Archives / 2019 / July

The Sassy Scientist – Hyde: Lithosphere Dynamics

The Sassy Scientist – Hyde: Lithosphere Dynamics

Every week, The Sassy Scientist answers a question on geodynamics, related topics, academic life, the universe or anything in between with a healthy dose of sarcasm. Do you have a question for The Sassy Scientist? Submit your question here.

Senna asks:

I’m torn between mantle dynamics and lithosphere dynamics as a research topic. Which shall I choose?

Dear Senna,

I don’t know what came over me when writing last week’s post. Must have been something I ate. Sometimes I just get some crazy idea stuck in my head. For example, I completely misconstrued ideas on the relative importance of mantle convection over lithosphere dynamics.

Last week I was babbling on about plate tectonics, and how the focus on the lithosphere shifted attention away from mantle convection. To be clear: this was for good reason. Granted, McKenzie and Parker’s (1969) concept of rigid plates on a shell was simplified. That was also the point: it explains some first-order observations like relative plate motions and seismicity. No, plate boundaries are not necessarily narrow zones of deformation (Kreemer et al. 2014). Yes, plates can also be flexed through loads on top, or radial mantle tractions from below (Watts 2001). We know that the Earth’s lithosphere can generally be considered to be a visco-elastic plate that is primarily moved by forces acting at its boundaries. Does this preclude seismicity or deviations from principal stress directions far away from such plate boundaries? Of course not.

It is obviously a great idea to base model predictions on non-unique gravity and geoid observations, seismic tomography, and Earth’s normal mode undulations for structures 2500 km away from the surface. The surface. You know, that place where we actually have direct observations. Sure, just extrapolate some rheology measured on a rock sample down to the lower mantle using some equations of state. Should be similar, right? Do you have a kinematic plate tectonic reconstruction based on loads of geological and paleomagnetic data? Just pop it into a mantle convection model to see whether it is feasible. You measured some SKS splitting data? That’s definitely a measure of present-day mantle flow. You infer instantaneous dynamic topography at the surface in the order of a couple 100 meters? Mantle convection models predict kilometers of instantaneous dynamic topography, so you must have made a mistake in your observations.

Do you want to understand what’s driving deformation of the lithosphere? Choose lithosphere as a research topic. Do you believe in fairy tales? Look into mantle convection: it’s full of magic.

Yours truly,

The Sassy Scientist

PS: This post was written after recovering from a major headache from reading last week’s post.

PS2: I won’t be surprised if there is going to another post on this topic next week. How well do you know your Scottish literature?

Kreemer, C., G. Blewitt, E.C. Klein (2014). A geodetic plate motion and Global Strain Rate Model, Geochemistry, Geophysics, Geosystems, 15, 3849-3889, doi:10.1002/2014GC005407
McKenzie, D., and R. L. Parker (1967), The North Pacific: An example of tectonics on a sphere, Nature, 216, 1276
Watts, A. B. (2001), Isostasy and flexure of the lithosphere, Cambridge University Press, 458 pp.

Iron volcanism on metallic asteroids?

Metallic asteroid

This week, Francis Nimmo, professor in the Department of Earth and Planetary Sciences (University of California Santa Cruz), tells us about volcanism on metallic asteroids! Around and after the formation of the solar system (4.5 billion years ago onwards), volcanoes on some of the gigantic bodies of the asteroid belt might have erupted … iron. Explanations.

Francis Nimmo

Francis Nimmo

One way in which we can learn about the insides of planetary bodies is by looking for signs of volcanism. Volcanism transports molten material from the interior to the surface of a body, where it solidifies. Something similar may happen on icy bodies, where cryovolcanism is thought to occur. Together with Coby Abrahams, one of my grad students, I recently wrote a paper proposing “ferrovolcanism”, which is the eruption of metallic iron. We suggested that this may have happened in the past on metallic asteroids.


What are metallic asteroids? Some asteroids, such as 16 Psyche are thought to be made mainly of iron and nickel, on the basis of their spectra and radar signatures. They probably originated from a catastrophic collision. The picture is that a proto-planet, which had already formed a dense metallic core, was broken apart so violently that its core was fragmented, and the fragments then cooled to form free-floating metallic bodies. Such an event would also explain the large number of iron meteorites in our collections. So far we have never seen a metallic asteroid up close, but NASA is planning to send an orbiter which will reach 16 Psyche in 2026.

Iron meteorite

Artistic view of an iron meteorite with ferrovolcanism (Elena Hartley)

How would “ferrovolcanism” work? Immediately before the catastrophic collision, the core would have been liquid (because the silicate mantle is a good insulator). After the collision, the surface of a free-floating iron blob would have initially cooled very rapidly. Analysis of iron meteorites tells us that, in some cases, the blob developed a strong, cold iron crust. The melt (liquid iron) beneath the crust would have been less dense than the crust – just like molten magma on Earth. And so, just like magma on Earth, the liquid iron would tend to rise to the surface – hence ferrovolcanism.

There are a few factors that make ferrovolcanism less likely than silicate volcanism. One is that, as the iron crust cools, it contracts, putting the whole body into compression and making it more difficult for melt-filled fractures to open up. Another is that iron is both ductile and strong, so that fracture propagation is harder. And iron is also much more thermally conductive than rock, so that melt-filled fractures will cool and solidify rapidly. Nonetheless, we don’t think any of these factors is fatal. For instance, Mercury also experiences global compression but still shows signs of volcanism. Iron eruptions might be aided by volatiles coming out of solution, just as with terrestrial eruptions, although the catastrophic impact may have removed some volatiles.

How do we test this hypothesis? One way is to wait for images of 16 Psyche! If we see things that look like iron volcanoes, that would be interesting. But identifying signs of ancient volcanism is notoriously hard – this paper documents several proposed volcanic or cryovolcanic features which turned out to be something else on further investigation. And our calculations show that the volume of erupted material will be only about 0.1 percent of the total volume of Psyche, so these features may be hard to spot – especially after 4 billion years of erosion by impacts.

Another way is to look at the iron meteorites in our collections, because ferrovolcanism should produce anomalous-looking meteorites. For instance, one might find vesicle-rich samples, or ones happening late in the crystallization sequence but showing extremely rapid cooling (because they were erupted to the surface).

How did we write the paper? It was more or less accidental. Coby was working on how iron asteroids might lose their volatiles early when one day he turned to me and said “I think they’ll erupt!”. I thought this was a neat idea, and so we started to look at how the eruptive process might work.

The strangest part of writing our paper was that at the LPSC conference someone came up to me and said “I just invented iron volcanism!”. It turned out that another group had been working on ferrovolcanism completely independently. They suggested that pallasite meteorites might be evidence of ferrovolcanism (something that I had completely missed). This simultaneous but independent development of ideas seems to be quite common in the Earth Sciences – but that is a topic for another day.


The Sassy Scientist – Jekyll: Mantle dynamics

The Sassy Scientist – Jekyll: Mantle dynamics

Every week, The Sassy Scientist answers a question on geodynamics, related topics, academic life, the universe or anything in between with a healthy dose of sarcasm. Do you have a question for The Sassy Scientist? Submit your question here.

Senna asks:

I’m torn between mantle dynamics and lithosphere dynamics as a research topic. Which shall I choose?

Dear Senna,

This is an easy one: mantle dynamics. Don’t you want to be part of cutting edge research, utilising state-of-the-art numerical modelling codes and high performance computing facilities? You can employ every solid rock rheology known to man and generate new lithosphere through volcanism. Nowadays the lithosphere and crustal layers are implemented so convincingly that we can easily match seismic tomography slices, passive margin architectures and even subduction zones to the scale of accretionary wedges. Inherited weak zones can lead to supercontinent cycles. The sky is the limit!

I am the first to admit that mantle convection has been overshadowed for quite a while by plate tectonic theory ever since the early days of Holmes (1931). Wilson’s (1965) tessellation of the Earth’s surface and the straightforward connection to relative plate motions (e.g., McKenzie and Parker 1967) and mid-ocean magnetic anomalies (Vine and Matthews 1963) resulted in a longstanding main focus on lithosphere dynamics (Forsyth and Uyeda 1975). Even though early work, by for example Morgan (1971), showed that simplified systematic mantle convection explained first-order observations as well, it took quite a while before mantle convection became more popular. To be clear, researchers from several fields studying the mantle didn’t help very much: lingering discussions on whole mantle convection vs. separate flow cells, whether slabs or plumes can penetrate the 660 km boundary, and problems with incorporating a realistic lithosphere with proper rheologies that produced surface deformation predictions similar to the actual Earth’s surface held back widespread acceptance that mantle convection can explain everything. Nowadays, computational prowess and numerical model sophistication is at a point where we have overcome these issues: once again, the lithosphere is simply the thermal boundary layer of the mantle convection system.

Yes, the lithosphere is tessellated and undergoes continuous reorganisation — we can model this with mantle convection. Yes, there are probably thermo-chemical piles in the lower mantle — we can incorporate these. Yes, we infer that India has moved faster than any continent does at present — just add a mantle convection cell. Yes, the surface has topography that we can’t explain with isostasy and flexure — it is induced by radial mantle flow. I’m forgetting other issues here, obviously. The point is: studying lithosphere dynamics is going to be an obsolete exercise in a couple of years. Join the mantle convection party and be prepared for the future!

Yours truly,

The Sassy Scientist

PS: This post was written after reading quite a bunch of lithosphere dynamics papers trying to explain surface observations without using the mantle!

Forsyth, D.W. and Uyeda, S. (1975), On the relative importance of the driving forces of plate motion, Geophys. J. R. astr. Soc., 43, 163–200
Holmes, A. (1931), Radioactivity and Earth movements, Transactions of the Geological Society of Glasgow, Geological Society of Glasgow: 559–606
McKenzie, D., and R. L. Parker (1967), The North Pacific: An example of tectonics on a sphere, Nature, 216, 1276
Morgan, W. J. (1971). Convection Plumes in the Lower Mantle. Nature, 230, 42-43,
Vine, F. J., and P.M. Matthews (1963), Magnetic anomalies over ocean ridges, Nature, 199, 947
Wilson, J. T. (1965), A new class of faults and their bearing on continental drift, Nature, 207, 343

It doesn’t work! (Asking questions about scientific software)

It doesn’t work! (Asking questions about scientific software)

Numerical modelling is not always a walk in the park. In fact, many of us occasionally encounter problems that we cannot directly solve ourselves, and thus rely on help from others. In this month’s Wit & Wisdom post, Patrick Sanan, postdoctoral researcher at the Geophysical Fluid Dynamics group at ETH Zurich, will talk about asking the right questions about scientific software. As an experienced scientific software developer, Patrick has often been at the “receiving end” of questions regarding numerical modelling and hopes to guide you through some important points that could make life for you as well as your ‘helper’ a lot easier.


Patrick Sanan is a postdoctoral researcher at the Geophysical Fluid Dynamics group at ETH Zurich

Numerical modelling is essential for geodynamics; since we cannot directly measure relevant phenomena, we partake in the magic of making a set of reasonable assumptions, setting up a model, and letting a system evolve to produce insight. It’s beautiful. We gain an understanding of the subtle-yet-fundamental processes which shaped our apparently-so-special Earth. We turn our eye beyond, to other planets.

I’m not here to talk about that, though. I’m here to discuss an ugly part of the job, which can bring all the profundity to a screeching halt: what to do when the code doesn’t work.

You know the situation. You’re stuck. There’s no output. Segmentation fault. Error Code -123. You didn’t sign up for this…

You don’t know where to start looking. Is it your model parameters? Your physical assumptions? Are you using the code as intended? Is it your compiler? Can it be the cluster? Your .bashrc? Is your keyboard plugged in?

You’re frustrated. No one around you has a good suggestion. Why are these cruel computers doing this to you?

What to do? Ask for help, in the right way. In this blog, I’ll point out some facts about scientific software, try to use them to formulate an effective email in which you ask for help, and then try to extract some guiding principles.

Some Points on Scientific Software

First, I’ll list some observations I, as a developer, have made about scientific software:

  • Scientific software projects are usually short on maintainers and time.
  • Software designers are not psychic, but they are often experts at deductive problem-solving.
  • Reproducible, bisectable problems are surprisingly easy to solve. Other problems are surprisingly difficult.
  • Working reference cases are very valuable.
  • Sufficiently-complex computing tasks must be treated like lab experiments: one must document the setup and control as many factors as possible.
  • The solutions to most problems are obvious, once found.
  • Numerical software is harder to test than generic software, because floating point arithmetic and parallel computing lead to acceptable differences in numerical output. Higher-level understanding of physics or (parallel) numerical methods is often required to know if something is a “real problem”.

With these as a guide, let’s consider how one might try to resolve a confusing issue.

No pretty pictures here, just error messages (if you’re lucky).

Asking for Help: The Bad Way and the Good Way

Email is a common way to ask for help, as often the person who can best help you is across the world. Let’s say that I’m using a regional lithospheric dynamics code called Rifter3D. I’ve come across an error I have no explanation for. I can’t figure it out, so I write an email to developers of the code. You might also write to a dedicated help address.

Hello - I'm using Rifter3D but on our local cluster I get this error:

/cluster/shadow/.lsbatch/1559505025.92314771: line 8: 951 Segmentation fault (core dumped) ./rifter3d -options_file options.opts

Do you know what I'm doing wrong?

The person on the other end wants to help, but has no information to work with and doesn’t know how much of their time you’re asking for. There is a better way:

Hello - I'm having some trouble diagnosing a problem using Rifter3D and was hoping
you could give me some pointers.

I've been working with Prof. XYZ and have modified a rifting scenario from XYZ et al. 2017 to study the effects of varying A on B. When I run a small case on my laptop, the simulation finishes as expected. I use the attached options_small.opts and run mpiexec -np 4 ./rifter3d -options_file options_small.opts. I'd like to run a bigger case (512 x 256 x 64) for 300 million simulated years, on 64 cores on our cluster.
However, my job fails, producing a segmentation fault almost immediately. I attached the job submission script (job.sbatch), and output from my run (lsf.o92314771). I am using the cluster's existing PETSc 3.11 module, and version 1.0.3 of Rifter3D. I can successfully run a simple isoviscous setup on this cluster - see attached job_small.lsf, option2.opts, and lsf.o92314681

Do you have any insight as to how I might be able to run this simulation?

E. Scholar

Attachments: options_small.opts options.opts job.lsf job_small.lsf options2.opts lsf.o92314771 lsf.o92314681

The recipient will likely respond with more questions as you work towards resolving the issue. Perhaps they’ll ask you for more information about how you built Rifter3D, or point out some unusual settings in your options file.

Why is this second email better? It is not simply longer, but

  • It clearly describes the problem. Just trying to precisely describe a problem has an almost-magical clarifying effect, and the solution will often appear. Software engineers call this rubber ducking.
  • It explains the true objective and how the problem is to be resolved. This is important to avoid the XY problem, describing a problem with a method to achieve a goal, without mentioning the goal itself.
  • It gives concrete information to reproduce the problem: the version of the code, input files, and launch commands/scripts.
  • It provides enough output to allow deductive reasoning, more than just a copy-and-pasted error message.
  • It’s polite
  • It shows some effort has already been put into investigating.
  • It notes similar, working cases.
  • It’s not too long, but it is detailed enough to allow for quick, intelligent follow-up questions. Supporting data are included as attachments or links.
  • It doesn’t make too many assumptions about the cause of the problem.

Boiling it Down: 3 Questions to Ask Yourself

When asking for help, consider these three questions. They will help with the central objective: clearly describing the problem.

  • Why do you need it to work?
    What is the context? What is the goal?
  • How do you show that it doesn’t work?
    What are the steps to reproduce your problem?
    How will you know that the problem is resolved?
  • What does work?
    How far are you from a working state? What similar cases work?

Here is a 1-page pdf which you can print out, with some of this advice:

Pdf and LaTeX source on GitHub

The “Real” Answer

To conclude, I will try to avoid an “XY problem” of my own. The most efficient way to resolve bewildering problems is to avoid them. To make an alpine analogy, the most important topic in avalanche safety is not how to dig someone out, it’s how to avoid risky terrain.

Real-life debugging. Avoid if at all possible (from Wikimedia commons)

First, borrow techniques from software engineering. Version control (e.g. git) will encourage you to save working states, amongst many other benefits. Next, leverage your intuition and experience as a geodynamicist. Always be able to quickly run and verify small, quick, simple cases, and test and visualize often. Look out for simple cases where you “know the answer ahead of time”: established benchmarks and analytical solutions.

The points in this article can help everyone save time (and not just running geodynamical models!): problems will be resolved more quickly, bugs will get fixed faster, and more time can be spent exploring more interesting questions than “Why doesn’t it work?”.