Geomagnetism.org

So back in April, the lab received our brand new 2G Vertical SQUID Magnetometer with RAPID automatic sample changer. It is now up and running and the first lucky people have had the chance to use it. One of the lucky few is our geophysics undergraduate Ed Sage, who in between measuring, has made a much appreciated spoof featuring the new equipment (if you are not familiar with the 1960’s show ‘UFO,’ then you may want to familiarise yourself with the opening sequence before watching this video).

Disclaimer: The dot matrix printer is not a part of the amazing new equipment we have received, if you couldn’t already guess.

The Summer School participants at the end of School dinner (Photo credit: Mostafa Ahmadzadeh)

So I’m a little late in posting this but it wasn’t too long ago that I attended the Institute of Rock Magnetism’s Summer School. Generally the School is only run every two years but, due to changes to the labs next summer, it was run this year as well as last year… and I’m glad it was. I highly recommend this School to any graduate students working in paleo-, archeo-, rock- and environmental magnetism. It answered a lot of the questions that I’ve had since starting my PhD and it was a fantastic opportunity to get to grips with some amazing magnetic equipment that I’ve not had the opportunity to use before. While the lectures by Bruce Moscowitz, Mike Jackson, Dario Biladrillo, Josh Feinberg and Lennart de Groot (who popped over from San Diego to give a talk on paleointensity) were all fantastic, what I really want to talk about here is the group project that I was a part of; looking at the suitability of Peruvian pottery for studying the Levantine Spike.

For those of you who are unfamiliar with the Levantine Spike, it is thought to be a rapid increase and decrease in field intensity to 3-4 times the background field intensity at around ~1000B.C. The spike was initially observed by Ben-Yosef et al. (2009) in the waste from Iron Age copper production in South Jordan but has since been seen in Israel (Shaar et al., 2011), Turkey (Ertepinar et al., 2012), Texas (Bourne et al., 2016) and probably the Canary Islands (de Groot et al., 2015) and China (Cai et al., 2014).

Since its discovery, there has been a great deal of controversy surrounding this geomagnetic spike. When the original paper was published, it seemed difficult to accept because a) the spike is not evident in the European record, which is the most complete archeomagnetic record available, b) the magnitude and the short duration of the spike (possibly as short as 30 years according to Shaar et al., 2011) could not be reconciled with core flow models and c) the copper waste products that the spike was initially observed in are quite different from the lavas and ceramics that are usually used in paleointensity experiments and so the reliability of the results was questioned. The geomagnetic spike has gained greater acceptance since, as it has been observed across other locations and in materials more conventionally used in palaeointensity experiments. However there are still problems reconciling the size and length of the spike with our models of core flow.

Even using the longest possible estimates for the spike, approximately 400 years (Bourne et al., 2016), the spike is only just resolvable with the maximum possible rates of field intensity change modelled, assuming that the spike is the result of a local flow feature at the outer cores surface (Livermore et al., 2014). These models also suggest that we should see the spike in the nearby Greek and Bulgarian records but this is not the case. Another possibility is that the spike may be a manifestation of columnar convection; this has been suggested based on evidence of paired lobes of relatively rapid geomagnetic variation in the northern and southern hemispheres (the Middle East and Africa) over the last 200 years (Shaar et al., 2015). However, a single column could not explain the expression of the spike at sites as far apart as the Middle East and Texas.

Another reason we are interested in constraining the nature of the Levantine Spike is because of the problems with Radiocarbon dating around this time. This technique works by measuring the amount of Carbon-12 (12C), the most common isotope, and Carbon-14 (14C), the radioactive isotope of Carbon, in organic materials. The 14C is produced in the upper atmosphere and then forms carbon dioxide that is taken up by plants and trees. When they die, the 14C decays and so the amount present in the plants/trees reduces over time. The technique assumes that the amount of 14C produced in the upper atmosphere is a constant, so the starting ratio of 12C to 14C should always be the same and, since the decay rate of 14C is fixed, you should be able to determine the age of the material from the amount of 14C left in the sample. However evidence from independently dated tree rings has shown that between 800-400BC the amount of 14C observed in organic materials doesn’t change during this period, suggesting something was affecting its production in the atmosphere; this period is known as the “Halstatt plateau.” For this time period, we need to rely

The team hard at work on our presentation (Photo credit: Mostafa Ahmadzadeh)

For the group project, we had a set of samples from a Peruvian site, from before, during and after the Levantine Spike, to check for their suitability for paleointensity experiments. Results from these samples could improve the paleointensity curve for this region, which would help with field modelling and archeomagnetic dating. They could also answer the question as to whether the spikes represent columnar convection as they roughly line up with the American site.

To assess their suitability, there are several criteria that the samples need to fulfil; the first of thing is to identify the magnetic minerals in the sample to check that their Curie temperatures are high enough to capture the primary remanence component. For this we used the Magnetic Properties Measurement System (MPMS) to identify the low temperature magnetic mineral transitions (down to 2° Kelvin), which allowed us to use the same samples for hysteresis in the Vibrating Sample Magnetometers (VSM). Both the hysteresis and susceptibility, done on the Kappabridge, were measured to high temperatures to give us the Curie temperatures of the samples and the temperature at which alteration of our remanence carriers was occurring. We also looked at the behaviour of the samples during a Pseudo-Thellier experiment (as it’s much quicker than thermal Thellier experiment); looking at the AF demagnetisation for any overprints and to check that the characteristic component goes to the origin and the ARM directions told us whether the samples were anisotropic or not. In the end, we decided we would recommend at least half of these samples for thermal Thellier experiments and this was our team’s conclusion when we presented our work at the end of the Summer School.

“That’ll do pig”

Finally, the Summer School came to its end and there was just one thing left to do. For anyone who hasn’t visited the IRM, one of the things it is well known for is the ‘pig book.’ This is a right of passage for any Visiting Fellow or Summer School participant who comes to the IRM where they must draw a pig while blindfolded. The original ‘pig book’ belonged to Sir (James) Alfred Ewing, the father of the hysteresis loop, who used to invite his famous guests such as Winston Churchill, Alexander bell and Sir Arthur Conan Doyle, to leave an entry. On this note, I will finish with my entry to the ‘pig, book’; see if you can get the movie reference…

I’d like to thank everyone in my team, “Team World Domination,” who were brilliant to work with; Patrick Arneitz (ZAMG, Vienna), Steve Victor and Michael Woodburn (Yale), Jackie Smale (Minneapolis) and our supreme (or at least team) overlord Josh Feinberg (IRM). I would also like to thank Mike Jackson for arranging the Summer School, all the lecturers, who I’ve already mentioned, and everyone who participated in the Summer School for making a thoroughly magical Summer School, especially my Roommate Courtney Wagner (Utah), who made sure that I had an amazing time.

Photo credit: Mostafa Ahmadzadeh

Delegates at the SEDI 2016 meeting – a prize to anyone who spots me…

All the standard text books say that Earth’s magnetic field is driven by convection. But what if it isn’t? I recently attended the SEDI (Studies of Earth’s Deep Interior) meeting in Nantes, France and was impressed by the work of Benjamin Favier and others looking into possible alternatives to the standard model of providing the power needed to drive the electric currents in Earth’s Outer Core that are responsible for our planetary magnetic field. These alternatives fall under the umbrella of “mechanical driving processes” and, as far as my limited understanding in this highly specialised field goes, the process can most easily thought of as “stirring” the fluid the outer core much as you would tea in a cup by doing work against resisting forces. However, clearly there is no tea-spoon involved. Rather, at the risk of pushing the analogy too far, the cup is being jiggled from the outside and it is the fluid’s interactions with the “container” (in this case the rest of Earth) as it sloshes about that is producing the flows.

So, what is doing the jiggling? We have already known the answer to this for a long time: interaction between the gravitational fields of Earth, the sun and the Moon cause variations in Earth’s shape and rotation. Because of the core’s elliptical shape, these variations are efficient means of converting gravitational energy into mechanical work (flows in the liquid).

The talk by Benjamin Favier (page 34 of this document) specifically focused on tidal forcing and on librations (small oscillations in Earth’s orbit). In it, he showed that these processes are very efficient at producing small scale flows within the outer core because of its very low viscosity and very high rotation rate. Together, he argued, they could easily account for the magnetic field we see today.  So, do they? After all, to paraphrase a comment made after the talk – “We KNOW the Earth is subject to such forcing, but we don’t KNOW  that the core is convecting”. Well, the jury is still a long way out. Benjamin confidently predicted the first fully self-consistent numerical mechanically forced dyanmos within a few years. The real challenge then will be to look at the magnetic behaviour they produce and compare it to that observed in the real Earth. The “Mechanical Forcing Brigade” (my name) face something of an uphill battle for the recent Earth because the convection-driven numerical dynamos are already doing a remarkably good job at replicating observed behaviour. My interest lies much further back in time. Before the inner core nucleated, it was much harder to generate buoyancy driven flow in the outer core; perhaps mechanical forcing had a (primary? Sole?) role to play in generating the magnetic field back then? Only comparisons between the next generation of dynamo simulations and newly updated palaeomagnetic records will tell us. I look forward to being part of that…

Pop-artist Andy Warhol famously stated that: “In the future, everyone will be world-famous for 15 minutes”. I suspect that yesterday may be the closest we will ever get to proving him right.

A paper on which I am lead-author claims that we have may have pinned down the point in Earth’s history when the inner core first started to freeze at the centre of the Earth to between 1 and 1.5 billion years ago.  I already thought this was big news so was a bit deflated when Nature decided not to run with the excellent picture (left) created by Kay Lancaster (cartographer at the University of Liverpool) on its cover.

Nevertheless, our excellent press officer at Liverpool helped produce a great press release which saw a story featured on the popular Phys.org website from the outset and an article in one of Spain’s top newspapers El Pais.

Things were a bit slow-burning for a while – except in India and Finland – before a  piece by Simon Redfern appeared on BBC news online. This was quickly followed up by a piece on the Daily Mail which our press officer tells me is the “most read online news site in the world”. A number of other things have followed including a post on one of my favourite blogs – IFLScience.

Then, just as I was packing up to go home, I received a phone call from the BBC World Service who wanted a short interview. I obliged in the evening and my nervous responses aired a few hours later. You can listen to the podcast here (it is the very last feature – “And finally…”). They refer before and after the interview to the finding as being that the magnetic field is much older than previously thought – incorrect in this specific case but relevant to another recent finding, albeit one that Liverpool people were not involved in making.

More informative is a piece I wrote for “The Conversation”.

A summary from our press office indicates that there are 39 news outlets and counting featuring the story  and tweets still coming through every few minutes. The coverage extends over at least 11 countries ranging from USA to China,  Argentina to Pakistan so, while I can, I am claiming (brief) world fame for our research…

Recently, I received an interesting email from Dmitry who wrote the following:

“According to the Dynamo theory Earth’s magnetic field is generated by rotating, convecting and electrically conducting fluid that maintain a magnetic field.

But what if it creates just by solar wind itself? In this case we don’t need to invent anything else to explain magnetic fields of planets. This process is similar to the principle of electromagnet.

Solar wind magnetizes planets. Now most of it just flies away of the Earth, but a small part of it passes mainly through the north and south poles.

This phenomenon may be also involved in heating the Earth’s core due to the conductor resistance, in the same way as wire is heated when electricity passed through it.”

Let me try to respond to this one piece at a time:

“According to the Dynamo theory Earth’s magnetic field is generated by rotating, convecting and electrically conducting fluid that maintain a magnetic field.”

Spot on – the fluid is in the Earth’s liquid outer core at a depth 2871 km and is convecting because of the release of light elements by the freezing of the inner core beneath it.

“But what if it is created just by the solar wind itself?”

This is already partly true. A small proportion (around 1%) of the Earth’s magnetic field (referred to as the external field) is produced by the interaction of the solar wind (charged particles from the sun) with similar particles within the Earth’s magnetosphere. This part is much weaker, more complex, and changes faster in time than the main part of the field.

“In this case we don’t need to invent anything else to explain magnetic fields of planets.”

Aah but we do. Venus is, in many ways, similar to the Earth and receives more solar wind (being closer to the source) and yet has a field that is 100,000 times weaker than Earth’s. The explanation? No core dynamo. And why not? Difficult to say for certain but likely because the core is not cooling fast enough to fuel a dynamo process.

“Solar wind magnetizes planets.” 

Yes it does but the effect is very small and relatively shallow. A subsurface exploration technique known as magnetotellurics measures the magnetic field induced in the ground by as a consequence of electric currents produced by the solar wind. The measured field variations are typically about 0.001 nT (nano-Tesla) over a period of hours and the resulting electrical currents go to a depth of a few tens of km. Compare this to the main internal field which has an intensity at the surface of around 40,000 nT and seems to come from the core with a depth of nearly 3,000 km. It is a small player.

“Now most of it just flies away of the Earth, but a small part of it passes mainly through the north and south poles.”

All true but the strongest external field variations are felt at the poles because that is where the existing internally-generated field is the steepest. Therefore, it is here that the solar wind particles can most easily travel down the field lines into the atmosphere.

“This phenomenon may be also involved in heating the Earth’s core due to the conductor resistance, in the same way as wire is heated when electricity passed through it.”

The principle here (Ohmic dissipation) is correct but, in practice, I would expect the core heating from this process to be vanishingly small. Pretty much all of the external field will be attenuated by the crust and converted (“dissipated”) to heat generated near the surface so none will be left to penetrate the core. Even close to the surface, the heating effect is very very minor compared to that coming from the Earth’s interior. A back of the envelope calculation gives, very approximately, a heating rate at a few tens of km down of about one-quadrillionth of a degree Celsius per second from these “telluric” currents. Compare this to natural geothermal gradients of tens of degrees per km and you get the picture that this is small-fry indeed.

I hope this post has answered your questions Dmitry and is useful to other readers. I would be very happy to respond to similar emails (or, even better, posts on this website) in the future.

In a recent post on this blog, Prof. Richard Holme guaranteed us that a field reversal won’t affect human life for “many lifetimes to come”.  This raises the question, how fast can Earth’s magnetic field reverse?  A recent article by an Italian-French-American research team (Sagnotti et al., Geophysical Journal International)  has garnered the attention of the global media over the past couple of weeks (Le Monde, Der Spiegel, phys.org).   In this study they calculate that the field could change very quickly, just less than 2 degrees per year, i.e. the field could completely flip polarity within a century.  This is very much quicker than the majority of previous estimates, which suggest it takes a few thousand years for a flip to occur, and quicker than the rate of directional changes observed today (a few degrees per century).  Indeed, there is only sparse evidence for such quick changes in the geological record, the most notable being a reversal recorded ~16 Ma in the Steens Mountain lava flows, SE Oregon, and a recent excursion (the Laschamp, ~41 ka) recorded in sediments from the Black Sea.  The Steens Mountain lavas were initially calculated to record changes in the direction of the field up to a phenomenal 3 degrees per day.  However, continued research on the Steens lavas over the last 30 years, headed by Prof. Rob Coe at the University of California-Santa Cruz, led his research group to recently conclude that an interesting phenomena in how the magnetization was acquired by these lavas gave rise to these apparent rapid changes in direction. The directional changes estimated from sediments from the Black Sea were quick (half a degree per year), but quite a bit slower than those from the Sagnotti et al. study and nothing like as fast as those recorded by the Steens Mountain lavas.

How did they reach their result?

The Sagnotti team investigated exposed lake sediments in the Apennines, Italy, covering the time of the most recent geomagnetic field reversal, the Matuyama-Brunhes reversal.  Some debate exists over the age of this reversal, but it occurred approximately 780 ka (this is an ongoing area of fervent research).  As sediments are continually deposited over time they can provide a detailed record of changes in the geomagnetic field depending on how much material is deposited per unit time.  Within these sediments are a number of tephras (ash deposits from volcanic eruptions), which can contain excellent material for radiometric dating, e.g., crystals of sanidine.  In this case researchers from two different labs used the ratio of argon-40 to argon-39 to estimate the ages of the ashes from a large number of experiments on sanidines.  They found the Matuyama-Brunhes reversal lay between two of the dated tephras and by assuming sedimentation was constant between the average ages of their newly determined tephra, estimated the age and the duration of the reversal.  The result: directional changes of just less than 2 degrees per day.

Does this mean the field could completely reverse in less than a human lifetime?

Changes in direction aren’t the whole story.  Although observations on the rate of directional change are interesting, the Earth’s magnetic field has another important component, namely its strength or intensity.  Evidence to date suggests the intensity of Earth’s magnetic field decreases during a reversal and this decrease brackets the polarity flip, but may last considerably longer than the flip itself.  It is likely the decrease in intensity is more intimately linked to underlying processes in Earth’s outer core than the directions themselves, although we don’t exactly know how the reversal process is initiated.  Work done at the University of Liverpool by me and Richard Holme (find here) suggests the length of the polarity flip could be variable depending on the location of the observation, e.g., the duration of the polarity flip could last twice as long in Australia as in Italy, yet the underlying pattern of intensity decrease could be similar. This observation is supported by numerical and empirical modelling of the reversing field.  Even if the polarity flip is short at some locations, the underlying process generating field reversals could be much longer.

In regards to the ability of Earth’s magnetic field to shield Earth from cosmic radiation, it is the strength of the field as well as its configuration that is important, and the field doesn’t fully vanish during a reversal.  Earth’s atmosphere also does a sterling job of protecting the Earth.  It is important to remember the genus Homo have lived through a number of reversals and excursions of the geomagnetic field, such as the Matuyama-Brunhes reversal and more recently the Laschamp excursion.

Is it better to burn out than fade away?  Some reversals may live hard and fast and others may like the quiet life.

ESA’s magnetic field mission Swarm
Image: ESA

Today’s (7/7/2014) Metro newspaper has a nice article in the Metrocosm section by Ben Gilliland entitled “Portends of magnetogeddon” . Apart from the slight alarmism, it’s (as usual with Ben) pretty good, and I’m pleased to see him talk up the magnetic field even more than I usually do! It seems to be driven by the current ESA satellite mission Swarm, which is measuring the near-Earth magnetic field with the best ever satellite instruments on three separate spacecraft, with a view to separating magnetic sources external to the Earth (like the sun, auroral currents, and the magnetosphere) and those internal (from the core, and from magnetised material at Earth’s surface).  However, much of the article is focussed on more general information, such as magnetic reversals (well known to readers of this blog!) and the possible effects of weakening of the field. That the field is dropping in magnitude is not a new result – we’ve known this for a century or so, and now have evidence that this has been happening for at least 400 years (and probably longer) but it gets a lot of publicity every time we have new measurements – I remember something similar happening when the Ørsted mission was launched in 1999. There was a nice documentary on this over 10 years ago, joint between WGBH Nova for PBS in America, and Channel 4 in the UK called “Magnetic Storm” (“Magnetic flip” in the UK) from which most of the information is still accurate. If you are interested, you can find it online. The issue is that the magnetic field provides us with a shield from harmful solar and cosmic radiation, and so if it weakens (and potentially reverses), this shield would be weakened. However, the answer to the initial question is

Don’t Panic!

(Readers of a certain age may recognise the source of both this response and the slight misquote of the title…..) The answer from the documentary is that the field will weaken, but not completely disappear, it isn’t going to happen for the next 1000 years or so if it does then, and even if it does, the protective measures required will be not to lie on the beach in Florida, and to wear a large floppy hat!

Whether the field is reversing is not known, and (to an extent) not knowable. We have models made from observational data and palaeomagnetic measurements (some created here at Liverpool) which show that in the last 10,000 years, the magnetic field has both decayed more rapidly than it is at present, and been substantially weaker than it is now, but then recovered. Therefore, the current fall in the magnetic field could be simply part of its normal variation. Things are a little more complicated, as there is a strong and growing feature of the “wrong” polarity at the surface of the Earth’s core under the South Atlantic, giving rise to the weaker area of field that we see as the “South Atlantic Anomaly” – which causes occasional havoc in the instruments of satellites flying overhead. This feature could be a signature of a major change in the field, but it could also simply reflect rearrangement of magnetic field at the core – if this becomes more complicated, it could look weaker from the surface without the field itself actually getting any weaker.

What Swarm is really for is getting a much more detailed picture of small-scale structures in the magnetic field, and its rapid changes on time scales of years – so magnetic “weather”, rather than the “climate” of a field reversal. Hopefully, this will give us a much better idea of these smaller variations – my interest is particularly in comparing these variations with changes in Earth rotation.  If this work is successful, we may yet also have something to say about the longer term decline in the field, although I can guarantee that this won’t affect anyone for many lifetimes to come!

Richard Holme

Today’s post was guest written by Richard Holme.   Richard Holme is Professor of Geomagnetism at the University of Liverpool. His research interests are in the behaviour of the magnetic fields of the Earth and other planets on timescales of milliseconds to billions of years, and variations in Earth rotation and their link to Earth’s core. He is involved with the ESA geomagnetic mission Swarm, is looking forward to July 2016 for Juno to arrive at Jupiter, 2017 for Cassini to begin close orbits of Saturn, and 2033 (?!) for the arrival of Juice at Ganymede.

Geochemistry, Geophysics, Geosystems (G3 or G-Cubed) is an online-only journal of the American Geophysical Union (AGU). It’s format means it is perfect for publishing papers with large supplementary information or appendices. That is definitely the case for this latest publication, lead-authored by Greig Paterson – a Scotsman in Beijing.

Measurements of the strength of the ancient magnetic field recorded in rocks  or archaeological materials (“palaeointensity” or “archaeointensity” measurements) can be very tricky to make. It would therefore be very useful to have some agreement amongst those people doing these experiments about what a reliable measurement looks like.  However, for one reason or another, such consensus has eluded the community for decades. This paper, we hope, is a step towards rectifying that. It uses the largest ever compilation of palaeointensity measurements made using only materials for which we know what the answer “is” (e.g. from lavas that cooled in recent times during which the magnetic field strength is independently known from observatory data). It then finds variants of currently used selection criteria that are better at picking out the reliable measurements than the originals. This stuff may not sound to the outsider as exciting as say, finding a new reversal, but it is the type of thing that underpins palaeomagnetism as a tool for understanding our planet better.

Philippe and Valerie taking archaeomagnetic samples from a Roman aged achaeological site in France

For regular readers of this blog, the term “archaeomagnetism” will have been seen a number of times, most frequently in posts by me or Andy Herries!

Archaeomagnetism is the study of burnt material found on archaeological sites. This can include everything from hearths, fireplaces and kilns through to tiles, bricks and pottery. Basically anything that has been subjected to heat at some point, either deliberately (e.g. the firing of a pot) or accidently (e.g. if a fire burns down a building, its foundations and walls become suitable for archaeomagnetic study as a consequence).

In certain parts of the world (for specific time periods), it is possible to date archaeological samples by comparing the declination, inclination and intensity values recorded in the archaeomagnetic samples (these 3 values describe the geomagnetic field vector) with the known changes in the geomagnetic field. One of the great things about archaeomagnetism (in my opinion) is the variety of ways in which you can use it. Associate Professor at Liverpool, Andy Herries, focuses on dating Hominid sites in South Africa by dating speolotherms (speolotherms are also known as flow stones and are created through the deposition of carbonate through time. Stalactites and stalagmites form in the same way.) see http://www.ncbi.nlm.nih.gov/pubmed/21392817.   It is worth noting that in speoltherms the record of the magnetic field is preserved as a chemical remanant magnetisation rather than a thermal remament magnetisation.

I myself am focused on trying to gather well-dated archeointensity data from Turkey for the dual purpose of enabling archaeomagnetic dating in the future as well as allowing us to accurately reconstruct changes in the geomagnetic field. This work is very interesting because there have been a number of recent papers linking sudden changes in the magnetic field with climate change and civilisation collapse e.g. http://www.sciencedirect.com/science/article/pii/S0012821X06002792.  I’m investigating whether I can see any evidence for this.

In this post I have only briefly mentioned some of the applications of archaeomagnetism: it is a fascinating field with new methods and ideas being tested every day.