Naturally speaking. Hello and welcome to Naturally Speaking. My name is Taya. I'm Athena.
And we're here today with Sophie Spitharis and Rebecca Mensi, who are two researchers within the Institute of Biodiversity, Animal Health, and Comparative Medicine, who are authors on a paper recently published in the High Impact Journal, PNAS. We at Naturally Speaking got the idea to do a podcast on this paper after head of Institute Dan Hayden mentioned it as his highlight of 2017. Well, you know, I think what I am most pleased about in some ways, Sophie Spitharis's paper in Ganes. Awesome.
Absolutely amazing. That pleases me more than I have many things to be doing. I'm almost proud to be sorry for the reason to do that. That paper.
So... So it sounds like a pretty important paper. The paper's intriguing title is, Lempe species coexistence arises robustly in fluctuating resource environments. Sophie, could you maybe start by telling us about what this paper is about?
Yes. Basically what we really wanted to see is a phenomena that are quite common in nature, like tide seasonality, that can cause important resources to fluctuate. For example, nutrients, which are relevant to phytoplankton, they could fluctuate during these events. How could that affect the assembly process of communities of species?
We used simulations to find out what happens under these conditions, and we found a very interesting pattern of clump formation, which means species with similar traits, so very similar competitive abilities to exploit these resources. Formed clumps, and we thought that this was quite intriguing. We published it. So basically you have these groups of species that are competing for similar resources, but they don't end up actually out-competing one another.
Instead they'll coexist on occupying the same niche. Is that correct? This basically follows a theory that was published before by two scientists in the Netherlands, Shafer and Vanessa. They found this pattern, but using completely different configuration and a much more generalized model.
But yes, the principle is that these species form clusters that are sufficiently different to each other to be competing between them for resources, but within each cluster of these species we have many species with similar traits, so kind of a redundant traits. Okay, so to summarize this paper is about studying the impact of resource fluctuations on interspecies competition. Can you guys tell us why is this important in relation to previous theories of species coexistence? So that is very important in our opinion.
I always had a hint that this theory can actually be widely applicable, and I'm happy that we actually proved it with this modeling approach, because first day it was a niche theory which basically postulated that species need to be sufficiently different in order to coexist in the environment. But then we had also had this theory which suggested that we can have species coexistence if we have a lot of similarity between traits, so neutrality. Now this theory really, what it does is bridges the two theories, because as I said before we have again the niche theory between the competing clamps, but we have neutral theory for the species that coexist within the clamps. So in your paper you mentioned the paradox of plankton, and I was wondering if you could tell us why phytoplankton is a particularly interesting system to use for investigating this broader ecological question.
So phytoplankton is interesting because it is a group of species that is incredibly diverse. When you see a drop of water under the microscope you might see 30 different forms, different species, and that is fascinating. Actually we were watching that with Rebecca the other day and- There are also very beautiful phytoplankton have very beautiful shapes that's really worth looking at some pictures just to see how pretty they really are. It has always intrigued scientists, especially Huttinson and his paradox of the plankton, why you see so many species coexisting in just a drop of water, if there are only a few limiting resources there in nature.
And this is what many scientists have tried to explain after him, tried to unravel the mechanisms for these coexistence. And this is what we also tried to do with this approach. So this study used computational experiments to address this question, and for listeners who are maybe less familiar with the use of mathematical modeling for understanding complex ecological processes, could you maybe explain briefly how this approach worked and the types of inferences that you used? What we assumed in our model was that we had two resources, nutrients in this case, let's say nitrous and phosphates, which are basically the fertilizers that we use for growing our plants in the pots.
These two fluctuated in an opposite phase. Then we also assumed that we started with a very rich communitio species. We assumed 300 species initially. We then started initializing our model and we saw that not all of these 300 species of course survived because many were competitively excluded because they were competing for resources.
Meanwhile, the number of species that survived had some specific traits and we basically saw what were the patterns in these traits. And this is what actually revealed to us that there was this clumpiness in the traits, this cluster formation, which was very intriguing. As Sophia's described, we set up a model, we parameterized this meaning that we set it up so that what we were modeling looked like a phytoplankton system, starting off with our 300 species, etc. And what we mean by computational experiments is that we ran that model, we solved that model on different circumstances.
So for example, we changed the rate at which the resources fluctuated over times. Each fluctuation cycle might have lasted two weeks to represent tides or a year to represent seasons. And then we looked at how the output from that model differed according to those changes that we made. I have a question for you because you're always talking about fluctuations of resources.
So fluctuations occur naturally and it looks like there is not such a thing as stable conditions in nature. What is the contribution of the resource fluctuations to a system resilience? We found that when resources fluctuated more frequently, then we had the formation of less clusters of species, but they are more redundant species within these clusters. And then we found an opposite trend of more clusters and less redundant species when we have fluctuations of greater periodicity.
So we found that early on the competitive species within this community, this assemblage that we set up consumed resources quite quickly, and this set off a pattern of cyclical fluctuations in the remaining ambient resources that were available for all the other species. What this did was it created favorable windows in which the remaining species could persist. So when that happened, we found the species that had the traits that corresponded to those favorable windows formed clusters and they were able to then persist. So clustering leads to resilience in the sense that because within each cluster the species are very similar, it's not catastrophic if a species within that cluster goes extinct because the ecosystem functions can continue because other species can compensate for them.
Very interesting. Thank you. And then I have another question. Assuming that most of the condition in nature lead to clump formation, is a higher or lower number of clamps preferable?
So I would say that higher number of clamps is more preferable because that would mean that these clamps can cover a wider range of ecosystem processes. So within an ecosystem we would have more functions being carried out by different clusters of species, so more functional diversity, basically. Which is kind of fun because I suppose I would argue that actually it might be better if we had fewer clusters because then you have more redundancy within each cluster for the same number of total species. So that means the extinction of any one species is less problematic.
So the jury's out on that one. Could you both maybe tell us a little bit about your scientific backgrounds and how this particular work ties into your research interests more broadly? So I've got a background amongst other things in mathematics and computing science and one of the things that I'm particularly been interested in as part of that is meta population ecology. So this is when a large population, if you like, is split into subpopulations and we're interested in how the processes within those subpopulations contribute to what happens at the larger scale.
And we call that larger scale the meta population. I'm particularly interested in how populations change over time, but one of the things that was particularly interesting in this project for me was that I've worked on meta populations which deal with a single species. Here we were doing lots and lots of species together, so this was really fun. So my background is on marine ecology and my main question is how do communities of species come together?
What are the processes that help maintain the large diversity of species that we see in nature? So what are the competition processes? What should be the species traits? What should be the conditions of the environment that help maintain this amazing diversity that we see in nature?
This paper was basically spot on on what I do and I really enjoyed the process of actually making it with my colleagues. And it was written as well with collaborators from University of Dejan, Greece and from Texas. How did you end up working with these partners? Three, so you think?
Yeah, these are ongoing collaborations. George was my PhD supervisor that we keep working on these topics together. Daniel Rolke was the person who created the modeling framework for this, kindly shared it with us and we extended it to address these questions. Athanasia is a master student in the University of Dejan who revived these questions for us and we ended up with this fascinating project.
And with Rebecca we work in the same office. We keep we keep breaking our heads up on these questions every day. Well not every day but yeah, because it's fun. I've been part of it for a couple of months.
Very much fun. And then just to finish our interview I would love to know more about the action of creation of this work. For instance how did you develop an idea that is on the basis of this paper? What was the mental process of you and your collaborators while formulating this new theory?
So the idea was initiated very long ago, five years ago actually, when I was still working in the University of Dejan. So we started basically me and George's running, Daniel's models, and we started toying with parameters and the parameterization and the fluctuation of resources for example. And we saw this pattern emerging and we weren't really sure is this an artifact, is this really because of a mechanism. So we were very tenderized about these courses discussing this with Daniel, but then we left it for quite a long time because we...
We got tied up in other things. But then the whole thing, as I said before, revived after Athanasia got this as a topic for her master's project. So we started all scratching our heads in finding the actual mechanism behind it and we found it. I think it was a really important catalyst to decide to run the paper about it, quite early on, which forced us to crystallize our ideas about what we thought was going on to try and figure out what's the right figures, the graphs, etc.
that we wanted to put in the paper what they would show. And we had on our whiteboard for a long time a picture sketch of all the figures that we wanted to go in and we were trying to see, well, okay, so if we want to tell this kind of story, what pictures do we need to draw to help the readers understand that? And then through the process of getting feedback from the reviewers, we refined both those figures, but also exactly what it was we were focusing on in the paper. And it was really helpful to force us to think in a lot of detail about the mechanisms that underpinned what we deserved and George, our University of Dejan was particularly critical in doing some work with a simplified project.
And we had to do some work with a simplified version of the model to try and understand the detail of all of those things and really understand the mechanism. And then we spent a long time trying to figure out how to draw the right pictures to make that accessible to our readers. One of the things that I found quite fun about this whole project was the title that you mentioned as Intriguing, earlier on in the interview, we actually wrote the day before the very final submission. PNAS had edited our title to fit with their grammatical guidelines or something like this and didn't really like what they've done with it, and so we came up with a completely new title.
So it went in with a completely new title two days before it got accepted or something. Well, there's the inside scoop on some of the complexities of the scientific process right there. It's fascinating. Thank you so much for taking the time to share this paper with us.
It was really fun working on it. I've been really happy to be involved with it. I think you all enjoyed really much working on that. We never thought it was only a creative process.
Yes, it was really good. And you made those other departments super happy. So, better than that. Yeah, Dan was actually, yeah, he knew what we were cooking.
He was really supportive. Very, very supportive. Yes, yes. Thank you, Dan.
In reading stuff for us and trying to make sense of even early drafts when it wasn't really there yet. So, yeah, it's been really good. Fantastic. Well, thanks so much.
And thank you all for listening to NaturallySpeaking. Until next time. Bye. Thanks for listening to The NaturallySpeaking Podcast.
If you've got any ideas for future shows or want to get involved, please contact us. We'd love to hear from you. Our email address is naturallyspeakingpodcast.com. NaturallySpeaking is a podcast from the Institute of Diversity, Animal Health, Parrot and Medicine at the University of Glasgow.
And for more information, you can find us on Facebook. You can find us on Twitter at theNSpodcast.com. And you can download future episodes and find more information at our blog at www.mattressspeakingpodcast.webpress.com.