Postdoc Spotlight: Modelling Mid-Ocean Ridges with Joyce Sim

Joyce Sim created this video abstract for AGU 2019 which describes her recent publication, 'The Influence of Spreading Rate and Melt Permeability on Melt Focusing at Mid-Ocean Ridges.'
Tuesday, June 23, 2020 

When exploring the dynamics of Earth’s deep interior, it’s often impossible to test your ideas in a lab. As a geodynamicist, Postdoctoral Fellow Joyce Sim rises to that challenge using mathematical modeling to study the internal workings of our planet and how they shaped the world as we know it today.

Sim specializes in figuring out the mathematical intricacies of two-phase flows, for example when liquid or gas flows through solid rock.  Specifically, Sim works on two fluids with very different viscosities, such that one—the Earth’s mantle—seems more like a solid. Over the years, her interest in the way materials flow has led her to study seismology, volcanology, geophysics, and, more recently, the chemical interactions of the deep Earth.

Sim recently published two papers that looked at mantle melting, magma, and crust creation at mid-ocean ridges. In one paper, entitled The Influence of Spreading Rate and Permeability on Melt Focusing Beneath Mid-ocean Ridges, she presents mantle permeability as a major reason melts become focused from a broad area beneath the mid-ocean ridges to the narrow axis where it eventually bursts forth to create new oceanic crust. In the second paper, entitled Time-Dependent Crustal Accretion on the Southeast Indian Ridge Revealed by Malaysia Airline Flight MH370 Search, Sim, and colleague Ross Parnel-Turner were able to show that oceanic crust formation varies in cycles of hundreds of thousands of years.

In this month’s Postdoc Spotlight, we sat down with Sim over Zoom to discuss these two papers, what it means to be a scientist, and what she hopes for her future and the future of geodynamics. This interview has been edited for clarity.

Carnegie Science (CS): Hi Joyce, can you please introduce yourself?

Joyce Sim (JS): Sure. Hi everyone, I'm Joyce Sim. I work at Carnegie Science on the Broad Branch Road campus. I'm a geodynamicist, and I am trying to apply two-phase flow modeling methods at tectonic boundaries, like mid-ocean ridges and subduction zones.

Joyce Sim derives equations in finite element weak form by hand while working from home during COVID-19. Photo credit: Joyce Sim

CS: How did you decide to become a geodynamicist? 

JS: I don't think I found geodynamics, I think it found me!

I studied mechanical engineering for my master’s degree. That's what really got me into all these different types of flows and how they apply to a lot of systems.

Then, when I was doing my Ph.D., I had a whole bunch of different projects going on. I was really interested in seismology and natural hazards, earthquakes, and hurricanes. And I just stumbled into studying volcanic eruptions!

So, essentially, because I love studying flows, I went from volcanic eruptions and their supersonic flows, to super slow flows. In the two-phase flows that I study, things are moving at a snail's pace, way slower than snail pace, actually.

That's kind of how geodynamics found me. I'm just excited about flows in general! And this was a really exciting topic for me that made sense to study.

In 2013, Joyce Sim took a field trip to Kamchatka when the Tolbachik volcano was erupting. This was during her Ph.D. when she was still in the subfield of physical volcanology and modelling volcanic processes. Photo credit: Joyce SIm

CS: So, when you’re talking about flows. What is doing the flowing?  

JS: Right. That's a great question.

I’m focused on areas where you definitely have a two-phase flow system—so, for my research, that’s magma moving in a mantle matrix. I’m trying to build numerical models for that movement. But, most of geodynamics is looking at the mantle movements on a broader scale.

Traditionally, geodynamicists think of the mantle as a solid entity that moves on a very long timescale. But in fact, on a very long time scale, it's flowing!

What I'm doing is actually taking it a little bit further and saying. “Okay, well, in some places we actually have melting that occurs.” When the mantle starts to melt, it makes magma that has a very different viscosity than the mantle it came from.

Now you have two very different materials that are flowing together but that have two very different viscosities: the very high viscosity mantle and the lower viscosity magma. This magma is more buoyant and has a lower viscosity. This essentially means that the magma can move around easier and that it sort of wants to float up above the less buoyant mantle.

CS: So you’re modeling the way that magma moves through the mantle. But, where does the actual melting occur?

Magma only occurs when the mantle starts to melt, right? And there are only a few places where that happens. Most of this activity happens at tectonic boundaries where two tectonic plates meet each other.

One of these places is at mid-ocean ridges where you get two tectonic plates separating. Here, the decompression of the mantle actually leads to melting. This is called decompression melting. Melting also occurs at subduction zones, where an oceanic crust is sinking beneath the continental crust. The exact mechanism is still debated, but we know we get some kind of melting because of volcanic activity. We also see melt at hotspots like Hawaii, which aren’t near any plate boundaries. You definitely see some kind of melting there.

CS: You recently published a paper in Physics of the Earth and Planetary Interiors where you show a new mechanism for melt focusing. First, what is melt focusing?

Okay. We know that melt is generated in wide areas from seismic studies. Seismologists can see a broad melt generation area in the mantle. That's interesting because on the seafloor we noticed that oceanic crust is only created in a very narrow region where two tectonic plates are pulling away from each other at the mid-ocean ridges. The melt area is much broader than where it ultimately ends up coming to the surface.

So, we know that the melt must be moving horizontally and becoming concentrated in that narrow area. That’s melt focusing.

In this work, we are trying to understand the physics behind that movement.

Melt focusing diagram as published in Sim’s 2020 paper, The Influence of Spreading Rate and Permeability on Melt Focusing Beneath Mid-ocean Ridges.

CS: What questions were you trying to answer in this study?

JS: We were trying to figure out two main things. First, since we know that there is a whole range of spreading rates on the ocean floor, we were trying to understand how this changes melt focusing. Second, we were trying to figure out how the permeability of the mantle matrix itself changes the focusing.

Now, permeability is a really hard problem. It's hard to do experiments in the lab that get at the permeability of the mantle system because it has to be done at very high pressures and temperatures.

In this case we decided, "Well, okay, let's see what happens if we just do it theoretically and solve it numerically." And it turns out that permeability is very important—as we kind of expected. We also determined that spreading rates actually don't matter too much, which also matches what we observe in nature.

So, seeing our numerical solutions match what we observe in real life gives us confidence.

CS: So, now that you’ve run the numbers, what do you think is causing the melt focusing effect?
JS: We found three different mechanisms of melt focusing.

One is a ridge suction effect. So, this is essentially where the tectonic plates pull apart and that movement sucks melt up to the ridge axis. A second mechanism is what are called “decompaction layers”, where you have the lithosphere cooling and thickening away from the ridge axis. It is basically creating a slope, and as the melt buoyancy rises it is channeled toward the center.

The third one is super interesting and really comes out of the equations themselves. We call it the "melting rate focusing effect."

This mechanism is moderated by how easy it is to squish that mantle matrix. Think of it like a sponge. If the mantle is easier to squish and it's in the melting region, the physics essentially dictates that the melt will pool where it's melting the fastest. So if the mantle is very permeable, and these other two mechanisms are causing melt beneath the mid-ocean ridge axis, the melt will keep being funneled from the very broad area to the very narrow area.

That's the newer phenomenon that we figured out that I think is really cool.

CS: Your background is in geodynamics and you’re interested in flows, but I understand you’ve been looking at questions of deep carbon cycling during your Postdoctoral Fellowship at the Earth and Planets Laboratory. Can you explain what your carbon research looks like? 

JS: I came to Carnegie to apply this modeling to subduction zones. So now, instead of geodynamics, I'm kind of diving into the realm of thermodynamics and what I think you could call theoretical chemistry. Actually, I'm not sure what the term is, but I'm looking at chemistry, which is weird. I hated chemistry when I was in middle school and high school. I can't believe I’m doing that, but it has been really fun because it's still solving equations and trying to figure out problems.

So, what I'm trying to do here is look at subduction zones. Traditionally we understand that there is a big slab of oceanic crust moving down into the mantle. That slab has a lot of carbon in it, megatons of carbon are being subducted each year.

It has been thought that a lot of that carbon is brought down with this slab and into the mantle. But more recent theoretical modeling and experiments have kind of shown that may not be the case. It turns out that carbon could actually be released from the subducting slab and dissolved in the fluids—like water—that escape the slab and are released back into the atmosphere by volcanoes. I’m trying to figure out how much of that carbon is getting dissolved out of the slab in subduction zones, and how much that could be happening on a global scale.

The ultimate question here is trying to understand how much carbon is being picked up by these melting fluids that go up into volcanic systems and how much is being subducted into the mantle. So that’s what I’m doing here. It’s still active research.

Joyce Sim put explains her carbon solubility research in this AGU 2019 video abstract.

CS: What would you say is the coolest part of your research?

think the coolest part is really, you know, probing the frontiers of science and human knowledge. No one really knows what I’m going to find and no one has the answer for you. It’s kind of amazing to look at your work and think, “Well, this is the answer to this question for now.” Because no one’s really looked at it before you!

It’s also really fun to be able to talk to someone about a question you’re working on, and they tell you something about their research. Then sometimes there’s a moment where it just clicks! You’re thinking, “Wow, we have very similar ideas but from very different perspectives. I wonder if we can work together”

That's actually something that happened in one of the very recent papers that I published. That was so exciting.

CS: Is that the recent paper where you worked with the Malaysia Airlines search data? 

JS: Yes, this is the paper that's led by Ross Parnell-Turner. He's a PI at the Scripps Institution of Oceanography, which is where I got my Ph.D. In this study, he looked at all the bathymetric data that was generated from the search for Malaysia Airlines flight MH370 to look for evidence of periodicity in seafloor spreading near the Southeast Indian Ridge. 

For the past five to 10 years, there has been quite a lot of debate about whether the Milankovitch Cycles have some kind of influence on seafloor spreading. If that were true, we’d see evidence in the seafloor at
41,000-year, 25,771- year, and 112,000-year cycles.

When he looked, he didn’t really see any evidence of the Milankovitch Cycles at all. What he did find in the bathymetric data were longer period cycles, around 300,000- 400,000 years. When Ross looked at my models he said, “Wow, your models are generating some very interesting waves? What is the periodicity in them?”

And it turns out that the periodicity in my models matched the periodicity he had found the bathymetric data from the ocean floor! It was the perfect person and the perfect conversation at exactly the right time! So we wrote up the paper pretty quickly. 

It's also a very personal paper for me because I am Malaysian and it uses this Malaysian Airline search data, which is kind of crazy.

CS: You’ve worked in so many fields of science, from engineering to geophysics to deep Earth chemistry. Did you know you wanted to be a scientist when you grew up?

JS: You know, for a long time, I didn’t know I wanted to be a scientist because I didn't know that you could be a scientist. No one ever told me that. What I did know was that I really wanted to understand the Earth.

Growing up, I loved looking at stars trying to learn about them. And, like, I don’t know why I’m obsessed with Rossby waves, but I wrote a whole paper on them when I was in high school and people thought I was crazy. I still remember my teacher being shocked. “You really want to look at Rossby waves? I don’t know why, but okay!”

I guess, to me, understanding the natural world is really interesting. It always has been. I want to figure out why it’s happening and why things are the way they are. I’ve always done that, but I didn’t know that I could do “science” as a job until I was already in college.

CS: Here’s a fun question. If you could meet one of your science icons, dead or alive, who would they be? 

JS: Probably Ada Lovelace. I mean, she was basically the first computer programmer. She was a mathematician who started trying to figure out how to use computing for science back in the 1830s. She was like, “Okay. We have this analytical engine. I have all of this numerical stuff. Let’s see if we can actually put this in there and get results!”

It was kind of the beginning of what I’m doing in some sense. It’s pretty amazing to look back now and think about how she was able to put together the starting of a puzzle that she didn’t even have all of the pieces to yet. How do you even do that if you don’t know that computers or the internet are going to be a thing!?

Diagram of an algorithm for the Analytical Engine for the computation of Bernoulli numbers, from ‘Sketch of The Analytical Engine’ Invented by Charles Babbage by Luigi Menabrea with notes by Ada Lovelace. Broadly considered to be the first computer algorithm. Public Domain 

CS: Where do you think your field will be in like a hundred years from now? What do you think will have changed?

This is a super fun question! There are all these possibilities. We still don't know that much, you know?

I think that sometimes as humans, we are kind of set in our ways. We see things in front of us and that's how we want them to be. For example, we don't know what an alien will look like. Hence you draw them in ways that reflect us:  two eyes, a nose, and a mouth. But that’s probably not going to be what alien life looks like.

So, I think in 100 years, new technology that allows us to explore all these different regions of the universe is going to really expand our ideas about life, our own Solar System, and Earth.

As for my research, it would be really cool if we have some kind of method to go down and illuminate the whole ocean floor. We have satellites right now, but they still give us pretty rough measurements of the ocean floor. In 100 years, I would love something that has slightly better capabilities that can pick up different layers or different carbon carrying life or a non-life that is just floating around down there. That would open up a lot of understanding of our world.

CS: What research do you hope to pursue in the future? 

JS: What I'm excited about in terms of future work is applying my research to understand the tectonic system and when the tectonic plates started existing. It’s still debated when Earth’s tectonics and atmosphere actually started to look more like how it looks today. There’s a lot of really exciting questions like: how and when did continents form over time? When did we get an exosphere? Will our plates ever stop moving?

Those are the questions I’m really excited about!

CS: What do you like about working at the Earth and Planets Laboratory? 

JS: Everyone's door is always open. I know if I want to understand something about flat slab subduction I can just run down the hall to Lara Wagner and she’ll be happy to talk with me. It’s a great place for that. Everyone wants to talk about science, debate, and figure things out. Plus, everyone is brilliant! So it’s super fun to work here.

From left to right: Amol Karankidar, Joyce Sim, and Wan Si Tang pose for a photo at the Earth and Planets Laboratory fall picnic. Credit: Peng Ni

CS: Thank you for taking the time to do this interview!