo. No matter where you look there are new layers on old. If catastrophic melting were a problem there would be a very thick darkened layer that all subsequent layers would have melted down to. That isn’t happening. The ice core is nothing but evidence of ice staying ahead of the soot.