Peter Driscoll: What Goes On Deep Inside Earth’s Magnetic Dynamo?

Peter Driscoll studies the dynamical evolution of Earth and planetary interiors as a DTM Staff Scientist. Photo: Roberto Molar Candanosa, DTM.
Thursday, November 01, 2018 

Earth's global magnetic field shields us from the dangers of space, including harmful levels of Solar radiation. This shield is one of the major reasons why life has survived on Earth, and it all starts in the core, where the geomagnetic field is generated by the turbulent motion of liquid iron deep inside the planet. It's the so-called geodynamo. Scientists like DTM's Peter Driscoll devote their career to understanding the nature of such planetary dynamics.

In 2016, Driscoll's research suggested that in ancient times Earth may have been significantly different, with prolonged periods of weak field intensity and strange multipolar geometry with many poles in contrast to the modern north-and-south pole orientation. Now, in a recent paper co-authored with DTM's Cian Wilson, Driscoll finds that geologic evidence for such a complex magnetic field will be very difficult to identify. In preparation for his Neighborhood Lecture on November 8, 2018, Driscoll answers some questions about his latest research and why studying the geodynamo is critical to understanding the history of our planet.

Q: What exactly did you set out to do in your latest paper?

A: We were looking at the results from computer simulations of Earth's magnetic field, or dynamo, from an observer's perspective. The idea is that we have all these simulations on one hand, and all this paleomagnetic data, records of ancient rock magnetism, on the other. But we're not sure how to compare them.

So, we took a paleomagnetic approach to "observing" our dynamo simulations, where the "observer" is a rock on Earth's surface containing magnetic minerals that record the ambient magnetic field at the time they solidified. Rocks that contain magnetic minerals (like those with significant iron or other conducting elements) are oriented by Earth's magnetic field much like a compass needle as they solidify over thousands to millions of years. If they are well preserved, these magnetic minerals become locked in place, revealing the orientation and intensity of the ancient geomagnetic field. To simulate this kind of observation, we sample our simulations at points on Earth's surface and average them over time.

Schematic of how magnetic minerals (inset) inside rocks are aligned with the ambient geomagnetic field as they solidify. These rocks can preserve the direction and intensity of the geomagnetic field over billions of years if they avoid high temperatures and magnetic contamination. The geomagnetic field fluctuates in geometry and intensity over thousands of years. Such magnetic rocks may cool quickly and preserve the geomagnetic field at an instant in time, or can cool more slowly over thousands of years and preserve a kind of time average of the varying geomagnetic field. Illustration: Roberto Molar Candanosa and Peter Driscoll, DTM.

For paleomagnetic rock samples, we don't know their location on Earth's surface at the time of their magnetization/solidification because the tectonic plates that make up Earth's crust move slowly over millions of years. The locations of samples younger than about 180 million years old can be reasonably well repositioned using the seafloor spreading record, but the locations of older samples, particularly older than 500 million years, are much harder to reconstruct. In fact paleomagnetists attempt to infer the paleo-latitude of a rock from the orientation of its internal magnetic field by assuming that the Earth's magnetic field is a geocentric axial dipole (or GAD), the simple dipole magnetic field oriented along Earth's rotation axis that we are familiar with today

It is expected that the geomagnetic field should conform to a GAD field when it is averaged over a long time, but this is an assumption we wanted to test with our models. How long is long enough? So, we generated synthetic "observations" of our simulations over millions of simulation years. From the local time-averaged magnetic orientation of each synthetic observation we computed an "expected" latitude using the GAD assumption. Then we compared this "expected" latitude to the known latitude. What we found was that the they do not always agree, implying that the GAD hypothesis does not always hold! In fact, it only holds under certain conditions: at mid-latitudes (around 45 degrees) and only when the underlying magnetic field orientation is rather stable. We found that synthetic observations at high (>60 deg) and low (<30 deg) latitudes tend to produce errors (or "biases") in the observed orientation and intensity of the magnetic field. By "bias" we mean that the observations were systematically and predictably different than expected. Similarly, for simulations that underwent regular polarity reversals, when the dipole orientation changes 180 degrees, we found significant differences between the synthetic observations and known values.

Schematic comparison of magnetic field lines for a non-dipolar field in ancient Earth (left) and a dipolar field in modern Earth. The geomagnetic field spends most of its time in a dipolar state but occasionally slips into a non-dipolar state. Most continental reconstructions (inferring the location of tectonic plates back in time) assume a geocentric axial dipole (GAD) field, but could be misleading during times when the field was less dipolar. Driscoll and Wilson (2018) found that non-dipolar fields can also lead to misinterpretations of the paleointensity (magnetic field strength back in time). Illustration: Roberto Molar Candanosa and Peter Driscoll, DTM.

Why is the initial solidification of the inner core the singular most important event in core history?

Well, this I suppose is a somewhat subjective statement. Perhaps there are other equally important events in core history, but the initial solidification or nucleation of the inner core minimally changed the structure of the core. It also likely caused a big jump in the dynamics of the core, as latent heat and light elements rejected from the solid were all of a sudden introduced at the base of the core. These sources likely generated new kinds of fluid flow and magnetic induction in the outer core that may be observable in the paleomagnetic record. Scientists have been searching for such a "signature" of inner core nucleation for years without much luck. Some datasets show simultaneous jumps in magnetic intensity, but some datasets don't. It's not particularly clear from the paleomagnetic data that there ever was a global simultaneous jump.

If we were able to identify this inner core "signature", why would that be important?

Pining down the timing of inner core nucleation would give us an incredibly valuable data point, a temperature at the center of the Earth at a particular time. This data point would be a new constraint on the internal structure and thermal evolution of the Earth, which is critical to understanding the history of the Earth. If this signature is old (several billion years), that would imply that the Earth initially cooled rapidly and has been cooling slowly ever since. On the other hand, if this signature is relatively recent (around 0.5-1 billion years) as has been recently proposed, that would imply that Earth has been hot for a long time. It is also possible that the timing of inner core nucleation may correlate with other large scale changes in Earth history, such as the timing of large igneous provinces or the emergence of complex life. (Yes, there actually is a line of argument saying that complex life required a strong magnetic field to shield the surface, but let's not go there just yet.)

What exactly goes on in your models?

The dynamo models are numerical simulations of a fluid (analogous to liquid iron) convecting and transporting heat in a rotating and heated spherical shell (analogous to Earth's outer core). The convecting fluid sustains complex three-dimensional turbulence, which generates magnetic fields by the motion of the conductive fluid. This magnetic field is then extrapolated from the top of the dynamo model (outer core) to Earth's surface in order to produce the synthetic observations. The biggest caveat of the dynamo models is that they rotate much, much slower than the Earth due to numerical constraints. The numerical impediment is so severe that there aren't enough computational resources in the world to simulate the actual conditions in the core for any reasonable amount of time, so we are forced to rely on slower models and then make efforts to scale the results towards Earth-like conditions the best we can.

Animation of magnetic field lines during a simulated magnetic reversal. Magnetic field lines emanate from the core-mantle boundary (grey circle) and pass through Earth's surface (black circle). During the magnetic reversal the field lines fluctuate wildly, a substantial departure from a geocentric axial dipole. Animation courtesy of Peter Driscoll, DTM.

Why do you study terrestrial magnetism?

The ultimate goal is to understand why the Earth has evolved the way it has, and why it is different from other planets. Why does Earth have a large dipolar magnetic field at all? Has it changed over time? The paleomagnetic field can give us unique insights into the thermodynamic evolution of Earth's deep interior, which has huge implications for everything above it, from mantle to crust to atmosphere and ultimately to life.

What are the day‐to‐day effects the geodynamo?

The geomagnetic field is very helpful for several reasons. Way out into space, about 10 Earth-radii above the surface, the geomagnetic field balances incoming charged particles from the Sun, known as solar wind, at a boundary called the magnetopause. This boundary is helpful by deflecting much of the charged particles streaming towards Earth, kind of like how air is deflected around your car by your windshield. However, in this case the "air" is high-energy charged particles that we would consider harmful radiation and could potentially disrupt our atmosphere and man-made electronics. Some solar wind does make its way down into the atmosphere, but this is presently concentrated near the geographic poles where the geomagnetic field lines are nearly vertical. These particles can trigger auroral displays in the atmosphere. But if the field were to undergo a magnetic reversal, which we believe it is not at present, then these radiation "funnels" could wander to lower latitudes and have a more direct effect on civilization.

Earth's magnetic field, generated by the motion of liquid iron deep inside the core, reaches out into space until it is balanced by non-stop flows of Solar charged particles, also known as solar wind. This balance occurs around 35,000 miles above the Earth's surface (or about 10 times the radius of the Earth). The planet's magnetic field closely approximates an axial dipole at present, where the magnetic and geographic (rotation) poles are coincident. This is the simple dipole magnetic field oriented along Earth's rotation axis that we are familiar with today. This magnetism conveniently funnels most of the incoming charged particles into polar regions and sometimes generates visible aurorae. Image: Roberto Molar Candanosa, DTM.

If there's a thing you wish people knew more about in terms of Earth's magnetic field, what would that be?

There are many things, but perhaps the one thing I think people should keep in mind is that even though it is difficult to image and envision what is going on in Earth's deep interior, it is the major controlling factor that keeps Earth's surface habitable over long time scales. Without a magnetic field or plate tectonics, Earth might not have been able to retain a habitable surface and a water-rich atmosphere. Life as we know it might not exist. That sounds more like a threat than I intended, so let's also say that the Earth's interior is just super interesting!

—Roberto Molar Candanosa



Join DTM's Peter Driscoll at 6:30 p.m. on Thursday, Nov. 8, 2018 as he discusses physical mechanisms that produce the geomagnetic field, conduct a brief survey of planetary magnetism, and explore Earth's rich paleomagnetic history preserved in ancient rocks, including polarity reversals, superchrons, paleointensity, and the impact of the solidification of the inner core.

For event details and registration, visit: