Skip to main content
Loading Explore guides…
Plate Tectonics · Acedra
Home
Explore
Geography
Plate Tectonics
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
171 segments
Copy
Hide video
0:03
SPEAKER: All right. Great. So I think without further ado, let's hand it over. ROBERT STERN: Thank you. Well, good morning. It's a real pleasure to be here, honor and a privilege to talk to such a remarkable assemblage. And it's a special pleasure to be out of Texas in July, so. [LAUGHTER] Last night, I was like, I should have brought a warmer coat. It's wonderful. So today, I want to try and get through some of this stuff. And I hope to be interrupted frequently.
0:36
Don't throw any rotten tomatoes. I hope that it elicits some concern for my sanity, but I think that you'll find some of these discussion items interesting. What is plate tectonics? Basically, we all know what it is. But let's talk a little bit more about it because as we go through the question of what makes the plates move, this issue of what plate tectonics is, it becomes important. And then we do have to address this issue. We are uniformitarianized into thinking
1:07
that what's going on today is common and expected. The present is key to the past. If we understand our planet, we understand other planets. But I think this question, is plate tectonics common or unusual, is critical for understanding the history of our planet. And then we do need to discuss a little bit about, OK, who cares? So what? When plate tectonics began, it really does turn out to be very important, in terms of understanding continental crust,
1:37
to understand, first of all, when the processes that are now building it and as you'll see, destroying it started happening. And then we do need to talk a little bit about how planets cool. And then this is a particularly good place to talk about the Urey ratio, since Harold Urey got his PhD in chemistry here almost a hundred years ago. And then finally, if we have time, we can wander through what I think are important petrotectonic indicators of plate tectonics.
2:08
I think each of us has an opportunity to say what we think are the important hallmarks of plate tectonics and start that discussion. So without further ado and with the full expectation that you'll interrupt me without raising your hand and Cin-Ty will have to run around or Richard's going to have to run around with the microphone, just chime in. I may get halfway through this talk. Maybe we'll get lucky or unlucky and get through the whole thing. [LAUGHTER]
2:39
So let's talk first about what is plate tectonics. And just to remind you, Earth has 11 large plates and many more small plates. And those of us that are interested in the continents have to put plate tectonics in a separate box because most of the areas where the plate tectonic approximation doesn't work are on the continents. This has got to do with the rheology of continental crust as opposed to oceanic lithosphere. But anyway, where we have plates?
3:13
They are basically a code word for lithosphere. Lithosphere, plates-- these are exactly the same thing. And it's one of the problems I have with teaching tectonics to undergraduates and graduate students, is they tend to equate the plate to the crust. And that's wrong. The plate is the crust plus the uppermost part of the mantle. Plates are typically 100 and 200 kilometers thick, depending on how you define them-- mechanically, thermally, whatever-- up to many thousands of kilometers across.
3:44
And they're moving at speeds that are, fortunately, we can relate to, centimeter to 10 centimeters a year. That's how fast our expendable body parts, nails and hair, grow. Some of us, it's more expendable than others. [LAUGHTER] So these are the fundamental features of plate tectonics. We have destructive plate boundaries and constructive or divergent plate boundaries-- mid-ocean ridge spreading centers, subduction zones-- and then second order features like transform faults,
4:21
which probably isn't going to be second order when this one fires off. But in the grand scheme of things, this is where the action is, in the spreading ridges and the subduction zones. And this is an old, old representation of the plate approximation. And it's a pretty good one. But what you notice is this flow of the asthenospheric mantle is just the opposite of what we often see. That direction of flow is one of the most important things that we really need to resolve.
4:54
So these are the top 14 plates in area-- the Pacific, by far the largest and getting smaller all the time, North America keeping up there, Eurasia, Africa. And down the list, top 14, down to the little Cocos plate and then many others, ranging down in a fractal, down to your microplates or whatever. So one of the most important characteristics or definitions of plate tectonics is this geometric approximation that we can describe plate motions
5:27
by motion around an Euler pole, that the relative motions between the two plates is defined by this angular separation around a pole of rotation. And there's different methods for determining the Euler pole transforms, earthquakes, all sorts of things.
5:50
And this is our modern understanding of the major plates' relative motion. What I want to draw your attention to is not the plates, because they behave so well, but these areas in the continents where plate tectonics kind of goes out the windows, where you have broad zones of deformation that the plate tectonic representation is almost meaningless. What's the plate boundary between India and Eurasia? That's a fool's game, really, because it's-- all of this
6:20
strain is distributed all through basically China and many other places-- the East African Rift, the Andes, and the Western Cordillera. When you're on the continents, you're not as interested in plate tectonics as you are if you're in the oceans. And it's not really a surprise that plate tectonics came out of marine geology more than continental geology. So we've got a really good understanding of how the plates move and where the plate tectonic approximation fails us.
6:56
And that's a very useful thing. But where we're struggling and maybe we don't appreciate what we don't understand is the difference between kinematics and dynamics. So kinematics is a very simple description of the motion. And in plate tectonics, that was the great breakthrough, was that the plates-- that the Earth's lithosphere was broken up into a number of plates. And the motion of these plates could be described. And it was an amazing thing that happened in the late
7:30
'60s and early '70s that we were able to do that. But what we left in the dust was understanding what caused the plates to move. So we celebrated our success in a kinematic description, just describing what motion. And maybe we didn't do as good of a job with the dynamics. That's where the real future opportunities I think are, is coming up with dynamic interpretations. So just to make it more intuitive, I like that word, talking about making things intuitive,
8:08
a kinematic description of a car's motion would center on the direction, speed. And maybe in a little bit of detail, you'd say, well, let's talk about what the tires are doing. Those would be kinematic descriptions. And a dynamic description of a car's motion would maybe also discuss gasoline, engine mechanics, thermodynamics, transmission, wind resistance, things like that. Not to degrade kinematics, but dynamics is a different kind of understanding
8:40
and perhaps a more useful understanding of the way things operate if we can get there. So we've got a number of very powerful kinematic models for the plates. It's been a remarkable demonstration of plate tectonics. With the development of satellite geodesy, we are able to check what we came up with in terms of plate motions based on magnetic anomalies and Euler poles and all of the things that we were limited to in the 1960s, 1970s, and '80s. And then with this huge breakthrough
9:29
in satellite geodesy, we were able to go back and say, now let's see if that was right. And the plate tectonic approximation is remarkably consistent between these geologic indicators of plate motions and the geodetic indicators of plate motions for the larger plates. The smaller the plates, the less things agree. But these are some of the standards. NUVEL-1A, these are geologic models of kinematics. And the more recent version of this is called MORVEL. I don't even remember what those--
10:07
that acronym stands for, Modern Relative Velocities or something. But these are based on seafloor magnetics, transform orientations, earthquake slip vectors, and are useful for back 3 million years or so. And in contrast, the geodetic ones, beginning with REVEL and the most recent one, GEODVEL-- I'm not even sure how to pronounce that-- these are nearly instantaneous plate motion from GPS, good for the last-- or mostly GPS, other kinds of satellite geodesy. Only good for a few decades, but these
10:43
are both remarkably consistent for the large plates, remarkably. It's very confusing when we start using these different techniques to understand what's going on in the zones of continental deformation. But this is a nice summary. Plate kinematic models. The geologic models are with the yellow arrows. And the satellite geodesy are with the purple arrows, a little bit hard to see the purple. For the most part, they agree, except for the smaller plates. AUDIENCE: Sorry, can I ask--
11:19
ROBERT STERN: Yeah, yeah. AUDIENCE: So when you said these two, the geodetic and the geological plate motion model agree with each other for large plate. The small one, what are the plates that don't agree? ROBERT STERN: Well, the most spectacular, I would guess, would be the Tonga-Kermadec. What's the name of that plate? Doug, do you remember what you call that, the Lau Basin plate? The Tonga-- the thing that's on the east side of the multiple spreading centers of the Lau Basin, that was moved.
11:51
With geologic, it was argued to move much more slowly. And now with GPS, we can see it's-- AUDIENCE: The reason that I ask this question is I can't see the geodetic ones may differ from the geological ones due to processes like post-glacial rebound, for example. And that may actually cause some modern deviations from the geological-- ROBERT STERN: Probably-- AUDIENCE: I guess it's more whether the plates like the Antarctic, there, you may actually see more influence from the rebound.
12:22
ROBERT STERN: I don't think rebound is a big part of the inconsistency because that's vertical. And I think what these models are nearly all horizontal motions. By and large, these models for plate motions are purely horizontal. But Doug, maybe you have some-- AUDIENCE: It is kind of neat that in Antarctica, where you're looking at something very subtle, that the difference in velocity between east and west are certainly-- well, it's something like, say, half a centimeter a year.
12:55
The post-glacial rebound signals become important. So if you look at horizontal velocity now, yes, across Antarctica, it's really wild because you have a lot of local glacial [INAUDIBLE]. But that's a very slow, much less than half centimeter a year level. ROBERT STERN: Yeah. AUDIENCE: But there must be also some averaging over earthquakes, right? These motions along plates aren't smooth and uniform, ROBERT STERN: So. Do we have a geodetic plate measure expert in the audience? [LAUGHTER]
13:30
I'm not that person. This is a hugely vigorous field in our science. And these details of what I consider details in terms of the plate scale, what's happening vertically with post-glacial rebound over the earthquake cycle, these are huge concerns for things like the plate boundary observatory. AUDIENCE: So in some ways, it's a little surprising that they do that well. ROBERT STERN: Yeah. We don't get very many kudos from ourselves very often.
14:00
But I think as a community, we should congratulate ourselves on doing a pretty good job. We had no idea GPS was going to come along and test all of our models. And it's a remarkable demonstration of the strength of those arguments that happened 40, 50 years ago. Yeah? AUDIENCE: Chris Pierce. Just to note that some people have argued that there is a degree of mismatch to say something about a change in plate motions over the last 600 million years. This is all that.
14:37
There's some second order agreement and people have models for discrepancies in [INAUDIBLE].
14:47
ROBERT STERN: Yeah, there's even signal in the noise. Doug, you have another comment? AUDIENCE: I was going to say I'm not an expert in this area. But early on, when you guys just started, people were interested or wondering whether there's any temporal changes as you get towards the center of plates. And there's not. The plate rates are extremely constant as long as you're not within, let's say, a hundred kilometers and 200 kilometers of an active plate boundary, where
15:16
you start to get to see the effect of the earthquake cycles. ROBERT STERN: Yeah. What I take from this, not being an expert in either the geologic or the geodetic measurements of plate motion, is that there's not a lot of controversy. There's a lot of information in these issues of glacial rebound, the earthquake cycle, whether the plate approximation works. And my sense is that for the most part, the two large scale information sources are remarkably consistent. OK.
15:51
So we're kind of in a little bit of the position of the Native Americans when they first saw steam locomotive coming across. It's like, OK, what the hell is that? And the first thing they developed or any culture that was unassociated with-- unfamiliar with any technology would have been able to figure out would be the kinematics, how fast these things went, a number of issues about what the relationship between the wheels and the motion was. But the dynamics of a steam engine
16:28
would have been something quite different. There would have been almost no way until they were trained in what a steam locomotive was-- how heating in here drove steam and how the steam drove the cylinder and how the cylinder drove the wheels-- those dynamics, that's where we are, in a way, with plate tectonics. We're in the same position as the Native Americans were in facing the iron horse. Kinematically, OK, we can figure that out. Dynamically, that's going to take us a little while.
17:05
So we have these kinds of things. We teach them. Most introductory geology texts show them. And what they say-- and when the students naturally ask, why do the plates move-- the answer is mantle convection. So here, we can see an interpretation. There's a million of these. I'm not picking on any one. I'm picking on our general ignorance about what makes the plates move. You see this. And this, you notice the direction of the asthenosphere is quite opposite the one that Isaac showed originally.
17:39
But it's this huge thing circulating from the core to the surface that makes the plates move. That's really a tautology because what plates are is part of the convection. So in a way, if you ask what makes the plates move and answer mantle convection, you're saying, what makes the upper part of the convection cell move, the answer is the convection cell makes the upper part of the convection cell move. OK. Maybe you can get away with that with a group of undergraduates.
18:10
But I don't think we can get away with this understanding. It's a tautology. A tautology is something very useful to recognize when you're-- the main evidence that you're making for your argument is your argument. And that's what the statement that the plates move because of mantle convection is. It's a tautology. So let's talk a little bit about our understanding of what makes the plates move because we can't really understand the history of plate tectonics on this planet
18:45
unless we have an understanding of what are the forces that make the plates move today, because, fundamentally, the question of when did plate tectonics begin is a question, when did the forces that drive plate tectonics first become powerful enough to overcome the resistance of the lithosphere and start subduction or spreading or whatever causes them? So this is a summary from a really classic paper by Forsythe and Uyeda in 1975. They're almost up to a thousand citations.
19:24
But they're still being cited. It's one of these papers that's still keeps providing insights and motivation for studies. But they basically tried to summarize all of the forces that were involved in plate tectonics. Ridge push, we talked a little bit about that with Claudio. Maybe you could lump in plume push into the ridge push force. This negative buoyancy force, the sinking of the slab in subduction zones. Slab pull, which is-- the way they've shown it is transmitted
20:04
from something going on down here to what's going on in the plate. Trench suction, which is a mysterious force. And then the resistances-- the ridge resistance, the bending resistance, the fact that it takes a tremendous amount of energy to bend the thick lithosphere, the slab resistance, the fact that the slab has to penetrate viscous mantle, the overriding plate resistance, the mantle drag under the ocean. And they distinguish it from the mantle drag
20:40
under the continents, which is a theme we'll probably return to, that the plates seem to experience different resistances, depending on whether they're in the oceanic realm or the continental realm. And these mantle drag forces, both for the ocean or the continent underneath them, can be either positive or negative. We saw that with the two cartoons, the one of Isacks and the one of the whole mantle convection. You can have the mantle flowing either way. You can make strong arguments.
21:15
So let's see if we can put this into a nutshell rather than spend too much time on this really interesting problem, and talk about the two main things, which are ridge push and slab pull. And Claudio, yesterday at the end of the day, put this in a very nice synopsis, these tensions between, OK, what do you think is more important, ridge push or plume push, which is an interesting variation, versus slab pull? And so people have modeled ridge push in a number of ways.
21:52
It's as a body force and as a boundary force. So as a body force, it's attributed to the cooling and thickening of the oceanic lithosphere with age, so the fact that the ridges are higher and they're made of rock and they want to move down. We see this over and over again. In terms of what's happening with origins, the Himalayas can probably only get to a-- Tibetan plateau, can only get to a certain elevation because of the body forces that are making it want to collapse.
22:22
Ridge push as a boundary force is the idea that what's-- there's something that's forcing its way up into that region between the plates and elbowing its way in there so that it causes a horizontal pressure gradient.
22:44
So anyway, Forsythe and Uyeda and other people have come in and tried to quantify that. So the ridge push force, very simply, is just rho gh minus the rho g of the water overlying it. And you can simplify it just to get a order of magnitude number, which is on the order of 4 times 10 to the 12 newtons per meter. So what we can do is do the same thing with a slab pull. And this is the sinking of the lithosphere in a subduction zone. And this is basically reflecting the fact
23:26
that the density of the lithosphere, because of the coefficient of thermal expansion for mantle peridotite, is thought to be about 1% higher than the asthenospheric mantle. So that-- just that small difference provides a body force for the lithosphere to sink into the mantle. And we argue about what is the real change in difference in density between the lithosphere and the subduction zone versus the ambient mantle. What is eclogite do when you convert into eclogite? What does that add to it?
24:07
But in general, the slab pull force depends on the dip of the subducted slab, the length of it, the age of the subducted slab-- younger oceanic lithosphere has much less-- has a much-- well, somewhat slower delta rho than old oceanic lithosphere-- and the temperature difference between the slab and mantle, which controls the coefficient of thermal expansion and the density differences. So with estimates, you can do the same sort of thing, the body force that you get with a slab pull
24:42
and that, you can do that. And they come up with something on the order of 5 times 10 to the 13th. Other people have calculated at 3 to 5 times 10 to the 13 newtons per meter.
24:58
But what you see is that the ridge push force is an order of magnitude smaller, according to just about anybody's calculation, than the slab pull force. And at the present understanding of what's making the plates move, it's the sinking of the lithosphere in subduction zones that drives the plates. And this is a point that I understand there is some controversy. And I invite the junior CIDERites to stake out your own position, examine this. Don't take my word for it.
25:29
But this is basically my understanding of what the bulk of the geodynamic community has to say about what drives the plates. And I really invite the senior CIDERites to comment on this because if I'm leading anyone astray, this is the time to set me straight.
25:50
OK. Yeah? AUDIENCE: You haven't said anything about the pull, the traction. You've just compared the ridge push to the slab pull? ROBERT STERN: Yes. AUDIENCE: Do you want to say a little bit about that maybe? ROBERT STERN: I think I'd rather you said something about it because I really-- what I saw on those cartoons is that Isacks and Alvarez show the asthenosphere moving towards the ridge, away from the plate. And what I see other people showing is that they have the asthenosphere moving
26:20
the other way. So the first question I would have is, which way is the asthenosphere moving? Alvarez, I think, made a fantastic argument that as the Pacific shrinks, the asthenosphere is getting squeezed and it finds a way out in the South Scotia arc. And it finds a way out in the Antarctic-Australian discontinuity and maybe out through the Caribbean, that it's moving in neither one of these directions. But the traction force resulting from whichever way the asthenosphere is moving, I really
26:59
am uncomfortable saying much to anybody, other than this is confusing and it doesn't seem to be important. Would you agree with that? AUDIENCE: I don't have a particular opinion. I guess what I was getting at is you try to estimate the magnitude of the forces. Do we have any possibility of coming up with a similar estimate of the magnitude of the forces of traction? I think the answer is no. I don't know. AUDIENCE: Yes, the force of balance, they should be equal. ROBERT STERN: What's that?
27:31
AUDIENCE: The forces are all balanced, right? So if you know the driving force, then you must have residual force to balance that up. ROBERT STERN: Yes, but I think the question is, in this simple-minded description of what's driving the plates, have I left out an important point, which is the traction force? So for people that aren't familiar, the idea is that there may be a force acting on the bottom of the plates. And in fact, we think there is, because if you have a continent
28:02
on a plate, that plate is going to move more slowly than if you don't. This is one of the observations. And it's not too surprising, because continents have these thick keels. And so they extend down into the more viscous part of the asthenosphere or maybe below the asthenosphere. And they move more slowly than oceanic plates, which have thinner lithospheric keels. AUDIENCE: Above? ROBERT STERN: Yeah. AUDIENCE: I've heard just the opposite-- ROBERT STERN: OK.
28:30
AUDIENCE: --that the up-down sailboat bottom. The continents have these keels. They catch the convective forces and it pushes them along faster. And that's not observed. ROBERT STERN: No, it's not observed. You might think that one observation would be, OK, if there is a mantle wind-- and people have argued there is an asthenospheric wind-- that you might catch it with, what you said, upside down sailboat. But-- AUDIENCE: I had a [INAUDIBLE] several years ago that if you--
29:03
today, you have the impression that plates with continents moved slowly. But he argued on the-- using past the motions that this was probably just a snapshot and that when you look over time, there's no systematic dependence between the continents and the velocity that's not studied, at the same time as you process new data.
29:26
ROBERT STERN: So yeah, I get some questions over here.
29:31
AUDIENCE: The same way that you said at the beginning that we can't just say, well, the plates are part of the convection system, can we really totally isolate the plates and do a force balance on the plate without considering the rest of the convective system? Is it too simplistic to just say, here, I have my plate and what's pushing, what direction, and not consider the whole rest of the system? ROBERT STERN: Well, I can't answer that question, except that success in science
30:01
often goes with the ability to reduce a problem to its simplest parts. And if you want to make a big name, answer the question for all the plates. But if you want to make a name, doing it for one would be a good step in the right direction. AUDIENCE: I'm really talking about all the plates. I'm talking about is the fact that you have an entire convective system and not just the little thing, lithosphere at the top. I think you do have the large [INAUDIBLE]
30:30
associated with places that have the largest temperature differences, which would be the plate. But still, it seems like you're neglecting the rest of what's going on. ROBERT STERN: So Claudio-- AUDIENCE: I have just a comment on that. [INAUDIBLE] Now, of course, for a student who appears in the situation and in a lot of work done on the topic, if there is a slab pull or slab suction, where there are more important than the slab suction is the drag of the mantle or drag and drift
31:03
and how the lithosphere participate directly into the convection system. So for example, from the [INAUDIBLE] paper, we are talking about 40 years ago. And I think, for example, an interesting paper that students can look at is the paper by Conrad and Carolina Bertelloni, where they tried to take percentage of the importance of the pull versus slab suction, looking at the kinematics of the up plate and kinematics of down-going plates. This is later published in Science about the two.
31:44
I think to give an overview of the problem-- and that is, of course, they can have a number-- let's say that, for example, something can be [INAUDIBLE] or suction is [INAUDIBLE]. That is to say that it's difficult to discriminate where is one of the others and Adrian was pointing out that-- and also after our discussion. And it's clear that the final issue concerns that in some plates, you can have another-- one is more relevant and some other elements are more relevant.
32:18
And it's a room for work. ROBERT STERN: No, and I think that, as this discussion has shown, there's a tremendous interest in understanding in greater detail what makes the plates move. So my issue here was, let's back-- AUDIENCE: I was going to ask him if you could put that on the Wiki, that paper? AUDIENCE: Pardon me? AUDIENCE: Could you put that on your Wiki? AUDIENCE: Yeah, sure, sure, I would. Also, I hear a comment from Adrian. AUDIENCE: Another way that you could
32:46
think about this is think about induction of an energetic point of view. And there's two issues. Either you've got cold stuff going down or you've got warm stuff coming up. What we think about the heat fluxes through the mantle, it's probably mostly the cold stuff going down rather than the hot stuff coming up. So that really leaves you two options, either the cold stuff, which are the plates, are driving the motions by the stresses that get transmitted through the plates,
33:10
or they're driving the plates through the large scale circulation of those seeking motions actually drive actually, one of the other and probably a bit of both. ROBERT STERN: And there was another-- yeah. AUDIENCE: I guess I just want to add to that. One way I look at it is, something that is kind of helpful for myself, but I'm not sure it's going to be useful for the rest of the audience, that is also, just by looking at energy-- so we know amount of convective heat flux is about 36 terawatts.
33:41
And then you can actually just take the plate motion that we observe, we have. And you estimate the energy flux associated with the downgoing slabs. And you actually see that the downgoing slabs really account for about 80% of the surface heat flux. So that, to me, that is actually very good evidence that the convection of plate motion is largely driven by the plates sinking slabs, not the other way around. And so I guess we will be hearing this top-down tectonics or bottom-up tectonics.
34:21
I cannot see, I cannot envision how this bottom-up tectonics can ever work. So I think this is one-- I guess what I'm saying is I actually agree with what you summarized. I think that we shouldn't really confuse ourselves by bringing all these kind of various forces. They're probably there, but probably not very important.
34:47
ROBERT STERN: Question? Yeah? AUDIENCE: I think this problem you brought up, which way the asthenosphere is moving, is very important. Nobody here has addressed it yet. I'm wondering, do we have an answer to that? Something's wrong. Our geology, beginning geology textbooks, just like you point out, some of them show up going one way. Some of them show them going the other way. ROBERT STERN: They went that way. [LAUGHS] AUDIENCE: So the question was, does anybody know here?
35:11
ROBERT STERN: No, I don't think there is a community that actually has considered this in detail. And maybe this is part of the problem of keeping up with the literature, which the junior CIDERites, I think, will appreciate this, that this expert discussion has happened. And we really wish they would come in and explain it to us in a way we would be able to put into the introductory textbooks or make the argument, hey, you guys need to stop reusing those old figures,
35:37
just adding color to them, and update them with some of our modern science. [LAUGHTER] AUDIENCE: So I gather we don't have an answer to that question, because nobody said yes. ROBERT STERN: No, we do, we do. But let's hold this off. I think we do have a discussion about this over coffee. But yeah-- AUDIENCE: I will just very quickly say that one of the observables that we have that can tell us about shear at the base of lithosphere is seismic anisotropy. We'll talk about that on Friday.
36:03
AUDIENCE: But you can't tell which direction-- AUDIENCE: Well, you can't tell which direction. You can, in theory, get a sense of shear if it's pushed on the base versus pull from above. It's a tricky observation. But I think there are observations that I think definitely compare to this problem. ROBERT STERN: Very exciting, that's a very exciting area of new research, is this idea of getting shear wave anisotropy at depth. So you can start to put together more powerful arguments.
36:34
So I'm going to move on, if you guys will let me, although I'm very tempted to stay here. [LAUGHTER] But as you'll see, I'm less than halfway through my lecture. And I did want Pauline to make her appearance. Pauline decides against a career in business there, so we all of feel like that, but. [LAUGHS] All right. Let's talk about how planets cool off, change gears a little bit. And here, we have the five planets that we understand. And we promote the moon to be a planet because it's
37:12
almost the size of Mercury. Absolutely one of the most exciting things that's happened in the last few years has been this mission to Mercury and some of the surprises that it's developed. But Mars is absolutely an attractor for all kinds of research. Venus is-- I hope we start to understand Venus better because Venus is our twin. And when the Magellan mission started, they fully expected to find plate tectonics on our twin. And one of the disappointments to parts of the community
37:46
was that it doesn't have plate tectonics. So let's just talk about how these planets cool off, because if you think about plate tectonics, it's just a way of cooling off the planet. It's a special way. It's a remarkable way. And as our small sampling shows us, it's a very unusual way. But in terms of understanding Earth history, if we looked at those five planets and all of them had plate tectonics, we'd say, hey, that plate tectonics, that's what planets-- that's what silicate planets do.
38:17
And if you look at five and only one of them has plate tectonics, maybe you go, well, maybe plate tectonics is unusual in space and it's also unusual in time. So things get a little more interesting when we can bring the other-- our neighbors into the game. Part of the problem is that some 60% of our-- or 40% of our sample are not doing anything. Really, they're just dead. So the moon and Mercury are both tectonically dead. Certainly, they're cooling off or heating up or whatever they're doing.
38:52
But they're not having any magmatic activity. They're not having any significant deformation, other than when something collides with them. Mars, Venus, and Earth, on the other hand, the larger ones, still have enough energy to put on a show, to erupt once in a while, to deform once in a while. So obviously, you have to have this size filter in order to use these. So these last three-- Mars, Venus, and Earth-- give us some idea of how the planets cool off, how do planets get dead.
39:25
So one of the concepts that came out of Venus was this concept of the stagnant lid. So we have plate tectonics. This is from Korenaga's very recent summary of plate tectonics, when did plate tectonics begin, in Annual Reviews of Earth and Planetary Sciences. And for the junior CIDERites, if there's one place to go to, to try and keep up with our science, look at Annual Reviews. They always have the big areas identified. There's always something interesting in there
39:58
that provides you an up-to-date overview. If I had just one thing to read, it would be Annual Reviews. But anyway, Jun Korenaga considered plate tectonics, which we know and love. And then we have this thing called stagnant lid. And what stagnant lid is, in both cases, we have this conductive layer-- we call-- it's the lithosphere-- and underlain by a convecting layer. And the difference is, the Earth and plate tectonics, that conducting layer is moving. Whereas, in these other two planets,
40:34
as we'll see Venus and Mars, the conducting layer just sits there on top, doesn't move. Well, it ruptures, magma comes up through it. But it doesn't translate horizontally. We could put all the GPS there we wanted to and we would be getting, yeah, still not doing anything. But this stagnant lid mode of cooling is very interesting for a number of reasons. We'll talk a little bit about them. Let's just take a look at one of the stagnant lid planets. This is one of my favorites, is because looking
41:09
back at how we saw reality, we can see things where we go, how could we be so blind as to draw these lines on here? Well, there was some relationship to some of these lines. I'm not sure exactly which of Lowell's canals would line up with the Valles Marineris. But it is remarkable looking back at what our maps looked like compared to what we think they are today. Mars is-- it's got-- everybody knows it's got the largest volcano in the solar system.
41:43
And this is something like what happens when you have reasonably low gravity and a mantle plume or some mantle instability that just keeps pumping away. It just makes these big volcanoes of the Tharsis region. It's one of the characteristics of Martian geology, is these big constructions that don't move anywhere. Obviously, there's a lithospheric strength issue behind the size of the Martian volcanoes. But this is one manifestation of stagnant lid planet. There's no plate tectonics on Mars.
42:21
And things are behaving like that. Venus is the other one. One of the remarkable accomplishments of the late 20th century was the mapping of the Venusian surface with the Magellan spacecraft. And basically, you've got all of these highs and lows. The reds are the highs, and the blues are the lows. And one of the arguments is that what you're seeing in Venus is a periodic resurfacing. So every 300 to 500 million years-- this is a very flexible interpretation because it's based on crater counting.
43:00
And the Venusian atmosphere is so dense that the big crater-- that most meteorites burn up. And then any impact is likely to get covered over with basalt. So the age of these surfaces are difficult to understand. But an understanding of this stagnant lid planet is that periodically, it resurfaces itself. And we can think about why that happens. AUDIENCE: Question really quick. ROBERT STERN: Yeah, yeah. AUDIENCE: So this may be really naive,
43:30
but it seems actually the analogy between Venus and Earth isn't super great, because we can put things pretty far in the Earth before they vaporize and we can't even get through the atmosphere of Venus. And so really, you would be doing a lot of the cooling at the top of the atmosphere. So then it should act more like a gas giant really, in a way, with a very big core maybe. ROBERT STERN: Yeah, and then that's a hydrodynamic question, which is way out of my comfort zone.
43:59
Obviously, there's huge differences. And one of the big ones is the density of the atmosphere and the temperature at the surface, which is ultimately controlled by that CO2 atmosphere. So yeah, it's-- those of us that are geologists-- there's some planets that geologists are very comfortable talking about, Mars being the most-- we love talking about Mars because, yeah, we look at that. There's a sediments or that's a rift or whatever. This planet makes us a little nervous.
44:27
It doesn't make us as nervous as Jupiter does. And Jupiter's like, do we have to talk about Jupiter? We have no intuition whatsoever, other than there are hurricanes. We like the hurricanes. AUDIENCE: Just one thing, most of it related to probably the atmosphere. With planetary science and sometimes the Earth, what we see is now. We don't know when that atmosphere kicked in, which relates to when the runaway greenhouse effect kicked in on Venus.
44:51
That's what the ESA mission is trying to figure out. But I will tell you, there are models that say right after formation. And there are models that say around that 500-million year time.
45:03
ROBERT STERN: Cool.
45:07
So this is Phillips and Hansen's model for Venusian tectonics. And it's stagnant lid. But it's an active stagnant lid. The first order separation I would make in stagnant lid tectonics is you've got dead, stagnant lid, like the moon and Mercury I think. I don't really much about Mercury, just what I read in the papers. But Venus has got this very active stagnant lid. So right there, you've got a separation and stagnant lid tectonics. And the idea is that you've got mantle plumes,
45:41
they're heated by the core the way we know and love. And then you also have delamination, that the base of the lithosphere sinks. And maybe that's what's causing the-- if these 300 to 500-million year resurfacing events happen, maybe these things are all linked somehow, that the plumes come up, the lithosphere collapses, and you get a big resurfacing event. I don't know. So let's move on. So let's talk about why do we care. Obviously, it's interesting, these discussions
46:22
about what moves the plates and what other ways of cooling the planet Earth could have experienced if it didn't have plate tectonics. But from the point of view of what we're trying to accomplish here in CIDER, understanding the continental crust, what difference does it make when plate tectonics began? Well, plate tectonics makes crust today. One of the magmatic expressions of plate tectonics is the formation of crust. And Earth is unusual because we have two kinds of crust.
46:53
We have oceanic crust, and we have continental crust. Continental crust is the topic of particular interest here. And where does plate tectonics make continental crust today? At convergent plate margins. Now, some people will argue that hotspots, oceanic plateaus, especially the initial outburst of when a mantle plume arrives at the surface, generates material that gets added to the continents. And they use the example of Wrangellia in the Pacific Northwest as a spectacular example of that.
47:32
But most people who have thought about how plate tectonics makes continental crust would agree that most-- what do we mean by that-- most of the continental crust today is being generated at convergent plate margins. Island arcs, the Sierran Batholith over here, the other side of Fresno, that's where it happens. And this region of crust formation is incredibly complex. So what we see is down here at depth, we have the subduction factory. And next week, I'll talk a little bit about subduction.
48:08
But basically, you have circulating asthenosphere and hydrous melting. And those melts, the mantle seems to like making this thing basalt that those hydrous basalts come up. And because plate tectonics convergent plate boundaries stay in the same place, the locus of igneous activity stays in the same places. In the mid-ocean ridges, the igneous activity-- you have an igneous event. But then it gets carried away. So you never get much of a vertical construction. You get that with plumes.
48:41
But they move away, too. Hawaii keeps moving away. In these arcs, the plate keeps going down, the material keeps getting added. And you thermally work this crust as you thicken it. So that's the main thing that takes-- that processes this deeper crustal material, is the fact that this region is heated up and heated up, magma heat magma heat. And it goes on for 50 or 100 million years, incredibly long periods of magmatic additions, the crustal filter acting and causing fractionation and anatexis.
49:25
And that's one of the key arguments about how we make the continental crust, is what's making this unusual rock type called granite. It's probably not found on any other planet. Granite seems to be the signature rock of Earth. How is it generated? Is it generated by fractionation or anatexis, melting of pre-existing rocks, crystal fractionation of melts from the mantle? It doesn't matter. All of this ends up being created. And then this somehow is the nursery of continental crust today.
50:07
AUDIENCE: Bob? ROBERT STERN: Yeah, yeah. AUDIENCE: Can I ask more of a general question? As we say, most of the continental crust today is being formed at subduction zones. And just as a crazy question, and you said water, what if there was not much water? Can you still generate arcs? ROBERT STERN: That's an interesting thought experiment and a topic of-- AUDIENCE: --tectonics [INAUDIBLE]. AUDIENCE: No, I'm talking about you have the background water in the mantle to which CPM.
50:35
But we're talking excess water becomes a wedge. So you can still have plate tectonics. You still have subduction. Could you get arcs? ROBERT STERN: And I suspect-- AUDIENCE: Most people are going to say no. But I don't know if that's-- ROBERT STERN: I'd rather not take a position. It's an interesting question that's worth further exploration. So the question really is, can you generate felsic rocks without water, right? So-- AUDIENCE: Well, relative to other water. ROBERT STERN: What's that?
51:07
AUDIENCE: Actually, we'd like to reverse-- this may be a bigger question. Do we maybe need water to have plate tectonics? ROBERT STERN: And that's a very good question. I think that will be a question that-- a comment that when it comes down to why is Venus-- why are Venus and Earth so different, it'll be, one's the water planet and one's the CO2 planet. Yes? AUDIENCE: I'll just point out again that there are felsic rocks that are generated in non-arc settings, which is nice,
51:34
a bit volumetrically minor. But they exist. There is little water there, I think, because there also have [INAUDIBLE]. AUDIENCE: 10% of oceanic crust is plagiogranite. AUDIENCE: Sorry? AUDIENCE: 10% of oceanic crust is plagiogranite. AUDIENCE: Plagiogranites are 10%? AUDIENCE: 10%, what is 10%? ROBERT STERN: Yeah. AUDIENCE: Well, that's the estimate that I got from-- I forgot where. [LAUGHTER] AUDIENCE: But that's pretty common. ROBERT STERN: Yeah, all right. AUDIENCE: [INAUDIBLE]
52:05
ROBERT STERN: Yeah, I would contest that, too. But-- AUDIENCE: Of course, granites aren't actually that common in arcs. ROBERT STERN: Well, true granites are not. And now we get-- now we start to go into the petrologic zoo. Granitic rocks are common. True granites are not, absolutely. This is an interesting point of discussion. Here is the sort of thing that would make a continental crust today. And these are natural laboratories, tremendous opportunities for petrologic research
52:38
to understand this thing. It's a cartoon. So let's talk about lost continents, because we mostly focus on the fact that continental crust is created at convergent margins. And we less commonly explore the idea that continental crust is destroyed at convergent plate margins. And it's not quite as exciting as this particular movie. We may have to wait a while longer for the real realistic lost continent movie to come out. So I just want to give you three plot devices that we
53:20
could use in this movie. One of them is the truncation of North America about 250 million years ago. The geology of this part of the world is dominated by a craton in west of Idaho and coming down actually to Los Angeles. And then a passive continental margin shelf that's been tectonically worked, dominating Idaho and Nevada. And a little bit of an island arc up here in Northern California-- all of these trends have been cut off. And we know when it was cut off because the first inklings
53:56
of the Sierran Batholith starts in the Triassic. And there was a new trend developed at the beginning of the Mesozoic along here. But what no one has ever figured out is where the hell-- well, A, how much of stuff was lost from North America, and B, where the hell it went. One very likely possibility is it went down a subduction zone, that it was sheared off during the opening of Tethys, which in this part of the world, we most firmly relate to the opening of the Gulf of Mexico.
54:31
But that opening event was accompanied by a shear margin along Western North America. Jason Saleeby has demonstrated this convincingly, that there was a shear margin here in Permo-Triassic times. That sheared this stuff away. And it went somewhere down the tubes. OK. Maybe you don't believe it? OK. Plot device number two, the subduction of Arabia since 20 million years ago. So geologists have argued that Arabia contacted Eurasia, the beginning of collision, perhaps, about 20 million years ago.
55:06
Plate motions reconstructed as best as geologists and geodesists can do show that there's been no change in plate motion of Arabia with respect to Eurasia over those 20 million years. So Arabia continues to be subducted beneath Eurasia. And we had a very interesting discussion at the end of yesterday about this. And some of you may find this plot device less convincing than other. But OK, it's something to think about. Continental crust, we have this sense that continental crust
55:39
is unsubductable. But that's demonstrably false from the third argument, which are ultra-high pressure metamorphic rocks. One of the most exciting things that happened in the last 20 years was the recognition that continental materials have been carried down to a hundred kilometers or more. And the diagnostic thing is the diamond, sphene, and zircon, so the Kokchetav Massif in Kazakhstan is the most spectacular example of that. The diamond stability field goes like that.
56:16
So it's got to come down to at least 2 or 3 gigapascals, so cut 100 kilometers at least. We don't how much more. The other argument is coesite and quartz. So coesite could be a little bit lower pressure, but very strong indications that continental materials are carried down and then come back up. Why? And once you're down to a hundred kilometers, we shouldn't see you again.
56:51
There's a very interesting poster out there about how these materials come up. I invite you to take a look at it. OK. We can argue about how much continental crust goes down. But there's proof that some of it goes down whether Arabia is going to be somehow turned into a big, ultra-high pressure metamorphic terrane.
57:17
[CHUCKLES] All right. So we tried to put together a budget. And this is a very useful exercise because it exposes our ignorance. The best exercises are the ones that make you say, hm, we need to study that. So there's been three efforts in the last few years. Dave Scholl and I did this in 2010. And we came-- so these are the things that add continental crust. And the units that are being shown here are Armstrong units. So this is a Bob Kay honoring of Dick Armstrong, who
57:55
argued against vociferous opposition 20 years ago, before he died, that continental crust is being recycled. And in honor of that, Bob Kay said, let's call the 1 cubic kilometer of either production or destruction the Armstrong unit. So 1 cubic kilometer, that's what these units are. So as best as we can work it out, this is what is added to the continental crust each year, 2/10. Now, these are two different estimates. One is by myself and Dave Scholl.
58:31
And the other is-- the one in italics is Peter Clift. So that gives you some idea of how much we agree and disagree. And he actually did it before we did. Didn't really know about that paper when we did it. But continental arcs, arc magmatism out in the ocean basins, there's a difference of 50% estimate. And then these are the losses, so things that are lost by sediment subduction, big difference of opinion. Subduction erosion, which is the overriding-- oh, yes.
59:07
AUDIENCE: Please provide us some uncertainty on these numbers. [LAUGHTER] ROBERT STERN: We're very uncertain. [LAUGHTER] AUDIENCE: This is outside of my field. And I would say that I have zero idea as to whether or not 0.2 as a plus or minus of 10-- or 100. [LAUGHTER] ROBERT STERN: Well-- AUDIENCE: I don't mean percent. ROBERT STERN: No, no, I-- [LAUGHTER] I take your question seriously. And the only answer I can give you is one that exposes our ignorance of these issues.
59:38
And how do you deal with the admission of ignorance, except with humor, so. [CHUCKLES] I-- I can't. I'd love to. We, as a community, should. Each of these represent stabs in the dark. We have more uncertainty about different ones of these than others. The delamination flux is a particularly perplexing one. Some of us will put a number there.
1:00:10
So I think Peter Clift was willing to put a number on the delamination flux. I'm so uncomfortable with estimating even an order of magnitude that I can't do that.
1:00:24
Yeah, I'm sorry, I can't give you an uncertainty.
1:00:30
This is an opportunity for the junior CIDERites to say, let's try and do that. Let's take one of these and try and nail the uncertainty and get serious about this game. This is a cartoon. This is a cartoon with numbers, a particular-- the worst cartoons are the cartoons with numbers. So you're absolutely right. But this is where we are. AUDIENCE: So could you tell me why Stern and then another Stern-- ROBERT STERN: Charlie, and no relation. AUDIENCE: --had come out with two significant and then
1:01:02
three significant figures? [LAUGHTER] ROBERT STERN: Uh, we're geologists? [LAUGHTER] We don't really pay attention to significant figures as much as we should. It's certainly a problem for my students, I know, maybe problem for yours, too, to a greater or lesser extent. And-- yeah, I'm not comfortable with much beyond one significant digit. Absolutely, you're right. AUDIENCE: I don't know if it's worth mentioning that some of these do have uncertainty. Subduction erosion from the Clift
1:01:33
study originally, that's coming from seismic reflection and things like that. ROBERT STERN: Good point. AUDIENCE: That one's got a good uncertainty to it if we want to put those papers up. So the subduction erosion, I think that's plus or minus in the 0.1. ROBERT STERN: Yeah? AUDIENCE: I maybe jumping ahead slightly. But what it seems like these studies are indicating is that we have a net constant balance of continent as we're removing and adding about the same amount.
1:01:59
So doesn't that get back to the question of why do we have these really super stable arcane regions? ROBERT STERN: Yes, it implicates a lot of things. You're absolutely right that once you start throwing these numbers up there, the thing that jumps out at you is-- well, we're not real sure on the total budget. The big question is, is there more continental crust being added or subtracted? That's a question that these differences between-- Clift is arguing that the total crustal losses approach
1:02:34
5 Armstrong units. Dave and I argued that it was greater than 3.2 because we didn't-- couldn't put a number on what's called lower crustal foundering, but more commonly used is the term delamination. I don't like that because it implies that there's a lamina, and the lamina is somehow peeling away like wet wallboard or something. And probably the real losses, if they occur, are Rayleigh-Taylor instability. So rather than link a particular description of how that happens,
1:03:13
I just call it foundering. And then we can discuss mechanisms, as well as fluxes later. But yeah, that's what I would be excited about, if the junior CIDERites said, hey, wait a minute. We've been-- maybe we should consider crustal losses as much as crustal gains. Cin-Ty? AUDIENCE: But along those lines, [INAUDIBLE] showed all these images of the crustal ages, they're very episodic. So this is clearly a snapshot today even, I don't know. But it's better to look at portions of these
1:03:51
because the absolute numbers of fluxes can be changing through time. And the proportion, you can do it by mass balance instead and composition. And that might be a way forward to refine these. ROBERT STERN: Yeah, no. And the supercontinent cycle makes you think about, OK, how would these things vary over supercontinent cycle? Once we did finally come up with some confidence in our numbers, then we could say, OK, the supercontinent cycle begins with rifting and then ends with collision.
1:04:23
And if we really do lose continental crust in subduction zones, then you'd expect the crustal losses to happen more at the end of a supercontinent cycle than the beginning. So there is a time element that has to be considered that isn't considered in this cartoon. This one just is trying to capture what the main losses and gains are and how we agree or disagree about what they are. And I think you can see that this is a very fertile ground for people that want to play this game.
1:04:59
And I think it's a-- really, that's one of the first questions we have to ask about, continental crust, is it growing or shrinking? Yes? AUDIENCE: I sort have on general question. I was wondering-- as you go over the slides, I was wondering if you could just walk us through some of the latest estimates that you made. ROBERT STERN: Can we do that offline and maybe as a group that-- because the first thing I have to do is reread our paper. [LAUGHTER] One of the things that happens as you get older
1:05:31
is your memory goes, yeah, yeah. Back there? AUDIENCE: Yeah, just wondering if these additions [INAUDIBLE] were of what you might consider juvenile additions or whether or not it was taking into account [INAUDIBLE] recycling or the [INAUDIBLE]? ROBERT STERN: No, that we were trying to identify juvenile, that if you're recycling existing crust-- and I think people have made the argument-- that the best place to make continental crust is at an Andean type margin.
1:06:01
The only part of the Andes that we would like to know about would be the addition from the mantle, not the remelting of pre-existing crustal rocks. That doesn't get into this budget. In the University, the money goes from the academic departments to the physical plant. It's not the same as when you have to pay for a service outside kind of recycling. Rick? AUDIENCE: One of the observation that's not involved in this calculation, I think, should be is the effect of recycled continent
1:06:34
on mantle composition. So if you think about the level of [INAUDIBLE] in the crust, a tenth of a percent of that mixed into the mantle will take the barium-niobium ratio, the 87/86 strontium of that mixture of the mantle outside the range you see in interplate oceanic basalts. So what that means is if you're recycling these quantities of crust, you have to do it in a way that by the time it goes through the mantle circulation system, comes back to the surface, it's been diluted by thousands to 1
1:07:05
by mixing with large volumes of [INAUDIBLE]. I think that's a problem. ROBERT STERN: And my addendum to that would be, over what period of time? We're talking about the plate tectonic effect-- the effects of plate tectonics on continental crustal growth and destruction. And you're right, we have a signal, an older signal, coming from the mantle. But if I say that plate tectonics has been doing this for a short period of time, your objection would lessen. Yes?
1:07:38
AUDIENCE: Sir, the comment on that, what Rich had just said, is that I think you're talking about something like upper continental crust. And we're talking about mafic material, it's not quite as big a problem as it would be for sediments. AUDIENCE: Yeah, losses would be an example of the numbers I was talking about, so an average subducted sediment. ROBERT STERN: But the stuff that you see in ultra-high pressure metamorphic rocks, that getting mixed into a mantle reservoir
1:08:05
should have a significant effect. Most of the ultra-high pressure metamorphic rocks, they have to have quartz or they've got diamond, which means they've been in a carbon-rich environment. So they probably were upper crustal materials. AUDIENCE: They came back up? ROBERT STERN: Yeah. AUDIENCE: We don't know what proportion? ROBERT STERN: Yes, that's right. Yeah? AUDIENCE: I think one of the great uncertainties is how well the mantle is mixed. We're going to recycle this stuff down there.
1:08:30
Rick is right, you mix in the whole mantle, you had a problem with it. It stays in discrete areas and plates here and there that accumulate in a double prime. And it's not well mixed back in the [INAUDIBLE]. That eliminates at least some of that problem. ROBERT STERN: Yeah. AUDIENCE: So just a quick comment, maybe another number actually would be relevant here. And this isn't a production rate of oceanic crust is about 3 AU. First time I see U [INAUDIBLE].
1:08:57
So that's a bad production rate for oceanic crust, 3 AU. ROBERT STERN: No, I think the oceanic crust is on the order of 20. AUDIENCE: No, 3. I wasn't so sure, I just checked. ROBERT STERN: All right. AUDIENCE: [CHUCKLES] ROBERT STERN: OK. AUDIENCE: Sir Bob, we have 15 minutes. ROBERT STERN: Oh, OK. Well-- [LAUGHTER] I was hoping this would happen, so yes. AUDIENCE: [INAUDIBLE] How do you obtain these numbers? For example, how do you estimate the lower crust foundering? [LAUGHTER]
1:09:27
ROBERT STERN: Well, I was proud-- I was proud because the one thing a scientist wants to do is put numbers up there. And Dave and I said, we can't, with a straight face, defend them. And what you've just done is those are our question marks. You basically validated my lack of courage, which was like-- I can't defend that. So I'm not going to put a number. AUDIENCE: The subduction zone, you have some sediment that go in subduction and other sediment that go to form the region.
1:09:58
So how can you estimate this to-- ROBERT STERN: Well, there's a literature. And I'm not-- so I have 15 minutes. But you haven't asked a question, so please. AUDIENCE: Is it the same lot? Is there a hope to estimate lower cross boundary then? ROBERT STERN: I hope that some of the junior CIDERites will show a way to move in this direction. I think this would be a great contribution if we thought about how we could spin off a project off of this cartoon, that would be one, that would be another.
1:10:29
So 10 minutes now. [LAUGHTER] But that's fine. That's what I was hoping would happen. [LAUGHTER] So have you guys seen these kinds of diagrams before? AUDIENCE: Oh, yes. ROBERT STERN: OK. AUDIENCE: There were more lines in the last one. [LAUGHTER] ROBERT STERN: So I want to throw-- I'm glad I got to this point because what I want to do is show you that. This is the range of crustal growth curves that has been shown with great diligence by-- so that's Earth history 4 billion years ago,
1:11:01
the age of the oldest crust. Another one to do it would be to consider the-- oops, now I'm coming apart-- that the crust once covered the whole planet. And what we're left with is about 7.2 times 10 to the 9 cubic kilometers cubed, that's the continental volume, continental crust volume. And that we've been losing it with the beginning of plate tectonics at something on the order of 2.6 Armstrong units per year. That's reasonable given the uncertainties in what plate tectonics does.
1:11:40
Remember that that crustal growth curve is just the rocks that are still here. What if the crust is disappearing? Those curves mean nothing about crustal growth. They only mean what is preserved. AUDIENCE: At first order, wouldn't you think that the loss was somehow proportional to how much was there? ROBERT STERN: No, I think it's proportional to the-- AUDIENCE: --very beginning, instead of TT bit. ROBERT STERN: That's-- AUDIENCE: You lose the 2.6.
1:12:10
ROBERT STERN: All planets, silicate planets, have crusts. And so what we see is the Earth is the only one with a bimodal crust, continental and oceanic. So one perspective and the dominant perspective is that that crust has grown with time in some way or another. But another equally valid position-- and I'm glad to get to this point because this is a challenge I'm happy to take-- is that the planet was originally covered in continental crust and has been losing it.
1:12:49
And tell me why that is the most ridiculous thing you've ever heard. Isotopically, I know what you're going to say. But-- AUDIENCE: There's another issue with it. So right now, the continents have half the uranium in the Earth. And if they're three times their present size, they have more uranium than the Earth has. So they can't-- AUDIENCE: Can't do the same composition. AUDIENCE: Yeah, they can't be the same composition they are now and be three times their volume, their current volume.
1:13:18
So basically, you'd have to push them back into being a basaltic crust. In which case, you're talking about losing-- Hadean basaltic crust is fine. We'd lose Cenozoic basaltic crust. ROBERT STERN: But what if that uranium has been processed through multiple episodes? It keeps going down, but it keeps coming up. AUDIENCE: But you're talking about removing it from their surface. ROBERT STERN: Yeah, but things like potassium and rubidium, cesium, uranium, they're going to find their way back
1:13:51
to the surface preferentially. AUDIENCE: Yeah, they're called continental crust. ROBERT STERN: OK.
1:14:03
Anyway, it's just a little-- maybe that'll open up a little imagination space that in trying to assess the significance of these curves that the realm of possibilities might be a little bit larger. So that's for the junior CIDERites as much as anybody else. AUDIENCE: And what would be the mechanism for giving the primitive Earth basalt continental crust? ROBERT STERN: No, but none of that crust is here. So as a geologist, what I really want is a sample. And with the sample, I can get data
1:14:37
and I can make an interpretation. We don't have any samples. We've got these wonderful zircons that have inclusions that have been perhaps overinterpreted. But they are the piece of that early history that we have to work with. The rocks themselves-- AUDIENCE: [INAUDIBLE] ROBERT STERN: Yeah, yeah. Thank you. [LAUGHTER] AUDIENCE: Well, since you brought up other planets, you said they all have continental crust-- ROBERT STERN: No, no, they all have a early-- they all have an early--
1:15:06
AUDIENCE: So we do have samples of the crust? ROBERT STERN: Yes. AUDIENCE: And it's not continental, as best we can tell on Venus. And so Rick's argument, I would say, is right. If the early stuff you're talking about is something basalt-like that got reworked and built up, then I will agree with you. I can't argue against this. I thought I could. I tried, I can't. ROBERT STERN: [LAUGHS] AUDIENCE: However, if you are saying it's primitive, then I think you have some problems.
1:15:32
ROBERT STERN: Well, I've got to bring this to somehow a halt. I'm about halfway through the talk. It's because you guys had so many questions, which is-- in this kind of a setting, what I hoped was that I wouldn't make it anywhere close to the end. I've succeeded far beyond my-- [LAUGHTER] --wildest dreams. But the point that I'm not going to get to is discussing when plate tectonics began. But a key piece of that understanding is the understanding that continental crust
1:16:04
is as much destroyed by plate tectonics as created. And I did get to that point. And with that, I'm pretty happy. Richard, you'd be best-- I can stop there. I can go on a few more slides. There's questions. What would you-- AUDIENCE: Why don't you give us the two-minute synopsis on when you think plate tectonics started or something like that? AUDIENCE: That was my question-- ROBERT STERN: Two minutes? [LAUGHTER] AUDIENCE: Two minutes, given uncertainties, as Bill has just outlined them.
1:16:35
[LAUGHTER] ROBERT STERN: OK. Is that-- AUDIENCE: Sure, yeah. ROBERT STERN: It's not going to be two minutes. AUDIENCE: Go ahead and have fun. [LAUGHTER] AUDIENCE: Thank you. ROBERT STERN: So OK, I'll just zoom-- I'll just zoom through the rest the slides.
1:16:55
Well, I like the cartoons. [LAUGHTER] So I hate to zoom through those. And the Urey ratio, which is a great thing to talk about because Harold Urey got his PhD here, what a wonderful story. And stagnant lid and evidence for a shutdown, delamination foundering-- OK. [LAUGHTER] We can theorize till the cows come home, what do the rocks say? So this is where everybody in this room has an opportunity. Open up your minds to the possibility that plate tectonics hasn't always been going on this planet.
1:17:31
That's almost assuredly true. And if you read Korenaga's paper, he has a very nice stratigraphic column of what the people that have taken a position on it in print say. And I am way up there towards the top. I think it's Neoproterozoic and most people put it in the Archean. But let me explain to you why, because I think we can't just rely on the models. We have to say to the modelers, here are what we see the constraints are. So we could talk about the Petrotectonic indicators that
1:18:07
are diagnostic of plate tectonics and go in and interrogate the planet and say, when did these rocks get produced? And there are several of them. So there's general plate tectonic indicator, which is ophiolites. Everybody knows what an ophiolite is. Anybody not? Happy to explain. It's one of my favorite rocks, should be the state assemblage of California. [LAUGHTER] Subduction indicators, jadeite is a metamorphic mineral that's associated with subduction zones.
1:18:41
Blueschists, glaucophane schists, another candidate for the state mineral of-- state rock of California. Lawsonite eclogites, a sort of subvariety of eclogites. Those are shown there as a function of time, so. AUDIENCE: Would it be possible to normalize that by the amount of exposure of the different ages? ROBERT STERN: Sure, I think that could be done. AUDIENCE: What would it look like? ROBERT STERN: Yeah, different? [LAUGHTER]
1:19:21
There's lots of old rocks running around, Australia, South Africa-- AUDIENCE: Yeah, but not nearly as much as there are new rocks. ROBERT STERN: Yes, but still if there was a proportionate distribution of blueschist through the geologic time-- and Kent will correct me-- there would be a lot of bars sticking up here. Is that a fair statement? AUDIENCE: Fair. ROBERT STERN: OK. And then collision indicators. These are two different types of plate convergence.
1:19:54
But because subduction is fundamentally different, healthy plate convergence and continental collision is pathological plate convergence, you have different indicators. One of the most spectacular, of course, are the ultra-high pressure metamorphic rocks, which don't appear until about 650 million years ago. Ruby, in particular, is a metamorphic gem. This is from a paper we just published in Geology, talking about plate tectonic gemstones ruby and jade. And you see that these--
1:20:28
what we find as a very powerful petrotectonic indicators of plate tectonics-- ophiolites, jade, blueschists, lawsonite eclogites, ruby, UHP rocks-- continental margin, I'm not going to talk about because this is much more interpretation-driven-- they all start in the Neoproterozoic. So I think I can leave with the comment to the junior CIDERites, in particular, find your-- the thing that is the most convincing to you of plate tectonics. It could be isotopes.
1:21:05
Some people find that the evidence that there was a circulation from the surface to the interior of the Earth from the earliest time of Earth history convinces them that plate tectonics began at that time. I don't find that. I think there's other ways to take surface materials to depth. But you need to decide for yourself. What do you find interesting-- convincing, not interesting-- convincing about the indicators of plate tectonics? And with that information, construct,
1:21:40
defend, argue about a history of when plate tectonics began. To me, this is one of the most exciting opportunities in the solid Earth sciences, is to start to go after the history of our planet. And it's really exciting because in this fantastic search for exoplanets, people are looking for other people out there. And one of the things that might-- is clearly critical is life. You need to have life on a planet. Somehow the one planet in our solar system with life has got plate tectonics.
1:22:16
Is there some kind of a relationship between how we got where we are today and when our planet started doing this unusual way of losing heat? And I think probably I ought to stop there, huh? [APPLAUSE]
1:22:39
SPEAKER: So on the off-chance anybody didn't get to ask their question, you can answer a few more questions right now. so that's it. AUDIENCE: I'll just mention there's a couple of possible datums that are not on your plot there. One is that there's been a proposed ophiolite at 2.5 billion years in old China craton. It's highly controversial. But Tim Kusky and his coworkers have proposed that. The other is that-- you may not be aware of this work,
1:23:04
but Pierre Cartigny studied some diamonds from a kimberlite that I think is 2 billion years old and said that these were UHP diamonds from continental crust. So that would put those back to 2 billion if it's correct interpretation. AUDIENCE: Mike Brown doesn't believe that at all. I've talked to him about that. AUDIENCE: Yeah, I know. I'm the one who told him about it. And he doesn't believe it because he wants plate tectonics not to be operating in the Archean.
1:23:30
ROBERT STERN: I have to comment on the Archean ophiolite because there have been two papers in Science, one of them, the North China craton, the other in Greenland, the Isua Belt. Both of them have been challenged and I think convincingly so, not just because I'm happy to see them disappear, but I think the scientific community that's looked into this has said, nah. AUDIENCE: Bob, another way of looking at this is if you just look at erosion rates, they're about 5 of your Armstrong units a year.
1:24:04
And probably 75% or 80% of that is recycled sediment and doesn't really come into the thing. But if that's the case, then you've got maybe roughly a cubic kilometer, approximately 1 Armstrong unit of erosion of real crust. And then that implies a very long residence time for the crust that remains to be processed through that. So if you want to erode crust and then subduct some of that and then some even smaller fraction, that's a residence time longer than the age of the Earth, right?
1:24:39
I'm not sure what that gets you, because it would imply that much of that crust must be stable for longer-- have a expected lifetime longer than the age of the Earth. And that's not necessarily really what we observe in the geochronology. We don't have lots of 4-billion-year-old terranes out there. We have a few. But I'm just-- that's an argument there that I think is telling you something about the dynamics of that recycling.
1:25:12
AUDIENCE: Anybody else? ROBERT STERN: Greg? AUDIENCE: So one other thing that is a fun puzzle is to ask in the models that you talked about of the formation of crust and the evolution of the Earth, the Urey number, all of this, what kind of rocks would you expect if the Earth was a little hotter? Would you make blueschists or would the blueschist be something else? Would they be green schists or-- ROBERT STERN: That's a really good question. So one of the arguments--
1:25:42
the blueschist observation has been around since before plate tectonics. People talked about the fact that there was no blueschist in the Precambrian. Well, they found some. But it only goes back to about 800 million years. So one of the arguments is, well, the Earth was hotter. So the thermal pressure conditions weren't conducive to forming blueschists. Now, this brings up the question of, how do blueschists form? Blueschists form in subduction zones. Subduction zones represent regions
1:26:16
where a huge tract of lithosphere are injected, cold material is injected into hot mantle. And the geodynamicists have shown us convincingly that the thermal structure of a subduction zone, to the most part, is controlled by the thermal parameter, the convergence rate times the age of the oceanic lithosphere being subducted. At the slab surface, the slab surface is controlled by the temperature of the ambient mantle. Most of the estimates for Earth's ambient mantle
1:26:52
are maybe a couple hundred degrees hotter. People have done this using Cin-Ty's geothermometer for basalts back into the Archean. And they're not that much-- komatiites do have-- you have to do have to have a hotter Earth. But it's not a thousand degrees hotter. So you have a modestly or moderately hotter mantle. But the main thermal control on subduction zone is that material that's being injected. And in fact, if you think about the thickness of the oceanic crust, blueschists
1:27:31
are mostly subducted oceanic crust. You've got a thicker pile of oceanic crust to turn into blueschists. So it actually should be more favorable for the formation of blueschists in the past than at present. That's taking an extreme argument. But the simple statement that the Earth was hotter, therefore there were no blueschists, is misleading because blueschists don't reflect the temperature of Earth's mantle. They represent the temperature in the oceanic crust,
1:28:09
a kilometer or two kilometers below the slab interface in the subducted oceanic crust. So that would be my response to people's argument that blueschists-- you don't see blueschists because Earth was hotter. Yes? AUDIENCE: How were they returned to the surface? ROBERT STERN: Yes, good question. [CHUCKLES]
1:28:31
I think Cin-Ty's and his students poster out here shows that very nicely, that there's two arguments. One is that it comes up back up the subduction channel, which is something Mark Cloos worked out for the Franciscan here, that somehow there's this elevator that goes, mph, like that. And the Franciscan is the most spectacular example of that subduction channel thing. And the other is that these things develop as low-density diapirs above the slab interface.
1:28:59
And then they organize and rise diapirically. Is that a fair statement, Cin-ty, that it's those-- it's got to be somewhere between those two possibilities? AUDIENCE: Ask Monica. ROBERT STERN: Monica? Where's Monica? AUDIENCE: Yeah, those are the two arguments. ROBERT STERN: Would you add anything to my simple-minded-- AUDIENCE: No? ROBERT STERN: OK. Diapirs and subduction channel. AUDIENCE: But that might be another interpretation that somehow the aspect of those return mechanisms
1:29:29
could be somehow a little bit different-- AUDIENCE: I suggest-- ROBERT STERN: Certainly. AUDIENCE: --tomorrow in my talk, if you could wait to do that. ROBERT STERN: Yes, and advertising, Kent Condie will give you a very different interpretation tomorrow so. SPEAKER: OK. I think we should stop there so we can have some coffee. Plenty of time to continue the conversation after 30 minutes. Thanks again to all you. ROBERT STERN: Thank you. [APPLAUSE]
1:29:56
SPEAKER: All right. Great. So I think without further ado, let's hand it over. ROBERT STERN: Thank you.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat