Original subtitles

In the first video about Flood Basalts we have learnt what Flood Basalt Eruptions are

and what scope they had. We explored using the example of the Siberian Traps, which are

the remnants of one of the largest of these eruptions in history, what impact this kind

of volcanism can have on our planet and its inhabitants and we have learned where and

when else these so called large igneous provinces arose.

One question, however, remains to be answered. Where did they come from? Or more specifically,

what causes these cataclysmic eruptions. That's what we will explore today.

The exact processes are of course still subjecte to heated debate – but the principal model

for their origin which is widely accepted today, is the so called Mantle Plume Model.

Before we can discuss what exactly this is, however, we first have to understand how volcanoes

are typically formed. Almost all volcanoes on Earth are the result of tectonic processes

and are formed either by two colliding plates or two plates drifting apart.

Subduction zones are convergent boundaries where the edge of one plate (usually a denser

oceanic one) is forced under the edge of a less dense plate and then pushed into the

mantle. During the subduction the down-going plate experiences increasing pressure and

temperature. At depths of around 100 km or 60 mi the pressure is great enough that water

inclusions trapped in the rock are freed and released into the overlying mantel which lowers

the melting point of the already hot mantle rocks resulting in partial melting. This process

is called Flux melting and the magma it produces then slowly rises into the crust above and

finally to the surface forming a chain of volcanoes alongside to the Subduction zone

In contrast divergent boundaries are zones where two plates move apart. Here a spreading

ridge – either in form of a rift valley or a mid ocean ridge is created – through

which hot mantle material can rise to the surface. This also reduces the melting point

of the rocks, this time as a result of decompression melting caused by the reduction in pressure

during the ascent. The resulting magma then leaks onto the surface, cools and creates

new ocean floor in form of giant undersea mountain ranges.

But how do mantle plumes fit into this picture? - Surprisingly little. The volcanic processes

we just covered are all the result of mechanisms inside the uppermost layers of the earth – roughly

the first 100-200 kilometres. Mantles Plumes on the other hand have a much deeper origin.

They are columns of enormous quantities of hot rock, upwellings that rise to the surface

from the depth of earth’s mantle – 2900 kilometres or 1800 mi below our feet. As such

they are part of the on-going convection processes that take place in the mantle in which hot

material from inside the earth continuously rises up below the tectonic plates, cools

and moves back down again: Similar to what happens inside a lava lamp, just way more

complex. Because the mantle consists for the most part out of solid rock that only behaves

like a fluid over a geologic timeframe these processes are of course very slow and take

millions and hundreds of millions of years. The formation of a mantle plume starts at

the core-mantle-boundary. Here, in the thermal and chemical boundary layer at the base of

the mantle which separates the liquid outer core from the solid lower mantle temperatures

rise rapidly, faster than in any other layer. The temperature of the outer core is already

approximately 1,000 degrees Celsius higher than that of the overlying mantle just a few

kilometres above. This causes large amounts of heat to be transferred into the mantle

through conduction where it heats up the rock causing it to start rising: A Mantle Plume

forms. As it ascends through the mantle it slowly

start to take on a mushroom shape because the hot material rises faster through the

plume than the plume itself rises through its surroundings – not unlike during the

explosion of a nuclear bomb. Most plumes never make it to the surface before they cool down

again and lose their momentum but some of the largest can rise all the way through the

mantle and below the lithosphere. When such a plume hits the tectonic plates, which act

like a barrier it flattens out and deforms into a thinner and wider disk. As we have

discussed before the reducing pressure will eventually reduce the melting point of the

hot solid rock so much that it begins to melt. This produces enormous quantities of liquid

basaltic rock, basically a giant bubble of magma with a diameter of multiple hundred

kilometres directly beneath earth’s plates. From here it will start to rise into the crust,

build up in countless magma chambers and ultimately produce large scale flood basalt volcanism

on the surface. This can go on for a few hundred thousand

to a couple million years but eventually the plume head will cool down so much that the

large scale volcanisms stops. What remains is the much more narrow tail of the plume

which will periodically continue to transport magma to surface for 100 million years or

more until it dries up too. This model is of course a very simplified

view of the process. The reality is much more complex and chaotic. But it gives us at least

a basic understanding for why and how these massive eruptions have occurred, why they

are so rare and why their chemical composition is so different from regular volcanoes. Beyond

the formation of large igneous provinces this models also allows us to explain the dozen

or so volcanic hotspots that you can find all across the planet. These regions of continuous

volcanic activity are unusual because they are often far away from plate boundaries - in

some cases thousands of kilometres - and thus can’t be explained through tectonic processes.

Their chemical composition is also notably different from other volcanoes and more in

line with that of flood basalt eruptions. The most well-known of these hotspots are

probably Hawaii, Yellowstone and Iceland. Because Mantle Plumes aren’t the result

of plate movements they aren’t tied to them. Quite the contrary: Because each plume is

anchored at the core-mantle boundary and is therefore relatively stationary in relation

to the core the hotspot is constantly changing is position on the plate– not because the

hotspot is moving but because the plate is moving. As the plate moves across the hotspot

over the course millions of years it creates a chain of volcanic structures. This explains

for instance the formation of the Hawaii-Emperor Chain, a chain of around 130 dormant sea-mounts

stretching over 5800 kilometres across the pacific plate like a string of pearls with

the Hawaiian Islands at the end. Other hotspots follow similar patterns. Yellowstone

for instance has over the last 15 million years slowly moved east-ward as the north

American plate moved westward over the hotspot. Follow this trail of breadcrumbs and you can

see an image of the volcanic activity through time. When you reach the point in time 16

million years ago the position of the Yellowstone hotspot on the North American plate overlaps

suspiciously well with the Columbia River Basalt Group. This suggests that the hotspot

was responsible for the enormous flood basalt province that formed between 16-9 mya when

the Plume head hit the lithosphere and that since then the tail of the plume was responsible

for the regular supervolcano eruptions In Hawaii’s case however you can’t trace

the breadcrumbs back a large igneous province as the subduction zone off the coast of Russia

has already destroyed all evidence of its existence. On first glance you might think:

Doesn’t the trail point perfectly to the Siberian traps? This is true, of course, but

only a coincidence. Once you cross plate boundaries you also have to consider the position of

the plates relative to one another over time and Eurasian Plate was of course at no point

in time over todays Hawaii, at least not in the last 500 million years or so – In fact

no continental plate was – The Hawaiian Hotspot was always surrounded by oceanic plate.

When exactly it happened and how destructive its first eruption was, we will therefore

probably never find out. For other hotspots this is however still possible.

The Reuinon Hotspot off the coast of Madagascar for instance can be linked to the Deccan Traps.

When 66 million years ago the Indian Plate on its way to its current position moved across

the hotspot apparently right as the plume head hit the lithosphere the resulting eruption

then formed the massive lava province that even today covers nearly a third of India.

The some 4,300 km long chain of seamounts produced by the Louisville Hotspot may point

towards the Ontong Java Plateau while the Iceland Hotpot was likely responsible for

the formation of the North Atlantic Igneous Province. But connections like these aren’t

always possible – The Siberian Traps for instance can’t definitely be linked to a

present day hotspot. Iceland as well as the Ural mountains are proposed locations of the

plume today but it’s also possible that its corresponding hotspot already went dormant

in the past 250 million years. But not every large ignous province can be

explained through the mantle plume model - there are exceptions.

One such exception is the Central Atlantic magmatic province whose remnants today stretch

across 4 continents. This flood basalt province lacks some of the key features usually attributed

to mantle plumes for instance the characteristic chemical composition of the rock. Evidence

suggests that its formation instead was the result of the rifting and breakup of the supercontinent

Pangaea. This rift which later formed the Atlantic Ocean could’ve freed enormous quantities

of magma trapped inside and beneath the continental crust produced by the nearly the 360° subduction

zone around the supercontinent. Despite this many researchers also believe that the arrival

of a plume must have at least played some role in the initial breakup of the Continent

- which goes to show that there is still a lot to learn about the geology of our planet.

One tool that we are now using more and more is so called Seismic tomography. This method

– which is basically a CT scan of the interior of our planet - makes use of seismic waves

caused by earthquakes. Because these waves travel at different speeds in various types

of rocks and rocks of varying temperatures we can measure the velocity of the waves at

various points on the planet to make conclusions about the physical properties of the subsurface.

While the models these scans produce have at the moment still a very poor resolution

they have at least finally confirmed what for a long time was only conjecture – namely

deep-mantle plume-like structures under most major hotspots like Yellowstone, Hawaii and

Iceland. But these are obviously only still images of convection processes that take millions

of years. To fully understand the formation of mantle plumes and flood basalt provinces,

it therefore still requires a lot more research. But large ignous provinces are of course not

only of geological importance but also of biological interest. What catastrophic impact

the formation of such provinces can have on the planet we have already discussed in the

last part using the example of the Siberian Traps – But this was by no means an isolated

case – quite the opposite. If you illustrate the formation of the largest

flood basalt provinces of the last 500 million years graphically, you can see a striking

temporal correlation with the boundaries of geologic time periods. These boundaries are

defined by abrupt and significant changes of earth’s biosphere and climate and the

planet as a whole and often mark the point of catastrophic mass extinctions.

The formation of the Deccan Traps for instance falls almost perfectly on the Cretaceous-Paleogene

or KT boundary that marks the extinction of the non-avian dinosaurs, the formation of

the central Atlantic magmatic province overlaps with the Triassic Jurassic boundary and the

formation of the Siberian Traps coincides as we have discussed with the Permian-Triassic

Boundary and the End-Permian Extinction. At the end of the Devonian Period the Earth

also experienced this kind of volcanism - in fact an exceptional amount of it. It is speculated

that this was the result of the arrival of a super plume under the east European platform

resulting in the formation of at least 4 flood basalt provinces over a relatively short 20

million year timeframe. Although the timing and importance of the individual events is

in this case still poorly understand it might be an explanation for the series of smaller

extinction events typically combined into the Late Devonian extinction. Particularly

the formation of the Viluy Traps seems to correspond well with the so called Kellwasser

Event – the first and most severe of the Late Devonian extinction Events.

Together these four boundaries describe 4 of the 5 largest mass extinctions in history,

each responsible for the loss of 70% or more of all species and probably 99% or more of

all individuals. Even the fifth and earliest of these so called “Big 5” can be linked

to large-scale volcanism due to elevated mercury concentrations in the Ordovician rock layers

although a corresponding lava province seems to no longer exist.

After the discovery of the Chicxulub crater in the mid-20th century and the realisation

that an asteroid likely was what had wiped out the dinosaurs the consensus in the scientific

community subsequently became that asteroid impacts must the main driving force behind

mass extinctions in general. Today however, thanks to intensive research

on the formation and role of large igneous provinces over the past few decades we now

know that this is likely not the case. Asteroids certainly had an important impact on this

planet but It simply can’t be a coincidence that all 5 of the largest mass extinctions

in history as well as many more - more or less - dramatic changes of our planet all

happened at the same time as some of the most catastrophic lava eruptions the world has

ever seen. Advancements in high precision dating both

of the mass extinction events and the formation of the corresponding flood basalt provinces

during the last 20 years have in almost all cases only strengthened this temporal link

and the examination of the events individually has in most cases confirmed volcanism as the

main cause for the extinctions and not as previously thought, asteroid impacts.

Even for the case example - the extinction of the non-avian dinosaurs – we now know

that large scale volcanism has at least played a role: Through the formation of the Deccan

traps that started erupting about 400,000 years before the Chicxulub impact and continued

for another 600,000 years after expelling a total of over half a million km³ of lava.

Although this eruption didn’t nearly have an as catastrophic effect on the climate as

previous ones, probably due to the lack of large scale sill intrusion, it still caused

alternating episodes of warming and cooling for multiple hundred thousand years. This

must have put significant stress on the biosphere and likely weakened many animal populations

as a result which is why many researchers believe the later impact might have not been

enough to drive the dinosaurs to extinction if it hadn’t been for the contemporary volcanism.

If you take all this into consideration it can no longer be denied what impact flood

basalts had and likely will continue to have on the development and evolution of life on

earth. It seems no other natural disaster has on such a scale decided who lives and

who dies. We owe these catastrophes our existence. Humans,

just like every other animal alive today are the ancestors of the few survivors of these

mega eruptions. The ancestors of the species that managed to survive literal hell on earth

for sometimes hundreds of thousand years only to then rise of the ashes to bring new life

to the planet that we call home.