Volcanoes affect life. That is as true for volcanoholics as it is for other life forms. As Bjarki pointed out, the puffins on Bogoslof are not going to be impressed, when returning to their nesting holes to find them all gone, blown up to bits or filled with ash and lava. They will be affected by the errant eruption. The penguins of the South Sandwich Islands also live at the mercy of their resident volcanoes. As do the Icelanders.
But in all of living memory, no species is known to have gone extinct as a result of a volcanic eruption. Volcanoes affect life – they do not wipe it out, at least not beyond their immediate environment. The tortoises of the Galapagos have evolved and survived in their volcanic home, in spite of not being known for their ability to outrun a pyroclastic flow. If any species would be at volcanic risk, it would be them. And still they live.
Going back further in time, things change. Five major mass extinction events have been identified in the fossil record. The most famous of these is the so-called ‘K-T event’, 65 million year ago, which is linked to the demise of the dinosaurs. This was a bad one: few creatures bigger than a small dog survived. But it is not the only one, nor is it the worst one on record. And volcanoes carry at least part of the blame. Your local volcano, friendly, albeit explosive, is not a risk to the species. But flood basalt eruptions are a different matter. They are infrequent, distant – and deadly.
The main extinction events often fall at the boundary between geological eras, as is reflected in their names. They are the Ordovician-Silurian extinction, 440 million year ago, the Devonian extinction, 360 million year ago, the Permian-Triassic extinction, 250 million year ago, the Triassic-Jurassic extinction, 210 million year ago, and the Cretaceous-Tertiary extinction: 65 million year ago. Three of these coincide with major flood basalt eruptions: the P-T event, the T-J event, and the K-T event. (It is obvious what these abbreviations mean, apart from KT where the K is from the German name for the Cretaceous.)
The blue lines shows the rate of extinctions over time, clearly showing several mass extinctions. The red bars show the known flood basalt eruptions, most recently that of the Columbia River. The largest extinctions coincide with flood basalt events. The relation is not conclusive: better dates would be needed how well they coincide. But it is suggestive. Original at https://www.le.ac.uk/gl/ads/SiberianTraps/Images/LIPsExtinctions_jpg_files/LIPsExtinctions.jpg (University of Leeds)
Extinction
So what happens during a mass extinction? A global catastrophe reduces the number of species by half or more. It affects not just one environment, but all or almost all of the globe, and impacts every ecological niche.
In the O-S extinction, life was still based in the seas. These were hammered (ok, perhaps ‘hammered’ is not the best expression for a sea, but you get the gist). 85% of life in the sea died, possibly in two different episodes a few hundred thousand year apart. Trilobites were badly reduced, by 90%, although they survived as a group. The event coincided with a major ice age but it is not clear whether this ice age is responsible. It has been suggested (somewhat speculatively) that the ice age was caused by explosive plant growth which reduced greenhouse gases in the atmosphere so much that it led to the ice age.
In the Devonian extinction, 75% of all species died out, with especially the shallow seas devastated. The corals of the day died out, and reefs took 100 million year to recover, when new types of coral had finally evolved. This extinction may be linked to a sharp drop in sea temperatures. Animals had only just made it on to the land, but apparently the newly evolved land-conquerors did not survive. The next evidence for land-based amphibians occurs ten million year later, and they may have had to re-evolve. How different life could have been!
The P-T event was the worst, with over 90% of species dying out. This was the only mass extinction which also affected insects. The ‘Great Dying’ is the closest life came to be eradicated. It will be the topic of the next post.
The T-J event is a bit of a strange one, with sea life and animals badly affected but plants not. The domination of the dinosaurs was undisputed after this event. Beforehand, some mammals were still competing with the dinosaurs on equal footing.
The K-T event is famous for being the one that removed the dinosaurs from the Earth. But much other form of life was also affected: about 50% of species disappeared. Surprisingly, of all the dinosaur-related life, only the birds survived. Birds are not the most robust form of life, so how did they survive something that eliminated so much else? The one thing that stands out is that birds can traverse long distances over sea. They could have survived on a remote island and re-spread. Dinosaurs could have survived in such a place as well, but unable to make their way to the continents, they finally disappeared when the island went down – Atlantis of the dinosaurs. (This is of course extreme speculation.) Birds ruled the world, until mammals became competitive. (New Zealand kept out the mammals, and until human settlement was the last vestige of this post K-T world-of-the-birds.) The dominance of sharks in the seas came after the K-T event removed their competitors.
These are the five major extinctions in the fossil record. There may have been older events, but there is little fossil record pre-dating the Cambrian, and so it is not possible to find evidence. There are also some 18 minor extinction events known, which affect a smaller percentage of species, and may only affect one major group of life forms.
Flood basalt eruptions
Normal volcanoes only affect their immediate environment, or if they are so large that there are global effects, those effects are not long-lasting. Tambora’s year without summer was worldwide, but only one year. Even Toba, the worst bang on record, an eruption like a Trump tweet, did not destroy the world.
Flood basalt eruptions are different. They erupt along long rifts, draining huge magma reservoirs in the process. Volumes are enormous, and individual flows can be hundreds of kilometers wide, 500 kilometer long and 50 meter deep. And coming at you faster than you can run. The most recent one was the Colombia flood basalt, in the western US, 16 million year ago. Huge amounts of sulphates are released. There may be some explosions, but the eruptions are largely effusive, like the Icelandic fires but incomparably larger. Locally, they are not as dangerous as explosive eruptions (although not without danger – just the heat from the lava flows can still lift debris into the stratosphere, and therefore cause secondary pyroclastic flows even without explosions). But the enormous volumes can give large distant impacts.
Large flood basalt eruptions (or LIPs: large igneous provinces) are infrequent but not rare. The official LIPs database lists 47 such events over the past 500 million year, of which 36 are in the past 250 million year, so on average there is one per 7 to 10 million year. Many are not well dated, and the age may only be known to the nearest 100 million year. Some are related to mantle plumes, whilst others are part of continental or oceanic rifting events. There is one within the UK: the North Atlantic Volcanic Province, 60 million year old, also found in Greenland and covering an area of 1.3 million square kilometer. It is associated with the formation of the North Atlantic Ocean, and perhaps still continues as Iceland but that is disputed. Severe flood basalts form much of their volume within a fairly short period, perhaps 1 million year. Many LIPs have formed more sedately, over a much longer period of time. Iceland can be considered a LIP but is not such a short-lived flood basalt. The Colombia flood basalt was a short-period event.
It is notable that there have been relatively few of these events over the past 100 million year. As many (but not all) occur during the formation of new oceans, it makes sense that you get more flood basalt eruptions during a supercontinent, when they are in the process of breaking up (which supercontinents do most of the time). We are living in quiet times, as far as flood basalts go.
Lists of the largest flood basalt eruptions can readily be found on-line. However, these should be used with caution. Often, the true size of a flood basalt is not well known as most of it is covered under later sediment. So the ‘largest’ ones are really the ‘most obvious’ ones, or even worse, the best studied. As an example, one of the largest known is the Kerguelen LISP. But this benefits from being on the ocean floor, where it is readily visible as a huge dome (with an island on one end). But it may have formed over a 30 million year time span which is far too long, so it may actually be a combination of several separate events. The largest one by area, as far as known, is the Ontong Java event, also below the ocean, but also with more than one age (125 million year and 85 million year).
Now we can ask the question whether the flood basalts coincide with the main mass extinctions. The answer is ‘Yes, but..’. There is a good correlation between the P-T extinction and the Siberian Traps, between the T-J extinction and the Central-Atlantic Magmatic Province, and between the K-T event and the Deccan Traps. But other flood basalts, including the largest, do not correlate with an extinction peak. And one has to wonder about the accuracy of the dates. The most accurate dating has been done for those already suspected of involvement. Plots of the correlations, such as the one above, exclude events with uncertain dates (or worse, pick the ‘best’ date among several possible). This makes the correlation look better than it really is. And with one every 10 million year, you do really need very accurate dates.
For the three events listed above, there is a very good coincidence in the timing. But whether this is by chance or not, and why some would cause extinctions but some other equally large events would not, is open for discussion.
The causes of mass extinctions
To explain a mass extinction, three things need to be defined. First, the type of event that caused it. Second, the product of this event that did the damage. Third, how that product caused death. As a specific example, the main proposed causes are impact and flood basalt eruptions. The product that (potentially) causes mass death can be dust, which can cause a decade of darkness and winter. The cause of death can be starvation, or freezing, or more. Since different types of events can lead to similar effects (i.e. both impacts and volcanoes can cause dust), it can be difficult to ascertain the original cause, even if we know everything else about the extinction!
To get a mass extinction, there has to be a global change to the environment. There are several possibilities: the climate, the air, or the water. But what can cause a change so severe that half or (much) more of all life dies? It can be a severe ice age, much worse than the Earth has seen recently. A volcanic eruption may poison the entire world with sulphate and fluorine. But it is actually difficult to envisage an event that is so bad that it impacts the habitability of the entire world.
One geological principle says that only Earth can affect Earth. Mass extinctions must therefore be due to something internal to the Earth, such as volcanic eruptions. But this principle is incomplete, because we know that external factors cannot always be ignored. The extinction of the dinosaurs coincided with a big impact, the ultimate external factor. This led people to go to the other extreme, and suggest that all extinctions should have an external, impact origin. But clearly, things are not as straightforward as that. For the K-T event, there was both a major bang (impact) and a boom (the Deccan Trap eruption), so which one was to blame? And this is not the only such case. The Eocene (minor) extinction 30 million year ago has been blamed on a flood basalt, but recently the Popai impact crater in Russia was found to have exactly the right age. Both flood basalts and major impacts are not that common. It seems that statistically, both correlate with extinction events, and in several cases they both coincide with the same extinction event. Pick your choice: are you an internalist or an externalist?
Two solutions are possible. First, it has been suggested that a big impact can trigger a flood basalt eruption. There is no clear reason in physics why this should be, but that does not mean it is impossible. (At one time it was thought a big impact could cause some effect at the opposite point on Earth, the antepodal point. But the Chicxulub crater and the Deccan Traps are not antepodal enough for this, and again there is no physics model behind this idea.) The fact that the Deccan Traps eruption began well before the impact is a bit of a blow to this theory. Impacts do big things, but they can’t make time run backwards, no matter how attractive living in the past may seem. That pretty much killed the idea that impacts caused the flood basalt eruptions.
The original Chicxulub crater, before being buried by sediments
The other solution is that it takes two to tango. An event can lead to a mass extinction if the environment is already stressed for some other reason. In this model, the Deccan Traps set the scene, so that the impact could wipe out the dinosaurs. Neither the Deccan nor the Chicxulub impact could have done this by themselves. But together, they killed of T. Rex. Impacts the size of Chicxulub may happen once per 50 million year, perhaps more often. The Deccan Trap eruption may have lasted a million year, so the chance of such an impact happening during the stress time is 1 in 50. That is bad luck, but not impossibly bad. And seeing that there is a flood basalt eruption every 10 million year, over 500 million year you are talking about even chances that one such impact will coincide with one flood basalt eruption. There is no need to suggest that impacts can cause flood basalts: quite apart from the lack of a base in physics, flood basalts are frequent enough that it is more than likely that one will coincide with a big impact. A decent risk analysis would have flagged the possibility to the dinosaurs.
Now one should not panic too quickly. Nowhere on Earth is there any indication of an imminent flood basalt eruption. The signs would be hard to miss: prior to the eruption, a huge bulge would form, caused by inflation from the magma chamber and the hot mantle underneath. There are many eruptible magma chambers on Earth but none in such a location. They occur on average 10 million year apart – that is a very long time. On geological time scales, they are frequent. On human time scales, they are non-existent. Adding to this that only some flood basalts cause mass extinctions, when different circumstances come together, and you’ll soon find other things to worry about! Neither is there a major impact due. If there was an extinction-size asteroid heading for us in the next century, chances are we would already know about it.
It appears that life is robust enough to deal with minor mishaps. It takes something big to get a mass extinction, and only huge events can give massive mass extinctions. This can be one individual event, or it be a sequence of unfortunate events. What is the role of volcanoes? Are they always contributors in part? Are they innocently accused? Or are they the main cause of mass extinctions? It seems they have a major role to play in most mass extinctions. That friendly neighbourhood volcano has a dark secret to hide.
Next: the Permian extinction