Discoelastic shear

 I think I originally created this for a comment on a structural geologist's blog back in the blogging heyday of the zeros. But I can't find any trace of it on the internet now, so I'm reposting for the younger generation. The following figure is an example of discoelastic shear:

The Greenland impact crater

This is a brief note on the recent Science Advances paper on the Hiawatha Impact crater, a large, recent crater which lies under the Hiawatha glacier in extreme northern Greenland.

In the past, I have bagged out impact crater scientists and being alarmist and even dangerous. However, this discovery is the real deal. Similarly, I have occasionally criticized the “glamour-mag” approach to scientific publication, but in this instance, a big splash is appropriate, because it is a big deal, and the evidence is overwhelming.

The short version: Ice penetrating radar and analysis of glacial outwash sand show a large (31km diameter), recent impact crater under a Greenland glacier, complete with central peak. The outwash shows shocked quartz, probable melt glass, and PGE anomalies consistent with an iron (or stony iron) impactor. This is not one of those ancient, deformed, maybe-if-you-squint-you-can-see-a-circle crators, this is in your face and completely obvious to anyone who has studied even a little geology.

Like many short format papers, a lot of the details are in the supplementary materials.  For example:

This is a recent discovery because of global warming! Prior to 2012, the outlet glacer emptied into a lake. It is only ni the last 6 years that it has retreated onto land, so that the sediment they sampled and found the shocked quartz, impact glass, etc in was only exposed from beneath the melting ice sheet a few years ago.

They are planning on running conventional gravity surveys to look for rebound, but because all the ice is melting, the melt signal dominates the GRACE gravity signature.

No known impact ejecta is known from any of the North Greenland ice cores, making the crater likely to be older than the oldest of them (about 100ka). Ice cores are regularly checked for volcanic debris, and it is unlikely that they would miss something this large and close (quick math suggests the ejecta volume should be about 200-600 km3, making it a medium to large VEI-7 equivalent).

The crater overprints pre-glacial river valleys, and this is (as the authors state) probably Pleistocene in age (10ka-2.5Ma).

The melt glass should be datable via Ar/Ar dating, but it is not clear if they have recovered a large enough volume of the material to date at this stage.

I would expect a tektite field from an impact this size, but it isn’t clear where those tektites would end up. If they fell on ice (By definition, the Arctic was mostly ice-covered during most of the Pleistocene), then they would get carried to a moraine (on land), or float around until the ice was exported through the Fram Strait and melted somewhere in the NE Atlantic Ocean.

There is a controversial Younger Dryas impact hypothesis, which basically calls for an Arctic impactor as a trigger for the Younger Dryas cooling and extinction of the Clovis culture in North America. I would be careful connecting this crater to that event, as the NEEM ice core, less than 400km away, doesn’t have any reported ejecta, as known tephra are mostly basaltic. 

Finally, they report carbon in the silicate impact melt. That seems odd to me, as neither crustal gneisses nor iron meteorites have much carbon.  they should do ion probe d¹³C to get the isotopic composition. Who knows, maybe the impactor hit a peat bog.

  

The wrong kind of Bang

In science education and popularization, there is a delicate balance that must be struck between overcomplicating and oversimplifying. Insufficient simplification can result in overly obtuse deviation into secondary details, which confuse and distract the readers and derail the flow of the prose.  Excess simplification can be wrong.  And this is where the Medium article by Ethan Siegel of “Starts with a Bang” fame has ended up. 

Dr. Siegel argues that the recent Philae comet lander would have more successful if it had been powered with a ²³⁸Pu RTG device instead of solar panels.  However, his simplified argument ignores the reality of ²³⁸Pu fuel production, the definition of “we”, and the nature of comets.

²³⁸Pu is a byproduct of the nuclear arms race between the USA and the USSR. It is created by neutron activation of ²³⁷Np, which in turn is a byproduct of ²³⁹Pu production for nuclear weapons. With the nuclear arms deals of the 1980’s the superpowers stopped building nuclear weapons by the tens of thousands, and the cheap source of ²³⁷Np disappeared.  The USA stopped ²³⁸Pu production in 1988, all subsequent material has come from Russia, which has almost depleted its stockpiles.

This brings us to the definition of “we”.  As the battleground over which the USA and the USSR fought, Europe never developed its own mass nuclear warhead production facilities; the UK and French arms supplies are only a tiny fraction of the size of the 20^th century superpowers.  As a result, Europe has never had its own large scale ²³⁸Pu production facilities. 

Philae was a European mission, not a USA or Russian one, so the ESA (European Space Agency) did not have access to ²³⁸Pu needed for RTG production.  NASA (USA) and the ESA (Europe) are separate space exploration entities, a point that was very unclear from this article’s frequent discussion of NASA and Philae.

Finally, RTG’s are hot, and comets are cold. The Philae lander was a very risky mission- there was a significant chance that it would not succeed at all, and in the end the lander ended up bounding off an unexpectedly hard surface several times before ending up on its side in a crater.

Comets, by definition, evaporate at low temperatures- this one is jetting out gasses despite being way out beyond the asteroid belt. So landing a heat-producing source on it, especially on a lander that ended up tipping over, would end up in a situation where the lander could drastically alter the local environment of the comet through thermal contact.  The whole point of the mission is to sample a comet in as pristine condition as possible, so potentially cooking the comet due to a landing mishap is not really a sensible design choice.

Dr. Siegel is correct that ²³⁸Pu is crucial for missions that operate beyond the orbit of Jupiter.  But the fuel used on previous missions was subsidized by the nuclear arms race.  It, and all the wondrous outer solar system exploration it allows, was an unintended byproduct of Mutually Assured Destruction, and the tens of thousands of nuclear weapons that policy produced.  Since the arms race ended, production of this isotope for the sole purpose of planetary exploration has been deemed too expensive to pursue by all the world’s governments.  Until we collectively decide to blow ourselves up again, this barrier to outer solar system exploration will continue.

Bayesian Geochronology

 Bayesian statistics is a growing field of exploring data-rich, complicated, explanation-poor phenomena. In a nutshell, you start out with an initial assumption, called a “prior” and then modify it with new sets of independent data. The Bayesian approach is one of the hot new techniques to come into Earth Science in the past decade, as it allows us to sensibly integrate disparate sets of data about the same physical process derived from different proxies or fields. Long time geobloggers will know that James is an expert in applying this technique to climate; see his blog and papers for more information.  However, despite the decade-long use in climate science, Bayesian statistics have not often been applied to more interesting areas of geoscience.  E.g. Geochronology.  Until now.

In Geology, Vleeschouwer & Parnell present a new Bayesian method of constructing the geologic timescale.  They point out that the previous timescale, which were just spline fits between selected stratigraphic/geochronological time points, had the perverse effect of yielding smaller error bars in the gaps between data points than near to where there are data. So they try a Bayesian approach instead.

For their Prior, they start off with the rule of superposition- which states that lower stratigraphy is older than higher stratigraphy.  They then apply all the high precision geochronology dates (high-precision CA-TIMS U-Pb in this case), and generate a probability density function for stratigraphy vs time. Their resulting chronology is, sensibly, more erroneous in areas where there is no data, compared to those where data does exist.  But in addition to being more sensible, this approach is far more useful.

Firstly, from the modeled uncertainties in the interpolated regions, one can guestimate the precision (both geochronological and stratigraphic) that is necessary to help constrain the timescale.  So, for example, in some of the more data-poor regions, one can probably usefully constrain the timescale using a less precise microanalytical method such as SHRIMP, if the only available zircons are not suitable for CA-TIMS.  On the other hand, as you approach a high precision date, lower precision techniques become less useful.

In fact, you can tell whether or not a particular age is useful for the timescale simply by looking at how the inclusion of that date alters the PDF.  Previously, the inclusion or rejection of particular ages has been a non-transparent potentially politicized process.  The use of this model allows us to replace the argumentative old men with statistics.

Of course, one needs to be careful in adding data.  A potential flaw would be adding lots of data with a systematic error (for example, of there are still lingering systematic issues with U-Pb vs Ar-Ar dating, or the offsets in some zircon ages dated using laserICP by Gehrels et al. 2008). But one thing this method does allow is the addition of different types of data.

Vleeschouwer and Parnell (2014) demonstrate this in the second half of their paper, using astrochronology.  This is the appearance in the sedimentary record of depositional changes that are related to Milankovic cycles- the changes in the Earth’s orbit that govern, for example, the extent of ice sheets in the modern Earth.  Astrochronology cycles do not add additional tie points to the geochron/stratigraphy tie curve, as they are relative dates with no fixed reference.  However, what they can do is constrain the slope of the line.  So changes in stratigraphic distance between periodic eccentricity cycles tells us how the stratigraphic progression changes with time, often on a much shorter timescale than what we can access using traditional U-Pb geochronology.  The Bayesian approach allows us to integrate the strengths of both methods, giving a more accurate and reliable timescale.  

The first rule of microanalysis

 There are lots of rules in microanalysis. In the sort that I practice, elemental and isotopic in-situ mass spectrometry, most of these are more what you’d call guidelines than actual “rules.” But there is one rule that is not so malleable; the first rule. And that is this:

Don’t run out of atoms.

Figure 1. The protagonist in this figure has insufficient iron atoms for his proposed procedure.

In ordinary life, we think of atoms a something so gobsmackingly small that they are essentially innumerable in any visible object (like a rock).  And for things you can pick up and throw at crows, that is generally correct.  In fact, even for things that need a handlens or a low power optical microscope to see, there are still a lot of atoms there.  But the combination of small scales (particularly nanoscales), low concentration of trace elements, and high required analytical precisions can all multiply rather quickly to leave one atomically short handed.  Fortunately, atom counting is fairly straight forward.

How many atoms do you have?

A good ballpark number of atoms in crystalline solids is 100 atoms per cubic nanometer.  For the mineral zircon (the Mick Jagger of mineralogical microanalysis), the actual number is about 92.  It is higher in densely packed phases like corundum (119), and lower in loosely packed ones like sanidine (72). Of course, the great thing about cubic nanometers is that you get one BILLION of them per cubic micron.  So a smallish SHRIMP spot (e.g. 15x10x1µm, or 111 µm³) would have about 10¹³ atoms in it, while a large (160x160x50µm, or 1 million µm³) laser ICPMS spot would have 10¹⁷ atoms.

These are heaps of atoms.  

How many atoms could we possibly need?

 This depends on three things.  The concentration of the analyte, the precision of the measurement which is desired, and the detection efficiency of the instrument. Since the presence or absence of individual ions of interest in an analytical volume is a poisson process, the minimum number of atoms needed in the least abundant species measured is the square of the inverse of the desired precision.  So 1% precision  requires 10,000 of the least abundant atom.  A permil requires a million, and 0.1 permil requires 10⁸.

For isotopic ratios, the abundance of the least abundant isotope must be multiplied by the chemical abundance of the element of interest in the sample. And finally, the detection efficiency of the analytical equipment must be considered.

So, the atoms you have are:

~100 x # of cubic nanometers

The atoms you need are: 1/ (required precision^2 * lowest isotopic abundance x volumetric concentration * detection efficiency)

Detection efficiency cam be further broken down into useful yield * dwell time.

Some examples:

A small (100µm³) SHRIMP spot for 3 permil ²⁰⁷Pb/²⁰⁶Pb ratios in a 1.5 billion year old zircon with 50ppm (atomic) ²⁰⁶Pb.

You need:

1/ (0.003^2 (precision) * 0.1 (²⁰⁷Pb/²⁰⁶Pb ratio)  * 0.00005 (²⁰⁶Pb concentration) * 0.02 (Pb useful yield) * 0.15 (dwell time on ²⁰⁷Pb) = 7.4E12

You have:

100 (atom/nm3) * 10⁹ (nm³ per µm³) * 100 (analytical volume- see above) = 1E13

So there are enough atoms, but only barely. All of a sudden, ten trillion atoms seems rather stingy instead of extravagant.

If you want a 0.1 permil ⁴⁶Ca/⁴⁰Ca ratio in a  multi-collector SHRIMP spot of the same size on calcite,

You need:

1/ (0.0001^2 (precision) * 0.00004 (⁴⁶Ca abundance)  * 0.2 (Ca concentration) * 0.1 (estimated useful yield) *0.9 (estimated multicollector dwell time).

1.4E14 atoms required.  

You have:

100 (atom/nm3) * 10⁹ (nm³ per µm³) * 100 (analytical volume in µm³- see above) = 1E13

So you need 14 times more atoms than the measly ten trillion atoms the small spot contains.

Insufficient atoms available.

Even though the target is a major element with great ion yield, the high precision required, combined with the low isotopic abundance of ⁴⁶Ca, means that a larger volume is needed.

Alternatively, a lower analytical precision could be desired, or a more abundant isotope could be targeted. Or, you could choose the option which every instrumentalist wants the inquiring scientist to propose:

“Just use a bigger spot”

A 2000 µm³ Spot is about the largest analytical volume that most sane ion probers would generally use. So although a patient man could do that with SIMS (It’s hard to work with SIMS craters of more than a couple thousand cubic microns), there’s always a shark willing to sell laser beams to the impatient scientist.  Would LA-MC-ICPMS would work?  While the volume of the big laser spot above is sufficient, the order-of-magnitude lower useful yield needs to be considered.  So you need 1.4E15 atoms.  Luckily, a the 1 million µm³ blast hole yields 10¹⁷ atoms.  So this problem can be solved with a bigger laser. And really, who doesn't want a solution like that?

Argon-Argon Dating: The Simple Version

  Sciency Thoughts has a post up on recent high-precision ³⁹Ar/⁴⁰Ar dating of the Toba supervolcano in Indonesia.  Unfortunately he seems a bit confused about the technique.

Argon has three naturally occurring isotopes: ³⁶Ar, ³⁸Ar, and ⁴⁰Ar. Potassium also has three isotopes, ³⁹K, ⁴⁰K, and ⁴¹K.  One of these isotopes, ⁴⁰K, is radioactive, with a half life of about 1248 million years, and one of its stable decay products is ⁴⁰Ar.

 In the universe, and in Jupiter and the Sun locally, ³⁶Ar is the most abundant argon isotope, followed by ³⁸Ar.  In the cosmic scheme of things, ⁴⁰Ar is so rare that we don’t even know what its overall abundance is.

However, Earth is a rocky planet.  It was not able to hold onto much gas during its formation, so there is very little ³⁶Ar and ³⁸Ar here.  Earth has lots of potassium though, so almost all the Ar in the atmosphere is ⁴⁰Ar, which is the decay product of ⁴⁰K.

In a potassium-bearing mineral, the ⁴⁰K decays into ⁴⁰Ar, so you can measure the ratio of these two isotopes to figure out how old the mineral is.

The problem is that it is technically very difficult to measure a potassium argon ration accurately, because one is a reactive solid, and the other is an inert gas.  They require different sorts of ion sources, different mass spectrometers, and there are all sorts of chemical effects that complicate the measurement.

If you want an accurate ratio, it is much easier to measure isotopes of the same element.

So for ³⁹Ar-⁴⁰Ar dating, what happens is that the mineral of interest is put into a nuclear reactor and bombarded by neutrons.  Some of the ³⁹K (the most abundant stable potassium isotope) absorbs a neutron, ejects a proton, and transmutes into radioactive ³⁹Ar.  ³⁹Ar has a half-life of a few hundred years, and is virtually non-existent in nature.   So as long as you know your nuclear ³⁹K to ³⁹Ar conversion ration well, this method allows you to use the ³⁹Ar as a proxy for ³⁹K.  The handy thing is that because it is argon, not potassium, it behaves chemically just like the other naturally occurring argon isotopes, so you can measure it in a gas source mass spectrometer much more accurately than you can measure the chemically different ³⁹K and ⁴⁰Ar.

The initial ³⁹K-⁴⁰K ratio doesn’t very much in nature, and is taken as constant (I think- I’ve never actually done Ar-Ar). But the take-home point is that ³⁹Ar-⁴⁰Ar dating is not its own decay system.  It is the ⁴⁰K-⁴⁰Ar decay system, but using a nuclear reactor to change some of the potassium into an unstable argon isotope to make the nuts and bolts of measuring it easier. 

Odd-shaped lakes in Google Earth

I was goofing around in Google Earth this evening, performing an activity that started out as meaningful and quickly degenerated into a Game of "Ooh what's that", when I came across the following:
Note the very strange shoreline on this lake, with numerous straight line borders. The first time I saw this, I thought I was looking at some dams I didn't know about, but I quickly realized that such an interpretation made no sense.

 Instead, what I believe this image is showing is a mosaic from pictures acquired several years apart. One of those years was a wet year, while another must have been after a period of extended drought. As a result, the lake is ~90% full in some of the images, but almost empty in others. And the straight-line lakeshores are just the tile borders, which Google's new color autocorrect makes less obvious.

 I have no idea where WoGE is up to these days, but I left the co-ordinates off in case anyone wants to chase up the Reservoir.

Migrating dinosaurs and oxygen isotopes

A recent paper made a claim that dinosaurs must migrate, based on the oxygen isotopes in dinosaur teeth. This paper is both awesome and flawed.

The awesome part:

It is hard to measure oxygen isotopes in teeth. Teeth are a mixture of organic matter and two minerals: calcium carbonate, and hydroxyapatite. The organic matter contains oxygen, the carbonate portion contains oxygen, and the hydroxyapatite contains oxygen in two different parts of the mineral; bound water (the “hydroxyl” part, and phosphate ions.

None of these phases are stable in groundwater, and they generally get replaced by other minerals such as silica during the fossilization process. That is why most of the dinosaur teeth you see in museums are black. Even if the teeth aren’t fossilized, the different components will exchange oxygen with groundwater at different rates. Depending on the groundwater chemistry. So simply finding appropriate samples from the Jurassic- 150 million years ago, is not easy.

Secondly, teeth are hard to analyse for oxygen isotopes in lab, because you need to make sure that the oxygen from the different materials isn’t mixed, especially if some of the oxygen has been compromised. This can also be tricky.

Most oxygen has an atomic mass of 16, from having 8 protons and 8 neutrons. However, about 2 in a thousand oxygen molecules have 2 extra neutrons, giving a mass of 18 amu. This oxygen-18 ( abbreviated ¹⁸O) evaporates slightly more difficultly and condenses more easily than normal oxygen-16 (¹⁶O), so rainfall is generally depleted in ¹⁸O. This gives what scientists refer to as a negative δ18O value, which basically means that the rain water has a lower ¹⁸O/¹⁶O ratio than seawater. As air cools, more and more ¹⁸O rains out, so that snow has a strongly negative δ¹⁸O deviation (figure 1).


Figure 1. Tropical lowland rainfall is generally slightly negative in δ¹⁸O (left), while mountain snow in generally highly negative (right).

This leads to the flawed part of the paper.

Fricke et al. (2011) state that because their dinosaur teeth have a variation in δ¹⁸O, the dinosaurs must have migrated from lowlands to uplands (figure 2).


Figure 2. oxygen isotopic variation in dinosaur teeth in interpreted as arising from migration.

But as the low δ¹⁸O snow melts, it forms low δ¹⁸O rivers (figure 3). In an environment with seasonal rainfall, local, tropical rain could give a modest δ¹⁸O depletion, white water draining from high mountains would have a strong δ¹⁸O depletion. This is exactly what Lambs et al. (2005) see in modern day India: The Ganges river, has a δ¹⁸O value of -5, while the Bramhaputra, which flows into India from Tibet, has a δ¹⁸O value of -11. Despite their different sources, both rivers empty out into the same river delta. So in this case, the water is migrating, by flowing down hill.


Figure 3: Water can move as well.

An animal which drank from locally fed streams and ponds during a wet season, but retreated to a river with a distal source in the dry season, would also have a δ¹⁸O anomaly like that of a migrating dinosaur. This would also explain why the dinosaur had more negative δ¹⁸O values when it died; the rock which contained the fossils was a river sand.

This is seen by Dettman and Lohmann (2000) in rocky mountain oysters (fossilized bivalves, you pervs). Shellfish fossils have a δ¹⁸O value that ranges from -5 to -23, all in the same sedimentary sequence. Nobody interprets this as evidence for oyster migration. Rather, it is thought to be caused by rivers with very different source characteristics feeding the same depositional setting. Just like the modern Ganges delta.

So my opinion is that the analytical work and sample selection were very good, but the interpretation is a bit simplistic.

Fricke, H., Hencecroth, J., & Hoerner, M. (2011). Lowland–upland migration of sauropod dinosaurs during the Late Jurassic epoch Nature DOI: 10.1038/nature10570

1. David L. Dettman and, & 2. Kyger C Lohmann (2000). Oxygen isotope evidence for high-altitude snow in the Laramide Rocky Mountains of North America during the Late Cretaceous and Paleogene Geology, 28 (3), 243-246

Lambs, L., Balakrishna, K., Brunet, F., & Probst, J. (2005). Oxygen and hydrogen isotopic composition of major Indian rivers: a first global assessment Hydrological Processes, 19 (17), 3345-3355 DOI: 10.1002/hyp.5974

Orbital cycles, Australian lake levels, and the arrival of aborigines

Australia is a dry country. It is so dry, that the largest drainage basin on the continent has rivers that only occasionally carry water, and drains into a salt pan. Imagine if the Missouri only flowed every third year, or if the Zambezi was a generally a sand-filled channel that crossed a nondescript cliff at what we know as Victoria Falls.

Admittedly, the Lake Eyre basin is smaller than either of these drainages, but only slightly. But for the last 150 thousand years, it has told geologic tales which rival the best Swahili stories or Souix legends. It describes the movement of the Earth against the stars, and the coming of the first people to Australia.

The reason it can tell these stories is that, as a closed basin, the water level of lake Eyre varies dramatically with the amount of water flown in from its major tributaries. So, although the part of central Australia around the lake and the southern part of the drainage is a desert, tropical rainfall in the northern rivers fills it occasionally today, and has in the past allowed a lake many times larger than the current lakebed to exist. Magee et al. (no relation), have carefully and painstakingly reconstructed the history of the lake level over time, and it tells a fascinating tale of alternating floods and aridification over the last 150 thousand years.

What they found is that there have been five periods where a large, permanent lake replaced the current playa. Comparing the lake record to the changes in the Earth’s orbital tilt and eccentricity shows that the lake filling is consistent with wetter conditions- and a more powerful Australian monsoon, being correlated with high sea levels, low ice mass, and high northern hemisphere sunshine.

The exact reasons for this are not discussed in great detail. One is that the outflow from the Asian winter monsoon might be pushing moist tropical air towards Australia more than Australia’s modest monsoon sucks air in. Another point (made mostly in related, referenced publications) is that the warm sea north of Australia- the Gulf of Carpentaria- is shallow, and during times of low sea level, was land. So the northern edge of the Lake Eyre basin was a thousand kilometers from the sea instead of 150, due to the retreat of the Gulf shoreline.

But the other big feature is that the lake-filling events that occurred after 50,000 years ago were much smaller than those which occurred before. Climactically, the conditions 10,000 years ago should have been the same as the conditions 115,000 years ago. But the lake was only a fraction of the size. The authors find no natural causes which can explain this. So they suggest that the aridity starting around 50,000 years ago is related to the reduction in forest and increase in grasslands which occurred at this time. This vegetation change was a result of a huge increase in the frequency of fire in central Australia, which allowed fire-adapted plants to prosper at the expense of moisture-retaining forest. The increase in fire at this time is generally associated with the arrival of the first people on the Australian continent. IT is known that of Australia’s megafauna went extinct at this time, but Magee et al. (2004) show that even the tropical rains were effected by human migration, with drastic changes to the continent’s largest river basin.

---

Magee, J., Miller, G., Spooner, N., & Questiaux, D. (2004). Continuous 150 k.y. monsoon record from Lake Eyre, Australia: Insolation-forcing implications and unexpected Holocene failure Geology, 32 (10) DOI: 10.1130/G20672.1

---

p.s. A few Gene Expression commenters asked a month ago if I could summarize this paper. I hope this helps.

Mass–independent isotopic fractionation

The whole point of geology is to figure out what happened in the past based on the rocks from that time which are still around today. It isn’t actually about the rocks. It’s about the story. The rocks are just the publishing medium. And the craft of geology is learning to read the language of stones.

Similarly, the purpose of geochemistry is to determine the story told by a rock’s chemical composition. The way we do this is somewhat counter-intuitive. We generally search for chemical relationships- that are hard to change. The reason for this is that a ratio that is easy to change doesn’t tell us very much. The potassium/platinum ratio, for example can be changed by just about any process, so measuring it doesn’t tell us what process was occurring.

This is why geochemists like to study systems like noble gasses, rare earth elements, and isotopes. These things are generally changed by only a few processes, so if a change is seen in a rock, there are relatively few processes that could have made the change.

For example, isotopes are nuclei of the same element with different masses. They generally have similar chemical properties- all sulfur isotopes are still sulfur- so only a few processes can change them: evaporation, digestion by bacteria, and diffusion, are some examples. This is the basis of all stable isotope geochemistry; to use the limited number of possible processes to pin down a story by looking at isotopic changes.

In general, when isotopic ratios change, that change is mass dependent. That is, the change is a function of the difference in mass. For sulfur, for example, the change in the ³³S/³²S ratio should be about half of the change in the ³⁴S/³²S ratio. Mass-independent isotopic fractionation refers to a process that fractionates the different isotopes by a ratio that is not strictly mass-dependent. So instead of the ³³SS/³²S change being half the ³⁴SS/³²S change, it might be 0.6. Or 0.3.

The number of causes of mass independent fractionation is exceedingly small- way smaller than the number of effects that cause normal mass dependent fractionation. So if mass independent fractionation is observed, you pretty much know that a particular unique process must have happened.

Most mass independent isotopic work at present is done in sulfur. This is because mass-independent fractionation of sulfur is ubiquitous in rocks from the first half of the Earth’s history, but is rare to nonexistent since that time. So this is a powerful tool that tells us that the Earth’s surface was fundamentally different in Archean time; a process (photolysis of atmospheric SO₂) was occurring from 3800 to 2450 million years ago, and hasn’t happened since. SO₂ is not stable in the presence of oxygen, and photolysis requires UV light that is currently blocked by the ozone layer, so the sulfur isotopic record is the best tool we have for determining just how different the early atmosphere was from the one we breathe today.

Why fuel rods are radioactive

As everyone knows by now, there is a nuclear crisis in Japan. One of the reactors, despite shutting down correctly, has had a number of explosions and has leaked a frightening amount of radioactive material into the surrounding environment. We all know that nuclear reactors produce lots of highly radioactive waste, so hopefully I can clearly explain why.

Every atom has a nucleus, which contains protons and neutrons. The number of protons determines what chemical properties the atom will have, and thus which chemical element it is. Thus, every nucleus with 20 protons is calcium.

In order for the nucleus to be bound together, it must contain roughly equal numbers of protons and neutrons. However, large nuclei need more neutrons than protons.

color coded decay schemes for nuclei

This can be shown in figure 1. This is a chart of the nuclides. The X axis is the number of neutrons, and the Y axis is the number of protons. The little black squares are stable elements. All of the colored squares indicate radioactive nuclei; the color determines the type of radioactive decay.

Radioactive decay is a process in which an unstable nucleus changes into a more stable nucleus, releasing energy in the process. The type of energy released depends on the type of decay.

In figure 1, the yellow and green colors, which are most common in the very large nuclei, indicate alpha decay and spontaneous fission. In both of these cases, the nucleus is too big, and breaks into smaller parts. In alpha decay, the nucleus loses two neutrons and two protons. In spontaneous fission, the nucleus splits into two large pieces.

In figure 1, the blue and pink colors are types of radioactive decay where the nucleus has a ratio of neutrons to protons which is either too high or too low. In the blue scheme, one proton turns into a neutron in one of two processes, called “electron capture” and “beta plus” decay. In the pink area, the nucleus has too many neutrons, and one of the neutrons transforms into a proton by a mechanism known as beta decay.

When a neutron (which has no electric charge) turns into a proton (which has a positive electric charge), it also emits a high energy electron (which has a negative charge), so that the total charge does not change. A particle known as an antineutrino is also produced, but this is harmless. However, many beta decays also release additional energy in the form of a gamma ray, which is a photon, or type of light, that is basically the same as a high energy X-ray.

Large stable nuclei have a higher ratio of neutrons to protons than small stable nuclei. Nuclear fission, the process that powers nuclear reactors and bombs, involves splitting a large uranium (or occasionally plutonium) atom in two by hitting it with a neutron. This releases lots of energy, more neutrons, and creates two much smaller nuclei. If one or more of the neutrons created by fission then causes another atom to fission, then the reaction is self-sustaining, or “critical”.

²³⁵U fission products, color coded by abundance

Fission is a messy process, so the new nuclei produced are not always the same. Figure 2 shows the probability of various possible fission decay products. purple-red is the most likely, while yellow is less likely. As you can see from figure 2, Most of the nuclei most likely to be produced by fission lie to the right of the stable elements, and this have too many neutrons. This means that they must decay via the beta decay process, turning protons into neutrons (and emitting beta particles and gamma rays) without changing the total mass number, until the resulting nucleus become stable. Note that the distribution field is somewhat dumbbell shaped- there is a heavy purple blob and a light purple blob, with a red connector in between.

Figure 3. Fission decay products in the main heavy area of fission yield. Decay shifts atoms up and to the left.

Figure 3 shows the closeup of the area near the centre of the heavy purple blob. As seen here, the elements that are far from stability generally decay rapidly, but in some cases the last decay to stable can take many years. The result is that even if the fission process is stopped, the radioactive decay in the recently created fission products will continue to heat the fuel in the reactor, leading to all the problems we’ve seen on the news.

A few thoughts on the Hungarian Red Sludge disaster

The news has been showing pictures of an environmental catastrophe in Hungary, where a huge spill of toxic sludge from an alumina processing plant has killed several people and lots of fish. I am not an expert on alumina refining, but I’m going to take an educated guess at what happened here using geochemical first principles.

Aluminum is mined from bauxite, which contains lots of aluminum hydroxide plus other contaminant elements. These need to be removed when the ore is processed into pure aluminum oxide.

Aluminum is very insoluble under moderate pH (pdf), dissolving only in very strong acids or bases. So any purification procedure based on aqueous chemistry is probably going to use either very low or very high pH. So their bauxite ore was treated with sodium hydroxide to increase the pH to over 13.

High pH solutions are extremely caustic, and will chemically burn people fish and other organisms just like strong acids do. That is what makes this dangerous.

In addition, most toxic metals are generally more soluble than aluminum is, so any dangerous elements originally present in the ore sill stay in solution. Their concentration will depend in part on whether they accumulate during mineral processing, and how much was in the original ore. That is why they probably don’t have a good heavy metals estimate yet. The same reasoning applies to radioactive U and Th.

As this sludge reacts with things and gets less basic, the solubility of aluminum and iron (a major contaminant in bauxite) will drop, and gibbsite and goethite can be expected to precipitate. The goethite (a.k.a. “rust”) and related iron minerals give the sludge its red hue.

This is all conjecture, however, so I’d love to hear from anyone with actual knowledge.

Astronomical risk calculation

One of the greatest challenges facing media saturated democracies is the inability to correctly gauge the relative importance of catastrophic, visually impressive, rare events vs. common incremental ones. Here are some examples:

If, dear reader, you happen to die in transit, I personally guarantee that you won’t be killed in an airplane blown up by terrorists. The overwhelming odds are that you will die in your own car instead.

Similarly, if you are killed by our energy infrastructure, it won’t be nuclear meltdown that wipes you out. Chances are, you’ll have an asthma attack, or get emphysema from air pollution instead.

And as a geologist, I make the following prediction with even more confidence.

Nobody reading this blog will be killed by a meteorite impact.

It just isn’t going to happen. Fatalities from geohazards are a rare form of death anyway, but if you are unlucky enough to succumb to one, it will probably be a flood or some sort. If not, then an earthquake, or a debris flow. Asteroid impacts are far more spectacular than muddy rivers, but the big ones just don’t happen often enough to create any significant health risk. But that doesn’t stop bolide researchers from scare-mongering in order to get funding.

Just look at the second picture in this Planetary society post from last year. It is an overlay of the Tunguska event on the city of Los Angeles, showing how huge swaths of the city could be incinerated by this relatively modest sized impactor. No mention is given of the likelihood of this event. Luckily for us, though, it is easy to calculate.

Impacts are equally likely on any point on the globe. Thus, the likely hood of an impact in the city of Los Angeles is simply the area of LA divided by the surface area of the Earth. For those who don’t have the numbers on hand, this is about one in 400,000, or 2.5 out of a million. And LA is a very large city; the city limits are only slightly smaller than the state of Rhode Island. Assuming one Tunguska event every century, we’d expect to have to wait for something on the order of 20 million years for a 50% probability of hitting LA. And the rocks LA is built on aren’t even that old.

The most likely scenario for the next Tunguska-style event is for it to happen over the open ocean. And if that would occur, the effect we would see would be…
Absolutely nothing.
A satellite might see such an event. Otherwise, we wouldn’t even know it happened.

So, just for kicks, I pulled an equal area map off of the web, and hit it with 100 impacts, just to see what would happen. I’ll get into the gory details in another blog post, but just to whet your apatite, the picture is here.

The dots are about 130 km across. This is the approximate kill radius for a 400 meter rock under standard assumptions. A rock this big lands about once every 100,000 years, according to the theoreticians, so this is 10 million years worth of impacts.

In future posts I’ll look at some of the details, and what they might mean.

The lead isotope systematics of pregnancy and lactation

Isotope geochemistry is useful for so many things that its application to fetal and maternal health can be overlooked. However, this does not diminish the value of isotopic studies to this field. Consider, for example, the groundbreaking ‘Russian Bride’* experiments of Gulson et al.

Determining calcium loss and lead exposure in pregnant women is not easy. Most of the bodies calcium is stored in bones, and because lead substitutes for calcium in most minerals (biogenic or otherwise), bone hydroxyapatite can also be a potential source for lead, if bone resorption occurs. But detecting this is not always easy.

Bone density measurements can be done by X-rays, but in pregnant women this risks giving the unborn child super powers. Bone biopsies are painful and intrusive. So isotope geochemistry is a much less harmful way of measuring whether or not bone resorption is occurring, and whether or not it is contributing lead to the bloodstream.

In the study reported in these papers, European immigrant women who became pregnant a few years after migrating to Australia were studied. This is important- soft tissues exchange lead rapidly with the environment, whole bone matter can take decade to equilibrate. Most environmental lead in Australia comes from the Proterozoic Mt. Isa and Broken Hill mines. These lie on (and help define) the earth isotopic evolution curve, so have a ²⁰⁸Pb/²⁰⁶Pb ratio of about 2.23. Most Eastern European women who grew up in the Eastern Bloc were primarily exposed to Paleozoic lead with an isotopic ²⁰⁸Pb/²⁰⁶Pb ratio of about 2.10. So their soft tissues and bones should have different isotopic signatures.

Figure 1. An Australian immigrant will have different Pb isotopic signatures in bones and soft tissues.

If such a person starts to resorb bone calcium, this should also liberate Pb substituting for calcium, and the blood Pb isotopic ratio should move towards the skeletal value. This is exactly what Gulson et al. saw during pregnancy and lactation in women who did not take calcium supplements.

Refs
B.L. Gulson, C.W. Jameson, K.R. Mahaffey, K.J. Mizon, M.J. Korsch, G. Vimpani; Pregnancy increases mobilization of lead from maternal skeleton; The Journal of Laboratory and Clinical Medicine, Volume 130, Issue 1, Pages 51-62 (July 1997)

GULSON B. L. ; MAHAFFEY K. R. ; JAMESON C. W. ; MIZON K. J. ; KORSCH M. J. ; CAMERON M. A. ; EISMAN J. A. ; Mobilization of lead from the skeleton during the postnatal period is larger than during pregnancy; The Journal of laboratory and clinical medicine 1998, vol. 131, no4, pp. 324-329


GULSON, B., MIZON, K., KORSCH, M., PALMER, J., & DONNELLY, J. (2003). Mobilization of lead from human bone tissue during pregnancy and lactation—a summary of long-term research The Science of The Total Environment, 303 (1-2), 79-104 DOI: 10.1016/S0048-9697(02)00355-8
* For what it's worth, every Russian bride I knew during the time period of this study had an advanced degree in physics, and was working in Australian academia or industry to win their family's bread.

A different take on the PepSciBlog scandal

Much has been said of the PepSciBlog scandal, both by PepSciblings, the MSM, and the rest of the internet. I won’t repeat or comment on any of those opinions. Instead, I will try to prove that corporate shilling can be useful to basic science education, by using the Pepsi logo to explain the Rossiter-McLaughlin effect. Steinn should be ashamed that he didn’t beat me to this.

Imagine that in the days before spin, the Pepsi logo was designed by physicists, and thus was a featureless, luminous white sphere.


Figure 1. The pre-spin Pepsi logo.

This is fine for the Precambrian, but here in the 21st century, spin is very important. So we will spin the logo.


Figure 2. The spinning logo.

Spinning the sphere means that the side spinning away from us will be red shifted by the Doppler effect, while the side spinning toward us will be blue shifted. This is well illustrated by the modern logo.

This is fine, as long as the Pepsi logo spins alone, in the vastness of space. But interactions with other logos are important. Consider, for example, a transit, or partial eclipse, of the Pepsi logo by the Diabetes Australia logo.


Figure 3. Beginning of a diabetes transit of Pepsi.

If the direction of the transiting logo is the same as the direction of spin from the pepsi logo, then the light from the blue-shifted portion of the logo is blocked first. This makes the average observed Pepsi light somewhat redder. Later, as the Diabetes Australia logo moves to block the red-shifted part of Pepsi, the average light becomes bluer.


Figure 4. Late stage of a Diabetes Australia transit of Pepsi.

Like our hypothetical pre-spin Pepsi logo, stars are (approximately) luminous white spheres. So if they are rotating, the same effect can be observed when planets transit in front of them. More details of this effect can be found at the systemic blog. Thus education triumphs (until the evil lawyers shut me down).

Update:
A former Frink Tanker tells his story.

Why investors need stoichiometry

I was browsing the internet recently, eyeballing some of the ASX releases of various other exploration companies working in the NT, when I noticed the following release (pdf).

The headline announcement is “MgO values of up to 37.9% after removing LOI.”

A brief aside:
Magnesium is commonly mined from magnesite, MgCO3, and a brief description of several mine types can be found here (pdf). Magnesite decarbonates at a few hundered degrees C to MgO + CO₂. As shown in the article above, ore grades are often reported as %MgO after LOI (so pure magnesite would be 100%).

Magnesium is a very common element, though, and is found in all sorts of other less economic minerals, such as pyroxenes, dolomite, hornblende, talc, etc. But it is generally less economical to extract Mg from these minerals. If, as this AXS release suggests, their exploration model is a Kunwarara-type sedimentary magnesite deposit, then they may want to demonstrate that their magnesium is actually in magnesite, and not some other less economically useful mineral.

So I wondered, is the magnesium content of dolomite after LOI greater or less than 37.9%? Dolomite, CaMg(CO₃)₂, decomposes to CaO + MgO+2CO₂, which is lost. Using masses of 15.9994 for O, 24.3050 for Mg, and 40.078 for Ca (via webelements), we get a LOI-removed MgO content of 41.8%. So any bulk analysis greater than this amount must contain some phase that is more MgO rich than dolomite (like magnesite, the stuff we are looking for).

Trouble is, their highest reported value is 37.9% MgO, from what they describe as “dolomite nodules”. Assuming no other Mg bearing minerals, 38% MgO corresponds to a rock that is about 90% dolomite. Now, finding dolomite could be very exciting in some parts of the world where carbonates and magnesium are rare. But the Georgina basin is full of dolostone.

Of course, if there is no calcium in the rock, then their can't be any dolomite, and it could very well contain magnesite. We don’t know what the other 68% of the rock is made of. But given that they report dolomite nodules, and dolomite has a higher post-LOI MgO content than what they found, there is no reason to believe that they have any magnesite.

Read this part carefully:
I am not an investment advisor. Anyone who buys or sells shares based on my advice will probably end up broke. But I am a geochemist. And based on the calculations above, these reported results are consistent with having found a common rock type in the Georgina Basin, and not magnesite.

Happy International women's day

Here's a very simple reminder of the contributions that women have made to the understanding of the Earth:

Laser ICPMS imaging

One of the big developments in geology over the past decade or so is to turn everything into a pretty picture. If it isn’t imaged, it’s mapped, and if it ain’t hyperspectral it needs to be saturated in coffee.

In one sense, this is just a continuation of a trend that started around 1800 with the first geologic map; it’s simply the best way to present complex spatial information. What makes the last decade or so different is simply the scale, speed, and breadth with which this has been applied.

For me, the turning point was EPMA element mapping. This is where the geologist goes out on the town, and while he or she is partying all night long, the electron probe maps out the elemental variation of a thin section. In the morning, the hangover is vanquished with a pretty picture showing such crazy things as REE zoning in optically homogenous garnets, or alteration features that render a year’s worth of work pointless.

In order to map things easily and rapidly, a few things are required. First, you need a fast analytical method. If you’re taking tens to hundreds of thousands of data points, you can’t spend all day on each one.

Secondly, you need to know where you are. You can’t make a map on any scale without positional information.

Thirdly, you need a computer to process the data so that all data is somehow normalized to a useful value, and all the bits are put together in the correct places.

Woodhead et al. show how all this can be done using laser ICPMS. The key for 1 is rapid washout time, and for 2 and 3 are computer-controlled stage operation. They did a bit of extra homework, though, and ran several experiments to show that ejecta resampling produced negligible signal contamination on the scale of their measurements. This is important with destructive mapping techniques, because the sample quality can be compromised by annihilating it with laser beams.

As their example experiment, they show some lovely stalactite images, with U-Mg-Sr spectral maps. You’ll have to read the paper- or guess- to learn which element the vertical axis represents.

Jon D. Woodhead, John Hellstrom, Janet M. Hergt, Alan Greig, Roland Maas (2007). Isotopic and Elemental Imaging of Geological Materials by Laser Ablation Inductively Coupled Plasma-Mass Spectrometry Geostandards and Geoanalytical Research, 31 (4), 331-343

Sponges lived on the snowball Earth

The results I briefly mentioned hearing the preliminary talk for in Melbourne 3 years ago have just been published in Nature.

Short version: Sponges evolved before the end of the Marinoan “Snowball Earth” glaciation, and were common from then through the Ediacaran and into the Cambrian.

How they did it:

As anyone who is conscious about healthy food knows, animal fat contains lots of cholesterol, while plant fats don’t. This fact has been used for years by people worried about their cardiovascular health. But it also lets us explain the evolution of animal life.

At the risk of slightly oversimplifying, there are three kinds of cholesterol: Good cholesterol, which is found in fit people. Bad cholesterol, which is found in bacon rinds. And 24-isopropylcholesterol, which is found in the cell membranes of certain kinds of sponges.

When organisms die in anoxic environments, the organic remains can sometimes accumulate and alter to form petroleum. 24-isopropylcholesterol is altered by this process to form a similar chemical called 24-Isopropylcholestane. 24-Isopropylcholestane is not known to form in any other way. So Love et al. extracted oil from a bunch of Cryogenean and Ediacaran sediments, fed it through their fancy mass spectrometer, and identified lots of 24-Isopropylcholestane.

Of course, oil can migrate- the gasoline in your car migrated from its source rock to a reservoir, and then to a well, to a supertanker, and to a refinery, before entering your gas tank via a pump. So the cool thing that these authors did was to use hydropyrolysis to show that the kerogen also contained 24-Isopropylcholestane.

Hydropyrolysis uses hydrogen and heat to break hydrocarbons apart. Kerogen is basically a large, insoluble organic compound of some sort. If a sterane is bonded to that kerogen, it cannot migrate, until broken off in the lab.
Hot hydrogen attacks the bond between the molecule and the kerogen:

to release the molecule:

Love et al. use this to show that the bound hydrocarbon population is similar to the unbound population.

When organisms die, they don’t just leave characteristic bones behind. They also leave molecules. And those molecules can be just as diagnostic as more traditional fossils.

For more details, read the paper. Or see Callan's take.

Gordon D. Love, Emmanuelle Grosjean, Charlotte Stalvies, David A. Fike, John P. Grotzinger, Alexander S. Bradley, Amy E. Kelly, Maya Bhatia, William Meredith, Colin E. Snape, Samuel A. Bowring, Daniel J. Condon, Roger E. Summons (2009). Fossil steroids record the appearance of Demospongiae during the Cryogenian period Nature, 457 (7230), 718-721 DOI: 10.1038/nature07673

Mercury fly-by

If all went well, the Messenger spacecraft is 50 minutes past closest approach of its second flyby, and is currently 20,000 km above the fastest planet and receding rapidly. The spacecraft has turned away from Earth to perform its mission, so it will be another 19 hours before contact with the spacecraft is regained. I wonder if anyone on the mission will manage to get any sleep during that time. Here is the penultimate approach picture taken before the spacecraft turned away: