Gregory Henkes

Stable isotope paleothermometry of human origins in the Turkana Basin

The stable isotope geochemistry of terrestrial carbonates is among the most important recorders of paleoclimate and paleoenvironments in East Africa. In this talk, Greg Henkes will describe our efforts to apply clumped isotopes - a novel isotope thermometer - to reconstruct ancient climates in Turkana, and he will discuss future advances and research directions.

FULL TRANSCRIPT 

Get started and going to note some of my institutional affiliations before I really dig in. So, I'm in the Department of Geosciences here at Stony Brook University and sort of formally as part of my hiring here in 2016, was hired with the understanding that I, okay, this thing's a little weird, that I would work very closely with the Turkana Basin Institute, and I'll come back to that a few times during my talk. I also thought it would be a good chance to mention that colleagues in the School of Marine and Atmospheric Sciences (SOMAS) here sort of graciously in the first couple of years I arrived on campus allowed me to sort of gain affiliation. As Thure mentioned, stable isotopes are everywhere and including in the ocean and so lots of applications and wonderful student collaborative projects have come out of SOMAS. And Paul Shepson, the dean of SOMAS was here, and I see a few SOMAS faculty in the audience and students, and I'll come back to some of the work that we're really starting in collaboration with SOMAS scientists to study the actual lake, right? Everyone's been focused on things that aren't the lake, but almost everyone has shown a picture that includes the lake surface, right? So big lake in the middle of the basin, we need to study it. 

Before I dig in, I want to make a little bit of a disclaimer, which is on what's today, the fourth day, midway through the fourth day of the conference, it strikes me that it becomes really hard to say something original, especially when your talk follows one of the handful of stable isotope geochemists in the National Academy. So thank you Thure, for giving an introduction to the talk. But I will say that my one advantage is with maybe the exception of President McInnis, maybe Pat and now you all who've given a talk up here. I've actually been up here once before. It was during Covid, so there was no live audience. I actually gave a talk to a bunch of Zoom screens in the audience, but nevertheless, I was up here. So I'm used to the bright lights and as part of that I received some training in giving a public talk. 

I'm not going to claim that I'm an expert in it, but nevertheless, I was trained by a wonderful group of colleagues and people here at Stony Brook University through the Allan Aldo Center for Communicating Science. So I have some tricks. So they told me that I need to move around, so I'm going to try and do that. Most people have come up here and just stood over there. So actually I gave my talk mostly last time from over here. They also told me to use devices that will engage you all, keep you all engaged. So thankfully everyone just filled up on coffee, but if you are tired in day four, you would be excused, right? So what I'm going to do, or what I've decided to do is say dope like Homer Simpson every time or just before I say something that has already been said before in the last few days. So dope. 

I think it was Johannes on day one. I don't know where Johannes is  who pointed to the sort of immortal words of Richard that live on the Turkana Basin Institute website. And my referencing them or sort of snippets from the full text comes with the story and the story goes that the search under which I was hired was started relatively late in the cycle of academic faculty hires. So, I was a sort of ambitious postdoc and had a bunch of applications in and I received an email from a friend and now very close colleague, Troy Raspberry saying, “Hey Greg, you need to apply for this job, and I thought, cool. And like many of you, I mostly knew of Richard through stories that I had heard from my PhD advisor, Ben Passey, and folks like Naomi Levin and occasionally Thure, but I had never met Richard and so I only really knew Richard from the pages of National Geographic. 

So, the first thing I did after I can literally picture me sitting in my office as a postdoc receiving the email from Troy, clicking on the Turkana Basin Institute website and navigating to the page that says, I think it says about TBI or a message from Richard Leakey, I should click that first. And I pulled it up and was, I mean I didn't really need much convincing to apply to the job, but I read these words, a science that is on the move each year, more discoveries are made as a consequence of our understanding that changes and that TBI, the promise was that there was an opportunity to facilitate new explorers discovery of treasure buried in the Turkana Basin and that was Richard Leakey's dream. So I found it just the cherry on top in terms of motivation to apply and put every effort into the interview process. 

So thank you Troy. Thank you Richard. And thank you to the whole sort of TBI family who's welcomed me in. So, in my mind, Richard's legacy was embodied by another sort of big figure in the TBI ecosystem. Isaiah Nengo, who like Richard and Lawrence and others at TBI sort of welcomed me. The first time I met Isaiah he was sort of this unsure figure hanging out outside of my lab door, which many of you saw the lab on Monday morning and it sort of came in and he busted in the lab and talked all about the Miocene and how he wanted to reinvigorate research. And so, it was obviously a really tragic loss and sort of a double whammy to have Richard Pass and Isaiah soon thereafter. And so, a lot of the work that Kevin described yesterday, and you'll hear more about this afternoon and tomorrow is the brainchild of the Turkana Miocene Project in the memory of some of our work recently is done in the name of Isaiah. 

So I guess dope. So, Kevin showed this slide yesterday, which really just articulates this sort of ambition of the Turkana Miocene Project, some of the leadership or I guess all of the leadership team and articulates or makes the point that there are 50 members from 15 to 20 institutions. Some of that is changing actively. We're adding and losing new institutions. It's a really fun, highly collaborative effort. I don't know if Kevin noted yesterday that we're sort of present online. So you can find us on Twitter and Instagram and we'll try and update things as we embark on this summer's field season. Here's that big group. As Kevin said, you can sort of find your friends, find your colleagues in this list, but I wanted to show it and it maybe is a little hard to tell, but there are names that are sort of being highlighted here in bold and italics who are going to be featured in a lot of the things that I'll talk about today. 

So just to give you a sense of temporal scale, and some have talked about various parts of the Miocene, but sort of the circled sites, this is borrowed from Craig Feibel outline these sites that are outlined sort of show the sequence of our project and the various ages of sites in the Miocene in the Turkana Basin that we've been visiting. And with this Venn diagram, I was trying to articulate this sort of tools. So, I'm talking about stable isotope geochemist as there said, we like tools, I forget the analogy that Thure used, but we like cars, Ferrari that are supposed to make the right turn on Thure's sort of left-hand, right hand turned road picture before. And so, these are some of the tools that our project is bringing to bear on the Miocene climate earth and life record. And this sort of idea with this sort of highly overlapping Venn diagram is that many of these tools are answering questions that really span the disciplinary boundaries of the project. And of course leave it to the isotope geochemist to pull down something and show it to a bunch of anthropologists from Encyclopedia Britannica, 

Maybe ask some people. But this is the motivation I don't need. I've given this a form of this talk before to my geologist friends and they don't know that I'm sort of pulling one over on them. But really this is the motivation and I'm preaching to the choir here. And then there are things like Isaiah's discovery of alesi at Napudet sort of Mid-Miocene site. And if you read that paper carefully, many of you were on that paper, you sort of pick up words and sort of ideas that really call on the need to bolster and reconstruct paleoenvironment during this time period. So we can also look to the existing data as we've seen many examples in the last several days of climate, both at the global scale. So, what I'm showing on the left here is pCO2, the sort of paleo CO2 reconstructions side by side with a very local as Kay was sort of alluding to in her talk, this sort of local expression of some global climate phenomenon that is unfolding over several million years. 

And the thing I like to point out with this, this is from a paper by Naomi Levin, a review paper that I think puts a nice stamp in time in terms of what the record looked like from 2014 to 2015. And what I point out when I show this slide is that the record here has shown stops right before all of the interesting stuff I showed in the previous slides and some of which was talked about earlier in the conference. Okay? And Thure did, you heard it from the guy, right? So, I don't need to explain the virtues of oxygen isotopes, but Thure did a very nice job of explaining some of the problems. So we have this sort of big global record of paleo climate that comes from deep sea cores and little shells that you can extract from those cores. And they tell you something about how the ocean is a big huge heat sink with a lot of thermal inertia. 

And so it takes a lot to change the temperature of the bottom of the ocean and to sort of superimposed squiggles on this line here. And so we can do this sort of scale hopping over to the record for where all of the action is happening over the last 10 million years in East Africa are area of interest. And so this is shown in soil carbonate oxygen isotope records that have been generated prior to 2015. And again, that scale stops at 10 million years. So when I first started working on this project, it was like, okay, well it seems like the need is pretty clear and all we need to do is sort of extend these records as if it were that simple. But Thure also really nicely laid out, and now I don't have to talk about it, that the oxygen isotope record can be interpreted two ways. So it can be interpreted as a paleo temperature record if you were to assume the oxygen isotope ratio of the fluid from which these carbonates were forming. Or it can be a record of the oxygen isotopes of the fluid if you assume a constant temperature. So to reset it, right, all we need is a paleo thermometer that is derived from these same materials but is independent of a temperature signal independent of the oxygen isotopes. 

I was going to make a joke about you all now becoming students in my stable isotope geochemistry class. But so what you're seeing here, I'll spare you a joke. What you'll see here is a carbonate lattice that I've sort of populated with all of the elements that are involved in a calcium carbonate. These calcium carbonates are making up those soil minerals. So the greens here in this lattice are calcium groups. The oxygens found in groups of three are the oxygens in the carbonate groups and then the carbons are at the center of the carbonate group. And if you're a chemist, you'd look at this and say, oh, cool, maybe you're interested in the structure or the immunology of it. But us isotope geochemists see the world slightly differently. So now what I've done is taken those carbon and oxygen groups and sort of color coded them where they are substituted by a heavy rare isotope, either oxygen 18 or carbon 13. 

So you can see that in some molecules we have a carbon 13 bound to just a regular old oxygen, most of which is oxygen 16 or an oxygen 18 bound to a regular old carbon 12. Most of the science in the last 60, 70 years has focused on just understanding, just sort of counting up all of these isotopes in a material and making the carbon 13 to 12 carbon ratio. But the advent that I'll be talking about today and what we've been spending most of the last five years on is the cases where you find a carbon 13 and an oxygen 18 bound together. And as Thure said, this is a signal that is entirely dependent on thermodynamics. It doesn't care about how many thirteens or twelves or sixteens or eighteens are in the mineral. So to give you a sense of the challenge that this presents, so you've got two rare things together in the same group. 

So to give you a sense for how challenging that is, Thure mentioned patience to measure carbon 12 to carbon or carbon 13 to carbon 12. That's measuring the amount of nitrogen in this room in the air. Most of the air that we're all breathing is nitrogen. Thankfully some of it is oxygen or else you'd all really be asleep. To measure oxygen 18 to oxygen 16 that would be trying to measure the approximately 1% of the air that we're all breathing. That is a noble gas argon. That's about 1% to measure these clumped isotopes. The CO2 in this room is probably with all of us in here, is probably something like a thousand parts per million. One 10th of that would be measuring one of these carbon clumped isotopes, carbonate clumped isotopes. So they're really challenging. And part of Richard's legacy through TBI was sort of giving the Department of geosciences this sort of excuse to buy into some of this very exciting science. 

And Craig Feibel yesterday showed a picture of Ian McDougall next to his mass spectrometer. And I was like, yes, okay, now I can show this sort of modern version of that. So this is the mass spectrometer that we've sort of built, constructed in our lab and a prep line that is specifically designed. You can't go to Fisher or VWR and just say, give me one of these things that can tell me what the clump isotopes are. We have to make that ourselves and spend lots of time purifying CO2 so that we can pick out that very, very rare combination. 

Okay? So I'm going to prove to you very quickly that it works. My time is vanishing. So the details of this line are unimportant, but what  you should see or take away is that there are data that fall on the line and there's not a lot of scatter around the line. And the relationship here is between temperature on the Y axis and the clump to isotope ratio that we measure, excuse me, temperature on the x axis and the clumped isotope ratio on the Y axis. And this is effectively a transfer function. So we can take this to various modern or contemporary soils and there are a number of us in this sort of very small subfield who've been working on this and ask the question, if we generate a clumped isotope temperature or oxygen isotope estimates of the water, do we get the right answer? And so the fact that lots of this data falls along a one-to-one line gives us some indication that we're safe in applying this to pal paleo applications. 

Thure already mentioned some of the work that he and his student Ben Passey and Naomi Levin did in 2010, which really put a mark in both the clump isotope community with Turkana being the first application of this sort of highly specialized method. And they sort of made the case that Turkana perhaps not surprisingly, has remained hot for at least the last 4 million years. I will also make an observation that not only is Turkana one of the hottest places, but in a series of papers that came out in the last year or so by Callam Munday and have really pointed to the fact that Turkana is unique in its aridity or dryness as well. So, this little red dot Chalbi can stand in for sort of northwestern Kenya here, but this is the latitude of the deserts on earth of the major deserts on earth. And you'll see the sort of lowest latitude desert in the tropics is there's really only one, and that's the sort of part where we all are drawn to. 

Okay, so we're extending the record back into the Miocene. I'm almost nearly out of time, so I'm going to skip through this relatively quickly. The first place we worked at was Lothagam. We are selecting pedogenic carbonates and developing a sort of chronology of where those are from in the sequence. Most of the samples were collected during the nineties effort by Craig Feibel. And here's the record. So, the blue data are the data from Ben and Thure's work published in 2010. And the red data are our effort to both sort of prove that we can replicate their results in the Plio-Pleistocene but then extend, effectively extend this record back into the early or the late Miocene. And the bottom line is that it's sort of the temperature story at least isn't all that interesting? Temperature appears that the Turkana basin appears to have been hot for at least the last 8 to 9 million years, or as hot as it is now. 

And therefore any changes that we see in the carbonate oxygen isotope record must be driven by changes in soil, water, oxygen, isotope ratios as opposed to temperature. And so here is this sort of back strip, this bottom figure here is the back strip fluid Delta 18 O values. We'll come back to that. I'm going to skip through this for sake of time and jump over to a sort of compilation where we have temperature on the top panel here, fluid oxygen isotope ratios in the middle, and then the soil carbonate, carbon isotopes, which come along for the ride. When we make these measurements on the bottom panel, I'll sort of assert and we can talk about this later today, that I think this sort of record makes a lot of sense with the work that was published by John Wynn in the Lothagam monograph on these sorts of distribution and types of soils at Lothagam. 

And if you sort of squint your eyes and fall along with me, we can sort of start to superimpose some of the global climatic events. So, there was a change in the carbon isotope ratio of these benthic foraminifera in the late Miocene associated with an increase in marine productivity during that time. The onset of that is sort of coincident with the earliest part of the record at LFI gum and the local maximum, at least in carbon and oxygen isotope ratios, coincides with a change point in that and a statistical analysis of that benthic Foraminifera record. 

The values decrease thereafter in both carbon and oxygen isotopes and reach a sort of local minima during the height of mid-Pliocene warming and sort of increase thereafter implying a sort of more increasingly arid and increasingly grassland dominated landscape. But here's a little dirty secret if you look at these samples in cross-section. So this is a cross-section through one of these carbonates. They're not just sort of beautiful, you could see all these, they've got inclusions in them of detrital minerals and then they're sort of crossed through with all of this secondary carbonate that presumably grew in sometime after the original nodule grew in the soil. So I'll draw a little line and it may be hard to see a little circle on there that's about the scale of our sampling. And so, one of the things that we're working on currently is a method to essentially take out the secondary overprint from these pedogenic carbonates. 

So we're working on statistical methods for segmenting these images. I said I'd walk around to segment these images of these petrographic images. And this is largely the work of Mae Saslaw my graduate student who's unfortunately not here today, but hopefully you got a chance to meet her, and Mae is working on a sort of new method, a new approach to try and make this sort of fidelity of these carbonate records even better. So instead she makes sort of Swiss cheese out of these instead of just sub sampling. Once she makes kind of Swiss cheese out of these carbonate nodules. And then we can map on where she sampled, we can map onto the petrographic image to come up with individual measurements of the relative proportion of primary versus secondary carbonate in these samples. And Thure talked a lot about mathematical methods. I paid attention in my linear algebra class and a simple linear inversion of the data. 

This is just an example from a pedogenic carbonate from Buluk shows that we can sort of successfully remove what we interpret as a primary signal. So, a paleo temperature of 37 degrees and a carbon isotope ratio of about -9/mil from this sort of diagenetic overprint, which shows a much warmer temperature and slightly heavier Delta 13 C and Delta 18 O values. I'm going to skip through, but I'm happy to talk about the sort of climate modeling comparisons that we're going to try and make as a part of this project. But I wanted to end by showcasing a couple of things that are kind of on the horizon. And Troy Raspberry this afternoon is going to talk about the sort of application of pairing these carbonate clumped isotope measurement with dating of things that aren't pedogenic carbonates, but that are nevertheless useful for understanding this sort of stratigraphic chronology in the Turkana basin. 

One of the really exciting projects that is just beginning with seed funding from Stony Brook University and the School of Marine and Atmospheric Science is led by a graduate student, Madison Mule, Josie Aler  , who I believe is here, and our colleague Mike Frisk, trying to extract isotope information from fish teeth so that we can develop a record of paleo temperature and oxygen isotope ratios from Lake Turkana itself. So here are some of our preliminary results. They're actually quite exciting because Mae my graduate student has also been working on a massive compilation of water, oxygen isotope ratios from the Turkana Basin, which show a big difference between. So temperature was about the same 1.6 million years ago in Lake Turkana, but the oxygen isotope ratio of that lake water was five per mil, lower than it, excuse me. Yeah, five per mil lower than it is today. 

What determines the lake water oxygen isotope ratio, and it will be of no surprise to you all that it's evaporation. So evaporative processes, fractionate oxygen and hydrogen isotopes. And so, what you're seeing here is a compilation from Mae of lake water oxygen isotope ratios showing the enrichment that follows from this sort of closed basin evaporation. And then finally, as Thure alluded to, there's sort of the third oxygen isotope, which is even more elusive than the oxygen 18, oxygen 17 predictably fractionates also with evaporation. And so, one of the exciting things, this is work I wasn't involved in, but led by a colleague Julia Kelson, showing the triple oxygen isotope ratio of pedogenic carbonates from all over the world mapped onto their aridity index. And I've been able to generate some in Ben Passey and Naomi Levin's laboratory at the University of Michigan. We've generated some preliminary data that shows East African modern East African soils plot along where you might expect them to. 

These are data from Mpala collected last summer and interesting some of the soils from the Miocene sediments at Lothagam plot much lower, which is implying this sort of more arid landscape. With that, I'll wrap up flash. This picture dope, Kevin showed this photo previously. I'd like to thank NSF and the Turkana Basin Institute for providing funding for a lot of this work. And I'll make a plug. So, one of the exciting things I think that should come from this conference is more opportunities for all of us to gather and so Mae Saslaw is leading along with Emily Beverly and Bill Lukins. Those names should be familiar to many of you. A session at GSA this coming fall. I was told by Mae to note that Kay has been invited as the speaker, but I don't believe Kay has accepted. And there 

Is no pressure. There is no pressure. It was very explicit. 

There's no pressure. But if you feel like you have something to contribute, I think we would all welcome to hear it at GSA in the fall. Thank you.

View all Human Evolution videos