How Australia became an island
The oceans were once less known than the Moon. Perhaps this is true no longer. Gravity maps released last year show remarkable details of the ocean floor. Chains of (ex-)volcanoes are everywhere. Spreading ridges and transform faults are visible. Under water plateaus around islands are revealed. A new world has opened up.
A large, deep underwater plateau in the far southern Indian Ocean shows up, with an island on top that looks like an inverted version of Iceland. A quick search shows that the island is already discovered (by Kerguelen – twice), named (by Cook – twice), owned (by France), and occupied (by a colony of scientists). The main island is surrounded by an archipelago of perhaps 300 much smaller islands (I am not sure whether anyone has actually counted them). The archipelago is called Kerguelen Islands (Cook’s second name was Desolation Island). Kerguelen is supposed to be the most isolated piece of occupied real estate in the world. It is reachable only by boat across a storm-swept ocean. Questions arise – why is it there? Why the underwater plateau? And why does it look like an Iceland of the South?
As so often, the first stop for more information leads to Volcanocafe. A post by Alan gives fascinating detail on Kerguelen. The plateau is a ‘micro continent’, rather larger than shown above, traces a hot spot/mantle plume — and it was responsible for breaking up Gondwana. The kind of post that makes you want to know more!
The full underwater plateau is shown below (image credit: Australian government). It is over 2000 km long, covers an area of about 1.3 million square kilometres (four times the size of the British Isles, one sixth of Australia), and was created by volcanic eruptions. The southern end of the plateau formed 120 million years ago. The northern end, with Kerguelen itself, is younger, at 40 million years. But the age structure is complex, and in fact the central part, with Heard Island and McDonald island, has the youngest, and active, volcanoes (the only active volcanoes owned by Australia). Drilling has found evidence for charcoal: at times part of the plateau, now 2 km deep, was above water and covered in forests.
The large area qualifies it as a large igneous province (LIP). By definition, this is an eruption area covering more than 100,000 square kilometres. Kerguelen is larger than the Deccan Trapps. Eruptions this size, even if spread over as long as 50 million years, are often attributed to hot plumes melting the lithosphere below the eruption site. The head of such a plume can extend over 1000 km. This has lead to the hypothesis of the ‘Kerguelen mantle plume’. One problem with this hypothesis is that ocean floors move, and one place would not be expected to stay over a mantle plume for that long. It is hard to think of another explanation, though.
The map below shows the location of Kerguelen in the Indian Ocean (click to see full detail). Faintly visible is the SouthEast Indian Ridge, the ocean spreading centre between Kerguelen and Australia (the transform faults are better visible than the spreading ridge). Much clearer is the spreading ridge toward Africa. A straight line extending almost due south from the Andaman Islands is called the Ninety East Ridge and may be a hotspot trail. Just to the right of where it ends is the Broken Ridge. This is now best known as the area where the search for the missing MH370 flight is taking place, but Broken Ridge was actually part of the Kerguelen Plateau, separated by ocean floor spreading. Go back more than 100 million years, and the Indian Ocean didn’t exist. All of this was at the heart of East Gondwana, the last supercontinent.
Reeves and DeWit have calculated how the various spreading centres, active at different times, would have acted to form the Indian Ocean, starting from ancient Gondwana (click on the image above to see more detail). The ‘K’ indicates Kerguelen. ‘RT’ is the Rajmahal Trapps in India, near the border with Bangladesh, dated to 118 million years ago: in the simulations the Rajmahal Trapps, Broken Ridge, and Kerguelen belong to the same LIP, separated near birth. When the SouthEast Indian Ridge formed, 43 million years ago, it disected the Kerguelen Plateau; Broken Ridge broke off and became stranded on the India-Australia plate. (‘DT’ stands for the Deccan Trapps which formed, almost as an afterthought, 65 million years ago.)
Once separated, India moved very rapidly across its namesake ocean, leaving a trail of debris behind. One of those bits of debris is the Ninety East Ridge. In the picture below, it is shown beautifully as the line roughly north-south just above the centre. The top end is about 80 million years old, the bottom (where it connects to Broken Ridge) 43 million years. Re-connecting it to Kerguelen (following the transform faults, roughly southeast-northwest) shows that it came from the northernmost end of Kerguelen plateau — where the Kerguelen Islands are. It seems a typical hotspot trail. However, note how the SouthEast Indian ridge is offset towards Kerguelen in this position, with the hotspot located at the top end of this offset. This suggests an overactive transform fault rather than a hotspot trail. Perhaps it is a combination of both?
How did Gondwana break up? Africa/South America split off from the rest 180 million years ago. India separated from Australia between 150 and 130 million years ago. Next, India separated from Antartica. A sliver of the continent became the Kerguelen micro-continent. Finally, perhaps 100 million years ago, Australia separated from Antartica. India migrated to form a new partnership with Asia, leaving Australia isolated — the island ark it is now.
In the traditional picture, continental break-up is caused by mantle plumes, affecting large regions. India could have required several plumes. One is associated with the Comei-Bunbury LIP (connecting southwest Australia to current Tibet) which was probably erupted within 5 million years, around 132 million years ago. 15 million years later, the Kerguelen plume continued the division. Was this an accidental coincidence, was it the same plume, or was it in some other way caused by the earlier plume? It also coincided approximately with the Ojong Java Nui, the largest LIP in the past 300 million years (it may have covered as much as 1.2% of the Earth), emplaced below the sea 125 million years ago. Was the Earth a tad active at this time, or are we too quick to ascribe all LIPs to mantle plumes?
Although mantle plumes are the more likely, there are other models. In one of these, the thermal insulation by the thick supercontinent causes the mantle underneath it to warm up. The heat lowers the viscosity and can cause sideways convection. Once it reaches the edge of the continental plates it can melt through. This would explain why the breakup starts at the edge, and appears at multiple places in close succession. A warmer upper mantle can also suppress mantle convection underneath the continent, reducing the heat flow. To compensate, the heat flow and convection around the edges of the continent would increase.
In the case of Gondwana, a possibility is that as the Comei-Bunbury plume forced India and Antartica apart, the spreading caused the lithosphere to thin. The reduced pressure, together with heat from the distant mantle plume, allowed de-compression melt, and it was this melt which caused the Kerguelen-Rajmahal LIP, pretending to be a mantle plume.
What happened afterwards is also interesting. Kerguelen remained ‘hot’ and subject to eruptions. A spreading centre developed across it, probably not by accident but due to the fact that the lithosphere was weaker here. Most of Kerguelen ended up on the Antarctic plate, pulling the spreading ridge with it. The resulting transform fault lead to Ninety East Ridge. Whether Kerguelen stayed hot because of residual heat and melt, or that it accidently managed to stay exactly above a remnant hotspot, is a good question. A shallow hotspot (residual heat) could move with the oceanic crust, fixing the relative location. For a deep hotspot this is much harder.
The Kerguelen spreading centre finally let Kerguelen go and moved north east. A new plateau developed on the SouthEast Indian Ridge. It currently breaks the surface in two places, Amsterdam Island and St Paul Island, both volcanic — one of a few places in the world where mid-ocean spreading ridges come above the surface. Under water, a chain of extinct volcanoes points towards the Ninety East Ridge, with the unlikely name of the Chain of Dead Poets. The gravity map shows that this chain continues all the way to the end point of the Ninety East Ridge. Thus, if there is a hotspot, it is not even clear whether it is now underneath Kerguelen or underneath Amsterdam/St Paul!
Is Kerguelen like Iceland? Both sit where continents once split, and both are in fact the last phase of this split (in the case of Iceland, about 50 million years ago, when further south Europe and America had long separated). Both captured a nearby spreading ridge, causing transform faults, and both stayed ‘hot’, not letting the hotspot move away. They even have the same climate. Iceland and Kerguelen are not identical: they differ in age, level of activity, and presence of a continental crust. The Indian Ocean is not the Atlantic. But the similarities are fascinating.
- Kerguelen Plateau
- Ninety_East_Ridge (wikipedia)
- More on Gondwana (De Wit)
- More specialist: