Clark Johnson on the Banded Iron Formations
Transcript
Note, transcripts are not fully edited for grammar or spelling.
Oliver Strimpel
This is Geology Bites with Oliver Strimpel. In the early Proterozoic, about 2 1/2 billion years ago, enormous thicknesses of iron-rich rocks were deposited on continental shelves. Striking cliffs of these rocks form the walls of many canyons in Western Australia and South Africa. They're called the banded iron formations because they show vivid banding between the reddest, most iron-rich layers and ochre-colored layers containing silicious materials such as chert. What is the origin of the banded-iron formations? For many years, the prevailing theory was that large amounts of dissolved iron were present in the early ocean, and it was the oxygen from the first oxygen-producing life forms in the water that oxidized this iron into an insoluble form, which then precipitated out of the oceans to form the banded iron formations. But it appears that this is far from the whole story. Clark Johnson has been studying the banded iron formations for decades, and especially the role that microbial activity may have played in their formation before the oxygenation of the oceans. Clark Johnson, welcome to Geology Bites.
Clark Johnson
Thank you. Happy to talk about these unusual rocks.
Oliver Strimpel
Can you describe the banded iron formations?
Clark Johnson
So the interest in banded iron formations, many of which are banded, but not all of them (so, the more general term would be iron formations), they're essentially restricted to the Precambrian. So, do they represent something unusual in terms of a view of the ancient earth that we don't see, for example in the Phanerozoic? And to describe them or define them, people have commonly said it's a chemical sedimentary rock that has greater than 15% iron. That's unusually rich in iron. And they're also comprised of silica, about roughly half silica.
Oliver Strimpel
Now these bands… On what scale do we see structure within these rocks?
Clark Johnson
They vary a lot in the scale of banding. At the finest scale, they're sub-millimeter, and in terms of what defines the bands, this is commonly an iron oxide-rich layer followed by a silica-rich layer, so you can see them quite clearly in the field. And that has been proposed, perhaps annual depositional variations varve-like features. And then there are centimeter scale and then meter scales of these alternating bands of relatively iron-rich layers and relatively silica-rich layers. And there's been lots of hypotheses because for why we would have this alternation of bands.
Oliver Strimpel
Is it thought that most of the structure in these rocks is original, or is some of this thought to have happened after they were deposed?
Clark Johnson
That's a really interesting question. So what you can say is that you can go out to some of the famous very large iron formations like in Western Australia (they're about 2.5 billion years old)and you can follow these fine-scale layerings, like the bar code,if you will, in one locality, and you can find the exact same scale of banding 100 kilometers away. So, it's really impressive that this banding can go over great distance. How did we get that banding? Well, we have to go back to where we think these sediments were originally deposited on continental shelves, and most people have envisioned these to be very gelatinous, very dilute silica and iron oxide gels that formed on the ocean floor. They probably had to be a couple of 100 meters depth, because, to have that banding not being disturbed, if it's primary sedimentation, if that's what they reflect, you'd have to be strongly below wave base. And so some of the banding is thought because of that continuous nature to be primary. It would be unusual if you had some different lithification or compaction in one area, but 100 kilometers away you get the exact same compaction. That doesn't make sense. So that would be an argument that their primary features, but upon that you can see remobilization of silica and iron at the outcrop-scale that varies quite a bit laterally. And that probably reflects the great dewatering of this soft silica iron oxide gel. And people have made estimates that maybe we have a factor of 10 or maybe even 100 to 1 compaction. And so you see a lot of features that reflect that escape of water, as you slowly compact these gels on its way to becoming a lithified rock.
Oliver Strimpel
Are those compaction features generally on the smaller scale then, because they would depend very much on local phenomena?
Clark Johnson
Yeah, exactly. That would be less than a meter. You know, maybe several centimeters in size.
Oliver Strimpel
So I know that people have been working on understanding the origin of these rocks for decades. What are the main theories been?
Clark Johnson
So when we think about the genesis of iron formations or banded iron formations, I think it's useful to break it out into two steps. First would be the primary sedimentation step. What happened in the ancient oceans to make these rocks? But then there's a lot of evidence that they had a long, diagenetic history. And by diagenesis I mean processes that would change the mineralogy and change the characteristics of the rock on its way to compaction and dewatering to become a lithified rock. So, if we start for the first step, we would start thinking about the photic zone. So, something happening in the shallow oceans. And for that, it's uniformly agreed that that's an oxidative step,so that we have, let's say, ferrous iron - reduced iron - iron with a charge of Fe+2, that's being oxidized somehow to Fe+3, forming iron oxides. The oceans in the Archean and Proterozoic were largely silica-saturated. This is prior to the evolution of silica-secreting organisms in the oceans, and so there was quite a bit of silica also involved. And so you had this precipitation of both silica gel as you might think about it, and iron oxides forming in the photic zone and then slowly raining down to the ocean floor. So, the big question on that first step is what are the oxidative mechanisms? And there's been three main proposals. One of them is entirely abiological. It's called UV photo oxidation. And if you expose ferrous iron to ultraviolet light at high intensity, you can oxidize that ferrous iron to ferric iron. The arguments against that have been that the experiments that discovered that, maybe didn't mimic the conditions of the early Earth, specifically in terms of the wavelengths of the ultraviolet flux.And the big key that we now realize inhibits UV photo oxidation is the presence of all that silica. OK, so that's one of three oxidative steps. The second one that was very popular for many decades was an indirect biological role, where we had evolution of oxygenic photosynthesis producing free O2 in the photic zone, in the upper part of the water column, and that when the ferrous iron encountered this O2, you had oxidation and then precipitation of the iron oxides. And then, more recently, as we understand more about possible ancient microbial pathways in the ancient oceans, there's an alternative oxidation mechanism that directly involves biology. And this would be also a phototrophic pathway or a photosynthesis pathway. But instead of using water as the electron donor, this used ferrous iron. And so there are anoxygenic pathways that you can oxidize that iron and then that would be precipitating out on the ocean floor. Now one of the products of that is organic carbon and the criticism for this pathway is that the organic carbon contents of iron formations are quite low. And so the argument would be, well, if there was all this biology doing this, where’s all the organic carbon. And the answer to that probably comes in the later stages, the reductive pathways, the diagenetic pathways, where that actual carbon has been respired later.
Oliver Strimpel
Just talking a little bit more about this third pathway that you just described. Do these bacteria correspond to the anaerobic bacteria that we have around still today?
Clark Johnson
Yes, absolutely. And so we can find this pathway in modern microbial mats that are undergoing photosynthesis, where on the surface layer we have the oxygenic photosynthesizers. But you don't have to go down very far, just a few millimeters, or maybe a centimeter into modern microbial mats to where you become anoxic. And that's where you see the anoxygenic phototrophs that thrive at a slightly longer wavelength of light because they're actually buried, and so the wavelength of incoming light is shifted towards the longer wavelengths. And that's where we can see active anoxygenic phototrophs today.
Oliver Strimpel
And the precipitation of the iron takes place without the oxygen really being involved at all.
Clark Johnson
Correct.
Oliver Strimpel
Let's talk a bit about what happens to all that carbon then, according to the theory.
Clark Johnson
So that is an interesting question. So, let's say that you're a proponent of anoxygenic photosynthesis, oxidizing the iron in the photic zone — the first stage — and then you have all this iron oxide raining down. You also have to rain down a lot of organic carbon, because you have to make that chemical reaction balance. So if you put all that organic carbon down into the lower part of the oceans, accumulating with all this iron oxide, you set up the opportunity for iron-reducing bacteria to then run those reactions backward. Essentially respiring that ferric iron using up that organic carbon. That organic carbon gets oxidized, so it's gone. We don't measure it anymore. That falls into then carbonates. And, if you look at the carbon isotope composition of iron formation carbonates — iron-rich carbonates like siderite or anchorite — they have extremely light isotopic carbon, which is a very strong signature for respired organic matter. So it means that we have to be more clever at looking at the rocks than simply saying, oh, there's no organic carbon, therefore there was no life. Ah, we see the isotopic fingerprint of that in the carbonates as respired iron. And I think that's a very strong argument in support of that second reductive process.
Oliver Strimpel
So we have these alternate theories. What is the evidence really that we have this anoxygenic pathway? And you've just said it's to do with the isotopic composition of the resulting carbonates. Is that really the main supporting evidence?
Clark Johnson
Yes. So I think what you mean is that last step where we are respiring the iron oxides and the organic carbon. So what would we look for? And one of the things that is really striking when we start to analyze iron formations at the micron scale, so, not just crush up a bulk sample, which is what people did back when instrumentation wasn't allowing us to do what we call in situ measurements, you know, spot by spot, which we can do by using let's say, an ion probe or by laser ablation attached to a mass spectrometer… When we look at individual grains, let's say a magnetite sitting next to a siderite, we see that none of them are an isotopic equilibrium with each other. And so, if they were all in equilibrium with the oceans, they should be in equilibrium with each other, isotopically, in terms of the compositions. And so, what we find when we look at iron formations at the micron scale is that disequilibrium rules, and whenever we see disequilibrium, we tend to think that that's a specific diagenetic or in this case biological diagenetic pathway that, for example, formed the iron carbonates and totally ignored what the adjacent magnetite or iron oxide was doing.
Oliver Strimpel
Coming back to the oxygen-based process versus the anoxic process, we have quite a lot of evidence from other sources that oxygen really only started to prevail on the planet sometime in the Proterozoic and that we really can't invoke that pathway before we have a significant amount of oxygen, because there wouldn't be any free oxygen left to precipitate that iron. So, how does the timing of the banded iron formations which we seem to be able to date pretty well, relate to other evidence for oxygen emergence on our planet?
Clark Johnson
You're absolutely right. That first large increase in atmospheric oxygen, which has commonly been called the great oxidation event at about 2.3 - 2.4 billion years old, that leaves us with an awful lot of banded iron formations that are older than that. And so, if free oxygen didn't exist in the atmosphere before that, that would lead people to say, oh, well, maybe oxygenic photosynthesis did not occur until then. How do we do all this oxidation with free O2? The problem with focusing on just the atmosphere is that it ignores what could be oxygen in the photic zone.
Oliver Strimpel
In the ocean.
Clark Johnson
In the oceans, and, we know from a lot of modeling that people have done that, we could have a slightly oxygenated photic zone underneath an anoxic atmosphere. And so, when we're talking about a chemical sedimentary rock that formed in the oceans, it's important to distinguish the oxygen contents in the oceans relative to what the atmosphere would be. If we take that perspective, we actually are finding evidence for free oxygen in the photic zone going back as far as 3.2 billion years old. And so, when we broaden our perspective of where was the oxygen prior to the quote great oxidation event, we actually are finding it by many different proxies for oxidation, be they isotopic or trace elements, going back much further.
Oliver Strimpel
Do these theories help us explain what controls the abundance of these iron formations where we see some very old formations, but then they really develop into these enormous thicknesses around 2.3 – 2.4 billion years ago, and then, as I understand it, gradually taper off to about 1.8 billion. And then we see a little bit more in the late Devonian. How do we explain this timing of the deposition of these iron-rich rocks?
Clark Johnson
Yeah. So if you just make a plot of the size, the tonnage, or the volume of iron formations through time, what you do see is that the oldest iron formations, these would be things in Canada or Southwest Greenland that are about 3.8 to 4.0 billion years old, those are very small, and they're actually, in the field, they're very thin, maybe a few meters in thickness. And then we get a steady increase in these, and then, as you said, they're really big ones, the famous ones, like in Western Australia or in South Africa of about 2.4 - 2.5 billion years old. Those are tremendously thick. Those are many hundreds of meters thick. And so how can we explain that? Was that a change in the iron content of the oceans through time? Was it increasing? Was it a change in where these are deposited? So it's important to remember that we find iron formations which are marine sedimentary rocks preserved on continental shelves. Could it be that there were so few continental shelves, or they're relatively thin in the early Archean, and that's why we have just a few scrappy, in a comparative sense, iron formations then. But by the time we had the continents being in a fairly large volume at about 2.5 billion, that we had a much bigger area to deposit them on. And so there's preservational aspects, but also the question of, well, do we need large continental shelves to evolve to have the large iron formations? Or do we just not have many continental shelves early in the Archean? Maybe there was a lot of iron formations deposited in the deep oceans, on oceanic crust that was all then subducted, and we don't know that record. So we're always wrestling with this preservational bias versus what the driver might be.
Oliver Strimpel
What about the original source of the iron? Could it be that there just wasn't that much iron around in the early Archean, and where does all this iron come from?
Clark Johnson
Yeah. So that's an interesting story. There was a famous paleobiologist named Preston Cloud who, in the 1960s, proposed that these iron formations, their source of iron, was anoxic weathering on the continental crust, producing rivers that were rich in Fe+2, delivering this to the oceans. Then he wanted to have oxygenic photosynthesis occur, and then do this oxidation by reaction with O2 to make the ferric oxides. Then, in the 1970s, as people started discovering these hydrothermal vent systems at mid-ocean ridges and sending submersibles down and actually sampling the vent fluids, people were struck that these trace element chemistries, particularly things like rare earth elements, were remarkably similar for the vent fluids and the iron formations. So then the idea shifted to this being a hydrothermal source. If it's a hydrothermal source, then you actually would think that there would be more ferrous iron early in the Earth's history, when we envision hydrothermal systems to have been much more vigorous because the heat flow was much higher coming out of the early Earth, much more radioactive decay-generated heat flow, and so it's the opposite of what we would see in the iron formations. More recently, by looking at other isotopic systems such as neodymium isotopes in these iron formations, we now realize that there actually is a continental input, not the riverine input, but maybe a microbial iron source on a continental shelf, and so that then puts the question back: is the large volume of iron formations in the more recent periods of the Precambrian due to just having larger continental shelves?
Oliver Strimpel
Wow, a lot of intertwined factors, but just going back to the very origin of this iron, presume it comes up from the mantle and forms in the crust with volcanic processes that produce basalt, which has olivine and peroxides in it, which contain iron, and that eventually through that weathering gets leached out, if you like, and goes into the oceans. Is it thought that that was the origin of all ths iron, and it just eventually accumulated in the oceans over time?
Clark Johnson
Well, in the broadest sense, yes, you're right. The ultimate source of iron in the oceans, I think almost everyone would agree, has to come from some process that leached it out of basalt, be it at a hydrothermal system at some mid-ocean Ridge or weathering of basalt crust on the continents, producing a riverine flux, as envisioned originally by Preston Cloud. (We've come back to appreciating his ideas.) So. Ultimately, it has to be some reaction between fluids and the salt to actually deliver that iron to the ocean.
Oliver Strimpel
But then we still need to explain the difference, then, between these absolutely enormous deposits we see in the Proterozoic, and we don't really see that, or maybe we do see that happening on the same scale today. I'm not sure. But we still have basaltic volcanism, \and we still have mid-ocean ridge spreading, so there's plenty of basalt being generated today. So are we still seeing iron formations today?
Clark Johnson
So the big thing that would change between what modern systems are and what we might envision an ancient low oxygen world hydrothermal systems is that the oceans have vanishingly small iron contents today, many orders of magnitude smaller than what people have estimated for the Archean oceans. And the reason for that is that hydrothermal vent systems today oxidize iron very quickly, and the particulates fall out very close to the vents. Some of it is iron oxides. We also have a lot of sulfides in the systems today that we probably didn't have in the early ocean systems. And so that precipitates out as iron sulfides in these very extensive iron sulfide chimneys that you can see all these wonderful pictures — Alvin [submersible] going down and taking photographs of these things. And movies, they're quite spectacular. So the iron is not far traveled in the oceans today, it falls out within hundreds of meters of most hydrothermal plumes. But in the anoxic oceans you would maintain that dissolved iron for some period of time, and could probably transport it for thousands of kilometers.
Oliver Strimpel
So some of the mechanisms that we have today for removing that iron were not present in the ancient oceans, and so the iron could just accumulate in its dissolved form until it reached quite a high concentration in those early oceans. And then, when the mechanism to precipitate them finally came along, whatever that mechanism was, and you described three possible oxidation steps. Then we just had this huge reservoir of iron, if you like, that could all fall out of the ocean and generate these hundreds of meters of thick formation. Is that the basic idea?
Clark Johnson
Yes, exactly. And, in fact, you can bring back a connection between oxidizing all this iron at some point when conditions became favorable to actually having the rise of oxygen in the atmosphere, because you can make a nice argument that if you have this hydrothermal flux that is reduced iron, you have to oxidize all that material first before you can then have the ability to actually rise free oxygen in the atmosphere, and so those could likely be coupled in that way.
Oliver Strimpel
You mentioned a moment ago that, according to one theory at least, we need oxygen in the photic zone within the water, which implies that we need to have perhaps continental shelves, because we need it to be deposited on rock that didn't then disappear through subduction. We don't have any really old oceanic crust, because that's all gone down a subduction zone. So what is our thinking about the rise of the continents and how much continental crust would have been available at that time?
Clark Johnson
Yeah. So the evolution of the continental crust seems likely to factor into this, because that's the formation environment for these iron formations, and so people have spent a lot of time looking at the evolution of continental crust. A lot of the evidence comes from isotopic studies of the crust itself, and there's a broad consensus that the continental crustal volume reached maybe 50% or maybe 60% of its volume by maybe about 3 billion years. And so that's potentially a lot of continental crust ground. For the purposes of an environment to supply iron or to deposit these iron formations, we want to know more. We want to know what type of continental crust it was. Was it basaltic and low lying, in which case maybe the continents weren't emergent? (Some people have argued that they were mostly submerged until perhaps the great oxidation event.) On the other hand, there's evidence that maybe the continents were emergent early on. That evidence comes from looking at the chemistry and the isotopic compositions of sea water proxies like carbonates. And if that were the case, that the continents were emerging early, it changes our biologic story because it potentially supplies a large flux of nutrients. And if there's a biological component to iron formation genesis, you're not going to do much without a whole bunch of nutrients. And we would think that that would come from emergent continental crust. So this feeds back into what we think on the broader mantle crust evolution and it's part of this holistic approach, if you will, to look at what the solid Earth evolution and the shallow surface environments out there intertwine.
Oliver Strimpel
That's fascinating. The basic picture, though, is that the amount of continental crust goes up overtime, because continental crust is a secondary phenomenon that results from fractional crystallization or partial melting, and so on, so that we know that the amount of continental crust has gone up over time. There’s debate about at what rate It increased and whether there was enough, or how much was around around 2.3 or 2.4 billion years ago, when we saw these massive iron-rich deposits, but constraining that using the evidence from the banded iron formations, you get into kind of circular self-consistency arguments.
Clark Johnson
Yes, we always have to be aware of that. And so let's say I'm gonna argue this line that, OK, we needed continental crust to evolve. We wanted an emergent, we wanted to supply nutrients to the oceans. We wanted this to happen before the rise of atmospheric oxygen, because we need that to have a biological mechanism to precipitate-out the iron. Well then, I pretty much have to go to an independent line of argument. And so I'm going to then go read all the mantle crust evolution papers that look at hafnium isotopes, lead
isotopes, and neodymium isotopes, and I'm going to plot those curves up and say, oh, look at that, that kind of correlates. And that's a relationship that might be permissible. Proving it as a cause, of course, is different and more difficult, but that would be the approach, and I'm a strong believer that you need just every single line of evidence that you could just grab your hands on when working in the early Earth, because none of us were there and we have great ambiguities. And so we looked for every independent line of evidence that we can throw together.
Oliver Strimpel
You mentioned that these iron formations can have 15% of iron content. That's huge. So presumably we've seen a lot of commercial exploitation of these formations.
Clark Johnson
Right. And that's probably why the ones in Western Australia are so famous, because they're a source of enormous iron deposition, and anybody who’s been out there, they'll see just the very large scale of mining, and this is all mined in Western Australia and then shipped by rail, generally up to a port called Port Hedland, which likes to claim truthfully that they are the largest port in the world, but no one has heard of them. That's the largest port in terms of tonnage. And so all that iron formation that's been mined is going to head off to be made into steel, and ultimately it finds its way into our vehicles, into our phones, our silverware, anything — a very large important commercial aspect to it.
Oliver Strimpel
What are you working on at the moment?
Clark Johnson
These ideas of continental growth and possible changes in nutrient fluxes are quite interesting and timely right now. The big deposition of iron formations in Western Australia are incredibly impressive, but they actually only represent the deep part of the basin. And so what we've been looking at is the oldest known well preserved marine sedimentary basin called the Pongola Wits Basin — Witwatersramd watering basin. It's about 2.9 billion, it's in southern Africa, South Africa and Swaziland are the main exposures, and the fascination there is we can walk from the shallow water and even braided stream deposits down into the deep water parts of the basin. And so we can see the whole picture, and if we're interested in what is the relative role of deep hydrothermal input versus continental input, looking at that transition is really valuable. We're interested in what the nutrient fluxes might be as we move towards the shoreline. What might be the hydrothermal influences as we move towards the deeper water, and so it's that interplay at that basin system that's what we're currently working on.
Oliver Strimpel
Clark Johnson, thank you very much.
Clark Johnson
Great. Thank you. It's been a pleasure.
Oliver Strimpel
To see pictures and illustrations that support this podcast, go to geologybites.com, where you'll also find transcripts and a subject-matter index of all the episodes. There you can also give me feedback, which I welcome, as well as sign up to get my emails about new episodes.