Kevin Uno

Miocene ecosystems and mammal diets in the Turkana Basin

Much of Dr. Richard Leakey’s research career focused on the fossil record of the Plio-Pleistocene (~4-1 Ma) deposits of the Omo Group. Here, I present new data on ecosystems and mammal diets from the Turkana Miocene Project that focus on the less well-studied deposits spanning the Oligocene to Miocene epochs.

FULL TRANSCRIPT 

Good morning everyone. Can you hear me? Okay, great. I'd like to just first thank the organizers, Lawrence, Fred, Alicia, the whole team for giving me the opportunity to speak today on some of the work that the Turkana Miocene Project has been doing in the Turkana Basin. There's a single name up here, but really, it's the work of many, and I'll describe the team as I go through it, but I would really like to highlight and single out Daniel Green, my postdoc at Lamont, who has really spearheaded the tooth enamel project or that component of the TMP, which I'll talk about. 

So, I thought I'd start by paying tribute a bit to Richard, who I didn't know well, I'm young enough that by the time I got to the basin in 2007, he was really putting his efforts behind infrastructure building and was less in the field. And so, I really got to know him through two people who were my mentors, Thure Sterling and Frank Brown. And so the way that I began to appreciate Richard was sort of indirectly. And so, in 2007, Thure sent me his youngest graduate student, wisely alone up to Turkana, but he linked me up with his most senior and Turkana tested graduate student, Naomi Levin. And so, we bumped up to Turkana in an old Land Rover with Paul Mulinge, one of the longtime drivers for Thure and Frank. And we rendezvoused with Frank in Lodwar and from there headed north and set up what I would call one of Frank's minimalist fly camps. 

And so, the morning would, this is my first time ever in Kenya in Turkana, in field work and the morning would begin with a few biscuits, some chai, and maybe we'd share a tin of fruit or something. And Frank would smoke a few cigarettes and then he'd say, all right, let's go. And we'd go out all day and maybe you get a few biscuits we share a packet of biscuits at lunch. This was new to me. Fast forward two years, and Frank and Thure were putting together a field geology course, and it was my first time on the east side, and Frank, somewhat begrudgingly, I think, said, we're going to stay at this new TBI place. And thank God for that because it was farewell to cold breakfast and hello to hot breakfast, bacon, and eggs and that bacon and eggs, hot breakfast symbolizes all the research and resource support that TBI has provided. And I'll come back to that in more detail. 

That's not to say I don't occasionally email Lawrence with complaints about TBI, but it's a wonderfully run place and it has really facilitated research in the Basin, and it's part of Richard's big vision. Okay, onward to research. So I guess the key focus of my group, my research group, is on terrestrial ecosystem change in the Cenozoic, primarily in the Neogene, the last 24 million years or so. And the most profound change is the emergence and expansion of grassy ecosystems around the world. And you can see in the map here, there are boxes with ages in the top corners that show in millions of years when the expansion of grassy ecosystems happen in that part of the world. And you can see in the equatorial region in Africa, it happened earliest around 10 million years ago. And there are a bunch of references at the bottom of the slide that show the papers that discovered this or published this. And then the various proxy, soil carbonate tooth enamel. And we have two pioneers of the tooth enamel proxy here, Julia Lee Thorpe and Thure Sterling more recently plant wax biomarkers, and then organic matter. 

And a fundamental question that comes from that is, how did Neogene climate influence major shifts in terrestrial ecosystems? And what were the evolutionary consequences? And we can further break that question down to these two on the left side here. What caused the emergence of the world's tropical and subtropical C4 grasslands in the late Miocene to Pliocene? and specific to Eastern Africa and southern Africa, did climate and vegetation play a role in the evolution of hominins or humans and other large mammals? And so the way that I think about this is in the schematic on the right side there we have three climate variables have precipitation, which can be looked at as moisture deficit or aridity. We also think about the seasonality of that precipitation. There's atmospheric CO2 and then finally temperature. And those three variables, which Curtis brought up as sort of a little bit distant from everything, interact with the ecosystem on the right there. And so you can see the feedbacks from those climate variables in terms of the proportion of woody versus grassy vegetation on the landscape, and then the impacts of other things like fire and herbivory. And so my research group is focused on reconstructing a lot of these things using geochemical techniques. 

And so one of the first things I did after arriving at Lamont about a decade ago as a postdoc, was team up with Pratigya Pollisar, an organic geochemist and Peter deMenocal a paleoclimatologist to develop a plant rex biomarker record for Eastern Africa using I guess Bernard, these are massive lampposts out in the Somali Basin, and they provide a very wide but somewhat diffuse light, meaning they're integrating the vegetation signal from across the horn of Africa and maybe even larger parts of the African continent. So, we looked at these two sites on the east side, DSDP cores 235 and 241, and we show with the carbon isotope record there in panel B, that C4 ecosystems emerged and expanded around 10 million years ago. And in the panel next to it, you can see many different primates including the origin of hominins at 7 million years ago. 

And the significant part here is that this means that our entire origin and evolution of hominins in humans occurred in Africa in the presence and on grassy ecosystems. So after producing this East African dataset, we went on to look at the West African side of the continent, central Africa using these cores 959 and 659 off the coast of West Africa. And remarkably to us, or surprisingly to us, the story was the same at about 10 million years ago. We see the expansion of C4 ecosystems, and throughout the talk I'll use this sort of tree and C3 grass or some kind of grass to show C3 ecosystems in sliding to the right C4 ecosystems. Furthermore, we found in panel C here that it didn't seem to be the result of hydro climate shifts, large scale aridity shifts there's no secular change there. And so we're still working on the causes of that, but it doesn't seem to be aridity in that case. 

And then just a couple months ago, the REACHE team published two big papers in science. Dan Peppe led this one on multiproxy evidence for C4 grasses in the early Miocene at the REACHE sites, which are in Western Kenya and Uganda. And you can see moving across from left to right, we have soil organic matter, Delta 13C plant wax, Delta 13C that came from my lab, pedogenic carbonates and then Carolyn Stromberg, Rahab Kinyanjui, Elise Novello produced a beautiful phytolith record. Okay, so these are sites that are somewhere between about 21 and 18, 17 million years old, and I'll come back to those throughout the talk. 

Okay, so I'm going to introduce the Turkana Miocene Project, although you've probably seen this decal like a thousand times already, which is great. The Turkana Miocene Project is an NSF funded project, and it's a very large team, and it's so large that we sort of got three overlapping subgroups. There's the earth team, which does field geology, the geochron team, which is headed up by Troy Raspberry here, and her students and Sid Heming at Lamont. We're doing some paleomeg. We're doing geochemistry of volcanics headed up by Sarah Armana. And one of the unique aspects of the TMP is that we're doing tectonic modeling, and that's headed up by Bill Holt and his group here. And then more recently we've added seismic data, which Richard and Isaiah were critical for linking the team working on the seismic data, that’s Chris Rowan and Anne Becel to Keith Hill and Africa Oil, which unlocked terabytes of seismic data for us. 

There's a climate team, which I'm a part of, Greg Henkes is working on clumped isotopes, I'm working on the biomarkers. And then we have a whole team doing paleobotany that includes Rahab Kinyanjui, and she's doing pollen phytoliths, plant macros, some of which you saw yesterday in Natasha's talk from Torpenawi. And then we have climate modeling, which is led by Chris Poulsen and his students. So the idea is, and finally the life team, we saw Natasha's part of the life team and Ellen's part of the life team, I'll get into this more, but they're doing the more classic vert paleo stuff at the sites. And TMP is really trying to integrate all these different sites where people have been working for years and years and years into a more common framework and adding modeling on top of that to fundamentally address the question of, what were the roles of climate and tectonics in the evolution of humans and other mammals?

So, this is the leadership team. Sadly, we lost Isaiah in January of 2022, and then the rest of us are carrying on much with Isaiah's inspiration behind us. These are our lampposts scattered around the Turkana Basin. They're small temporally and spatially, so tiny little lampposts compared to the Omo group deposits. In total, we have about 50 members. It always seems to be growing from about 20 institutions and four countries, they're listed here. I'll give you a second to read and digest, find your friends, your colleagues, your next potential mentor, whatever. And again, one of the things we're doing is trying to weave together the TMP with the various paleontological projects that are going on and those are in the bottom right corner in red. We're working with Ellen Miller a lot at the Buluk, the Lothagam research project headed by John Rowan, work at Nakwai by John Kappleman and Ellen Miller. Ellen, your name's everywhere in here. The Napudet research project with Gabrielle, Tara, and Ellen, the REACHE project and then Torpenawi.

So, this is a summary figure from our proposal. It's nice to be able to recycle it and reuse it, and it's actually growing and we're working on it. But here you can see the earth climate and life themes and how they're interwoven. And so, we are working on refining ages, timing of volcanism and tectonics in the Basin, linking that with large scale, but also regional and local climate proxies. And then filling in the sort of life side and trying to understand periods of turnover or transition in the faunal record. And in this talk, and my main job in the TMP, aside from as a PI, encouraging people to do things and being the complaint department, is to work on the biomarkers, which includes vegetation reconstruction, but also trying to reconstruct fire and other things. And then tooth enamel to look at diets, vegetation and hydro climate. 

So I'll start with the tooth enamel, because when you hire an amazing postdoc like Daniel, the job gets done very quickly. So we're pretty far along, almost finished with that. And the biomarkers we're still working through because it just requires a lot more bench time in the lab. So, I'll show a lot of XY plots with carbon on the X, oxygen on the Y. And on the left side on the carbon axis is C3 vegetation. And moving to the right is Grassier C4 vegetation. And on the Y axis for oxygen, it's wetter at the bottom, dryer at the top. 

And I forgot to mention that Daniel has this amazing team he assembled. This is Maria Kuzina. Wow, this thing is all, Maria Kuzina is a lab technician who ran through a lot of the enamel samples. Then we had these two undergraduates who Sneha took the sort of late, early, and middle Miocene sites and Ashley took the early Miocene sites, Oligocene sites, Torpenawi, and then both did a massive data compilation for most of the tooth enamel coming from Eastern Africa from about 40 to 5 million years ago. So that's what I'll be showing momentarily. Okay, so Natasha showed this yesterday, Torpenawi at about 29 million years is very different than the other sites. It's clearly in the C3 space. And we have these nice probability distribution. So this is the oxygen distribution here, carbon up here. And so we'll overlay those as we step through time. 

So that’s Torpenawi pretty low in oxygen as you'll see momentarily and decidedly C3 diets there. These are mostly higher axis, as you saw from the talk yesterday with a few samples from Lokone, something around 27, 24 million years. Add some data from Rusinga, so down in the Victoria Basin. Loperot is a key site that's very close to Nakwai at about 17 million years again, still in the C3 diet space, but we're starting to see in the early Miocene more positive oxygen. So probably more arid. Locherangan has a fairly large fauna. It hasn't been worked in many years, but we returned to that collection and we'll be going back there in the field in about 10 days. Then we've got a huge data set from Buluk, which Ellen showed yesterday. So I'm going at a very broad spatial scale here. This is all the teeth from the site. 

And I'll zoom in a moment, but again, at 16 million years, we still have predominantly C three diets and a wide range of oxygen coming from Buluk. And then this brings us all the way through the middle Miocene to Napudet where we have essentially C3 diets again. And then at Napudet we have something called the red series, which is not well dated yet, but it's probably corelative to the Lothagam sediments. So we're cautiously calling it something like 11 to 5 million years. Hopefully we'll have some updates on that in the coming year or so. And that's the first time we see mixed feeding and C4 dominated diets in the fauna. And these faunas look a lot like fauna at Lothagam. I have to check the pigs to see what they say. And then we have Lothagam, which is where we see a massive shift towards mixed feeding and C4 grazing in the late Miocene. 

So that's the story we've assembled from the TMP so far. We can put it in the context with the compilation that Ashley, Sneha and Daniel did. And what we see is we're folding in here lots of data from the Fayum, North Africa, Chilga. Basically, this is all data from the Eocene to the middle Miocene. The main point is we have a huge range in oxygen and carbon is relatively limited to C3 and perhaps a few mixed diets. This is all the published data that postdates 10 million years of Nakali, Samburu hills, Lothagam, again, et cetera. And then we just overlay all the TMP data over it. So it's a pretty consistent story with what's been published before. We really see the emergence of C4 in diets in a major way at 10 million years at Nakali. 

So then we can add this to the panel. This is the one missing panel I think in Dan's papers. We didn't have a lot of tooth enamel data and now we can put it there. So again, here's the previously published data. We have the Delta 13C enamel on the X axis, time on the Y, and then we can add the TMP data and then the sort of vegetation zones or the dietary zones. So again, very little evidence for any C4 in the diets until about 10 million years. So this presents a bit of a puzzle, right? Because there is clear evidence in the phytoliths, this is the pedogenic carbonate and somewhat in the organic matter for C4 on the landscape. And the biomarkers are some at Moroto in the purple dots there. And so, I guess this falls on the team of isotopes, we were called the other day to sort this out, but there are a lot of complexities to these proxies, and clearly, we need to dig into them.

And the tooth enamel data is exceptional because we can look at so many different levels. We can look over 40 million years at sites or we can compare within a site. So, here's Locherangan again, the anthracotheres, the water loving ones come in with the lowest Delta 18Oxygen, and the Afropithecus looks to be arboreal and folivorous plotting with the giraffe and oxygen space. As Ellen pointed out yesterday, we don't see that at Buluk so that presents a challenge or a puzzle for us. We can zoom even further into a single tooth, and this is some work that Daniel Green, Chris Poulsen and Daeun Lee, three TMP members, plus other members of REACHE, including Susie, Cody and others published last year in PNAS. 

And what Daniel did was sliced an Afropithecus molar, and then did he shrimped it, which is an ion microbe. It's like a 10 micronized dot, is that for me? 10-micron sized dot on the tooth, and it gives you an oxygen isotope value. So, if you go along the enamel dentin junction and make all these dots going, oh boy, it's hard right along here, you can create a weekly map of the body water of this Afropithecus from 17 million years ago from Kalodirr. And so, the data you see there is days in the tooth gross life, the tooth enamel formation, and then the delta 18Oxygen and the Y axis. And so, what Daniel's able to do is pull out this beautiful seasonal signal from Afropithecus that looks to be annual or sub annual. There's several patterns in there. And then our climate modeling team, Chris and Daeun, were able to come in and say, we can run a 17-million-year climate model and see what rainfall was like. And what they pull out in the dark blue is the biannual rainy season. You have months there on the X axis and delta 18O, on the Y. 

Okay, so the plant wax biomarker work is coming along. We have some results. And so biomarkers are anything produced by plants. And when we say plant waxes, generally we're talking about n-alkanes, which are linear hydrocarbons formed on the coatings of leaves. And we can make isotopic measurements on these because the carbon in that wax comes from photosynthesis so you can tell us just as a tooth or, so a carbonate does if it's from a C3 or a C4 plant. Okay, so far we've collected about 125 biomarker samples. That was in last summer's field campaign from the four sites listed there. Our wishlist this summer is underneath there. There's five sites we'd like to get to. And this is largely headed up by my graduate student, Ruth Tweedy. And so, here's the biomarker record that was published in Peppe et al. We recently added some samples from Buluk collected by Bill Lukens during the REACHE years. And then we have our samples that are in the hopper now. And the Buluk samples show C3 vegetation through and through. So that's the earliest and the only result we have so far, but we're working on it. Okay, so in the yellow on the right are the other sites where we either have samples or intend to sample biomarkers with the age ranges indicated by the height of the bar. So we have a lot of work planned ahead. 

In addition to just doing the carbon isotopes of the plant waxes, we're working on some other new approaches. One of them is the n-alkane distribution, so the different abundances of the homologs of waxes. So, we're actually looking at how tall each peak is for the various plant waxes and those seem to be a chemo taxonomic indicator. And I'll show a slide in a moment digging a little deeper into that. We're also working on a class of compounds called PTMEs, which are produced primarily by grasses. This will enable us to see in the past C3 grasses on the ecosystem, which are isotopically the same as C3 trees. And so this is a new way to look back in time. One of the students yesterday in the morning session, I think asked a really great question about differentiating C3 and C4 vegetation. And this gets us out of the isotopic framework into a different geochemical system. 

And then finally, PAHs are combustion products. They're out getting rained all over the landscape now as smoke blows down from Canada. And those combustion products are molecules that we can find in sediments in the past, so we can reconstruct fire that way. So this is a compilation of data by Ruth and a former undergraduate intern, Sarah Shi where we looked at woody vegetation, which seems to have a higher abundance of C29 and C31 homologs. Whereas the C4 grasses tend to have produce higher amounts or higher abundances of the long-chain homologs like C31, C33, and critically C35. 

So, what we're trying to do is analyze hundreds of modern plants, including C3 grasses, to look at their n-alkane distributions, and then use some machine learning techniques to try to do a chemo taxonomic approach to say, can we identify from a soil what the distribution of C3 grasses or C3 trees were in the landscape 25 million years ago? And so, this is a great note that Sarah Shi who was once an undergrad intern doing organic geochemistry, went away to Cambridge and did a master's that involved learning machine learning. And she came back and is now helping our group with that aspect of the project. 

Okay, so in summary, the TMP and other enamel records show no significant C4 and diets prior to 10 million years. And if C4 vegetation was there on the landscape earlier, why didn't animals eat it? And one of the things we've been talking about is looking into not just the large herbivores, but the micro mammals. And I started doing that actually with Thure many years ago at the site of Nakali. And so perhaps micro mammals have more limited home range. We should try to look for a graminivore or something that would be able to amplify that signal if it's there on the landscape. 

Enamel data, I hope I demonstrated, can provide diet vegetation and hydro climate information at different spatial and temporal scales, even down to the week if you can get time on one of these fancy shrimp machines, the preliminary biomarker data from Buluk indicates C3 vegetation at the site consistent with the diets. And there's lots more exciting TMP work to come in this conference this week, but also in the years to come. And then finally, you should think about the wattage of your lamppost. Okay, what kind of light is it casting over space and time? And I think Kay mentioned this when she put up the article, the Faith, that article about looking at the space time question you're asking in the space time of your proxy. Natasha, just a few more slides. I wanted to talk about the legacy of Richard and Isaiah, especially in the training of African students, particularly through the TUC program. And so my group has been involved in teaching lab methods and concepts. Daniel and I taught a short course on stable isotopes for the TUC students. We taught a wider isotope short course, and we've done workshops in the museum. And here's a list of students and some of the things they're doing related to the TMP. 

One of the fun things about the Turkana Miocene Project having a big project is the young people. And we have so many young, amazing researchers who are going to be the next generation and they're all featured here. And then finally coming back to the TBI field stations. For the last six years, I've been teaching an undergraduate course part-time at Mpala, and then we shift up to the Turkana Basin. And I don't think I would've had the courage to take a bunch of undergrads up to Turkana without knowing that the nest of TBI was there where we could have a place to take them, and we'd get that hot breakfast and all those other things. We do take them out to Lothagam for a few days, so they do get that Frank Brown fly camp experience a little bit, not really, but TBI has not just enabled research, but also I think teaching. And many of the students, when they fill out their course evaluation for a whole semester, I only teach them for a little part of it, say that going to Turkana was their favorite part because it's such an eye-opening and beautiful place. So, I'd like to acknowledge lots of people, but mostly my colleagues of the TMP. Thank you.

View all Isotope Biogeochemistry videos