This transcript was generated automatically and lightly edited for readability.

Oliver Strimpel: This is Geology Bites with Oliver Strimpel. The subfield of geology called geodynamics most commonly refers to the motions associated with convection in the mantle. These are slow by human standards and lead to phenomena such as plate motions, seafloor spreading, mountain building and volcanoes. But the Earth's interior actually undergoes motions on a whole range of timescales. The shortest of these occur within seismic waves, in which the vibrations triggered by earthquakes cause tiny movements of the material as they pass through. And the longest of these are the large-scale viscous motions of the convecting mantle, driven by the temperature difference between the bottom and the top of the mantle. But there are a whole range of intermediate-timescale motions, lasting hours, days, years and millennia. Harriet Lau is an assistant professor in the Department of Earth and Planetary Science at the University of California, Berkeley. She's especially interested in the motions of the solid Earth that take place over these intermediate timescales, and in what they can reveal about the Earth's interior. Harriet Lau, welcome to Geology Bites.

Harriet Lau: Thank you, Oliver. Thank you for having me. It's an absolute pleasure.

Oliver Strimpel: I don't think most of us even knew that the solid Earth is, in fact, moving all the time, on timescales of hours to millennia. What are the drivers of these motions?

Harriet Lau: Let's move from the short timescales to the longer timescales. You've already mentioned some. There are tidal forces; we call these solid Earth, or body, tides. When snow caps melt in the summer and grow in the winter, these changes in mass also warp the planet's subsurface. And longer still are ice ages, which we're currently cycling through. These redistribute mass on a global scale on timescales of tens of thousands to hundreds of thousands of years. Only 25,000 years ago, much of the Northern Hemisphere was covered entirely by ice during the last ice age. This has since melted, leaving only the Greenland ice sheet, but the solid Earth is still responding to it. Ultimately, Earth's gravitational field is being changed, either by masses shifting around on the surface, often to do with growing and melting ice sheets, or by external masses like the Moon and the Sun.

Oliver Strimpel: OK, let's talk more about each of these, starting with the shorter ones. We all know about the tides in the ocean caused by the Moon, but the solid Earth also has tidal motions.

Harriet Lau: In some ways, they're not too dissimilar from ocean tides. We have ocean tides because of gravitational variations between the Sun, the Moon and the Earth, and their effect on the ocean mass. These same forces also warp the solid Earth's surface, at exactly the same frequencies. Just as we have ocean tides that lap on and off the shore twice a day, we also have deformation of the ground beneath us, and this is up and down about twice a day, by as much as 40 centimeters. These types of motions have been predicted since the earliest theoretical geophysicists, like George Darwin in the 1800s, but measuring them, since they're very slow motions, as Darwin tried to do, took a little bit more time. So contraptions like tiltmeters, which measure deviations of the ground from the vertical, were first used, and still are nowadays.

We also look to space: satellite orbits will be slightly perturbed due to these forces. We even use a technology known as VLBI, or very long baseline interferometry, to measure this type of deformation. This involves really large radio telescopes; we can measure very slight differences in the distances between these radio telescopes, and these distances tell us how the Earth is deforming. More common now is the use of GPS, the very same technology in your smartphone that tells you where you are. Now imagine a really sensitive GPS station stuck to the ground, measuring how the ground moves over time. Some GPS stations have been there measuring this motion for decades now. And so, if the ground moves up and down by around 40 centimeters twice a day, every day, for decades, geophysicists can stack these measurements to get precise recordings of this deformation.

Oliver Strimpel: OK, so now that we have all these great measurements, does the exact tidal deformation that they reveal tell us anything about what's going on in the Earth's interior?

Harriet Lau: I mentioned we go up and down by around 40 centimeters, but think of that more as an average. You're calling from Massachusetts, and perhaps you might go up and down by 40.1 centimeters, and here in Oakland, where I am, I might go up and down by 39.95 centimeters. These are just numbers I made up, but it gives you an idea of what the GPS measurements have revealed: there are tiny, submillimeter-level variations that exist across the globe. What these deviations are saying is that some parts of the Earth's interior are stiffer and some are less so, and some are denser and some less so. So we can use these deviations to teach us something about the structural variation of Earth's interior.

One target in particular is two regions in the deep mantle, around 3,000 kilometers beneath our feet. These are two dome-like structures that rise from the base of the mantle to around 1,000 kilometers above it. So they're massive. There's one beneath Africa and one beneath the central Pacific, and we see these in seismic images of the mantle. These two regions are often known as the large low-shear-velocity provinces, and they're known that way because they slow down seismic waves. Understanding why seismic waves are slowed down within these regions is a really big goal for many global geodynamicists. Tides, weirdly enough, provide a new angle to get at the story of these LLSVPs.

Oliver Strimpel: That's fascinating. I remember that in an earlier podcast, Allen McNamara explained that the large low-shear-velocity provinces that we see in the seismic imagery could be caused either by higher temperature in those regions or, alternatively, by a different composition, such as a more iron-rich one. And he modeled how these large provinces would be expected to evolve in each scenario, to see which one might actually result in the observed shapes. Unfortunately, the jury is still out on that score, because the resolution of seismic tomography is not yet up to the job of resolving the differences. But I wonder, does your tidal analysis say anything about these large low-shear-velocity provinces, and perhaps help us distinguish between these two interpretations?

Harriet Lau: Very much like seismic imaging, we did a tidal version. In our case, the tides don't really care much about whether seismic waves are sped up or slowed down through these regions. Instead, we focused on the density, or more precisely the buoyancy, of the deep mantle: whether there are regions of the mantle that would preferentially rise or sink. And GPS measurements of the variation in tidal deformation seem to indicate that LLSVPs are negatively buoyant; that is, they are denser than the mantle surrounding them. To get that negative buoyancy signature, the LLSVPs have to be compositionally distinct from the mantle surrounding them, and in order to increase the density, this would likely involve some type of enrichment in very heavy elements like iron.

Oliver Strimpel: In another podcast, Gillian Foulger said that seismic speed is actually most sensitive to the presence of partial melt, more so even than to compositional or temperature variations. Can your analysis of tidal motions infer anything about the presence of partial melt in the lower mantle?

Harriet Lau: One caveat with this kind of tidal analysis is that the spatial resolution is extremely low. Imagine that the tides only see a really blurred density map of Earth's interior. So somewhere within the LLSVPs is a compositional difference, but the tides would find it very hard to pinpoint exactly where. With partial melt, the tides would be very sensitive to it, but if that partial melt is in tiny little pockets, then it might be invisible to the tides.

Oliver Strimpel: You mentioned that the marine tides and the solid Earth tides go through about two cycles a day. What other tidal periods are there in the Earth?

Harriet Lau: You might have heard of neap tides and spring tides. These also play a role in the solid Earth tides. So there are daily tides, there are fortnightly tides, and there are even tides every 18.6 years. Over these timescales, factors like Earth's tilt as it rotates, or other kinds of orbital characteristics of our planet, change slightly. What these can teach us is how Earth's deformation behavior might change with frequency. If we think about high-frequency seismic waves, which have periods of less than hours, and compare those to tides, which act on hours to weeks to years, we can really get a sense of how Earth's behavior changes on those timescales. Since GPS stations have now been out there for a few decades, we can really start detecting these signals, and so there's quite an opportunity to capitalize on this.

Oliver Strimpel: OK. Moving up to a timescale of months, we have motions caused by the seasonal variation in the size of ice sheets, as snow caps melt and grow, and that change in mass warps the Earth as well. And these are carefully watched in many places, like Iceland and across Scandinavia. And then the ice cover also varies over thousands of years, with the onset and ending of ice ages, and you said the Earth is still reacting to the end of the last ice age.

Harriet Lau: Over the last million years or so, we've had ice ages roughly every hundred thousand years, and what these involve is a slow growth phase for most of that hundred thousand years, and then a rapid melting over 10,000 to 15,000 years. We're currently in between ice ages. Even though that excess mass, which was mainly stored in North America, Scandinavia and Siberia, has long since melted, Earth is still responding. So these long-timescale forces drive a much slower kind of deformation. For example, Canada is still rebounding after being weighed down by what we call the Laurentide ice sheet, now extinct, and this motion is about centimeters per year or so, so not too dissimilar from plate tectonics. If you go to Hudson Bay, what you'll see are beaches that are hundreds of meters high, way above sea level, and they're up there because of this ongoing uplift.

Oliver Strimpel: So what do all these movements on the daily, seasonal and even ice-age timescales teach us? You already mentioned that they offer some new insight into what these large low-shear-velocity provinces are. But can they tell us anything else?

Harriet Lau: As you mentioned at the start of the podcast, Earth can behave in many different ways. At the shortest timescales, seismic waves are triggered by earthquakes and vibrate almost entirely in an elastic manner. Plates that move due to the flow of the underlying mantle are treated in a completely viscous manner, and those motions happen at the longest timescales, of millions to billions of years. The time range I focus on is in between these vastly different modes of deformation. Of course, Earth isn't a perfectly elastic solid; otherwise, seismic waves would keep going and never stop bouncing around. But as you move from these high-frequency processes to lower frequencies, the Earth becomes increasingly viscous, and you get hints of this on these hourly, weekly and decadal timescales, and finally on those ice-age timescales. The very fact that North America is still rebounding in response to an ice sheet that completely melted around 7,000 years ago is an indication of this: the motion occurring today is delayed because of viscous forces within the Earth's interior. So how the transitions between these end-members occur is precisely what these intermediate processes could provide insight into.

Oliver Strimpel: So does that give us the value of the viscosity on these different timescales?

Harriet Lau: One thing that I've been keen to play with is the idea that viscosity is frequency dependent. We don't really treat viscosity that way, but certainly, from our knowledge from materials science, solid mechanics and other fields, these ideas are prevalent. And so the idea is to bring those kinds of ideas to Earth science.

Oliver Strimpel: When you're starting to see these different types of data sets, it must be a real treasure trove, now that we've finally got to the right kind of timescales to be able to actually mine all this data for something useful.

Harriet Lau: There are some very clever geodesists, people who study the shape and the rotation of the Earth, who've spotted six-year variations in Earth's length of day. So we roughly have 24 hours in a day, but that changes slightly every six years, and these rotational variations are thought to reflect something, again, in the deep interior of the Earth. With six years, in order to really have that signal, you've had to have decades of GPS stations measuring this. We are at an interesting time to start thinking about these things. Actually, I've read papers from the '70s and '80s by folks who theoretically predicted a lot of these unique vibrations of the Earth, and we've had to wait this long to get good measurements of them. So it's a good time to revisit these problems.

Oliver Strimpel: So by measuring the amplitude and the spatial variations of all these solid Earth motions over all these many different timescales, it seems as if we can learn a lot, and we will learn even more, about the behavior and structure of the Earth's interior. But can we integrate these observations with any other types of observations or experiments that also have a bearing on the interior structure of the Earth?

Harriet Lau: A completely different way to explore this is in the lab. There are many people who work with rock samples about the size of your hand, and they conduct deformation experiments on these rocks, applying high pressures and temperatures to mimic the conditions of Earth's interior. In these studies, they're really focusing at the grain scale: as you apply these stresses, how do the grains that make up a rock interact with each other? How does temperature affect this? How does pressure? What if the rock is partially molten? What about its porosity? So all these sorts of factors go into these lab measurements.

Oliver Strimpel: But even if we can actually reproduce the extreme temperature and pressure conditions that prevail at depth, we can't really even begin to approach the very long timescales involved, can we?

Harriet Lau: Sadly, no, unless you're extremely patient. We've been talking about processes that act on hours to weeks to years to decades, right up to millions of years, and to get to these long timescales within the lab, we have to extrapolate. We mentioned that at these long timescales, the mantle convects; it flows like a very thick liquid. Two important parameters that dictate this flow are the buoyancy, which we've discussed, which dictates whether fluids rise or sink, and also the viscosity, which we touched on earlier: how runny a fluid is. Now, we have less understanding of the viscosity structure of the mantle. We don't have anything like the seismic images of Earth's interior. Seismic waves traversing the mantle are only slowed down or sped up by a few percent, which, practically speaking, makes the math and the computation much easier, and there's also a lot more data available. Viscosity, on the other hand, varies by orders of magnitude, both with depth and laterally. So if you have a little bit of hot mantle, like an upwelling plume, it might be 100 times less viscous than the ambient mantle, and this makes simulating mantle flow quite difficult. And so we really do want to combine all these different types of information we have, from the lab, from modeling and from observation.

Oliver Strimpel: When you do actually go about integrating all these diverse sources of measurements, and I guess applying the theory, is there anything you can say about what it's told us yet?

Harriet Lau: That's a continuously developing goal of mine: to really bring together the work done at the hand-sample spatial scale, focusing on the grains, at the timescales that are practical there, which are hours, with the planetary scale, at the timescales that are important for Earth history. If we can integrate these two perspectives coherently, we might have a pathway to understand how the Earth deforms continuously with timescale. And really, this kind of understanding is quite timely. For example, I mentioned the redistribution of mass driving a lot of these solid Earth responses. If we go back to that ice example and look to a place like West Antarctica, the ice sheet that sits there is simultaneously undergoing a plethora of processes at various different timescales. It's still deforming from the last ice age, 25,000 years ago. Unlike the Laurentide ice sheet, it's still here; it didn't completely melt, but it was significantly reduced. And so the solid Earth underneath the crust is still responding to that reduction in ice. The Antarctic ice sheet is also undergoing steady changes from post-industrial climate change; it's responding to centennial-scale melt. And in the news, every so often, we hear of rapid ice collapses, like the Larsen B ice shelf collapse in 2002, where chunks of the ice sheet fall off near-instantaneously. It turns out that an important part of how sea level changes when you melt ice and dump it into the ocean depends on just how the solid Earth deforms in response to this redistribution. In order to consider all these different types of processes that even just Antarctica is undergoing, it's important to understand Earth's properties and how Earth will deform in response to both short- and long-timescale processes, if we really do want to understand how to predict sea level properly.

Oliver Strimpel: That's right. Of course, sea level at any one place is a combination of whether there's actually more water in the sea and whether the land is rising or sinking at the coastline.

Harriet Lau: Exactly.

Oliver Strimpel: What are you working on yourself at the moment?

Harriet Lau: Well, tides are still going strong, as is my research on solid Earth tidal processes. I have a graduate student who's working to consider both tides and other types of Earth vibrations simultaneously, to really home in on that deep mantle structure. The kinds of vibrations he's working on are called Stoneley modes. They're a special standing wave that vibrates along the core–mantle boundary, and we can now detect those, also with seismometers. I'm also working with experimentalists who do those hand-sample experiments in the laboratory. We're working towards a theoretical framework that can consistently bring their microscale studies to the macroscale, so that we can use them for planetary problems. And a lot of my work is performed on computers, so I've been very fortunate to continue much of this research through the pandemic.

Oliver Strimpel: What new observational or experimental results would you most like to see?

Harriet Lau: Any way we can get either seismometers or GPS stations on the ocean floor would be amazing. Of course, there's a small obstacle known as the ocean itself, but such technologies do exist, and they will only improve. Continents are really only 30 percent of the surface area, and that's pretty much the limit of where most seismometers and most GPS stations are now, so we're missing a really big chunk of the Earth's surface. But as I said, these ocean-floor technologies have been deployed here and there; examples include the East Pacific Rise and off the coast of Hawaii. So that would really help us get increased coverage for observations.

Oliver Strimpel: Harriet Lau, thank you very much.

Harriet Lau: My pleasure. Thank you, Oliver.

Oliver Strimpel: For more about Geology Bites, as well as pictures and illustrations that support this podcast, you can go to geologybites.com.