Revising the history of big, climate-altering volcanic eruptions

New method, co-developed at UMD, refines the 2,600-year history of large eruptions that inject planet-cooling particles into the stratosphere


This photo, taken on June 12, 1991, shows the eruption column of Mount Pinatubo on Luzon Island in the Philippines. The eruption–the largest on Earth in the past 100 years–ejected particles into the stratosphere, more than 6 miles above the planet’s surface. New research uses ice core data to rewrite the past 2,600 years of large stratospheric eruptions like this one. CREDIT Dave Harlow/USGS

For all their destructive power, most volcanic eruptions are local events. Lava flows tend to reach only a few miles at most, while airborne ash and soot travel a little farther. But occasionally, larger eruptions can launch particles into the stratosphere, more than 6 miles above Earth’s surface. The 1991 eruption of Mount Pinatubo in the Philippines–the world’s largest eruption in the past 100 years–is a prime example of a stratospheric eruption.

When volcanic particles reach the stratosphere they stay aloft for a long time, reflecting sunlight and temporarily cooling the planet. By understanding the history of these big eruptions, researchers can begin to place short cooling episodes and other discrete climate events into the context of large-scale climate patterns.

Researchers working at the University of Maryland, the Université Grenoble Alpes in France, the Ecole Normale Supérieure in France and the Tokyo Institute of Technology have devised a new, more accurate system for identifying large stratospheric eruptions recorded in the layers of Antarctic ice cores.

Using their method, the researchers made some important revisions to the known history of big eruptions–correcting the record on several misidentified events while discovering a few as yet unknown stratospheric eruptions. The researchers described their approach, which identifies airborne volcanic particles with a specific chemical signature, in a paper published January 28, 2019, in the journal Nature Communications.

“I find it very exciting that we are able to use chemical signals to build a highly accurate record of large, climate-relevant stratospheric eruptions,” said James Farquhar, a professor of geology at UMD and a co-author of the research paper. “This historical record will be highly useful for climate scientists seeking to understand the role of large eruptions in climate oscillations. But there is also the basic marvel of reading a chemical fingerprint that is left behind in ice.”

Eventually, volcanic particles fall from the stratosphere, settling on the ground below. When they land on snow, the particles get covered up by more snow that gets compacted into ice. This preserves a record of the eruption that survives until the ice melts. Researchers can drill and retrieve ice cores in places like Antarctica and Greenland, revealing eruption records that stretch back several thousand years.

Because particles from large stratospheric eruptions can spread across the globe before falling to the ground, previous methods identified stratospheric eruptions by looking for sulfate particle layers in ice from both hemispheres–usually from Antarctica and Greenland. If the same layers of sulfate showed up in both cores, deposited at the same time in Earth’s history, researchers would conclude that the particles came from the same large, stratospheric eruption.

“For eruptions that are intense enough to inject material into the stratosphere, there is a telltale signature in the sulfur isotope ratios of sulfate preserved in ancient ice layers,” explained Farquhar, who also has an appointment in UMD’s Earth System Science Interdisciplinary Center. “By instead focusing on this distinct sulfur isotope signature, our new method yielded some surprising and useful results. We found that prior reconstructions missed some stratospheric events and falsely identified others.”

The study’s lead author, Elsa Gautier from the Université Grenoble Alpes, did a significant portion of the analyses at UMD while on a Fulbright scholarship to work with Farquhar in 2013. Following Gautier’s lead, the researchers developed their method using ice cores collected at a remote site in Antarctica called Dome C. One of the highest points on the Antarctic ice sheet, Dome C is home to ice layers that stretch back nearly 50,000 years.

Gautier and her colleague Joel Savarino, also at the Université Grenoble Alpes, collected ice cores at Dome C that contain records stretching back roughly 2,600 years, covering a large portion of recorded human history.

Time series of volcanic sulfate deposition at Dome C, Antarctica. a Sulfate deposition for volcanic events recorded in Dome C (Dome C volcanic index). Red colored symbols are stratospheric eruptions identified based on Δ33S proxy. Blue colored symbols are eruptions that do not display any sulfur isotope anomalies, and therefore are presumed to be tropospheric eruptions. Empty dots are uncertain events because the isotopic signal falls in the uncertainty of the method. Round shape illustrates the eruptions found to be bipolar signals in Sigl15, while square shapes represent the eruptions found to be unipolar (Southern Hemisphere eruptions) in Sigl15. Consequently, blue round dots and red squares are eruptions for which the isotopic and the bipolar method display different results. The isotopic records of Pinatubo and Agung are added from a prior study by Baroni et al.18,19. The flux is the volcanic sulfate deposition flux (cumulative sum integrated over each event), corrected from background, calculated from concentrations measured in this study. Dating is provided by Sigl et al.11. b Maximum sulfur anomaly for volcanic events recorded in Dome C. Color and shape code is the same as a. Values below 0.1‰ (in the gray area) fall within the variability obtained on background samples. They were therefore not corrected from background, to avoid false stratospheric signal (Δ33S > 0.1‰) due to correction, and are considered tropospheric or uncertain, if close to 0.1. Data are available in Supplementary Table 1. Error bars are 1 standard deviation (s.d.)

The researchers used their method to confirm that many events had indeed been properly identified by the older method of matching up corresponding sulfate layers in ice cores from both hemispheres. But some events, formerly thought to be big stratospheric eruptions, did not have the telltale sulfur isotope signature in their sulfate layers. Instead, the researchers concluded, these layers must have been deposited by two or more smaller volcanoes that erupted at about the same time at high latitudes in both hemispheres.

The researchers also found some big stratospheric events that contain the isotope signature, but were somehow constrained to the Southern Hemisphere.

“This is a strength of our approach, because these events would have a climate impact but are missed by other methods,” Farquhar said. “We have made a significant improvement to the reconstruction of large stratospheric eruptions that occurred over the past 2,600 years. This is critically important for understanding the role of volcanic eruptions on climate and possibly for understanding certain events in human history, such as widespread famines. It can also help to inform future climate models that will take large volcanic events into account.”


Via Eurekalert PUBLIC RELEASE: 

The research paper, “2600 years of stratospheric volcanism through sulfate isotopes,” Elsa Gautier, Joel Savarino, Joost Hoek, Joseph Erbland, Nicolas Caillon, Shohei Hattori, Naohiro Yoshida, Emanuelle Albalat, Francis Albarede and James Farquhar, was published in the journal Nature Communications on January 28, 2019. (open access)


0 0 vote
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
February 6, 2019 9:12 am

Look the 1270 drouth that wiped out the mesa verde community

John Tillman
Reply to  Hans Erren
February 6, 2019 9:31 am

The big (VEI7) Samalas eruption on Lombok Island was in AD 1257. The lesser (VEI6) Quilotoa, Ecuador eruption occurred around 1280. Dunno when or where the other apparent major late 13th century eruption shown in the graph was.

John Tillman
Reply to  John Tillman
February 6, 2019 9:45 am

Mt. Tarawera, NZ, did however erupt (VEI5) early in the 14th century or very late in the 13th, ie AD 1310, plus or minus 12 years.

Sweet Old Bob
February 6, 2019 9:17 am

Too much conclusion …too little data .

February 6, 2019 9:24 am

Willis is going to have to do another “Spot the Volcano’ post.

Dave Yaussy
Reply to  SMC
February 6, 2019 10:52 am

My thought as well. I’m going to wait for him to weigh in before getting too excited about this study.

Rich Davis
Reply to  Dave Yaussy
February 7, 2019 2:40 am

Well it comes to us from EurekAlert! so it starts out with three strikes against it by my reckoning.

I have always enjoyed Tokyo Institute of Technology studies, though.

February 6, 2019 9:44 am

Wait, wait! The warmunistas want and need volcanic eruptions to be big cooling events to explain past coolings. Don’t want intrinsic climatic changes to be the reason…..

Krishna Gans
Reply to  beng135
February 6, 2019 9:50 am

Don’t forget, vulcanic eruptions and earthquakes are climate change driven !

Reply to  Krishna Gans
February 6, 2019 10:35 am

Those must be volcanic eruptions in Vulcan, Spock’s planet. Its hotter temperature, stronger gravity, and thinner atmosphere make for an interesting climate.

Krishna Gans
Reply to  Javier
February 6, 2019 10:40 am

sorry for typo 🙂

Krishna Gans
Reply to  Krishna Gans
February 6, 2019 11:08 am

thx for the details !

Krishna Gans
Reply to  Krishna Gans
February 6, 2019 1:39 pm

That in Spanish it’s written with “u” is in so far interesting as in Frensh it’s written with “o” like in English, and both, Spanish and Frensh are Romance languages…

Reply to  Krishna Gans
February 7, 2019 12:07 pm

I often find myself typing vulcanology and have to correct it. Too much Latin in school, I guess. (4 years)

Krishna Gans
Reply to  Javier
February 6, 2019 10:42 am

German writing is always with “u”, so I didn’t pay enough attention 😀

Reply to  Krishna Gans
February 6, 2019 11:02 am

I also make the same mistake. In Spanish volcanism is also with u.

The origin of the word is the god Vulcan, with u.

Less known is that Vulcan was for a time a planet of the Solar System in orbit between Mercury and the Sun. It was proposed in 1843 by Le Verrier, but there were amateur astronomers claims since the early 17th century. Astronomers continued looking for Vulcan until Einstein explained the anomalies in Mercury’s orbit.

Heinrich Schwabe was allegedly looking for Vulcan when he discovered the solar cycle.

John Tillman
Reply to  Krishna Gans
February 6, 2019 3:34 pm

In Latin, the god’s name is spelt both Volcanus and Vulcanus, but his festival was the Vulcanalia.

As noted, it’s with an “o” in English and French, also Catalan, but with a “u” in other major Romance languages, Spanish, plus Portuguese, Italian and Romanian. English is unusual (possibly unique) among the Germanic languages, which are of the “u” school, probably thanks to our heavy French influence.

Germanic Icelandic however has “eldfjall”.

Eric Stevens
Reply to  Krishna Gans
February 6, 2019 3:40 pm


William Astley
Reply to  beng135
February 6, 2019 10:41 am

The cult of CAGW also require amplification, positive feedback of step changes to create CAGW.

The observations show after a tropical region volcanic eruption that there is cooling for only a few years rather than large oscillations of temperature which means the planet has negative feedback (cloud cover increase or decreases in the tropics to resist a forcing change), rather than positive feedback.

No amplification, no CAGW.

Limits on CO2 Climate Forcing from Recent Temperature Data of Earth
” …These effects do not have the signature associated with CO2 climate forcing. However, the data show a small underlying positive trend that is consistent with CO2 climate forcing with no-feedback.”

On the Observational Determination of Climate Sensitivity and Its Implications
Richard S. Lindzen1 and Yong-Sang Choi2

February 6, 2019 9:47 am

> Instead, the researchers concluded, these layers must have been deposited by two or more smaller volcanoes that erupted at about the same time at high latitudes in both hemispheres.

Please. This kind of wild speculation makes its way into a scientific paper?

Reply to  Rob_Dawg
February 6, 2019 10:41 am

It sounds like a reasonable explanation. I can’t think of another one, can you?

The key point is that when sulfate aerosols are ejected by a very large volcano high into the stratosphere, the UV light above the ozone layer changes the isotope ratios in the sulfur. That change is detectable in the deposited layers found in the ice cores.

So, according to the authors, if that signature is not found in the ice core layer, it must mean that the sulfate aerosols stayed below the ozone layer. That should mean the sulfate aerosols should not have been able to travel far enough for a single volcano to have created detectable layers in both Greenland and Antarctic ice cords. That, then, suggests two small volcanoes, with coincidental timing.

A more interesting question is how they could find layers which DO have the stratospheric isotope signature, yet appear in the ice core for only one hemisphere. Maybe if Greenland had one of its occasional surface melting events it could “erase” the evidence?

I like this! It is refreshing to read about interesting climatology done with actual measurements, rather than just speculative computer model runs misreported as “experiments.”

As for Pinatubo, its effects were even evident in the CO2 record:

comment image

Reply to  Rob_Dawg
February 6, 2019 11:21 am

I wrote, “A more interesting question is how they could find layers which DO have the stratospheric isotope signature, yet appear in the ice core for only one hemisphere. Maybe if Greenland had one of its occasional surface melting events it could ‘erase’ the evidence?”

If that’s what happened, then:

1. We should expect that the “unipolar” stratospheric eruptions would be seen in Antarctic ice cores, but not in Greenland ice cores. It shouldn’t be the other way around, because it never gets cold enough for such surface melting events in the Antarctic interior.

2. Each such unipolar (Antarctic-only) stratospheric eruption record suggests that there must have been a Greenland surface melting event within a few years after the eruption. That, in turn, suggests that (like today) Greenland was warm enough to occasionally have such events happen. So that suggests that we should expect to find those events mostly during the RWP and MWP, and not during the DACP and LIA. It also suggests that if that is when the occurred then it adds to the evidence that temperatures today are not very much warmer in Greenland than they were during the RWP and MWP.

Reading the paper, sure enough, apparently all of the unipolar stratospheric eruptions were seen in Antarctic ice cores, and not in Greenland ice cores, rather than vice-versa.

Three of the seven were during the RWP, one was at the beginning of the MWP, two were in the middle of the MWP, and one was at the end of the MWP.

So it does appear that the unipolar stratospheric eruptions could be evidence of warm RWP and MWP in Greenland. (That evidence is, of course, redundant, in the case of the MWP, since we have historical evidence of Viking settlers cultivating barley in Greenland during the MWP.)

So that leads to another question: Can those surface melting events be seen in the Greenland ice cores? If so, are they at the right dates to explain the lack of evidence in the Greenland ice cores for the seven unipolar stratospheric eruptions?

Rich Davis
Reply to  Dave Burton
February 7, 2019 2:57 am

It shouldn’t be the other way around, because it never gets cold enough for such surface melting events in the Antarctic interior.

I suppose that you meant to say warm enough?

Reply to  Rich Davis
February 7, 2019 9:28 am

Yes, I meant “warm.” Thank you, Rich!!

I hate it when my mistakes invert my meaning. ☹️

Even more annoyingly, my mouth sometimes does the same thing. My ears play back what my mouth said, and my brain hears the opposite of what it thought it told my mouth to say. Arrrgh!

February 6, 2019 9:47 am

According to what I found looking only at high latitudes volcanoes in the N. Hemisphere (Iceland, Kamchatka and Aleuti + Alaska) considering the V5+ there is a drop in temperature for usually one sometimes two years, followed with most often one year of temperature rise, then back to what might have been general trend before erruption. Effect of Iceland’s volcanoes stands out.

February 6, 2019 10:10 am

Volcanic eruptions affect weather ( a few years).
Climate change affects volcanic eruptions.

Krishna Gans
Reply to  Javier
February 6, 2019 10:22 am

Thanks that you added the link “to my post” 😀

February 6, 2019 10:23 am

Hmmm… Some eruptions were large but constrained to the Southern Hemisphere… Could it be that they happened nearer to the ice? Perhaps inside some sort of Polar Vortex? If only there were such a thing.

Krishna Gans
February 6, 2019 10:30 am

Instead, the researchers concluded, these layers must have been deposited by two or more smaller volcanoes that erupted at about the same time at high latitudes in both hemispheres.

It’s a somewhat unplausible to me

Reply to  Krishna Gans
February 6, 2019 12:17 pm

Why does two volcanoes going off at the same time sound implausible?

Reply to  Krishna Gans
February 6, 2019 1:03 pm

“At the same time” doesn’t mean at exactly the same time. Ice core records are somewhat “smeared” (though perhaps less so for particulate layers than for air bubbles?), so if they’re within a few years that’s surely close enough to be considered “at the same time.”

Bruce Cobb
February 6, 2019 10:33 am

Whew! It’s a good thing we have volcanoes, or else the oceans would be boiling by now.

February 6, 2019 10:46 am

Visual Volcanic eruptions are but a small aspect of volcanic activity generally. Admittedly they are very evident; but what is missing is the understanding of what is happening beneath our feet. Indeed not evident in weather perspectives; but very significant in matters of climate.
While we talk the continents move. As we conjecture, the magma broiles. On the effects of this on the thin layer of atmosphere in which we live to date we have little comprehension.

I may be wrong here but I gather some 70% of volcanic eruptions/activity occurs beneath the oceans often undetected, often discounted as a potential cause.

Adam Gallon
Reply to  Alasdair
February 7, 2019 12:42 am

Such eruptions, don’t inject sulphate particles into the atmosphere.

Kaiser Derden
February 6, 2019 10:53 am

our planet has 2 possible climates … iceball earth and not iceball earth … the weather can suck during both climates but iceball earth kills alot of things off … so I’m voting for the not iceball earth climate (current) …

Ben Vorlich
February 6, 2019 10:59 am

This gives me an opportunity to ask a question that Willis’ articles on volcano induced cooling or lack ot it and not to be too far off topic.

I cycle to the local shop most mornings at about 8am. At the moment it is just at sunrise, on 31st March the clocks are put forward one hour at which point I will be cycling at sunrise again. This triggered more important question: the manual readings of thermometers will be taken at a different time of day in summer and winter.

So the question is are the changes to and from summertime going to add to the adjustments done to compensate for time of day adjustments – TOBs adjustments? Or should it be taken into account if it isn’t?

Adam Gallon
Reply to  Ben Vorlich
February 7, 2019 12:44 am

Recordings will be taken at GMT & I doubt very much, if many used in the modern measurements, are read manually.

Loren Wilson
February 6, 2019 11:07 am

Novarupta in 1912 was bigger than Pinatubo. Is it in their data set? I can’t tell from the useless CE dating system.

Joel O'Bryan
February 6, 2019 11:18 am

The Pompey destroyer, Vesuvius AD 79, a VEI 5 event (similar to Mt St Helens in 1980) is not in their data at all.
The 1883 AD Krakatoa blast, a VEI 6, barely registers on their sulfate flux (at ~9kg/km^2), yet it was SH event.
And the only VEI 7 in recorded history, Tambora in 1815, barely registers in their sulfate flux data as well.

For reference, the 1453 AD Kuwae event registers strongly (at ~42 kg/km^2) in their data and was also a VEI 6.

Stay skeptical.

Reply to  Joel O'Bryan
February 6, 2019 4:12 pm

“I find it very exciting that we are able to use chemical signals to build a highly accurate record of large, climate-relevant stratospheric eruptions”

What they have done is identified sulfate ions.

I’m still puzzling how they identify Tropospheric ions versus Stratospheric ions.

Let alone, exactly what their sulfate ions mean graphed on a time scale.

February 6, 2019 11:31 am

People should know that VEI is not a measure of weather effect. The stratospheric sulfate that gets deposited in ice cores is the correct measure of the weather effect.

Mt. St Helens in 1980 had no weather effect because it did not inject much sulfate in the stratosphere. Other VEI 5 eruptions had a bigger weather effect, like El Chichon.

Joel O'Bryan
Reply to  Javier
February 6, 2019 11:38 am

Yet Tambora 1815, a Southern Hemisphere event, had measurable weather effects into the Northern Hemisphere for 12-18 months, but barely registers in their sulfate data.

Reply to  Joel O'Bryan
February 6, 2019 11:58 am

Tambora is one of the six biggest sulfate eruptions of the past 2500 years in Sigl et al., 2015 study. They used Greenland and Antarctic cores. I think their study is better, although the isotopic angle is interesting.

Reply to  Javier
February 6, 2019 1:18 pm

St. Helen’s blast was to the side.
With little vertical component to drive material “to the sky”.

Clyde Spencer
February 6, 2019 11:47 am

“The researchers also found some big stratospheric events that contain the isotope signature, but were somehow constrained to the Southern Hemisphere.”

It is well known that the Winter-time Antarctic circumpolar vortex keeps ozone from reaching the stratosphere over the interior of Antarctica. It should similarly act to keep volcanic injections from Antarctica spreading outward. The timing just has to be right to keep the aerosols confined for several months.

February 6, 2019 1:25 pm

St. Helen’s blast went mostly sideways–to the North.
Which is why I heard it that morning in Vancouver, B.C.
So relatively little stuff was propelled straight up, that might have left more of a record on this method.

Gordon Dressler
February 6, 2019 7:25 pm

From the article: “When volcanic particles reach the stratosphere they stay aloft for a long time, reflecting sunlight and temporarily cooling the planet.”

I would like to see some hard science data that supports that statement. Note that the first graph shows the sulfate flux from most large volcanic eruptions reaching the stratosphere is in the range of 5-30 kg/km^2 . . . that seems to equate to very little light interception/reflection since the sulfate particles are spread so thinly per unit area. Note: I assume these units reflect total ejected sulfate mass over the atmospheric column height to top of stratosphere . . . it would have been better to have provided values in units of kg/km^3 for several altitude ranges within the stratosphere/troposphere.

I suggest instead that there is actually an amplification of cooling caused by such eruptions by the mechanism of the sulfate aerosols acting as cloud condensation nuclei, and that it is the “seeded” clouds caused by such aerosols drifting down through the troposphere that increase Earth’s albedo for the time period associated with sulfates (and other particulates/aerosols) “settling out” from stratospheric ejection altitudes.

February 6, 2019 9:54 pm

Warner Hertzog did a documentary about volcanoes and their history called Into The Inferno. The Trailer is here. . I do not know if it is still available on NetFlix but Warner meets and interviews an esteemed volcanologist Clive Oppenheimer. Clive takes Warner to some of the sites of the largest eruption ever in the history of the world. Some of these make Krakatoa look like a non event with their sheer magnitude. The cinema photography is breath taking and the point of this documentary is we have not seen nothing yet because the past can be repeated.

February 6, 2019 9:58 pm

Has anyone out there got a link to David Suzuki hysterical blathering’s re the Mt Pinatubo eruption please? I have tried to find them on-line several times and cannot. I guess they have been “smoothed” out of existence…too much hot air for historical observance.

Krishna Gans
Reply to  bill
February 7, 2019 7:51 am
Krishna Gans
Reply to  bill
February 7, 2019 7:59 am
%d bloggers like this: