Martin Gibling on Rivers in the Geological Record - part 1
Transcript
Oliver Strimpel
This is Geology Bites with Oliver Strimpel. Rivers can seem very ephemeral, often changing course or drying up entirely. Yet some rivers have persisted for tens or even hundreds of millions of years, even testifying to the breakup of Pangea, the most recent supercontinent, about 200 million years ago. On the one hand, their courses may be determined by tectonic processes, such as the formation of mountain belts. And on the other, they themselves can affect tectonic processes by creating continent-scale features such as giant submarine fans. Martin Gibling has spent a lifetime studying rivers and river sediments around the world. He is emeritus professor in the Department of Earth and Environmental Sciences at Dalhousie University in Halifax, Nova Scotia, in Canada. This episode is the first of two of our conversation about rivers. In this episode we talk about fluvial deposits in the geological record and the impact of the breakup of Pangea on river systems. In the second episode, we talk specifically about the history of rivers of Europe and the Americas, as well as the impact of recent Ice Ages. We end by considering how humans have changed rivers and their deposits throughout human history. Martin Gibling, welcome to Geology Bites.
Martin Gibling
Thanks very much, Oliver. It's a great pleasure to talk with you.
Oliver Strimpel
How far back into the geological record do fluvial deposits go?
Martin Gibling
Go back a surprisingly long way. The oldest fluvial deposits about which we are really certain are in the Moodies Group of southern Africa, and those date back to about 3.2 billion years, and that's a long way back towards the origin of the Earth at 4.6 billion. They may well be older fluvial rocks, but up to now, they've been too altered and deformed for anyone to be certain about. However, it's worth saying that in rocks of Western Australia, researchers have found zircon grains that date back to about 4.4 billion years. That takes us shortly after the formation of the Earth, and, very surprisingly, they yield isotopic evidence for surface temperatures that were very much like those of today. That implies liquid water on the surface of the Earth. It in turn implies the presence of flowing water and rivers. So we have to assume that rivers have been with us since the very earliest days of planet Earth.
Oliver Strimpel
OK, but with those zircons, that's quite a complicated chain of inferences. How do we actually distinguish an intact fluvial deposit in the geological record from other waterborne deposits, such as those laid down on lake beds or on the seashore, and those deposited by wind?
Martin Gibling
Yes, you have to get into the details in the rock record in the strata. And there we're looking at the organization or architecture, as it's called, of the strata, as well as particular sedimentary features and fossils that are present. And, of course, if you're far back in time, you won't have the fossils to guide you. And if you've got very deformed or altered rocks, then you're struggling to make sense of the sedimentary features. But what you can often see in younger strata, is places where river channels have eroded down into the strata below, and you get surfaces that look very much like channel forms filled in with sand and gravel. What's filling them in is often cross beds laid down on the avalanche faces of ripples and dunes, and those allow you to measure the flow directions at the time the water was flowing, and in the case of river channels, they are unidirectional. They're all going in one direction down the channel, which is strong evidence that these were fluvial. And then you can find things like rain prints and mud cracks in the strata which indicate that these were exposures on the terrestrial surface, and then in younger beds you get terrestrial fossils, roots, woody material, freshwater fish, shells and things like trackways made by giant arthropods, and also by dinosaurs, of course. So all these lines of evidence can make you pretty certain that you're dealing with fluvial deposits.
Oliver Strimpel
Wouldn't most of those also possibly appear in wind-borne deposits? Or is it that it's unidirectional with rivers so that rules out wind because it can blow this way and that.
Martin Gibling
It can, but you can of course get systematic directions of wind flow that would generate unidirectional cross strata, but in those cases you wouldn't be seeing the channel bases, sometimes with gravel lining the surfaces as a kind of a lag where the channels eroded in, and that's where the architecture of the strata, and their organization come into play. You would also get mud layers in river channels that you wouldn't be expecting to get in dune systems, for example.
Oliver Strimpel
In that previous episode of Geology Bites, Neil Davies described the impact on the landscape of the first land plants during the mid Ordovician to the mid Devonian. Can you remind us as to how fluvial deposits changed when plants first colonized the land?
Martin Gibling
Yes, Neil has made a wonderful job of setting the stage for this. What happened when the earth was greened — the great Green Landscape revolution? As I think of it, the earliest land plants in the Ordovician are represented by spores, microscopic fragments of plants, and they came from very small plants, with simple, unbranched stems and minimal rooting structures. As we move on through the Silurian, they start to become a bit larger, complex enough that you get a bit of a macro fossil record, and then when you hit the Devonian, things really take off. Large rooting systems evolved forests by the mid Devonian and enough biomass for it to get preserved in the fluvial record, along with things like peat, which indicate that vegetation is really beginning to build up solidly. In the case of rivers, this led to a profound and dramatic increase in the amount of mud that was preserved in fluvial deposits. And Neil and his colleagues have very nicely documented this very, very dramatic trend. What the mud implies is the weathering was intensifying, almost certainly due to the organic acids, and the carbonic acids associated with plants and with the drawdown of CO2 from the atmosphere and groundwater systems. And when you had mud being produced in greater abundance, you also had more of it transported in river systems and then trapped and baffled by the standing vegetation. And so it tends to build up in the river plains and become preserved, rather than just bypassing through to estuary’s motions. The mud would have had the effect of making river sand more cohesive, strengthening the bed of the rivers, strengthening particular the river banks and they would have become more difficult to erode by flowing water. So you start to get more obvious channel banks and channel forms in the preserved record through the later Silurian and Devonian.
Oliver Strimpel
Are you suggesting that by the mid Devonian, the modern river had formed and that all the different plants that would have had an impact on rivers. How that was all sort of pretty much the way it is today, or — have there been further diversification and have certain regions of the planet been colonized that weren't really fully colonized by the mid Devonian?
Martin Gibling
The mid Devonian was when things really began to take off with a sense of very large forests. Trees that were a metre or even a couple of metres in diameter with very large root systems, and there have been recent discoveries in Upstate New York at Gilboa and elsewhere that have showed, very beautifully, some of these major forest systems. But that was only really the start, and by the Carboniferous, you're starting to get early gymnosperm plants, the ancestors of conifers, of the present day, and those were able to penetrate quite deeply down even into dry land areas. And so you start to stabilize not just the wetland environments, but also relatively dry open plains. And what's happening is that rivers diversify through to about the mid part of the Carboniferous, say the start of the Pennsylvanian, when they are more or less covering the range of things that we see at the present day. So, initially, we had large sand bed rivers, water, typically braided rivers with shifting sandbars and rather weak banks. And those progressively move into being more meandering rivers with muddy banks, where the meanders migrate much more slowly. You start to get abandoned channels. You start to get multi-channel anastomosing river systems, wetlands with black organic stained water. And the whole range of what we now know as river systems today is coming into existence through that period of time, especially with the riparian or bank vegetation that makes the whole system so much more intricate and complex, and where life organisms simply began to evolve at a much more rapid rate in these little niches along the lines of the river corridors.
Oliver Strimpel
Was there any significant impact on the fluvial record when the planet started to be covered by large tracts of tundra and open grassland?
Martin Gibling
This was a major change and people are still trying to work out how much that affected river systems. You have the angiosperm plants, the flowering plants coming in in the Cretaceous. They are typically greedy of water in many cases, and then you have the grasses coming in in the Cenozoic, with their very dense root systems. And the way they bound the floodplain surfaces must have had quite a lot of effect on river systems. It's just a bit difficult to figure out how exactly they affected things, but along with them, you have the rise of gigantic organisms like the dinosaurs and the big herds of ungulates that you might see in the Serengeti at the present day, and the buffalo before they were pretty much exterminated in North America, and those would have trampled through riverbanks and in places like hippos do today. They would have made trails down to rivers, which would have allowed rivers to break out along low points at times, and people have even made a case for Dino Evulsion, so-called, where dinosaurs might have trampled their way through to the rivers and promoted river breakout.
Oliver Strimpel
Wow, that's an incredible image of a dino or a hippo affecting the course of an entire river.
Martin Gibling
Yes. It's quite remarkable, and you mentioned the tundra areas and the rise of permafrost greatly affected Northern River systems eventually in the ice ages and those, of course, are starting to melt away somewhat at the present day, and you're starting to see modifications of rivers as the thermal water begins to erode down into the permafrost. There were complex effects related to the glaciation.
Oliver Strimpel
In your book entitled ‘River Planet,’ you have a wonderful chart showing the longevity of modern rivers. What exactly do you mean when you talk about the age of a river?
Martin Gibling
That's a very good question. Rainwater is always flowing downhill, so in reality the flow paths of water are always evolving, following tiny gullies, little rills moving together into larger channels, and trying to figure out the age of a river is really a bit of a chimera. It's like trying to say how old a rainbow is, for example, but it's helpful to think of the age of a river as the length of time that the river channel or course has been following pretty much the line that it has at present… Wouldn't be exactly the same course, because channels migrate, they break out, but the channel belt as a whole would be occupying within a matter of a few kilometres, probably about the same position. That can be very difficult to determine. You might think it would be obvious, but sometimes rivers cease to exist for a while. Maybe the river plane gets drowned out by the sea or lake, glaciers can overrun it. Lava flows, sand dunes can invade the area. The river may cease to exist for a while, but it then may reestablish itself in the same course, and a good example of this is the Columbia River in the western US. Probably originated about 17 million years ago in the Miocene, as far as anybody can work it out. But, periodically, the river was overrun by lavas and by pyroclastic flows coming off the Columbia Basalt Plateau in the East. And the Cascade Range of volcanoes in the West. And it seems that the river simply ceased to exist in its organized form for periods of time and then it reforged a new course, more or less where it's been before. So, it gets to be a little debatable whether those kind of gaps in the record matter or not, whether you reset the clock if for a little while the river ceases to exist. But in broad terms, I think we could say that with minor gaps there are rivers that have been present for hundreds of millions of years in their courses.
Oliver Strimpel
That makes sense. So how old are some of the oldest rivers, broadly speaking, that still run in their ancestral courses?
Martin Gibling
The oldest rivers that anybody has really documented running in ancestral courses go back to the Carboniferous about 300 million years ago, by no means back to the beginning of the Earth. The Earth is an extremely dynamic system, and rivers have been destroyed in the way that almost anything else has on Earth over geological time. But some of the world's oldest rivers are those of Eastern Europe, the Volga, the Don, the Dnieper. They may be as old as the Carboniferous, and they run down into the Black and Caspian Seas, which are the last remnants of the Tethys Ocean. Tethys got squeezed out when Africa came up against Europe. Arabia likewise, and then India against the southern part of Asia. And Tethys was eliminated, except for these minor oceanic remnants. And in that part of Eastern Europe, the deformation has been South of the Black and Caspian Seas in Turkey and the Caucasus. And so the rivers further north have not yet been disrupted by the rise of mountains to the South that affect them. And so in the course of time, they may very well reverse their directions to flow North as other rivers have done, but they haven't done that yet, so they're still in ancestral course, quite remarkably or broadly within them. And if you want an example in North America, there were rivers running along the line of the modern Mississippi west of the Appalachians back into the Carboniferous, about 300 million years ago, and this is a case where the Mississippi for a while in the Triassic and Jurassic ceased to be in its present position due to fault activity and marine flooding. But it's been running along its present line down to the Gulf of Mexico since the Jurassic.
Oliver Strimpel
A lot of these courses take us back to the time we had the supercontinent Pangea. Wouldn't the subsequent breakup of Pangea have completely altered the drainage patterns and therefore the causes of the rivers?
Martin Gibling
Yes, certainly so. The breakup of Pangea into smaller continents that drifted apart had profound effects on virtually all of the world's rivers. When Pangea was in existence, there were rivers of immense length, some of them probably longer than any in existence today. They would have flowed out of mountain chains, like the Appalachians in the Eastern US and the Variscan mountain chains in Europe, and those were equivalent in scale to the modern Andes. So we have to envisage rivers probably on the scale of the Amazon at times, running out across the adjoining plains. And in the case of the southern continents, Gondwana land. There's evidence from age dates, from again zircon grains, that there were very large rivers that flowed out of Antarctica, which was not at that time under ice, and they flowed out across Australia into the Sydney Basin, for example, on the New South Wales coast and then out into the eastern part of India. And these continents, of course, are now thousands of kilometres apart. But at that time, they were part of single, very large river systems that ran through. But once Pangea began to break up, all of those systems were dismembered and began to change.
Oliver Strimpel
Fascinating. So can we track a particular river that used to be contiguous, that now has been separated by thousands of miles of ocean? I mean, do we look at fossils or is it all the detrital zircons? How can we do that matching or is that even possible?
Martin Gibling
It's very difficult, and there aren't many cases where anybody has made a very precise match. These are not, for the most part, rivers that exist at the present day, so they're not rivers with names, in other words, that we apply to actual flowing water. In the case of the Sydney Basin in Australia, it's the grains that give you the age signature of an eroding upland in Antarctica. And the Sydney Basin has the Triassic rocks of the Hawkesbury Sandstone, a very, very big Brahmaputra scale, or nearly that scale of river system. It's flowing in across a fold belt out into what was the ocean at that time, and it has been matched to the Highlands in Antarctica. But it's very difficult to actually track it all the way back there. There are gaps. There are erosion, and so on and so forth. So it's these kinds of lines of evidence bits circumstantial that really build those stories. But I think some of those are the most fascinating stories, that you could pick up a river in one continent and potentially find its continuation in another continent, 5000 kilometres away. I think there's going to be a lot more coming through on that kind of thing in the future.
Oliver Strimpel
Yeah, it's amazing and can you go even further? You said earlier that fluvial deposits were characterized by being unidirectional, so can you infer the direction of current flow and say, OK, well, upstream there, there must have been a mountain belt at that time.
Martin Gibling
Yes, a lot of this comes from where the fluvial deposits are located and their flow directions. What people refer to as paleocurrent flow, the flow directions of the ancient currents. And if you go in the field, you can measure tens or hundreds of cross-bed directions and figure out which way the flows were going, and you can begin to build up patterns regionally of flow directions.
Oliver Strimpel
Did any major rivers actually survive the breakup of Pangea? Or do we have a whole fresh lot of rivers after that happened?
Martin Gibling
There were quite a lot of rivers that are relics. There are places on some of the plateau lands in Africa and in parts of Asia are up on the Tibetan Plateau where there are remnant river patterns. People have mapped them. They appear on satellite images, and they've been inferred to represent remnants of the original Pangean drainage. But they're not now connected to active systems. They've been broken through, and so they seem to be virtually all that we've got in the way of the original drainage, except for those rivers we talked about, particularly like those of Eastern Europe, which haven't yet had the effects of mountain building from collisions.
Oliver Strimpel
Were some rivers born as a direct result of the breakup of Pangea.?
Martin Gibling
Yes, they were. When you have a supercontinent like Pangea, heat begins to accumulate underneath it, and you start to get doming mantle plumes, volcanic activity that begins to generate domes, rifts of various sorts, and these began to affect virtually all the continents in different ways. You have things like the Karoo plume and the Parana plume across southern Africa and South America. These domed-up the crust. And there's evidence from Paleo flow directions that the rivers were radiating in places from those very large domes and then, in particular, once the continents began to separate, new rifts formed, which divided them, widening into oceans, and because the rifts were of relatively low elevation, they drew in the river systems and generated entire new drainage networks, and this is the case with the Niger. Also, with the Rhine very recently, and in the case of the big rivers of peninsular India, many of which occupy Cretaceous rifts. And you would think that then, once a continent is drifting away, nothing much would happen. But in point of fact, continents are drifting over thousands of kilometres of mantle, which is not uniform. It varies in its density, in its temperature, to other properties, and as the continents move slowly across at a few centimetres a year, they rise, they fall. They tilt up and over anomalies in the mantle below, and this is what's called dynamic topography linked to the mantle properties and, of course, rivers are very, very sensitive to elevation and gradient, and so you have good examples. For example in Africa, which is currently very high in the southern areas. It's anomalously high, a super swell as people have called it, and this has affected the history of the Zambezi, Limpopo and Orange Rivers, which have altered their courses and directions to correspond to the elevation of Southern Africa.
Oliver Strimpel
That's fascinating. We actually did have an earlier podcast with Carolina Lithgow-Bertelloni, who talked about dynamic topography, and another one with Becky Flowers, which included some discussions specifically about South Africa and what we can tell about the effect of dynamic topography there. So the ongoing collision of India with Asia that started about 50 million years ago must have had a dramatic effect on the river systems there.
Martin Gibling
Yes, this is one of the big stories of river modification through the breakup of Pangea, and it's absolutely fascinating. India is drifting northwards at quite a rapid rate, separated in turn from Australia, Antarctica, Africa, elsewhere, and by about 60 million years ago in the Paleocene, India is approaching the southern shores of Asia. The Tethys Ocean is narrowing down to a narrow seaway, and then you have the start of collision. And everything changes. In fact, there are not many rivers in this part of Asia that go back more than about 20 million years. And there are constant changes of direction and river capture, tributary systems that affect them all. And if you think of Himalaya and the Tibetan Plateau, there are rivers radiating out in all directions. It's called the water tower of Asia, and the thrust sheets are continuing to move southwards, and indeed India has pushed probably a couple of thousand kilometres into the body of Asia. Why wouldn't it have an enormous effect? And the kind of effects go a bit like this. Probably the oldest river in the system is still the Indus, which actually runs along the line of the collision westwards, and then cuts through the Himalaya. That's the Western Syntaxis, where Asia curls around the edge of India as it indented into the continent. And so it curls around and runs down into the Arabian Sea. And initially, the Ganges occupied the foreland basin, so-called on the south side of the Himalaya, running westwards into the Indus. So it was initially an Indus tributary, and then as India moved progressively northwards, an old body of rock in the Aravalli Hills South of Delhi began to impinge on the river basins and separated them off into 2 catchments. And the Ganges was diverted eastwards about 15 million years ago in the Miocene along its present line down to the Bay of Bengal. Also, up in the Himalaya flowing eastwards on the line of collision was the Yarling Sangpo River, and that initially flowed on into the Irrawaddy and out through Myanmar, and that is known because of fish populations that are relatively similar in the two rivers, but don't match elsewhere. But the Brahmaputra, rather steep, eroding headwords into the Himalaya cut through and captured the line of the Sangpo about 18 million years ago and brought it down into the Bay of Bengal. Those are massive changes in river and drainage systems. And give us time, and there will be more to come because there are several rivers like the Arun and the Gandak that actually cut through the Himalaya and are now within striking distance of capturing parts of the Yarling Sangpo System. They may eventually divert flow down into the Ganges and out to the Bay of Bengal by a different route. It's really fascinating.
Oliver Strimpel
Wow, that's incredible!
Martin Gibling
Meanwhile, down in Peninsula India, the rivers have not yet been affected, and you get big rivers still lying within Jurassic and Cretaceous rifts, as they have been since the break up, when India separated from other continents. But as time passes, and we're talking millions of years, they're going to come closer and closer to the mountain front and the thrust front. And so, as time passes, they are going to be modified, drawn under, with subduction, they're going to become tributaries to other systems or whatever. So there's a lot more still to come. And when we go further east into Asia, the big rivers of the Yangtze, the Mekong, the Salween, all began life as little coastal rivers on the edge of the rising Tibet, and they then eroded headward and started to capture the Red River, which may have been the biggest drainage system. It's thought that the middle part of the Yangtze was originally a tributary of the Red River that reversed its course, cut the three gorges in the Miocene probably, and then was locked into place, so there's been a jockeying of rivers all very close together as the tectonics of the Himalaya and Tibetan Plateau have squeezed them, modified them structurally, changed them, and these changes are still ongoing.
Oliver Strimpel
One thing that really struck me as you were giving all these examples is how you can have a river course that can be relatively unchanged for geological time, and then it can get, more or less in a geological blink of an eye, captured by another river and then flow to somewhere completely different on the planet, essentially.
Martin Gibling
Yes, yes, and you can get rivers that are captured and then diverted literally from 1 ocean into another. For example, some thousands of kilometres apart. In the case of the Yangtze Mekong and Salween, they actually all run in a series of gorges within about 100 kilometres of each other, and they all end up diverging into completely different oceans. They've been likened to a bundle of javelins in the hands of Zeus, or ancient pictures, where you have the javelins held by Zeus. But their tips are radiating out in different directions. A good example of this, by the way, is the Niger in Africa, Western Africa, which just began in a rift, and it was quite small, and then it took over an inland drainage very suddenly to the West, and then it went further west, when a river that was running out into the Sahara got blocked by sand dunes. And diverted into the upper part of the Niger. So very suddenly, you had a river that rises very close to the coast. Runs north, swings round in the Big Bend at Timbuctoo, and then runs for some 4000 kilometres and runs into the Atlantic on down the coast. In the original position. Absolutely fascinating.
Oliver Strimpel
The continuation of my conversation with Martin Gibling is in the next episode of Geology Bites. We'll be talking about the rivers of Europe and the Americas, as well as the impact of the Ice Ages and humans on river courses. For more about Geology Bytes, as well as pictures and illustrations that support this podcast, go to geologybites.com.