This transcript was generated automatically and lightly edited for readability.

Oliver Strimpel: This is Geology Bites with Oliver Strimpel. Several of the researchers featured in previous podcasts have talked about using the abundances of various isotopes in certain rocks to learn about aspects of Earth history. For example, Becky Flowers used the abundances of radiogenic isotopes in igneous rocks to infer the timing of their approach to the Earth's surface, and Rick Carlson used isotope abundances to probe the solar nebula out of which the Earth formed. In this podcast, it is the stable isotope abundances in carbonate rocks that are used, and used in an ingenious way, to reveal the rates at which various points on the Earth's surface were rising as mountain belts and plateaus were forming.

Carmie Garzione is Dean of the College of Science at the University of Arizona. Her research focuses on determining elevation and climate histories by combining evidence from many different fields, such as sedimentology, stratigraphy, isotope geochemistry and large-scale structural geology. In particular, her work has enabled us to pin down the elevation history of two geologically very active regions, the Tibetan Plateau and the Andes. Carmie Garzione, welcome to Geology Bites.

Carmie Garzione: Thank you, Oliver. I'm really glad to be here.

Oliver Strimpel: Why do people care about determining land elevation histories?

Carmie Garzione: Well, geologists have long speculated on the processes that form mountains, and early work in mountain ranges used observations of how mountains have deformed to measure how their shortening has taken place over time. That's something that you can easily measure without a fancy instrument. You can do that in the field, and so it was quite natural for a geologist to use that information to figure out how shortening of the Earth's surface led to thickening of the crust. We can consider crust floating on the mantle much like an iceberg floating in the ocean. And so when you thicken the crust, you can extrapolate what that thickening means in terms of surface uplift. And so it was quite natural for geologists to hypothesize that crustal thickening histories, determined from mountain belts through their deformation, related directly to very slow and gradual surface uplift.

However, there are other hypotheses about what's going on deeper in the crust, where lower crustal minerals are very, very dense, and in the upper mantle. And one hypothesis is that this dense material accumulates at depth and holds the surface of the Earth down. And so one of the questions that we need elevation histories to answer is: what happens to that dense material at depth? Is it accumulating over time and eventually dropping off, leading to a rapid rebound of the Earth's surface? Geologists have long been excited to be able to measure elevation histories, to test these very-difficult-to-see processes occurring in the Earth.

Oliver Strimpel: How hard is it to find suitable rocks that will enable you to determine elevation histories, to reconstruct past elevations?

Carmie Garzione: We need to find records of minerals or fossils that formed at the Earth's surface and were preserved over time. So this requires deposition and preservation of these in sedimentary basins, and sedimentary basins form in unique settings in mountain belts. And fortunately, we do have records of minerals or fossils that enable us to reconstruct elevation histories. So first, you have to understand the geology of the mountains, find these basins, and then find minerals that preserve a record of the basin's climate history.

Oliver Strimpel: I mean, just naively, you think the sedimentary basins would be in the valleys, and right up on the plateaus, that's where the source of the material for the basin is, not where the actual basin is. So are you kind of introducing a big bias there, potentially?

Carmie Garzione: Yeah. Sedimentary basins do tend to form where there are structural discontinuities, or even discontinuities between plates that collided in the past. And so at these zones where plates are suturing, sometimes you get basins forming that do store the history of climate there. There are also intermontane settings, sometimes within mountain belts, that naturally form a bathtub-like shape and preserve sediment. And so we also look for those types of places that may be a little bit less biased by a major structural discontinuity.

Oliver Strimpel: OK. So how can we actually go about determining elevation histories?

Carmie Garzione: There are a few methods, and the earliest attempts were really focused on trying to figure out past elevations from fossils. One of the earliest methods was called the nearest living relative method, and this relates a fossil to its modern nearest living relative. The idea is that if we know where that modern relative of the fossil lives in terms of elevation, we can apply that understanding to a fossil in the past. But as you know, evolution happens, and there are big assumptions going into relating fossils to living relatives, and so that method was considered to have low accuracy and large uncertainty.

So to address this problem, a more recent method relies on plant physiognomy. This uses the characteristics of leaves as they relate to atmospheric characteristics like temperature, and that allows us to estimate temperature. And because temperature decreases with increasing elevation, applying that method allows you to calculate elevation from the temperature information you get from leaves.

Over the past several decades, scientists have been focused on methods that rely on stable isotopes of water, that's hydrogen and oxygen in the water. And those isotopes have the same number of protons but a different number of neutrons, and so there's a mass difference between isotopes of the same element. So atmospheric processes like evaporation and precipitation that form rain or snow fractionate, or separate, these heavy isotopes from light isotopes, and we can use those processes and how they vary over Earth's surface to get at the elevation of rain or snow that occurred at the surface. Yet another method looks at temperatures from other isotopic characteristics, and again, relating temperature to elevation allows you to calculate past elevations.

Oliver Strimpel: OK, let's talk about some of these in a little bit more detail. Start with a plant physiognomy. Is there some particular characteristic in the plant fossils that we can use?

Carmie Garzione: So this method uses known traits from modern leaves, for example, leaf size or leaf shape or the margin characteristics of leaves, and we can use each of these measurable traits in comparison to air temperatures to define an empirical relationship between these leaf traits and air temperature. So we can then go to the fossil leaf record, and where we have an abundance of different types of fossil leaves, we can measure these same traits in fossils and use those traits to estimate past temperatures when those leaves were growing and were deposited in sedimentary rocks. And again, because temperature relates to elevation, that temperature can be used to estimate the elevation of where those leaves were living.

Oliver Strimpel: So I suppose that depends on there being leaves and/or fossils that have the relevant telltale characteristics. So I assume you need all these other methods because that's not always the case. You don't always have fossils. So let's talk a little bit about the isotopic methods. How does that work?

Carmie Garzione: As I mentioned before, heavy and light isotopes are fractionated by atmospheric processes. And by atmospheric processes, I mean evaporation or precipitation in the form of rain or snow. There are heavy isotopes, for example oxygen-18 in water, versus light isotopes, for example oxygen-16. And when water vapor is cooled, it starts condensing water droplets, and those droplets go to form rain. Because heavy isotopes tend to form stronger bonds, they form rain more readily than oxygen-16, and that leads to distillation, or preferential removal, of oxygen-18 from the vapor mass. So as a vapor mass moves up in elevation, it rains out more and more, and it becomes more and more depleted in oxygen-18 relative to oxygen-16, which stays preferentially in the vapor. There's a systematic relationship between those isotopic changes and elevation as vapor masses move up over mountain ranges, and we can use that relationship to calculate elevation.

The key is that we need some way to record rainwater. We really need minerals that preserve that rainwater composition. Fortunately, there is a very common surface mineral called carbonate. Listeners may know calcite, which is one form of carbonate, and this forms in various surface environments, including lakes and ancient soils, or paleosols.

Oliver Strimpel: But when you have a rock that forms, formed, say, in a lake, that lake is getting water that may have melted from a glacier of several thousand meters further up the mountain. So there are two different elevations, right? There's where the water came from and where the lake is.

Carmie Garzione: Yes. Lake carbonates are less reliable for paleoelevation estimates than are soil carbonates. A lake has rainwater occurring right on the lake, but most of its water comes in through rivers that feed into the lake, or maybe even groundwater, and so really we're measuring rainwater that fell at higher elevations relative to the lake. There's another problem with lakes: some lakes experience extreme evaporation, which also fractionates isotopes.

Oliver Strimpel: So that ancient soil, paleosol, is better because, I guess, that doesn't move around. That is where it is and stays put.

Carmie Garzione: Paleosols, or modern soils, have rainfall occurring directly on their surfaces, and that rainwater infiltrates the soil and forms carbonate at depth. And so paleosols in relatively wet climates that have carbonate are much more accurate recorders of the local rainwater composition.

Oliver Strimpel: You mentioned that you can also use isotopic methods to get a direct handle on the temperature, as opposed to the elevation. What principle's involved, then? How does that work?

Carmie Garzione: This is a relatively new stable isotope method called clumped isotopes. Those same calcite minerals, or carbonate minerals, that preserve rainwater compositions are made up of carbon and oxygen, and we use similar principles for this method, in that heavy isotopes of carbon and oxygen form stronger bonds than the light isotopes. At cool temperatures, when carbonate is forming, the bonded atoms of carbon and oxygen have lower vibrational frequencies, and so they tend to preferentially preserve these heavy isotope bonds between heavy carbon and heavy oxygen, and so there's a higher abundance of these carbon and oxygen heavy clumped isotopes. At higher temperatures, where vibrational frequencies are higher, you tend to break carbon and oxygen isotope bonds more evenly, and that results in a more even distribution of heavy isotopes of carbon and oxygen. And so what we're doing in this method is looking at the proportion of heavy clumped isotopes of carbon and oxygen as a measure of the temperature at which the mineral formed. And fortunately, this can be used in calcite, the very same mineral that we use to estimate rainwater composition.

Oliver Strimpel: Wow, that's amazing. Are you actually studying the crystal lattice of these calcite minerals?

Carmie Garzione: Rather than looking at the mineral lattice, we dissolve calcite in acid and we evolve CO2 gas from that. And then we measure the CO2 gas composition, and there's a direct relationship between the CO2 gas that comes from dissolved carbonate and the carbonate mineral lattice clumping.

Oliver Strimpel: So do you put tiny spots of acid? I mean, how big are these clumps?

Carmie Garzione: We can grind up a calcite or carbonate mineral and put it in powder form, and it still retains the clumping of isotopes. And we need a complete dissolution of that powder to get the absolute abundance of clumped isotopes. And the key is comparing that to the isotopic composition if all of the carbon and oxygen were randomly distributed. So we do need to know the carbon isotope composition of the mineral and the oxygen isotope composition of the mineral, independent of the clumped isotope composition, and we use the carbon and oxygen isotopes to calculate a random distribution for that specific sample. And that is the reference frame, or what we compare the clumping to.

Oliver Strimpel: Oh, I see. So the more clumped it is, the more of that heavy CO2 you're going to get when you dissolve it in acid, but you don't have to isolate the domain spatially. It just goes to the overall abundance of the heavy gas. So these measurements give us the temperature or the elevation at the time of the formation of the sample that you're examining on the surface. Are the results affected by subsequent burial and then exhumation back to the surface of the sample material?

Carmie Garzione: Yes, this is an important part of getting accurate paleoelevation information. So carbonate forms at the Earth's surface, and it's comfortable at the Earth's surface and stable. But when it's buried, it tends to reorder and reset isotopically. And so at depth, carbonate minerals can be recrystallized, or dissolved and reprecipitated. And when that happens, you reset the temperature of the carbonate clumped thermometer, but you also may see fluid exchange with the carbonate mineral that resets the oxygen isotopes as well. So a very important first step in applying these methods using calcite is looking at the samples under the microscope, to look at the history of diagenesis, or change in the carbonate mineral structure. So we do that before we even attempt making these measurements.

Oliver Strimpel: How does that work in the field? When you go collecting samples, are most of your samples good? Or is it like 90% kind of messed up by diagenesis?

Carmie Garzione: You can kind of look at how deformed the basin is and guess how bad the samples might be in terms of diagenesis. If it's a basin where the rocks are flat-lying and there's no evidence they've been buried deeply, then usually there's very little diagenesis. Now, if it's a basin that's been deformed by folding and faulting, often that deformation introduces fluids into the rock that can cause diagenesis. So sometimes you can see that the rock has been so altered that you wouldn't even try to analyze it.

Oliver Strimpel: So are these two methods that you mentioned, the meteoric water fractionation and the clumping in a carbonate lattice, are they totally independent measurements, even though they both derive from analyzing the actual same calcite material? And if so, I mean, it kind of reminds me a little bit of the two clocks that you have in uranium-lead dating, where you have uranium-238 to lead-206 and uranium-235 to lead-207, and only when you get the same answer for both of those clocks in a single zircon crystal, you know you're probably talking about a good date.

Carmie Garzione: Yes, that's a good analogy. So the oxygen isotope method records the water composition, from the water the carbonate precipitated from, and so it's an independent measure of rainwater. But as we also talked about, the clumped isotope method measures the clumping of heavy carbon and oxygen isotopes based on the temperature of mineral formation, and so that's a measure of a completely different atmospheric parameter. And so the power of using these two methods together is that you have completely different proxies for the same information, and that's elevation.

Oliver Strimpel: How do you actually extract the data from the rock samples that you collect in the field?

Carmie Garzione: There are a couple of different ways to measure stable isotopes in samples. One method that's rapidly becoming more used is called the laser cavity ring-down method, and that uses a laser to measure the spectrographic composition of gases, which tells you about the isotopic composition of those gases. The standard method that we commonly use is measuring the gases on a mass spectrometer. We put the sample in a vial that's at the front end of the instrument, generally, while connected with the instrument, so you can put acid on the sample. You have to do that in a completely evacuated sample chamber, a tube, so that all of the atmospheric gas is removed from the sample before you dissolve it in acid. The dissolution evolves CO2 gas, which you then run through the instrument. That gas is moved through the system and separated by strong magnets that allow for mass separation, and we measure the differences in those CO2 masses to get at the differences in carbon and oxygen composition, or clumped isotope composition. So it's all done on a mass spectrometer.

Oliver Strimpel: Let's talk about some of the results of your research, and let's start with the Himalaya and the Tibetan Plateau.

Carmie Garzione: The Himalaya and Tibetan Plateau are an interesting region on Earth. They represent one of the highest-elevation regions, at over 5 kilometers on average for the Tibetan Plateau, and one of the most vast mountainous plateaus on Earth. There's been a long-standing question of how regions like this form. We know that they somehow relate to the collision of India with Asia, but we have speculated on the processes that led to such a vast high-elevation region. And so applying isotope methods and other methods to resolving the elevation history has helped us look at the processes that have grown the plateau over time: whether there's been uniform uplift over the whole plateau, or it's been more distributed over time, based on the unique characteristics of the different fragments that make up the Asian lithosphere north of the Indian collision zone. Some of what we've learned in recent years, as we've compiled more and more data, is that the suture zones are really important boundaries that separate terranes that collided with Asia at different times over geologic history, and we see a lot of geodynamic activity at those suture zones, as well as unique surface uplift histories for the different terranes that are separated by those sutures.

Oliver Strimpel: I wonder if you can talk a little bit about how you found the rate at which the elevation increased and how significant that was, because this is an idea that Peter Molnar suggested when he spoke in his podcast. And he thought that the evidence, I think it was from your work, actually, suggested that the lower part of the lithosphere, in other words, the upper mantle, may have actually gotten hot and foundered and broken off, or dripped off, releasing the lighter, more buoyant part of the lithosphere to basically float back up and increase its elevation to what we see today. So is that actually what your measurements discovered?

Carmie Garzione: Peter Molnar's mantle foundering idea is something that we do think has occurred in portions of Tibet. By piecing together a lot of work done by many, many different scientists, we have been able to track surface uplift events, looking at the differences in elevation between different sites that record continuous time periods of surface elevation change, and there are unique volcanic events related to hypothesized mantle foundering. So the paleoaltimetry really does match what we believe we can see from volcanic history in the Tibetan Plateau. It's just a little bit more challenging, because we have to piece together histories from a lot of different basins that provide different snapshots in space and time.

Oliver Strimpel: But your measurements are consistent with a somewhat rapid increase in elevation at a certain point. And what did your research reveal about the history of elevation in the Andes?

Carmie Garzione: So the Andes has been a much less challenging problem to solve, in part because there is this beautiful region in the middle of the Andes, between the western volcanic arc and a mountain belt, or fold-and-thrust belt, in the east called the Eastern Cordillera. That region's called the Altiplano Basin. And you know, I like basins. I mentioned that these are the sites that preserve the minerals that we need. And so the Altiplano Basin has been accumulating sediment over a long, long time in the history of the Andes, and it provides a continuous record of surface uplift. And so we've found evidence of rapid pulses of surface uplift, all within the same record, and we can see surface uplift on the order of 1 kilometer per 1 to 2 million years at several time periods in the Altiplano Basin and local basins surrounding that region. So we can tie that directly to processes like mantle foundering, as well as another process called lower crustal flow. That process occurs when a region of very thick crust experiences heating because the lower crust is pushed deeper into the Earth, where it's warmer. That region of very thick crust tends to squeeze or push crustal material into lower-elevation and thinner crustal regions. And so it basically props up the crustal thickness and leads to surface uplift, again over relatively rapid time periods.

Oliver Strimpel: I think you said that you get about a kilometer increase in elevation over the order of a million years. Is that considered to be very fast in geologic terms?

Carmie Garzione: That is a very fast rate of surface uplift. When you look at the rates calculated from crustal shortening estimates that can be related to crustal thickening, those rates tend to be 10 times slower than a kilometer per million years.

Oliver Strimpel: Wow. So you're saying that really 90% of the elevation increase would be unaccounted for simply by crustal shortening.

Carmie Garzione: Not really. I'm saying that crustal shortening is actually very important in setting the elevation limit of a plateau. That crustal shortening occurs, and it's basically waiting to have that anchor at depth, lower crust that's very dense, that contains very dense minerals, as well as the upper mantle, to have that material founder and drop off the bottom of the crustal lithosphere. When that foundering occurs, the thickening that's already taken place by shortening realizes its full potential and rebounds to the elevations that we're seeing today.

Oliver Strimpel: So if I understand you right, you're saying that, in a way, the mechanism that enables the lower part of the lithosphere to heat up and then drop off is a consequence of the crustal shortening that thickens the crust to begin with, that pushes it down to that hot area where it gets soft and can founder. So I suppose the two processes work hand in hand.

Carmie Garzione: Yes, that's a great way of describing it. And there's an additional process that occurs when you thicken the crust and the lower crust is mafic in composition, more like the oceanic crustal lithosphere. When you push mafic lower crust deeper into the Earth by crustal thickening above it, you convert those mafic minerals to very high-density minerals that make up a rock called eclogite. And so in essence, you're actually making the lower crust more dense by thickening the middle and upper crust above it. And so that lower crustal material, which is very high density, can contribute to ultimate foundering of the lower crust and mantle lithosphere.

Oliver Strimpel: Do we see evidence for breakoff of this material in the mantle when we look at seismic tomographic images?

Carmie Garzione: We do. And in fact, sometimes we can see the remnants of foundered material, or we can see the remnants of slabs that have been subducted.

Oliver Strimpel: But do we actually see any blobs underneath the Andes or underneath Tibet that could be those foundered bits that are making their way down into the mantle?

Carmie Garzione: We do see that material. It's hard to image in all places of the Andes, in part because we don't have high-resolution geophysical images to resolve it in some parts of the Andes where we think this has happened, but we do see it in parts that are better resolved. Likewise in Tibet, we do see evidence of foundered slabs and what some people have interpreted as mantle lithosphere.

Oliver Strimpel: Did you go into the field yourself in Tibet and the Andes to collect your samples?

Carmie Garzione: I did. One of the joys of being a geologist who collects samples that we need to measure in the lab is that you have to go find those samples. I've been able to do fieldwork in a lot of different regions in the Andes of South America, and in many different parts, both within and on the periphery of the Tibetan Plateau.

Oliver Strimpel: Wow, what's it like to go to those really remote places to dig up? I guess you get carbonate samples.

Carmie Garzione: The rocks are not always where the road is. Sometimes we really need to get beyond the confines of roads that cut across these regions and go overland. It's really interesting to see some of these remote regions, very beautiful and sometimes untouched regions. Some of the challenges of doing that are that in the Tibetan Plateau, for example, there is permafrost, and in the summer, when the permafrost is melted, the region just turns into a gigantic bog, which you can't drive over. And so we've needed to do some of our fieldwork in the late fall and winter months, when the permafrost freezes. You can imagine that it's very cold doing fieldwork at that time, but it's also thrilling. It's a really unique experience and totally worth the effort.

Oliver Strimpel: You told me that you've been to Tibet thousands of times. Are you planning to go back?

Carmie Garzione: By geological terms, Tibet is a physiographic feature. I've been to different states that make up Tibet that aren't necessarily the state that China defines as Tibet. I hope to go back. It's challenging right now with COVID, and travel really hasn't been possible.

Oliver Strimpel: What are you working on?

Carmie Garzione: At the moment, I'm working on a couple of interesting projects that aren't necessarily focused on paleoelevation, but do ask important climate questions in high-elevation regions. A couple of the projects I'm working on in Tibet are looking at the isotopic composition of water in a region in the northern part of Tibet, as well as the temperature history of that region, to understand the history of permafrost. In Tibet, we're looking at a critical time period in the Pliocene, when climate was warmer and atmospheric CO2 was similar to what we have today, related to human combustion of fossil fuels. So the question we're asking there is: under CO2 compositions similar to today, was there permafrost in the Tibetan Plateau? And this has important implications for the stability of permafrost in Tibet today. What's interesting is that we're finding that during the Pliocene, when atmospheric CO2 was higher and similar to today, we don't have temperatures that are cold enough in the plateau to sustain permafrost. And this underscores the need to really study permafrost regions in Tibet to understand their stability, and what global warming may cause in terms of loss of permafrost.

Oliver Strimpel: Carmie Garzione, thank you very much.

Carmie Garzione: Thank you. It's been a real pleasure.

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