[MUSIC] Welcome. In this lecture, I will outline the history of atmospheric and oceanic oxygen and some of the most important lines of evidence found in the geological record. Most geobiologists agree that oxygen levels increased over Earth's history from a fully anoxic atmosphere in the Archean to the oxygenated atmosphere that we have today. A most remarkable oxidation event occurred 2.4 billion years ago. At that time, oxygen levels increased several orders of magnitude. This graph shows a sketch of how we think oxygen levels have increased over time. The y-axis shows atmospheric oxygen on a logarithmic scale in percent of modern-day levels. 100% PAL refers to the partial pressure of 0.21 bar in the atmosphere, the level of oxygen that we have today. After the Great Oxidation Event, atmospheric oxygen levels first increased, and then it may have declined for a while before it settled for a long period of stasis. In the Neoprotozoic, oxygen levels increased in the oceans, perhaps also in the atmosphere, and finally, oxygen rose to near modern levels with the emergence of vascular land plants in the early Paleozoic. So, how did we get to this figure? First of all, we have long known that all the world's major iron deposits formed in the oceans during the Archean and early Proterozoic. Here's a picture from the iron mine in Rio Tinto, in Spain. The next picture shows clear red hematite bands in banded iron formations of the Huronian Supergroup in Ontario, Canada. This particular sample is from the oldest banded iron formation in Isua in West Greenland. It formed approximately 3.8 billion years ago as a chemical precipitate in sea water. In combination with a massive extent, the banded iron formation suggests to us that the oceans contained dissolved iron, and that could be transported in solution. At seawater pH, iron would only stay dissolved in the absence of oxygen in the waters. Otherwise, it would form rust and precipitate out of solution. Thus, the banded iron formations are evidence that at least the deep parts of the oceans were anoxic at this time. However, the banded iron formations are quite spectacular and it seems that they formed only during periods of high volcanic activity when iron from the mantle was ejected into the oceans. The strongest line of evidence for an anoxic atmosphere in the Archean comes from sulfur isotopes in marine sediments. The sulfur in marine sediments older than 2.4 billion years carries a mass independent isotopic signal from an exotic photochemistry that can only take place with a UV light in an ozone-free atmosphere. Because the ozone layer is produced from oxygen, the sulfur isotope signal will disappear if the atmospheric oxygen level exceeded only few parts per million. Thus, this isotope record implies that the atmosphere contained less than PPM levels of oxygen in the Archean. Actually, over the last decade, it has become clearer that the exotic sulfur photochemistry also requires a high abundance of methane and hydrogen gas in the atmosphere. And that the oceans must have contained very little sulfate so the signal from the atmosphere could be transported to the marine sediments without being diluted with a large pool of sulfur that we see in the oceans today. So, high methane, high hydrogen, low sulfate, low oxygen are all distinctive features of the Archean Earth. Quite the opposite of the modern atmosphere and oceans. Overall, everything points to a very reducing situation on the early Earth. The transition from the anoxic world to the oxic world in the Paleoproterozoic is called the Great Oxidation Event, or just the GOE. This picture is from the Huronian Supergroup in Ontario. We are basically standing on sediments that were deposited before and after the GOE. Basically, the lower part of the section formed in an anoxic world without ozone and with methane and hydrogen in the atmosphere. The upper part of the section formed in the world with some oxygen, an ozone layer, and less methane and hydrogen. We know that cyanobacteria were the first to produce oxygen by oxygenic photosynthesis, yet it is still not known exactly when cyanobacteria appeared on Earth. We think that cyanobacteria evolved before the Great Oxidation Event because this biological oxygen production is several hundredfold faster than any abiotic reactions. So, the oxygen produced in abiotic processes would easily have been destroyed by reaction with reducing gases emanating from volcanoes and metamorphic processes, as I mentioned in my first lecture. Also, there was a large reservoir of reducing compounds in the crust that needed to get oxidized first, not to forget all the dissolved iron and manganese cations in the oceans at this time. Some scientists argue that cyanobacteria evolved immediately before the Great Oxidation Event, while others claim that cyanobacteria had been around for at least 500 million years at the GOE. They would say that the large time gap between the emergence of cyanobacteria and the accumulation of oxygen in the atmosphere was due to the large buffering capacity against oxygen rise in the Earth's system. In many ways, we should thank cyanobacteria for their novel innovation and also be thankful that their waste product, oxygen, ended up in the atmosphere. That said, it must have caused a catastrophe for those anaerobic organisms that were adapted to the Archean Earth. The Great Oxidation Event coincides with a period of dramatic cooling events. Actually, it seems that the event also affected Earth's thermostat. Four major glacial intervals are known between 2.45 and 2.2 billion years ago. And the mass independent sulfur isotope signal disappears between the second and third glaciation. Some of these glaciations were possible Snowball Earth events, which means that the glaciers extended all the way from the poles to the equator. It seems that even the ocean must have been frozen, yet some life forms did survive. Among them were the cyanobacteria. The first 100 million years after the last glaciation event is called the Lomagundi episode. During this period, Earth transited irreversibly from an anoxic state to an oxygenated state. The evidence comes from the carbon isotope record. The record shows that organic carbon was buried in the ocean at an unprecedented rate. As I explained in my first lecture, every mole of organic carbon buried in the ocean sediments leaves one mole of oxygen behind in the atmosphere and oceans. So long oxygen destruction does not keep up with the oxygen release from organic carbon burial, oxygen levels will increase, and this is what happened during the Lomagundi Episode. This is perfectly consistent with the disappearance of the mass independent sulfur isotope signal and the exotic chemistry in an ozone-free world. A thick ozone layer formed after the Great Oxidation Event, and that protected the Earth's surface against intense UV radiation from the sun. Many lines of evidence point to a more oxygenated Earth after the Great Oxidation Event. Banded iron formations also disappear soon after. There's also evidence from weathering profiles in Proterozoic soils that the atmospheric oxygen levels had reached 1% of modern-day levels. This is actually enough for a sponge to survive in surface waters, which they obviously didn't. But 1% PAL is still far too little oxygen to oxygenate the deep ocean. The ocean gets its oxygen from the atmosphere. Basically, with higher partial pressure of oxygen, more oxygen gets dissolved in the ocean. The uppermost mixed layer in the ocean is saturated with dissolved oxygen. Today, there's roughly 300 micromolar dissolved oxygen in seawater under the present atmosphere with a partial pressure of 0.21 bar. The proportionality between oxygen pressure in the atmosphere and oxygen dissolved in the surface water is described in Henry's law. This constant proportioning means that at 50% of present-day atmospheric oxygen level, the surface ocean would contain only 150 micromolar oxygen. Only close to cyanobacteria would the waters be supersaturated with oxygen and reach levels higher than this. Because 99% of all oxygen is in the atmosphere and only 1% is dissolved in the ocean, it is primarily aquiferic oxygen that controls how much oxygen is introduced to the ocean. Oxygen gets into the deeper parts of the ocean when cold and/or saline waters sink according to ocean circulation patterns. Below the photic zone, oxygen is no longer produced by photosynthesis, but it gets consumed by aerobic respiration because organic matter sinks from above. In general, this oxygen consumption is dictated by the amount of organic matter produced by primary producers, which in turn is given by the amount of nutrients available in the oceans. In the modern ocean, the organic matter consumed by aerobic respiration removes on average 150 micromolar oxygen in the ocean, some places more than others. On average, and as a rule of thumb, we can say that the deep oceans would go anoxic if the atmospheric oxygen level was below 50% of present day. There are actually many lines of geochemical evidence that oceans remained anoxic even though atmospheric oxygen increased after the Great Oxidation Event. Our current model suggest that anoxic water masses were still common in the oceans in most of the Proterozoic. Basically, we can think of how oxygen and life is distributed in a microbial map and then envision that chemical stratification for the entire ocean. Or just for the uppermost, say, 800 meters of the oceans that cover the continental shelves. The evidence for ocean anoxia in the Proterozoic comes from a whole suite of redox-sensitive elements, including molybdenum, sulfur, and chromium. These elements are chemically weathered and mobilized from continental rocks when the atmosphere in rivers become oxygenated. The metals precipitate in anoxic waters, so their behavior is exactly opposite to iron. We see enrichments of these metals in marine sediments after the Great Oxidation Event. That means the metals were first mobilized by oxidative weathering in soils on land. And then transported in oxygenated river water into the oceans before they precipitated in the anoxic parts of the oceans. I have worked on the development of a proxy based on molybdenum isotopes with which we can follow oxygenation of the global ocean. The concept of how molybdenum isotopes reveal ocean oxygenation is actually quite simple. With molybdenum isotopes, it's possible to trace how oxygenated was the entire ocean even if you had only a cupful of seawater or spoonful of sediment available. The seawater in this glass can tell us how much of the entire ocean is covered by anoxic and oxic waters. This idea is quite powerful because it will not matter where in the ocean I take this sample. Today we have both sediments and seawater available for analysis, and from the past, we have sediments. Thus, we can calibrate the method in the modern and apply it to the past. The concept used to interpret molybdenum isotopes is a basic principle that geochemists use to track the chemical evolution of the oceans. Once you understand this, you will also understand, for example, why carbon isotopes are just massive organic carbon burial during the Lomagundi episode. Today, the oceans are well mixed with respect to molybdenum. This is because molybdenum has a long residence time in the ocean of approximately 450,000 years, whereas, the ocean is mixing on a timescale of only 1,500 years. That means, on average, every molybdenum atom circulates the oceans 300 times before it finally settles into marine sediment. There are two major sinks for molybdenum in the oceans. Either molybdenum is buried slowly in oxic settings, or it's rapidly buried in anoxic and sulfidic settings. Today, the molybdenum isotope composition of the ocean is relatively heavy. Actually, it contains additionally two of the heavier 98 molybdenum isotopes for every 1,000 95 molybdenum isotope, compared to what we find in river water. We say that seawater is two per mil heavier than river water. This is illustrated in this figure. Let's assume that the oceans are in steady states so that all molybdenum supplied to the ocean is also buried in ocean sediments. The reason why seawater is heavier than rivers is that isotope fractionation occurs when molybdenum is buried in oxic settings. The oxic sediments preferentially remove the lighter molybdenum isotopes. For example, 95 molybdenum. Whereas little or no isotope fractionation occurs when molybdenum is buried in anoxic and sulfidic conditions. Thus, the isotope composition of euxinic sediments reflects the relative balance of oxic and anoxic molybdenum removal. For example, we can take this sample. It's a black shale. And we can use the molybdenum isotope method to analyze euxinic sediments like this. The molybdenum isotope signature reveals how common were the oxygenated areas compared to the anoxic settings in the oceans where this sample precipitated. Indeed, we find that all Proterozoic shales contain less of the heavier molybdenum isotopes compared to modern euxinic sediments. This means that oxic settings were less important, and euxinic settings were quantitatively more important back in the Proterozoic. There are also other lines of evidence that several Proterozoic marine sediments were deposited under locally anoxic and sulfidic waters. The molybdenum isotope composition of seawater is particularly sensitive to oxygenation of the deep ocean, because the strong isotope fractionation occurs when molybdenum is removed with manganese oxide in oxygenated settings of the abyssal ocean. Over the last decade, we have collected molybdenum isotope data from black shales, and we see that oceans first became a bit more oxygenated in the Neoproterozoic. But also that the deep oceans were first fully oxygenated some 400 million years ago, long time after animals evolved. Therefore, atmospheric oxygen levels may well have remained below 50% of modern-day levels in the Proterozoic. A powerful constraint on atmospheric oxygen levels in the past 400 million years comes from the fossil record of charcoal. Charcoal is a byproduct of wildfire and the oldest charcoal is found in the late Silurian. Here are some microscope images of that. After a gap in the Devonian, charcoal is essentially found throughout the rest of the Phanerozoic. This is strong evidence that atmospheric oxygen levels actually remained above 70% of present-day level. The reason for that is that wildfire occurs when fuel and oxygen is put together in the presence of heat. For example, fire can be ignited with a match by lightning, but the flame will go out immediately unless the mixing ratio of oxygen to nitrogen in the atmosphere exceeds 0.15 atmosphere. This corresponds to 70% PAL. Therefore, wildfire requires at least 70% PAL, and the fossil record of charcoal is essentially continuous over the past 400 million years. This means that atmospheric oxygen levels has stayed above 70% PAL. This is enough oxygen to sustain healthy human brain function, so we can conclude that humans could have survived already back then. And we should think that oxygen did not hold back the evolution of intelligent life on Earth even though healthy brain function requires particularly high oxygen levels. Other factors in evolution of human intelligence must have played a key role. Okay, so let's take a look at the history of oxygen again. Banded iron formations and an exotic mass independent sulfur isotope signal in the sediments suggest that both the Archean oceans and atmosphere were anoxic. That's in stage one. There was a remarkable shift 2.4 billion years ago when the atmosphere became oxygenated in stage two. The sulfur isotope signal disappeared, and oxidative weathering on the continents began to liberate redox-sensitive trace metals from crustal rocks, and sourced these metals into the ocean. This included chromium, molybdenum, uranium, and also sulfur. Then there was a long period of stasis with anoxia in the deep parts of the oceans, and euxinic conditions more common on the continental shelves. The first metazoans were sponges and they evolved in the new Proterozoic oceans, in stage four. There's no consensus of how atmospheric oxygen evolved at this time. But the atmosphere may have contained more than 1% and less than 50% of present-day oxygen levels. Molybdenum isotopes suggest that the deep oceans remained anoxic, and for the first time, became oxygenated in the Silurian and Devonian in stage six. When, at that time, vascular plants emerged on land. So, from this time and onward up to today, we also find fossil charcoal indicating that the oxygen levels were higher than 70% of modern-day levels. We're still not sure whether atmospheric oxygen levels increased when the first animals emerged, but the oceanic oxygen levels seems to have increased. Anyways, the first metazoans was sponges, and they appeared in the Neoproterozoic oceans perhaps more than 700 million years ago. They could have survived at 1% PAL. The first motile animals appeared after the Neoproterozoic glaciations, approximately 555 million years ago. Probably, they would have required more oxygen than sponges, but even that is not clear. Here, you see a picture of Kimberella. There's actually geochemical evidence to support the idea that the deep oceans became more oxygenated at this time. Intriguingly, the Ediacaran fauna lived from, 580 to 550 million years ago in rather deep settings, and there's evidence that oxygen was present in these deep upwelling waters. I would say that the jury is still out whether or not the emergence of motile animals occurred as a result of oxygen rise. In the next lecture, I will present some of the views on the coevolution of oxygen and life on our planet. [MUSIC]
