The Self-Similarity Of Land Or Pareidolia?

If you’re British, have good eyesight and are more than about fifty, the chances are you will remember this station ident. Back then of course, nobody knew the words “station ident” even though there were about eighteen of them, mainly for ITV companies, in this country. They’re an interesting topic in themselves, but not one I want to cover today.

What I want to illustrate probably works best with a short video clip:

As humans, we have a predeliction for discerning patterns everywhere, even where those patterns are not significant. This clip used to trigger that in me, but what I’m not clear about is whether they’re important. The black and white image emphasised the contrast for me, with black oceans clearly outlining the white continents, and it seemed to me that the Atlantic Ocean looked like a kind of triangular face looking west, which was also quite similar to how the Indian Ocean west of the Indian subcontinent looked. But of course, this could just be nothing. Maybe younger children are more likely to pick up on patterns compared to adults, but it’s still with me whenever I look at a map or globe of the world.

There are other examples. For instance, Afrika and Madagascar and the Indian subcontinent and Sri Lanka are both roughly triangular landmasses tapering towards the south with a somewhat rhomboid island to their southeast. Corsica and Sardinia are also somewhat similar islands. South America and Afrika, as well as fitting together due to having both been part of Gondwana in prehistoric times, are again both roughly triangular landmasses tapering towards the south. The same patterns repeat over and over again on the map. Another one I noticed recently which probably is spurious is the apparent similarity between the coastline of Europe and that of southwestern Great Britain. In this case, South Wales corresponds to Scandinavia, Land’s End to Iberia, the Lizard to Italy and the Isle of Wight to Crete. I’m pretty sure this one is nothing though, particularly when one considers the proportions.

All that said, one of the sources of fractal mathematics was the “Coastline Problem”. This was based on the realisation that measuring a coastline would vary according to the length of the ruler you were using. Great Britain on the crudest scale is roughly triangular, with a very approximate perimeter of 2 800 kilometres. A “ruler” able to produce a very crude but recognisable map of this island, with a length of about a hundred kilometres, yields a coastline of about 3 500 kilometres. According to the ‘CIA Factbook’, the perimeter is around 12 429 kilometres. At this point, one might find that if a ruler a mile, nautical mile or kilometre in length were to be used, different figures would be arrived at, so whereas it says 12 429 kilometres or 7 723 miles, which is correct as far as converting units is concerned, if the coastline had actually been measured twice, once in miles and the other time in kilometres, the results would not be convertible and the length in kilometres would’ve been greater. The same applies to millimetres, only much more so. One of the results of this is that it’s entirely possible to come up with similar figures for the length of the coastline from Aberystwyth to Hayling Island and Vingsand in Norway to Θεσσαλονικη by judicious choice of the right measuring sticks. Alternatively, even with the same yardstick the lengths could in theory be the same, and this is particularly plausible when comparing a fjord-rich coastline with a particularly smooth one. Also, if one considers the Mandelbrot Set, mini-versions of itself are found all over which are somewhat “morphed”, with larger regions minimised and smaller ones maximised, and this could happen to a coastline, but if that kind of thing’s allowed, it appears that anything could be made to fit.

The processes leading to the formation of the islands of Sri Lanka and Madagascar are entirely different. Madagascar is the result of rifting in the Afrikan continental plate causing it to calve off to the side like an iceberg, though one attached to the ocean bed of course. Sri Lanka has existed since Precambrian times in that position relative to the subcontinent and was joined to the land by an isthmus until recently, and is also in the centre of its plate. Nonetheless they both look superficially similar. Is it possible that there is another factor involved which leads to this kind of similarity?

As far as I can tell, although there are other continental archipelagos such as Indonesia and Japan, none are very similar to the islands I’m currently sitting in. That said, comparisons have been made between Japan and Great Britain in other ways, comparing Hokkaido and Scotland on the one hand and Wales and Shikoku on the other. Kyushu is also compared to the Six Counties on this map, but there isn’t an extra bit of Kyushu to be taken into consideration, and Scotland is no longer a different island to the one with England on it, although it was in the geologically distant past. Moreover, Sakhalin could be thought of as part of Japan geologically although there is no similar large landscape north of Great Britain which fills that rôle. That said, there are many similarities between Japan and Britain. Both have languages which are written non-phonetically under the influence of a powerful continental neighbour, both have a sense of reserve as part of their national character and both have an official state religion. To what extent, though, is this cherry-picking? This is what I’d like to get to the bottom of in all of these.

Probably the largest example of this in Earth’s geography is the fairly minor similarity between Afrika and South America. This is of course helped by the fact that they both used to be joined along an entire coastline before the breakup of Gondwana, but there are other factors. Both are roughly triangular continents straddling the Equator with a bulge in the north and a consequential human-impenetrable region in that bulge. This last point is stretching it a bit because hot deserts and tropical rain forest are opposite ends of the spectrum regarding rainfall, and the picture is complicated by the presence of rain forest in the Congo. There’s a possibly rather fruitless question regarding which bits of Afrika and South America correspond. The Amazon is in a sense the Congo of South America and there is no corresponding huge desert, although there are deserts in southern South America, none of them are that similar to the Kalahari. The Atacama is much drier and Cabo Polonio is a cold desert. The Andes are also very important for South America and there is no corresponding range in Afrika. One thing they do have in common is a thin piece of land linking them to a larger continent to the north. One might expect all of these similarities would have led to similar histories and ecologies, but it isn’t clear that this has happened. For instance, the Sahara Pump, if it happened, led to a distribution of the Afro-Asiatic language family across the Sahara and to the south about to the level of the Horn of Afrika and also into Asia in the form of Arabic, but nothing similar happened in South America. There are anteater-like animals on both continents in the form of anteaters themselves in South America and aardvarks in Afrika, and pangolins and armadillos are somewhat similar, but I don’t think there are any Afrikan “sloths”. Both of them, though, have been subjected to colonialism, although only Afrika lost people to slavery abroad due to imperial powers. Both have Hispanophone, Lusophone, Francophone and Anglophone nations, but South America is dominated by Portuguese and Spanish to a much greater extent. A lot of the differences are to do with South America being situated further to the south than Afrika, although there is of course a big overlap.

It’s conceivably instructive to look at Venus and Mars with flooded lower altitudes to see if the same kind of land forms can be identified. This is Venus:

and this is Mars:

(c) Aaditya Raj Bhattarai

There is a fairly clear problem with comparing either of these with Earth. In fact there are several. Both of these maps are based on the idea that 71% of the surface is covered in water as it is on Earth at the moment. However, this doesn’t mean they’re proportionately similar. That is, given that Mars is half Earth’s diameter, a proportionate amount of water would be about an eighth of ours, but that wouldn’t provide 71% cover, and on Venus, which is slightly smaller than Earth, the cover is provided by far less water because there’s less variation in altitude, possibly due to melting mountains (that’s me, not science). Taking the Martian map first, the planet is fairly neatly divided into highland and lowland regions in the south and north respectively, and again this is not a scientific judgement but I think of Mars as consisting of a single continent plus a single ocean due to its size not allowing for anything more complex. Also, the terrain illustrated depends on the absence of plate tectonics. Tharsis in the west of that map is a complex of huge volcanoes caused by a hot spot which doesn’t move and has caused a build up of a massive plateau heavy enough to have cracked the land to its east, which is the blue channel referred to as Noctis Labyrinthis followed by the great canyon, here below sea level, called Velles Marineris or Mariner Valley. Also clear is the large depression Hellas, visible on the eastern side and possibly the antipodes of Tharsis. Having a thin atmosphere, Mars shows obvious craters in the highland “continent” which wouldn’t be there if it genuinely had large bodies of water and rain eroding its surface. Therefore there isn’t really anything similar to the kind of land shapes found here.

Turning to Venus, at first glance the planet looks much more like Earth. The eastern side of the northern continent even looks rather like Siberia. However, there are many more approximately circular “islands” than there are here and the oceans are very different, being much shallower and lacking the oceanic ridges characteristic of Earth with its tectonic plates and continental drift. There are also a lot of archipelagos including relatively large islands and intermediately-sized masses of land between large islands and small continents.

Both these maps show no sign of water erosion, and there having been no continental drift the land is not like it is on this planet. Water also propels plate tectonics. It might be more informative to turn to Titan, since that alone in the solar system has both land and bodies of liquid on its surface like Earth, and it looks like this:

Like Mars, Titan’s size must be borne in mind. It’s a little smaller than Mars but its composition is very different, having substantial quantities of water ice in its make up. It’s difficult to work the consequences of this out because we’re used to water ice near its melting point, whereas Titanian water ice would be mixed with rock and therefore kind of “muddy”, and also much harder due to being well below the temperature of our South Pole in midwinter. Nonetheless, the solid surface material on Titan is eroded in a similar manner to how water erodes our rocks:

Those are pebbles, possibly of water ice (frozen water) which have been rubbed smooth by liquid ethane and methane. Titan always faces Saturn and takes just over a fortnight to orbit the planet. It’s also a lot colder than Earth and its atmosphere is thicker and therefore carries heat around the moon more effectively. Therefore, even though the surface temperature of Titan is below -180°C, it behaves as if its climate is tropical, bearing in mind also that the boiling point of methane on Titan, with its higher atmospheric pressure, is -155°C. Moreover, although we tend to think of temperature as a linear scale, it often makes more sense to see it as exponentially colder and to scale it relative to the temperatures we’re used to, because absolute zero, -273.15°C, is actually infinitely cold as it can never be reached and it always takes the same amount of energy to halve the temperature. Titan is only 5°C warmer than the melting point of methane but since its surface temperature is around a third of Earth’s, that’s equivalent to our whole planet being at 15°C. However, water is also a highly unusual substance because it expands when it freezes, which is not unique as bismuth and gallium, for example, also do it, but there are unlikely to be any celestial bodies with oceans of bismuth or gallium. Water also turns white when it freezes, which reflects heat and therefore tends to produce a feedback effect, cooling it further. Erosion and weathering from ice are therefore different from erosion would be from frozen methane. For instance, water can seep into rocks, expand on freezing and force rocks apart, and naturally glaciers can carve out U-shaped valleys and fjords. Titan would not have fjords or that kind of erosion. Nor does it have continental drift because that too requires water, meaning that there is more opportunity for erosion to smooth the surface without it being replaced by volcanism, which is also stimulated by continental drift. There are faults on Titan, but they don’t follow continental plates and there may also be volcanoes, but nothing like the “Ring Of Fire” around the Pacific Basin on Earth. Consequently something like the island of Madagascar cleaving away from Afrika won’t happen, although that doesn’t mean there wouldn’t be islands off larger landmasses on such a world.

I don’t know about you, but I find that map of Titan difficult to read. I presume that the darker patches are exposed liquid and the lighter patches land, but I’m not sure. This set of pictures, however, may make things clearer:

Despite what I said, that looks quite fjordy to me. There are also rivers running into it. This is Ligeia, a lake about a third the size of the Caspian and is the second largest body of liquid on the moon. This should be put in perspective in that in terms of proportion of coverage it’s more than four times larger and therefore bigger than the Caspian by scale. The largest body is Kraken Mare:

Kraken Mare has an approximate area of half a million square kilometres, making it almost twice the size of the Caspian in absolute terms but still only a small fraction of the size of the smallest ocean, the Arctic, even in proportion to the size of the moon. The essential difference between Titan and Earth is therefore that whereas Earth has land surrounded by continuous liquid, Titan has liquid surrounded by continuous land, meaning that comparisons of land forms are only meaningful for islands in its lakes and seas, although the shapes of the lakes and rivers are more meaningfully compared between the two worlds. The rivers of Titan seem to be much more subject to tributaries than those of Earth, which may be what gives the moon’s lakes and islands their fjordier appearance. The peninsula jutting out into the lake near the bottom left does have a larger headland to it rather than just being a finger of land, which is superficially like Iberia except that in that case the land used to be an island that collided with the rest of Europe. On the whole, it doesn’t look that familiar though. It also occurs to me that the density and viscosity of a mixture of liquid methane and ethane could be somewhat different, although the molecular weight of methane is very close to that of water. Gravity must also be a factor.

Without another Earth-like world to compare it to, it’s difficult to say what’s happening and what forms are likely or unlikely, but the considerations I’ve had to make here might narrow down the conditions somewhat. They seem to include the following:

  • Relative quantity and depth of liquid.
  • Coverage of surface. I’m guessing that more than fifty percent of the surface must be covered for there to be reliable continents and extensive islands.
  • Plate tectonics. A lot of what’s visible on the surface of this planet is a manifestation of continental drift, which allows land to be rebuilt via volcanism and islands to surface, submerge and so forth.
  • Density of liquid.
  • Weight of liquid. This is not the same as density because it’s related to gravity.
  • Coefficient of expansion. There may or may not be any fjords on worlds whose oceans are not made of water.
  • Difference in density of liquid and solid components of the surface.
  • Hardness of solid components.
  • Thickness of the atmosphere – a thinner atmosphere would lead to more craters which would probably flood, although they would also be eroded quite quickly.

I haven’t been able to come to any conclusions yet about this. I can see that a supercontinent might fracture into roughly triangular continents, that island chains form from both continental drift and volcanic hot spots moving around relative to plates and various other things, but to be honest I’m no closer to being able to decide whether the apparently similar land I saw on the BBC globe in the early 1970s was mere pareidolia or a pattern which exists independently from human perception.

History of the British Climate Part I

Yesterday I covered the last 400 000 years of British climatological history. Today I’m going to do something like the previous æon, and possibly all the way back to the beginning of the world. In fact, yeah I’ll do that.

4 543 million years ago, the future Solar System was a swirling disc of dust and gas orbiting a newborn Sun. Jupiter had already formed and was gradually pulling the particles whose times to orbit were in harmony with its own slightly towards itself, leading to them drifting slightly out of phase with it and clumping into fairly insubstantial rings of matter. I’m not sure how warm the belt which would become us was at the time, but it was probably well below freezing point, because if it hadn’t been, there would have been no grains of water ice. On the other hand, there were also comets, so maybe not, but the fact remains that the Sun was dimmer and weaker back then and there were no greenhouse gases in a position to warm the dust and gas which would become Earth. It took seventy to a hundred million years for it to form, and at the beginning it would’ve been slightly more massive, have no permanent moon and the atmosphere would have been briefly high in hydrogen and helium. Within ten million years of its formation, a Mars-sized body which has been christened Theia hit us and shattered the outside layers of the planet, causing them to go into orbit around us and fall together into the body I call Cynthia and most other English speakers call “the Moon”. Clearly there was no such place as Britain at this point and the entire surface of the planet was molten rock heated by the mechanical energy of compression and collision along with radioactivity. The atmosphere would have been substantially superheated steam. Shortly after being hit by a planet-sized body, the atmosphere would in fact have been vaporised rock. It’s possible to determine the climate of the entire planet at this point, as it was quite uniform, meaning that although it makes no sense to talk of Britain, it does make sense to describe how conditions were everywhere. This eon lasted about 500 million years, and during this period the vaporised rock atmosphere would have condensed and fallen onto the surface as drops of lava. Towards the end of the Hadean, life was present, which seems to imply that there was liquid water in at least some places.

The next period is referred to as the Eoarchean, when the pressure was probably dozens of times higher than it is today, more like the solid surface of Venus than today’s Earth. Temperatures were between 0 and 40°C and there may have been ice ages. To quote ELO, “the weather’s fine but there may be a meteor shower”, because this was the time of the Late Heavy Bombardment, when for 300 million years asteroid collisions and other large meteors would have rained very often from the sky, although this has recently been questioned. The atmosphere was high in methane and carbon dioxide, which being greenhouse gases may have ensured that this planet was warm enough for life to survive on it given that the sun was 30% weaker than it is now.

All of this is rather vague and applies to the whole world. The earliest known British rocks are found in Na h-Eileanan Siar, also known as the Western Isles, and have been dated at 3 000 million years old. It isn’t clear that anywhere can be meaningfully called Britain before that date, and there’s no trace of anything else. It was likely to have been a small piece of the surface of the planet with unclear neighbours. The rock concerned is gneiss, which is a common component of continental shields, which are bits of Earth’s surface that haven’t been affected much by continental drift, such as mountain formation or rifting. It would be a bit excessive to call the rocks in Na h-Eileanan “continental shield” because they’re quite small, the nearest substantial example of one being most of Finland and Sweden, but they are the original and only rocks in that small area of these isles.

Even long after this, the island of Great Britain would have been in several parts, making it difficult to describe the nature of its climate. It means imposing the current situation on the past when it’s actually quite transient on a geological time scale. Also, in some areas, including this one, Charnwood, sedimentary rocks were laid down at the bottom of the sea or ocean and the idea of this being Britain is almost meaningless. It also changes the significance of climate, and as far as being at the bottom of a really deep ocean is concerned, almost irrelevant.

In the Archean, which lasted fifteen hundred million years, the planet was shrouded in methane clouds and there was practically no free oxygen in the atmosphere. The sedimentary rocks surviving which had been exposed to the atmosphere show no glacial erosion, but they do show evidence of rivers and rain. Therefore it did rain. In fact, presumably there was an enormous rainstorm lasting thousands of years at some point in the late Hadean when the oceans were formed due to the atmosphere and surface getting cool enough for the steam to condense out and persist on the surface, but because the pressure was much higher this would have happened long before the surface temperature dropped below 100°C. It is actually possible to measure the surface temperature by looking at the proportion of oxygen-18 in the rocks. There are two stable isotopes of oxygen: 16 and 18. Because oxygen-18 is heavier, molecules containing it vaporise at a slightly higher temperature. Chert, which is a sedimentary flint-like rock, is silica, i.e. silicon dioxide, containing oxygen, and is present in some Archean deposits, making it possible to measure the temperature where it was laid down. This puts the ocean temperature at 70°C, but this is probably wrong because weathering once it was exposed to the atmosphere would influence this. The degree of weathering which occurred was unaffected by land plants, since there weren’t any – there weren’t any plants in fact – and suggests a surface temperature between 18 and 24°C, so semitropical. The fact that there was neither excessive heat nor excessive cold suggests various things about the planet such as the ratio of methane and carbon dioxide, a relatively transparent atmosphere and only limited land surface, so it seems that not only do we only have bits of Na h-Eileanan available but that may have been partly because there just wasn’t that much land.

The Archean was followed by the Proterozoic, which began around 2 500 million years ago. This was characterised by the evolution of blue-green algæ, which proceeded to release oxygen into the atmosphere and removed carbon dioxide. This may also have reduced the activity of methane-producing organisms, another greenhouse gas, and also oxidised the methane. Incidentally, this hedging language I’m using here is down to my ignorance more than scientists’. Anyway, the consequences of this were that iron began to rust in the ocean, depositing itself in bands of rust on the sea bed, and the temperature of the planet fell, triggering an ice age. It’s theorised that this planet has two relatively stable states climatically, which it switches between: icehouse and hothouse. Icehouse has generally not dominated but can do at certain times and in fact it is at the moment, anthropogenic climate change notwithstanding. The dominant state is hothouse, which is generally warmer than today for millions of years at a stretch. Even so, there does seem to have been an ice age in the early Proterozoic, and at the end of the Proterozoic there was another much more severe one. In between those times the world-wide climate would’ve been warmer than today.

The Cryogenian Period was a crucial time in our planet’s history. It appears that the land was mainly equatorial at the start of this period, which would probably have included the bits of land which were to become these isles. We were situated just south of the Equator, in Laurentia and Baltica, as part of the supercontinent Rodinia, meaning a hot, wet climate, except that we were below sea level, so a very wet climate! The oddity about this time is that glaciers are found at the Equator, i.e. the parts of the supercontinent which were equatorial at the time, and it’s thought that this means that most or all of the planet was covered in ice and as cold as Antarctica. My comment about tropical conditions applies to how things were before this arose. There are a couple of hypotheses about how this happened. One is that Earth may have had an axial tilt as high as 60°, meaning that constant night in the winter and the midnight Sun in the summer would’ve applied to everywhere further from the Equator than today’s Brazil or Israel. Very surprisingly, a snowball Earth can only happen if there’s a lot of equatorial land. Most of the Sun’s heat is absorbed near the Equator, meaning that if there’s a lot of land there the heat would not be absorbed as much, and this would cool down the whole planet.

By Ryan Somma – Life in the Ediacaran SeaUploaded by FunkMonk, CC BY-SA 2.0,

The Ediacaran follows the Cryogenian and is for this part of Britain very significant, because it’s from this time, lasting 94 million years from 635 to 541 million years ago, that some of the most famous fossils found in this area date. These can be seen in a local museum and include the feather-like Charnia as seen above, and Bradgatia linfordensis, a lettuce-like organism obviously (to locals) named after Bradgate Park and Newtown Linford, both in Charnwood. Charnodiscus concentricus is another. These are all thought to be “quilted” animals who left no descendants, although some people class them in their own kingdom because they’re quite unlike any animals or plants we’re familiar with. They appeared 600 million years ago and all died out before the Cambrian. They may have had symbiotic algæ in their compartments, meaning that since many of them were also attached to the sea bed, the water must have been sufficiently shallow to allow light to penetrate. Hence Charnwood was still underwater, but the ice must’ve been gone and the water wasn’t particularly deep.

Rodinia was breaking up at this time, so there would’ve been a network of shallow seas, which sounds like the situation as it was here. Rodinia was an unusual supercontinent because it seems to have formed by the landmasses moving all the way round the world and colliding with each other on the opposite side to where they originated, which meant they had a long time to erode and the land surface was quite flat. The network of seas would have increased rainfall on the land, since much more of it would’ve been closer to the sea. This may in fact have been part of what triggered the earlier ice age. The temperature of the Ediacaran was still around 2°C cooler than the average for the Holocene, so it looks like the weather here would’ve been cold, wet and rainy. Plus ça change!

The Cambrian was warmer, around 8°C warmer than the Holocene average, and in fact this set a precedent for the generally warmer temperatures of the Phanerozoic, our current eon. During the next period, the Ordivician, sea levels rose by a hardly believable six hundred metres. This ended as a new supercontinent, Gondwana, reached the South Pole and a new ice age started, lasting twenty million years. A gamma ray burst may then have cause the mass extinction at the end of the period, meaning that it may have rained concentrated nitric acid.

Around 400 million years ago, three mini-continents collided to form the British Isles as we know them today, and it begins to become more meaningful to talk about British climate. These were Laurentia, which is effectively all of Scotland, Avalonia, which is England and Wales, and Armorica, which is Brittany, Devon and Cornwall plus a lot of other land such as Iberia. Glen Mòr, the fault along which Loch Ness is situated, continues into Ireland and therefore I imagine Ireland was also in two halves before this. Avalonia began as a volcanic island chain north of Gondwana. Britain was about 30° south of the Equator then. It drifted gradually north, crossing the Equator about 300 million years ago, and over this time other land collided with the forming Pangæa, meaning that it was increasingly far from the sea. This is about the time the Carboniferous started and the future Britain became covered in the rainforests which would become the coal measures, so Britain was hot and swampy, and the oxygen content of the air was so high that lightning strikes would have ignited wet vegetation, so there would be many forest fires even though conditions were damp. Around 305 million years ago, climate got cooler and drier and sea level fell, leading to retreat of coal forests from higher ground and the emergence of fragmented rain forests, which were no longer able to maintain their genetic diversity and there was a lot of inbreeding, shrinking of the size of, for example, horsetails, to cope with the conditions and a new ice age started in the Southern Hemisphere, although not severe enough to make Britain cold.

By this time, Pangæa was forming, as were the Pennines. Hot dry desert conditions took over from rainforest, with presumably an intermediate phase which today would be like the Serengeti, although with very different flora and fauna the details are not obvious. The late Permian was a peculiar time climatically, as the interior of Pangæa seemed to have extreme temperature variations so that it was both very hot and very cold at different times of year, and it’s been suggested that this was a cause of the Great Dying, where almost all life on Earth became extinct. Britain was now in the northern tropics, and as such was in the same zone as the Sahara is now. The Scottish Highlands at the time would’ve been as high as the Himalayas and formed part of a range which extended southwest into the Little Atlas and Appalachians. There might also have been a rain shadow desert to the east, making it even drier than it would’ve been without them, but the monsoon conditions which prevailed to the southeast might make it heavily forested.

In the Triassic there were salt flats in Cheshire, hence the salt mines which existed there in historical times, and red sandstone forming in what is now the Southwest, hence the very red soils in that area. Towards the end of the Triassic, the sea level began to rise again, converting much of the isles into a subtropical shallow sea and many of the hills and mountains as they existed then into islands, such as the Mendips.

The following photo is taken from this website and will be removed on request:

This is the “Barrow Kipper”, or rather a monument to where it was found in 1851. Barrow-upon-Soar is about an hour’s walk from where I’m sitting and between 200 and 150 million years ago was underwater, over the entire Jurassic Period. This particular plesiosaur was formerly classed as a Rhomaleousaurus but now as an Atychodracon, from the Early Jurassic, looking something like this but with a bigger head:

It used to be thought that plesiosaurs had to climb ashore to lay their eggs, so this suggests that there was land nearby, but fossils have since been found of pregnant ones, and their limbs were arranged in such a way that they would’ve had to have dragged themselves along the shore quite roughly. However, although it isn’t from precisely the same time, a few miles away in Rutland, the largest and most complete dinosaur fossil ever found in Britain was unearthed, a Cetiosaurus, like a mini-“Brontosaurus”, suggesting that this area was an archipelago of smaller islands or just near a beach. There is a famous traditional song called ‘Ashby De La Zouch By The Sea’, which has often made me wonder whether that particular nearby Leicestershire village ever was.

I am of course a Southerner, and as such Leicestershire will always be slightly foreign to me. My mother is from Maidstone, a place sufficiently famous for its Iguanodon finding that the animal is actually on their coat of arms:

These dinosaurs, dating from 157 million years ago, are also found, along with very many others, on the Isle Of Wight. It’s tempting to telescope all these findings into an imaginary scenario where they’re all simultaneous just because they’re all Jurassic, but in reality the Jurassic Period lasted fifty-six million years, almost as long as the time since the non-avian dinosaurs became extinct, and the Isle Of Wight dinosaurs are mainly early Cretaceous. There were, however, coral reefs in Yorkshire. In the Cretaceous, the situation was once again one of rising sea level with lagoons and streams. To the extent that these isles existed at that point, they were substantially united. That is, Ireland and Great Britain formed a single island, which was intermittently joined to the mainland and still steadily drifting north.

The Late Cretaceous climate was warmer than today’s at the same latitude, which was about the same as Madrid and Rome, although it had been cooling for millions of years. When the Chicxulub Impactor hit, the widespread fires would have raised carbon dioxide levels tenfold and caused a greenhouse effect heating the planet by 7.5°C. In the Palæocene the climate was slightly cooler and drier due to dust in the atmosphere reflecting heat into space, but tropical forests then developed all over the world, even in the Arctic, where the water was lukewarm. The Eocene would’ve involved warm swamps in many parts of Britain.

At this point I’ll repeat something I said a few days ago about Europe. Europe over the Cenozoic, that is, since the extinction of the non-avian dinosaurs, has been gradually transitioning from an archipelago to a large peninsula, and the scattered islands of the region have shown a trend of joining together to build a subcontinent, for want of a better word. Looking at Great Britain and Ireland in this way, they are late developers, or outliers which show how the rest of the region used to be. There’s a common, and correct, idea that before the end of the last Ice Age and for several thousand years after that, Ireland and Great Britain formed a peninsula, and this is true, but there has been a kind of seesawing appearance and disappearance of sea around us and the level of the land at the moment has been pushed down by the recent weight of ice and is gradually springing back up. Hence it does make sense to speak of the British Isles, or perhaps an island comprising Ireland and Great Britain plus low-lying land in between, in the earlier Cenozoic, and moreover to see them as the westernmost members of a collection of islands a bit like the Caribbean or Indonesia in arrangement, although that may be a bit of an exaggeration. The North European Plain, though, was underwater for quite some time, Iberia ceased to be an island around the start of the Cenozoic and the Italian-Illyrian region was also separate for a long interval.

In the Neogene, Britain arrived in its present position and is no longer drifting north. Hence the climate began to approach how it is today although it would’ve been somewhat warmer still. Finally, the Pliocene saw a general drying out and the Pleistocene brings me to the start of yesterday’s post.

I can’t completely guarantee that all of this is accurate as I know a little, but some of it is disputed and I’m probably in the Dunning-Kruger trough at this point where I haven’t reached the point of realising how little I really know and how wrong I’m actually being. Nonetheless, it’s nice to imagine how our climate could’ve been more Mediterranean or Caribbean in particular in the geological past, and also, wouldn’t it be nice to holiday at home but do it using a time machine so we could get to the really sunny and warm climates which this part of the world, so to speak, used to experience?

Extinct Oceans

This is a map of the world as it was at the start of the Late Cretaceous Period. Although the Cretaceous is primarily associated with being the last period of non-avian dinosaur dominance, and eventually the arrival of the Chicxulub Impactor near Yucatan, it has another distinctive feature which is less often mentioned. The name “Cretaceous” derives from the Latin creta, meaning “chalk”, because it’s associated with extensive chalk and other limestone deposits. This is because foraminifera, protozoa with calcium carbonate shells, fell constantly onto the ocean floor, laying down what would become chalk, and, not to let foraminifera alone get the spotlight, there were also radiolaria, similar protozoa with silica exoskeletons, raining down and becoming flint. Neither set of organisms was new to the Cretaceous, but they’re particularly widespread in Cretaceous sediments because so much of what is now land was underwater at that time. Oceans covered a maximum of eighty-two percent of the planet’s surface during this period, meaning that there was only sixty-two percent as much land as there is today. Another distinctive feature of this time, which partly follows from that, is that much more of the water was in the form of shallow seas, all the light blue on the above map. Today the shallow seas are quite fragmented and not particularly extensive. They consist of the waters around Britain, the Red Sea, the Caribbean, the Persian Gulf, Indonesia, Hudson Bay, eastern Eurasia, some of the Med and not much else, although the Arctic is rather shallow but atypical due to being icy. Today’s shallow seas are also quite new, because many of them have only been flooded since the end of the last Ice Age or more recently, and being disconnected from each other don’t form particularly large biomes. By contrast, in the Jurassic and Cretaceous, particularly later on, something like an eighth of the water-covered surface of our planet was in this condition and much of it was also interconnected. This is for two reasons. One is that there were no permanent ice caps, so sea level was at least ninety metres higher – actually higher than that because of the thermal expansion of the warmer water. The other was that much of the land which is now mountain ranges had not only not yet become those ranges but hadn’t even got above sea level yet, although it was on its way to doing so. From the map, it can be seen that there are no Alps or Himalayas, but the surface which became those is not far underwater and has yet to be folded, therefore occupying a larger area. Besides that, in particular the future Sahara is submarine, the future Europe consists mainly of islands and the sea and the area east of the Rockies in North America is a sea referred to as the Western Interior Seaway, all of which is shallow. Since that’s quite an awkward name, it’s also referred to as the Niobraran Sea, which is how I’m going to call it.

With the mention of the Niobraran Sea, the question might arise of what distinguishes a sea from an ocean. The Niobraran had a maximum area of around three million square kilometres, which is a sixth larger than the Mediterranean. The Med is clearly bordered today by the Straits of Gibraltar and is easy to think of as a separate, smaller body of water than an ocean, but the Niobraran was more like the German Ocean (“North Sea”) in that it was open to the wider ocean, in its case at both ends. Seas are usually thought of as nearer land and smaller than oceans, and the Niobraran, big though it was, was still less than a quarter the area of today’s Arctic Ocean. Their proximity to land would generally also make them shallower but not always. In a way, in talking about prehistoric seas and oceans at all we’re imposing our own ideas on a world which was very different. Although the distinction we make today attempts to project an order which is more blurred in reality, that doesn’t mean that oceans and seas are not natural kinds. Making this assumption though, the underwater world of the Cretaceous is still geographically very different to our own because rather than consisting of much smaller shallow bodies of water and larger deeper bodies which could be called seas and oceans, its world includes vast seas as well as vast oceans. That said, the Pacific was still by far the largest ocean, and was in fact larger than it is now because the Atlantic was just forming and the continents were closer together.

All that said, there is on this map a body of water which would have made it possible to circumnavigate the globe without leaving the tropics. The future Isthmus of Panama is absent, the Mediterranean has yet to form, India is still near Antarctica and so is Australia, and then there’s the Pacific, which is already in existence. This is what is variously known as the Tethys Ocean and Tethys Sea, and at this point I will pause to consider the naming of ancient seas and oceans.

The Atlantic is named after the Atlas Mountains, which are in turn named after the Titan Atlas, who holds up the sky according to Greek mythology. The word “atlas” also refers to the first cervical vertebra, and the joint between the first and second, the axis, is called the atlanto-axial joint, indicating the unusual way in which the word “atlas” is inflected. Atlantis is also named after Atlas of course. Several of lost oceans are also named after titans, by analogy with Atlas. To an astronomically-tuned ear, this sounds rather odd because so are many of the moons of Saturn, but in any case they are the Tethys, after Τεθυς, goddess of the primal font providing Earth’s water, the Iapetus, named after  Ἰαπετός, the father of Atlas because the Iapetus is the forerunner of the Atlantic Ocean, and the Rheic, after Ῥεία, sister of  Ἰαπετός.

I’ll start with the Tethys. The Tethys began as a gulf on the eastern side of Pangæa in the Permian period:

(This is actually a map of the early Triassic world). The Tethys Sea (at this point) is the light blue area just northeast of the centre. It’s likely that because of the extreme climatic conditions on Pangæa, temperatures were high enough to render the Paleo-Tethys, as it’s called at this stage, quite saline due to evaporation. The existence of a single continent more or less implies the existence of a single ocean, the World Ocean, which is referred to as Panthalassa, which would have covered around seventy-one percent of Earth, or 510 million square kilometres. This is thrice the size of the Pacific, and it included the Pacific too, since the Pacific is that bit of the planet’s surface which has its own continental plate but no continent. Just as the Solar System has been said to consist of the Sun, Jupiter and assorted débris, it could equally be said that the surface of the Earth consists of the Pacific Ocean plus a bit of land and some other oceans, and this was already true back then. The Panthalassa, also known as the Panthalassic or Panthalassian Ocean, formed 280 million years ago at the end of the Carboniferous. Sea levels would’ve been fairly low at the time because there was an ice age going on in the Southern Hemisphere. Here’s a map of the world centred on Panthalassa with modern coastlines:

By Fama Clamosa – Own work, CC BY-SA 4.0,

The borders show the edges of the continental plates, which are however devoid of land masses here in three cases. The western part of the ocean has large islands at this point. The fact that it occupies more than an entire hemisphere and is on both sides of the Equator gives it a fairly simple set of ocean currents, one northern and one southern, but it also has a temperature gradient and is coolest in the east and warmest in the west, with cold water upwelling along the east coast (i.e. the west coast of Pangæa). Panthalassa is an example of a superocean, as it surrounds a supercontinent, although it can also be defined as any ocean larger than the Pacific. Mirovoi was the ocean surrounding Rodinia, which existed between an æon and 750 million years ago:

Mirovia existed during the extreme Cryogenian Ice Age, and may have been frozen to a depth of two kilometres, although only shortly before it ceased to be.

In general, superoceans are distinguished by extremely large waves with considerable destructive power when they finally break on land and tend to be wamer in the west than the east because of the Sun warming them in a fairly simple way. They may also have had powerful monsoons. Another known superocean was the Pan-African, which surrounded the supercontinent Pannotia 600 million years ago, around the time the first animals with hard parts evolved:

10 April 2012
Own work
Kelvin Ma

This superocean closed before the start of the Cambrian and was replaced by what would become Panthalassa. It can be seen from this globe that there is no plant life visible on land yet. A new superocean and supercontinent are due to form in about 200 million years.

Returning to the subject of the Tethys, this has an established detailed history which came to an end only fairly recently in geological terms. Eurasia today still shows traces of where it used to be. The Mediterranean and Black Seas, for example, were both part of it. Just a little further east lies the Caspian, which unusually for a lake, possibly even uniquely, has oceanic rather than continental crust under it, because it was once part of that ocean. Further east still are the dying Sea of Aral and Lake Baikal, the world’s largest body of freshwater though not its largest freshwater lake by area, which is Lake Superior in North America. All of these used to be joined together and connected to the sea, and were cut off by the ongoing collision of Afrika with western Eurasia and of India with southern Asia. Before that, it was the Ceno-Tethys, the last phase of that ocean.

Supercontinents break up because magma wells up beneath them and pushes the plates apart. One example of this which happened on the southern coast of the Tethys Sea when it was little more than a giant bay was the emergence of the continent of Cimmeria, a long, thin continental mass which was often broken up into islands. This moved north and collided with part of the future Asia, and this is taken as the division between the Palæo-Tethys and the Neo-Tethys, meaning that for a while they coëxisted. This continent has now become a series of mountain ranges including, among others, the Pyrenees, Alps, Caucasus and Himalayas. For some time during the history of the Neo-Tethys, Europe, known at this stage as the Para-Tethys, had yet to become a continent and consisted of an archipelago, including the famous Haţeg Island with its giant pterodactyls feasting on pygmy sauropod dinosaurs, though that was gone by the middle of the Palæogene Period after the Chicxulub Impact. The dinosaurs and pterosaurs were gone a lot sooner than that of course. The Tethys can also be divided into eastern and western halves, separated by a narrower channel between Laurasia and Gondwanaland:

By Lennart Kudling – This file has been extracted from another file: Laurasia-Gondwana.png, CC BY 3.0,

The thin isthmus between Afrika and Laurasia in the middle of that map would part, forming a seaway stretching right round the tropics. This would have had a longitudinal current flowing round much of the world. The Tethys was also full of reefs, formed not from coral but rudist bivalves, who were vertically oriented conical pipe-shaped animals who died out at the end of the Cretaceous. It probably hasn’t escaped your attention that I’m rather fixated on the Tethys. I don’t really know why, but I like to imagine a kind of vastly extended Mediterranean with its characteristic human culture, which is utterly absurd because humans didn’t evolve until after it ceased to exist. The Tethys would also have rendered the Suez and Panama Canals redundant. This would’ve been a world where it was possible to sail all the way from Southeast Asia to Columbia without having to venture far from shore. In a way, although we are creatures of our time, it sometimes seems a shame we hadn’t evolved when this was feasible. On the other hand, knowing us it would also have facilitated a lot of naval battles.

The Iapetus Ocean is another that tends to spring to mind. In a way, the Iapetus was like the Arctic in that it was a polar ocean, but in the Southern Hemisphere and not as closed in by land. Iapetus was the father of Atlas in Greek mythology, and although a lot has happened since then to scramble the jigsaw pieces, the lands bordering it are similar to those bordering the Atlantic today, hence the name. This is the “father” of the Atlantic. The ocean was detected when it was noted that bottom-living trilobites in Scotland and parts of America were similar but those in Southern England were different, suggesting that they had been separated by something, which at the time was thought to be a trench. The ocean existed from around the time of the first trilobites until 420 million years ago, and mixing of the trilobite fossils increases as time goes by and the ocean closes to form Laurasia, a component of the later Pangæa.

Then there’s the Rheic Ocean, once again an ocean named after a Titan. This was the water between the two major components of Pangæa before the supercontinent was formed, namely Laurasia in the north and Gondwanaland in the south. It formed in the late Cambrian, reached its maximum size of four thousand kilometres across in the Silurian and closed as Pangæa was forming. The Rheic was probably gone by the the enormous injection of carbon dioxide into the atmosphere from volcanoes in Siberia acidified the oceans and killed 96% of all life underwater and seventy percent above it. It closed in the late Devonian and Early Carboniferous, some time before the growth of the vast rain forests, when South and North America met for the first time, though they would separate again later.

By that time, the shallow Ural Ocean had formed where the future Urals would be, separating Scandinavia, then part of the same continent as the eastern part of North America, and Siberia. This was preceded by the nearby Khanty Ocean.

Other named oceans include Medicine Hat, Mongol-Okhotsk (between Siberia and China), Bridge River, Cache Creek and a number of oceans which existed in the Proterozoic, the era before the Palæozoic, such as the Poseidon and Pharusian. I find it difficult to get excited about these though, because at the time there would probably have been nothing swimming about or growing in them that could be seen without the aid of a microscope. Important stuff was going on, to be sure, but that far back Earth was like an alien planet, with an unbreathable atmosphere, irradiated surface and a day lasting around eighteen hours. Earth would only recently have come out of the period during which it was being pelted with meteorites 18/9.

Essentially the process of ocean construction and destruction is that they form as continents move apart, reach a maximum width, then become shallower, fill up with islands and turn into mountain ranges as the lands around them collide with each other. The constant seems to be the Pacific, and I’m no geologist but it seems to me it’s been around for a very long time. It used to be thought that Cynthia (the “Moon”) had been ripped out of the Pacific when it first formed. I doubt this is still believed, but since it seems to have been around for so long I’m not convinced that isn’t what happened. However, the further back in time you go, the vaguer the history of this planet becomes, because unlike Venus and Mars this is an unquiet planet constantly remodelling its surface over a time period of millions of years. In future, oceans will continue to appear and disappear, but within an æon or so this will cease as the Sun heats up and evaporates them. I presume this will leave a planet with vast low-lying plains separated by plateaux where the continents used to be, a little like Venus but still with the traces of the influence of water and a dense atmosphere of superheated steam. This is all rather sad, but it won’t happen for many hundreds of millions of years, so we’d better make the most of the oceans while they’re still around.