Connection — Volcanoes & Hurricanes

Volcanoes and Hurricanes: Mortal Enemies, Best Friends?

The plume from a lateral blast at Pinatubo in the Philippines seen on June 15, 1991. The eruption may have helped stifle hurricane activity in the Atlantic for three years afterwards.

We have had many discussions over the years here on Eruptions about the relationship betweenvolcanic eruptions and weather/climate (remember, they are different things). Most of the time, the concern is how weather will become worse (i.e., much colder or much hotter) due to volcanic aerosols or ash that are kicked high into the atmosphere during large eruptions. Remember, ash plumes from manyplinian eruptions can tower over 35-50 km up, so material can be injected into the upper atmosphere and spread around the world in a matter of weeks. It would be very surprising if these sorts of eruptions – which are relatively rare, only occurring maybe once or twice a decade – didn’t effect weather and climate for years until the aerosols can all settle out.


So, I was quite interested when I saw a new paper in the Journal of Geophysical Research titled “Atlantic hurricane activity following two major volcanic eruptions” by Amato Evan. My instant thought was I actually wasn’t sure what to expect – I mean, how would a large eruption effect the activity of such major, hemisphere-spanning events like hurricanes? Would it make hurricanes worse? As it turns out, this study suggests that major eruptions in the tropics (or close) might actually subdue Atlantic hurricane activity for years after the eruption.

Figure 2B from Evan (2012) showing the drops in sea surface temperature (SST) in the Atlantic after the eruptions of El Chichón (1982) and Pinatubo (1991)

Evan (2012) looks at two eruptions in particular – the1982 eruption of El Chichónin Mexico and the 1991 eruption of Pinatubo* in the Philippines. Both were large eruptions, ranking as VEI 5-6. Both eruptions injected large volumes of aerosols and ash into the upper atmosphere in the tropics, reducing theoptical depth of the atmosphere to 0.1-0.2 (normally it should be closer to 0.01). To give you an idea, that is almost as bad as other large eruptions such as Krakatau in 1883, famous for the vibrant skies it produced worldwide. All these aerosols in the atmosphere increase the albedo of the planet – that is, the planet will reflect more sunlight back into space. This means less sunlight hitting the surface of the Earth, and in particular, less on the oceans in the tropics. This produces colder surface and near-surface waters in what is called the Atlantic Main Development Region (MDR) for hurricanes – between 8-20°N/20-65°W (see right). This decrease in sea surface temperature, in turn, lead to an increase in vertical wind shear in the MDR.

What Evan (2012) found was that the total number of hurricanes in the three years before each eruption and three years after the eruption were markedly different – ~12 per season prior to the eruption and 6-8 per season after the eruption. Not only that, but the storms in the three years after the eruption were weaker and didn’t last as long as prior to the eruption. Even beyond this, the location that hurricanes formed changed as well, where before the eruptions most hurricanes were found in the MDR, after the eruptions there were dominantly found along the eastern United States. So, the long and short becomes large volcanic eruption leads to lower sea surface temperatures and higher vertical wind shear in the locations where hurricanes form, thus fewer hurricanes occur and those that do are weaker.

Now, bear in mind, this study only looked at two major eruptions in the last 35 years – and unfortunately both coincided with an El Niño, so one can’t conclusively link the eruptions and the change in hurricane activity. Evan (2012) mentions that there are at least three other major eruptions that could effect hurricane activity – Agung in 1963**, Santa Maria in 1902 and Krakatau in 1883. However, no pattern emerges from these eruptions as hurricane activity did decrease after Krakatau, it wasn’t effected by Santa Maria and appeared to increase after Agung. Evan (2012) suggests that the Agung eruption might have cooled the South Atlantic preferentially, causing the increase in North Atlantic hurricane activity.

Hurricane Irene off of Cuba and Florida, seen on August 24, 2011. Can volcanic eruptions help or hinder hurricane activity? It is still unclear.

Clearly, there is still a lot of noise in these correlations of hurricane activity and volcanic eruptions. The eruptions that Evan (2012) examined are the big ones – so, what if any effect would smaller eruptions in the tropics have (such as Merapi in 2010 or Nabro in 2011). Taking a look at the hurricane counts for the past century, you can see a number of periods of lower hurricane activity – can these all be correlated with eruptions like Katmai in 1912 (well out of the tropics) and what is causing the low hurricane counts in 2005-08? There are many unanswered questions here – but clearly, a closer examination looks to be in order – or, as the author of the paper suggests, maybe we need a large eruption in the tropics to test this theory out.

* Lockwood and Hazlett (2010) note that a typhoon/hurricane might have helped cause the cataclysmic eruption of Pinatubo in 1991. The lowest atmosphere pressure from the Typhoon Yunya passed over Pinatubo just 3 hours before the largest eruption. It likely didn’t cause the eruption (that was an injection of magma into the system over the prior few weeks), but it could have played a role in pushing the volcano pass the “tipping point” for an eruption.

** This eruption is listed in the paper as 1964, but the activity lasted from February 1963 to January 1964.

{Hat tip to Alex Witze for pointing out this article to me.}

Image 1: Pinatubo erupting in 1991. Image by Richard Hoblitt/USGS
Image 2: Figure 2B from Evan (2012), Journal of Geophysical Research
Image 3: Hurricane Irene in 2011. Image from the NASA Earth Observatory.



Ubehebe in Death Valley – Volcano Risk

As if Death Valley wasn’t dangerous enough… geologists discover that one of its volcanoes is due to go off


Death Valley in California has plenty of hazards, ranging from searing temperatures to flash floods, rock falls, rattlesnakes and scorpions.

Now geologists say that one of its volcanoes is actually far younger and more active than previously thought and is due to go off, because it last exploded in 1200 and has an eruption cycle of 1,000 years or less.

A team based at Columbia University’s Lamont-Doherty Earth Observatory found that the half-mile-wide Ubehebe Crater, formed by a prehistoric volcanic explosion, was created just 800 years ago – and not 6,000 years ago as previously estimated.

Explosive: The half-mile-wide Ubehebe Crater in Death ValleyExplosive: The half-mile-wide Ubehebe Crater in Death Valley

The researchers used isotopes in rocks blown out of the 600-foot crater to show that it formed around the year 1200.

That geologic youth means it probably still has some vigour, with the scientists certain that there is still enough groundwater and magma around for another reaction.

Ubehebe is the largest of a dozen such craters, or maars, clustered over about three square kilometres of Death Valley National Park.

The violent mixing of magma and water, resulting in a so-called phreatomagmatic explosion, blew a hole in the overlying sedimentary rock, sending out superheated steam, volcanic ash and deadly gases such as sulphur dioxide.

Study co-author Brent Goehring says this would have created an atom-bomb-like mushroom cloud that collapsed on itself in a donut shape, then rushed outward along the ground at some 200mph, while rocks hailed down.

Any creature within two miles or more would be fatally thrown, suffocated, burned and bombarded, though not necessarily in that order.

‘It would be fun to witness – but I’d want to be 10 miles away,’ said Goehring of the explosion.

Study: Geochemists dated the crater by analysing rocks thrown out when it exploded. Pictured is researcher Peri Sasnett contemplating a sampleStudy: Geochemists dated the crater by analysing rocks thrown out when it exploded. Pictured is researcher Peri Sasnett contemplating a sample

The team began its work after Goehring and Lamont-Doherty professor Nicholas Christie-Blick led students on a field trip to Death Valley.

Noting that Ubehebe remained poorly studied, they got permission from the park to gather some three to six-inch fragments of sandstone and quartzite, part of the sedimentary conglomerate rock that the explosion had torn out.

They pinpointed the dates to when the stones were unearthed to between 800 and 2,100 years ago and noted that this happened in clusters.

The scientists interpreted this as signalling a series of smaller explosions, culminating in the big one that created the main crater around 1200.

A few other dates went back 3,000 to 5,000 years – these are thought to have come from earlier explosions at smaller nearby craters.

Christie-Blick said the dates make it likely that magma is still lurking somewhere below.

He pointed out that recent geophysical studies by other researchers have spotted what look like magma bodies under other parts of Death Valley.

‘Additional small bodies may exist in the region, even if they are sufficiently small not to show up geophysically,’ he said.

He added that the dates give a rough idea of eruption frequency – about every thousand years or less, which puts the current day within the realm of possibility.

‘There is no basis for thinking that Ubehebe is done,’ he said.

The scientists stress that there are currently no signs of it waking up, which would be preceded by shallow earthquakes and the opening of steam vents, events that could go on for years before anything bigger happened.

The study appears in the current issue of the journal Geophysical Research Letters.

Read more:

Plate Tectonics & Earth’s Magnetic Field

Plate Tectonics May Control Reversals in Earth’s Magnetic Field

ScienceDaily (Oct. 21, 2011) — Earth’s magnetic field has reversed many times at an irregular rate throughout its history. Long periods without reversal have been interspersed with eras of frequent reversals. What is the reason for these reversals and their irregularity? Researchers from CNRS and the Institut de Physique du Globe(*) have shed new light on the issue by demonstrating that, over the last 300 million years, reversal frequency has depended on the distribution of tectonic plates on the surface of the globe. This result does not imply that terrestrial plates themselves trigger the switch over of the magnetic field. Instead, it establishes that although the reversal phenomenon takes place, in fine, within Earth’s liquid core, it is nevertheless sensitive to what happens outside the core and more specifically in Earth’s mantle.

This work is published on 16 October 2011 in Geophysical Research Letters.

Earth’s magnetic field is produced by the flow of liquid iron within its core, three thousand kilometers below our feet. What made researchers think of a link between plate tectonics and the magnetic field? The discovery that convective liquid iron flows play a role in magnetic reversals: experiments and modeling work carried out over the last five years have in fact shown that a reversal occurs when the movements of molten metal are no longer symmetric with respect to the equatorial plane. This “symmetry breaking” could take place progressively, starting in an area located at the core-mantle boundary (the mantle separates Earth’s liquid core from its crust), before spreading to the whole core (made of molten iron).

Extending this research, the authors of the article asked themselves whether some trace of initial symmetry breakings behind the geomagnetic reversals that have marked Earth’s history, could be found in the only records of large-scale geological shifts in our possession, in other words the movements of continents (or plate tectonics). Some 200 million years ago, Pangaea, the name given to the supercontinent that encompassed almost all of Earth’s land masses, began to break up into a multitude of smaller pieces that have shaped Earth as we know it today. By assessing the surface area of continents situated in the Northern hemisphere and those in the Southern hemisphere, the researchers were able to calculate a degree of asymmetry (with respect to the equator) in the distribution of the continents during that period.

In conclusion, the degree of asymmetry has varied at the same rhythm as the magnetic reversal rate (number of reversals per million years). The two curves have evolved in parallel to such an extent that they can almost be superimposed. In other words, the further the centre of gravity of the continents moved away from the equator, the faster the rate of reversals (up to eight per million years for a maximum degree of asymmetry).

What does this suggest about the mechanism behind geomagnetic reversals? The scientists envisage two scenarios. In the first, terrestrial plates could be directly responsible for variations in the frequency of reversals: after plunging into Earth’s crust at subduction zones, the plates could descend until they reach the core, where they could modify the flow of iron. In the second, the movements of the plates may only reflect the mixing of the material taking place in the mantle and particularly at its base. In both cases, the movements of rocks outside the core would cause flow asymmetry in the liquid core and determine reversal frequency.

* — Laboratoire de Physique Statistique of ENS (Ecole Normale Supérieure/CNRS/UPMC/Université Paris Diderot) and the Institut de Physique du Globe de Paris (CNRS/IPGP/Université Paris Diderot)