 So, if you're a little bit like me, you're worried about the environment, or actually very, I would say even more, I'm actually deeply concerned about what's happening all around us. And I increasingly frustrated with what, what the lack of action or so from the politicians, what's going on here. In this upcoming talk, Teodor is going to tell us how data science can try to find what points, what pain points need to be addressed with priority. And I'm really looking forward to hearing this, because I think this is going to affect all of our lives. Over to Teodor. Thank you very much for this kind introduction, Sifu. Yeah, good evening. I heard that I'm the last talk of the day here, so I'm not going to be extra quick, but I'm not going to be extra long either. Before I start, though, I would like to know a little bit about you. So I'm going to ask you some questions, and I would like to have some answers from you just by a show of hands. So who of you does self-identify, maybe, as a hacker? Who of you does? So those are like some. Okay, thank you very much. Who of you works with larger amounts of data, or would call themselves data scientists? So that's most of the guys here and the people in here? Perfect. And who of you is some kind of natural scientist, and it's maybe just here to refresh their memories of ecology? Okay, so that's... And who of you is in none of these categories? Wow. Okay, that's also quite a much. So my name is Theo Rosch-Berlea, I'm right now doing a PhD in Marburg, where I also studied. I studied biology and focused on the genetics of microorganisms. And now I'm doing a PhD in bioinformatics, and I'm studying ecology of the lake microbiome. To let me start, let me tell you a story. The story begins 4.5, 4.4 billion years ago, when Earth started to be cold enough, so that water could start to accumulate in first oceans, foreign. Shortly thereafter, amazingly, shortly thereafter, the first form of life emerged. We know this, or we assume this because 3.5 billion years ago, some billion years ago later, we find the first fossil evidence of similar celled organisms. Not like this, but this is the only creative commons image I found, which looks nice enough for microorganisms. They are hard to give an image to. So these first forms of life emerged, and with this it started a new age, or a new way of things developing. Because right from the get-go, you had a process of change and adaptation, of mutation, of diversification, which then emerged from that. So through evolution, as we call it, you have a diversification and life starting from single, simple cells, or something like this, started to diversify and conquer all of Earth. First the watery terrains, and then got more diverse and more special than the land and the air. And every niche that is available, every way of living that is thinkable, or not even thinkable, but rather possible, is currently more or less there is something, some organism that lives that way. And this went on and on, and organisms changed, and some of them look rather scary, some of them looked rather nice. We know all of this from fossil evidence, so we dug in the ground and found some fossils or something, and looked at it, and we said, oh well, it must have looked like this. But from time to time what happened, and what happened five times in Earth's history, was a catastrophic event called mass extinction, where 75 to 90 percent of all species went extinct in a blink of an eye. The most well-known of this is the last one that happened that killed the dinosaurs, which happened through meteorite, as I think most of you know, right? So this ended the age of the dinosaurs, but then also this enabled something else to happen, which is that every time such an extinction event happened is that new forms of life could have used that space that was freed through the extinction to foster. The extinction events usually, again, I told you, 75 to 90 percent killed almost everything. Some things were left behind, so when the age of the dinosaurs ended, the age of the mammals started. This is a rather late picture for mammals, but there were a lot of different mammals before that. Also the age of the birds started, but common birds. So the age of the mammals then led to, well, the age of the mammals then directly led to the age of the humans, the Anthropocene. This is a very special age in geological terms, because it's the first time that a single species shapes the whole face of the earth. When it said the age of the dinosaurs, it was dinosaurs, not a single species of dinosaur, but now it's just us. And what we're doing is we're building crazy stuff. When you came here, you have either used a car or a train or a plane or maybe a boat, I don't know, but you used a lot of infrastructure and things that were built and created by us, which no other species really does to this extent. And what we're doing by this is we're changing, again, the face of the earth, a lot of the land that is surrounding us. We're building cities, large cities, that remove all the living, most of the living species from that land at all. We create streets that cut through forests or plains and separate biomes and species, and by that we reduce their gene pool and make them more likely to go extinct. We're using plastics, which are, by natural physical forces, turned into microplastics, and those are all over the planet. We, of course, also produce a lot of CO2 and create a climate change that is man-made and is a lot of problem. I'm not going to talk about climate change here. I really want to stress that, because these are linked problems, the problem we have with ecology now and the climate change problem, but they're different, nevertheless. Also something very interesting, and I don't know what you thought about this, is the amount of light we produce, which really changes a lot of processes in nature, because usually light was only there at day, and then we create a lot of light in cities at night. This really changes everything, kind of. What we in a certain sense do is we take ecosystems, and out of these we create techno systems. We remove everything that is ecological about them. Actually, when you think about it, in your house, in your room, this is an ecological ecosystem-free zone. You only want those organisms in your house, in your room, that you put there. If there's something else, a spider or something, you really think about killing it just to get rid of it. You really don't want ecosystems around you. This is what we do to a big part of Earth. But I might ask you, is this a problem? So what we know, we don't know a lot. There's a lot we don't know. But one thing that we know is, for example, what you might have all heard in the last year or two years ago on the radio, no, actually, last year, not in 2019 yet, but soon will be in 2019, then this will be in two years, two years ago. So over the last 27 years, the insect biomass in Germany was reduced by 75%. This means when you take a box and you go through the air and you collect all the insects that you can find there and you put them on the way, this will weigh 75% less than 27 years ago, if you have done the same thing. This, of course, is only one publication, but more and more publications confirm this finding. And now you might think, if we are reducing the amount of insects so radically, this might have an effect on other organisms. And we see something like this for birds and for plants, of course, flowering plants. But actually, we see this in all species, all kinds of animals. So in the last 100, 200 years, extinction rates rise rapidly. This means that organisms that were there before, species that were there before, are not out there anymore, either extinct completely or extinct in the wild. This should, I think, really worry us. Because the question now is, how close are we as a source for an extinction? To those mass extinctions I told you earlier. And this is quite close. This is a nice figure that I found somewhere and had to redo it. Just look at the, for example, the Mamalia. Those are mammals, reptiles, some of the names you can understand. But the most important thing here is that the space from the bars to the red line, which would then qualify as a mass extinction, it's not that big anymore. And I hope you can see that this will only rise. So scientists are now arguing, and they're actually not really arguing, but they're coming together to the idea that human beings, human kind, might be the next, the sixth mass extinction. So we are as bad as this meteorite that killed all the dinosaurs. How do you feel about that? I don't know. Actually, I don't think that I'm telling you much new here. And I think I need to tell you this anyway to refresh your memories, because this is our motto, right? And then the other thing is that we need to frame it. On purpose I didn't put many images of cute looking animals in here, but rather took the part of the data. And what I found most interesting when I prepared this talk, I thought about the movies I saw in the last years, and so many of these movies actually are about this problem. And it's not only about amazing beasts and where to find them, where a guy goes out to save as many animals as he can, but even the bad guys in our movies are doing this somewhat better than us. When you think about the last Avengers movie, Thanos really tries to save ecology by killing half of it. That's a bad idea, but at least he wants to try it. And I don't know whether we are at that point yet, at the point that Thanos is about. And what's even more problematic, I think, you know the Deutsche Bahn, most of you, and this Deutsche Bahn has a magazine, and the last issue of the magazine has this issue on its cover. So if the Deutsche Bahn knows about a problem and really wants to put something on it, we really should do as well. So let me get some, please clap on so that I have some time to drink. Water tastes so much better when someone is clapping. And so to get more into it, I would just now want to talk about what an ecosystem, how we can think of an ecosystem, not really what an ecosystem is. So for you computer scientists and data scientists and so on, you know networks, right? So you can think about an ecosystem as a network. You have nodes, which might be animals or species or groups of species, and you have interactions between them. Here I have a very simplified version of this because of two reasons. On the one hand, it's quite hard to fit a very detailed network on this slide and I won't be able to tell you about it because it's too detailed. On the other hand, it's because we don't know it for many ecosystems in a very fine detail. So I just put some plants in there, algae, insects, fish, and so on and so on. And I put only two kinds of interactions, positive ones and negative ones. I hope you forgive me for this simplification. So this ecosystem is a network and it behaves somewhat like a chaotic system, somewhat in some other senses it does not. So it's not really the case that a small change in input. So no, oh well, I don't know really. But it might be the case that a small change in an input leads to a drastic change in the output. But what's even clearer is that you have a sense of robustness in these systems. So you can change most of the parameters for quite some and it doesn't really break. But then if you make more changes, this system will break. So this is a lake ecosystem most of all because I study lake ecosystems. And I don't really study the whole lake ecosystem, but only the small part on the left bottom of this. And I just want to go to this a little bit more because we don't really know. I don't know, I never know what I should say. We don't know anything, but we know next to nothing about this. We know that there are a lot of organisms in there abbreviated with Greek letters. And we know that there are a lot of abiotical factors that have something to do with these bacteria and that interact. And that bacteria, certain bacteria can only live with certain abiotical factors. But we don't really know many of the interactions there. But let's go back to our ecosystem and let me tell you some, to give you some idea of how these ecosystems are getting worse Let me go through three ways of disrupting an ecosystem. For example, what you could do is you could put too much nitrate on your acre. And this nitrate will get into the rivers and get into the lakes and the seas. And this will lead to a lot of nutrients, which you could think that it is a good thing because for your plants on your acre, it is a good thing. But on the lake, in the lake, if there are too many nutrients, this will lead to a lot of growth for algae and plants. And these algae and plants will then, on the one hand, reduce the amount of light that goes into your lake in the lower parts of your lake and the amount of oxygen that is available in this lake for other organisms. This in turn will lead to a giant wave of death for insects and fish. And then, of course, frogs and birds will die too because they don't have anything to eat. And now your ecosystem is dead. A second way of doing the exact same thing is to bring in an alien species. So now an alien species is, of course, not from Mars, but maybe from North America. And maybe you bring in another frog species. And this frog species, which is, this is a real case, brings with it a new disease, like a fungus or something. Again, this is a real example. And what happens is that these fungus will kill many of the frogs that are here. The new frog species will be resistant to the fungus. No problem there. But our frog species that live here will not be, so they will die. Also, the new frog species might just eat too many insects. So the insects will die too, let's say. This will again turn the birds and the fish to dead birds and dead fish. And the algae will have a good time. And now you have a dead ecosystem again. And the third thing you could do is you could change the ecosystem on a geographical level. And here's where our system, our model of an ecosystem as a network, really starts to not work anymore. Because if you build a dam and you enlarge the lake, you don't really, it's hard to see from this model what will happen. But what could happen is that the fish have more place to live. So maybe they need less and they mate less. So then you have less fish. Or maybe the algae will have more space and they will spread out more. And this will again lead to the same thing that happened earlier. It's quite hard to say with a dam what happens. And again, this is a reminder that there are other layers of information, geographical and evolutionary or historical data information that is needed to really model an ecosystem correctly. So let's talk about what we can do about that. And I think we'll not talk about like what we can do in our everyday life. This is a very important topic. But maybe for another talk or for another discussion. I will talk in this talk more about more systemical things to do. And I think there are two main paths that we can go about. And I think these two paths are not exclusionary but they are needed both. And one thing is just leave earth alone. But maybe not the whole earth but just half of it. There is actually a project which is called Half Earth Project. And there is a great book that is called Half Earth written by E.O. Wilson where he argues and this organization argues that you should keep half of the area, surface area of the earth away from human impact. Half of it. We humans, we live on the other half of the earth and nature will thrive on the other half. The half will not be like that straight in the middle but it will be mathematically half of it in patches. And these patches should be as large as possible and so on and so on. Because right now humans have impacted 77% of land and 87% of ocean area and of ocean areas that are not covered by ice. That's a lot and this is really a problem. If you ever get the chance of arguing for a largeening natural reserves or safe spaces for nature please do that because in these nature can recover and nature can do its own thing without us interfering anymore. Especially if you think about the Amazon in Brazil I think this is something that many of you might have in mind that this, I think this will not get better just by itself and we really should, as humanity, we should work that most of the Amazon is still in untouched shape. This is very important again, leaving as much of the earth for nature as possible. But then you can only do this so much and we also need to think about the other half of the earth because we need food, we need places to live and this will still be a lot of impact on these areas. So the other suggestion that I would make to maybe you in this room and on the internet of course to do something about a sixth mass extinction is to hack it. Because we are now of course in a hacking conference. This is of course what we should do, just hack it and this will get better. Earlier today I was in the talk of Frank Liga with hacker ethics and I thought, oh well, I should have looked this up before I prepared the talk because there are many, many parallels that I would have taken if I had known this but I didn't, so I didn't. But I think when you go to the general mindset as far as I understand it, oven hacker, it is you have a system there, it is kind of a black box. You cannot really see it, how it works from the outside. You'll find a way in, you open it up and you use the information that you have gained about this system to either destroy it maybe or leverage it or make it better maybe. And is this something similar that I, that should be our second solution but instead of hacking a computer, if you're this guy sitting in absolute darkness in front of your computer, instead you should hack us. And of course you shouldn't know something about the black box system to destroy it or to leverage it but to keep it alive as best as you can. So to do this we need more information how to put this into data because data is what we're working with, right? So let's go back to our lake ecosystem and on this slide again we have this network model in front of us. You can put many of these things into numbers. You can of course count the number of organisms that are in there and you can measure the interactions between them. You can then use these numbers to do some modeling. And this is a very, very simple model of just a predator-prey relationship which and how it changes all the time. And what's really interesting is that it has such a shape when you have on the x-axis the amount of prey and on the y-axis the amount of predators. This will be over time, this will circulate around a certain number. Of course these models can get as complicated as you want and I will not go much more into this detail but of course we could. So the best thing that you could have in terms of data when it comes to ecosystems would be interaction data because ecosystems as a network are defined by the interactions between the members of this network. These interactions should be then very detailed and in very between single organisms and so on. The problem is that it is very hard to get this data because you need to get outside and you will need to sit in the forest or in your ecosystem of choice and you would have to wait and look and see who interacts with whom. And while this might be feasible for larger organisms, let's say elephants, wolves or something like that, this is very infeasible for, for example, say insects. How do you want to measure the interactions between them? Of course there are ways to do them, to do that but still it's quite hard. But another way to get something like interaction data would be to track movements. So this is an image of organisms that are living in the sea, different organisms and from the movement data you can, as you know as hackers, because you say, well we don't want any movement data from humans, we do want movement data for animals. You can say who's connected to whom, who's living with whom, what kind of work does he or she do, same thing you can do with animals. But it's quite hard again to collect movement data for animals. What you can do is to put a GPS sender into the bodies of the animals. This has become easier and easier right now. Just a few weeks or months ago, a new module was added to the ISS called Icarus which makes it easier to track movements of organisms with a GPS sender. But still you would have to bring the GPS sender to the animal and the animal cannot be too small because it needs to carry this GPS sender without being harmed or anything by it. So another way to get some data is to simply more or less count the amount of organisms of every species that you have in your ecosystem. I told you that I work with bacteria and microorganisms so I want to present to you how we do this for microorganisms. We take a sample for example from lakes but you can also do this for soil or any other ecosystem for example, your skin is an ecosystem just as a lake or something so you could take a sample of that. Let's say you have enough bacteria in this sample, you can extract the DNA from these microorganisms quite easily actually in the lab. And then the next step would be to sequence it. So you take the physical object, this DNA and you turn this with a quite expensive machine into data, into text data actually. And what's more interesting is this text only contains of four different letters. So this is quite easy to work with for you as a informatician. This DNA you can then use to find out what organisms lived in your sample before you killed them by just looking up in some database what DNA sequences are present in what organisms. Again, this goes for microorganisms but of course as well for larger organisms. Then you can do some not so fancy analyses for example, simple correlation between many samples and you can say of course if these if two different organisms correlate over many samples they might be associated and that they might interact in nature. So that's what I do right now. I have two kinds of news now when it comes to data, data science, data ecology and how this might help us to save us all from a six mass extinction. And the good news is there is a lot of data out there. Before I looked for the creative comments of my images I had this Toy Story meme in there with data everywhere because there's a lot of data out there in public databases where it's really easy to download just right click it, it's public and so on and so on. So there's a lot of data out there people create a lot of data. Most of this data is from smaller surveys where they took two or three sites where they collected some data and put them on the internet. And the other thing is that this data is in a lot of shapes and a lot of different data is out there which brings me then to the bad news is for example, batch effects. I don't know whether you know this term but it simply means that if you have a different method of extracting the DNA from the samples or extracting the sequence from the DNA and so on this will change your result. So most of the data that is out there currently is not easily comparable because it was generated with mild differences. So this is a problem and this is like the bad news of data ecology. With that bad news actually I want to sum this up and I want to tell you that I, what I told you. I told you that we are facing right now a six mass extinction and what I didn't tell you but what I should tell you is we should not make the, we should not think that we as humans will survive this six mass extinction. We're just too large of an animal. We just use too much other organisms around us that will most likely go extinct so we will go extinct as well. We're facing a six mass extinctions. There are I think two different ways of avoiding this mass extinctions. The two ways need to go together to really do the job. The one thing is that we should leave our hands away from as much of nature as possible and then those places where we cannot keep our hands away from it we need to understand more and a good way to understand it is to use more data, more data and analyze this data more thoroughly but there is a problem which is bad effects. But before I really want to finish I want to give you a small invitation because when I prepared this talk I noticed that I will end on a bad note. It's never good to end on a bad note, isn't it? Especially when you're the last talk of the day. You will all go to bed with a bad mood and I don't want to do this. So I want to invite you to organize and to form a new kind of community which is right now called Open Digital Ecology Community and I'm really not happy about that name. But the idea is to, and this is not, I have thought a little bit about it. I think I know what would be necessary but I have not thought about this more. There is no infrastructure ready yet. So I invite you to form this with me to come together with you as generally people that are somewhat interested in data that maybe have a hacker mindset that maybe have some background in ecology but maybe not. Maybe you're just working with numbers because you like it and just to come together regardless of background as open. Also do this as an open source project just as you would do an open software project. Do this but as data science. And also do this openly so that everyone can use and reuse the data in every step of the way. Digital of course, as in data centered use all the data that is around and try to figure out how to use it and bring it together. And lastly a community as in citizen science. I think this would be a very interesting approach to bring in people. Usually citizen science projects are everyone should go out and take a water sample and send this water sample into some experts and these experts will analyze them. This is wonderful, this is very much needed. I think also we can do citizen science on the other levels of the analysis so we can do citizen science at the analysis step and we can bring experts and non-experts together there and work together and distribute work over more shoulders. Please think about this and I'm really not happy with the name. We need to change that name and I don't know how. Thought about this two weeks. So, with that being said, no, I want to give you one example because I come from Lake Ecology, this is what I'm always thinking about and I told you that I think that the most pressing problem right now are batch effects. So I think what this open digital ecology community, what a very good name, isn't it? Could do as a pet as a first project is to bring together all this different sequencing data from different sources, try to figure out how to remove the batch effect or how to analyze data with the batch effects in mind. There are some first steps in literature that could go there but we still need to figure out more about this then pre-process them and process them and then analyze them and maybe at the end find something new about ecology. In this case maybe Lake Ecology. Also might be Ocean Ecology. There is a lot of data out there for this. Might be soil ecology. There are some difficulties with that but we'll discuss them maybe later and so on. So this is my invitation to you to do this together with that being said now. I want to thank you. I really want to thank the organizers and everyone that's behind the scenes that they accepted this talk. I was very happy to hear about this. I'm very happy that you all have stayed here so long. This is the last talk. Amazing, isn't it? This is my mail address. Please write me via mail or maybe via Twitter if you have any more questions or meet me around. And now of course I want to take questions in the room and I'm very happy to start a discussion with you. Thank you. So that was a call to arms if I've ever heard one before. Excellent. We can take questions. We've got plenty of time for that. We have microphones throughout the room. There's one here, one on the other side two in the middle. There's also internet questions as well that will go through the signals angel. Please, a question is one or two sentences followed by a question mark and not your life story. So with that let's go to microphone two please. Yes, thank you for your interesting talk. And at the last bit you were talking about removing or dealing with the batch effects for data that has already been collected. Are there also initiatives to maybe standardize upcoming data collection? And can you tell a little bit about that? Yeah, there are a lot of, there are some, there for some fields there are a lot of them for some fields there are some of them. This would be very helpful if there were standardized methods and again they're starting to be more in place. But what I told you about all this DNA sequencing stuff is a very, very new field. Not two years, but 10 years. And things have not come together yet. This is the one thing that would be great. But the other thing is the other data is already out there and it's free, it's out there. And if we just throw it away now that would be a shame. Let's go back to microphone two again. So after analyzing the data, what's your idea of hacking ecosystems? So improving them or rebuilding them or something else? Yeah, this has been, I've been on purpose a little bit, didn't talk much about this because there's still something I think about a lot because I think, I mean, the first thing is to keep these ecosystems alive while still having some of the pressure they are facing right now. So the one thing is to reduce the pressure on the ecosystems but in case we cannot do this fully in a short time we need to figure out a way of maintaining the ecosystems while still having some of the pressure or having all of the pressure. Is that a good answer? Okay. So let me do one or two sentences more. So we, there are all these variables, right? Every node in the network I showed you could be regarded as a variable. What we are doing right now with, so in German you would say Jäger, in English it's more Forester. These guys are in the best cases doing maybe what I'm thinking about of going into the ecosystems while having some of the pressure relieved or having more pressure and manually trying to rectify the ecosystems in the best cases, in the worst cases they're killing for fun, which is bad. But in the best cases there are sculptors of ecosystems that stay alive although they're not in perfect shape. So with that let's go to the back of the room, microphone three. Well first thank you for this talk and this call for action. My background is database programming and I find it just very assuring to hear this idea from you as a data scientist. Is there, do you know of any website or so to connect database analysis specialists with data scientists or do we need to build this? For what exactly? For connecting hackers with data scientists who have data to analyze. In the, I don't think that this, I don't think this exists for people outside of academia. Inside academia in your field you know because you've been in that field for some months maybe. You know where to go to get the data, you know where to go to get the databases and so on. So you have enough tech support? I, for now maybe yes, but let's not, let's not start now by excluding the database guys or people, sorry, the database people. Thank you. Microphone two in the front. Thank you, very inspiring talk. Let me ask you, you mentioned toward the end of your talk citizen science if you had the choice to name one or two of your favorite methods or activities or tools to enable and encourage this kind of community activities and science. What would that be? Oh, that's a good question, isn't it? And I have to boil it down to one or two. I think the main problem here is that the whole workflow is very, will have a lot of steps and these steps will be from different fields of academia. So there will be some standard bioinformatics steps that include a lot of algorithmics. Then there will be steps that use a lot of statistics. In my work I'm doing some machine learning to maybe get something more out of the data. So it's not that easy to just say, okay, use this one program and all your fears will be relieved, all your aches will be gone. This is why I think a community is needed, a group of people that talk and that can organize in different steps and that can maybe then still specialize but as non-experts. Thanks. Microfront three in the back. Yes, thank you for your interesting talk. I was wondering if you would have all data about ecosystems and have all this data interconnected. How can this contribute to stop the mass distinction of all the species? And should we not also think about the link to decision-making and the world it, that has in this the political field and, yeah. Yes, of course. There are two things I have thought about this and there are two thoughts I have. The one thing and both are not positive. So the one thing is we know so much about climate change. There is a single thing that we need to do to stop climate change and this is reducing CO2 emissions. It's very easy actually and it has been known for 20, maybe 30 years and decision-making is lacking by a lot. So I'm really, I don't know what else to do, what else scientists should do to change this because I think scientists usually they're very late back and they say, well, we don't know really. It's hard to say if you ask me, well, maybe, maybe not. And scientists have escalated their language over time, over time and now scientists are saying, okay, if you don't solve this in three years, we're all gonna die. This is very not something that scientists really want to do but there is not much more that scientists can escalate in their language to tell policymakers. That's one step where I think very, I'm not really happy about things are going. Now the other hand, while climate change is very easy, there is only one variable that is the, if you pull this lever, everything will be okay. This is not the case for ecology. There are many levels, there are many variables, there are many things that would need to come together. So I think discussion with policymakers is very important. I have here focused on the science of the knowledge because I think this, because I myself, I'm very depressed about policymaking. This is my hope that maybe here, if we now find the solution, then maybe policymakers will listen, although they didn't for climate change. Microphone two in the front. Hi, I wondered you showed several examples of catastrophic failure in an ecosystem by say one parameter change or one new species introduced. And I wondered now, we have this system and we analyze it and maybe know very little of it and then we think, oh, we only have to change one thing and the whole forest dies. Is there a way to say like reduce the risk or say build like a petri dish for a whole lake or something where we can try? Or is there just a, well, we must try on the lake and see what happens? There are a very, very good question. There are many, I think, and I hope that there are some steps from a very, there are many people working right now on very small pet ecosystems where they just have two or three or four or five organisms that or species that interact with each other, which totally has not the complexity that real ecosystems have. There is a huge gap in this and I don't know whether it's really possible to replicate ecosystems in the lab at all because there are geographical parts of it so you would have to rebuild the lake in its entirety and also there are historical features so it does matter whether this species arrived there earlier than the other species. This will maybe not be possible. It will be necessary to do as much data crunching as you can to say, okay, I think we're 50% sure maybe we can, do we have another chance, choice, okay, let's do it. Sorry, sorry to tell you that. They're getting us all depressed here. Is there a microphone one or so? Is that a question? Thank you very much for this interesting talk and the amount of answer they give to our questions. According to your knowledge, your current knowledge, do you think we have any chance as a species to stop this extinction or only delay it? Yeah, I would really like to say some positive things. I would really like to say, oh yes, we can do it. Is this so doable? It's not easy, but do it. I don't know. I think maybe not, but I think this is not really an option. You know, there are so many places where you can say, okay, we can't do it, but still we have to try. So we cannot, as a scientist, I know that I cannot attain knowledge itself, but still I have to try. So this is maybe something like that. Maybe we cannot do it. Maybe we know right now that we cannot do it. Maybe the way to do it is very bad, but we still have to do it. So the microphone at the back, number three. I'd like to go one step back. I seem to understand that understanding the ecosystem is a big thing. So what can be done to understand the ecosystem better and to be able to predict what consequences decisions have? As a data scientist, of course I tell you, collect more data, analyze this data, and we will find perfect insights. Some other ecologists will tell you, okay, we need to go out and we need to categorize every single species with all the behavior in very fine detail to really understand the nice teeth of an ecosystem. So there might be very different answers. I think the data-centric point of view for an ecosystem is the one that scales the best to use that word. If you, if you, does this answer your question? Perfect. Microphone two. Yes, thank you also for the idea with the citizen data. I forgot the name already. Courage T. Good name I heard. Yeah, I think just for this idea to get drive, I asked myself it's maybe not better for workshops or that maybe people were interested just get her in the front of the stage after the talk and question and answer. Yeah, that would be great. Please do that. Let's gather in front of this stage later until unless we get kicked out of here. I don't know about that. When do we get kicked out of this room? Not at all. So let's stay here after that. No, please really do this. Let's form a small circle and let's talk about the basics of this and maybe we'll meet again in the next few days. So microphone one. Hi, I'm interested in soil. You mentioned there's a problem with soil. And I'm 100% sure there's over 9,000 people who will send you soil samples, like maybe a million people who are just like farmers or gardeners or just eco enthusiasts who, if they know on what to test, they share it. It's like I'm 100% sure, but what's the problem with soil in your, what you tease it like, what's the problem? Okay, please let me ask you a question back. What in soil are you interested in? Just generally are you working on soil? Building humus, maybe sequestering carbon as a solution. I don't know if it works. I really don't know. And I don't know if the science is like, if there's a consensus in science that it's bullshit or it's not. No, no, no, I don't want to go that far. There's a big, I'm coming again from a microbiological point of view. And in microbial ecology, there is the question whether everything is everywhere, but the environment selects. So you could say that if you take similar ecosystems, all the bacteria that are in the one ecosystem would also be in the other ecosystems with some changes because of there is more salt or less salt and so on and so on. For lakes, this is still under discussion and it seems to not be the case for soil. It is absolutely not in discussion because every soil is a little bit different and they're very hard to compare. So what I said, when I said soils are problematic is that they are harder to compare soil samples than lake samples, water samples from freshwater would be in theory. We might figure out if we form this beautiful new group that lake ecosystems are also very hard to compare or we might figure a way out to compare soil ecosystems. So this is the hopeful answer of the day or the other answers were very bad, but this is one. We might really see that soil is easy and not hard. We're starting to run out of time but let's take those last two questions if we can make them quick. And then after that, if you want to come and meet the speaker, that would be great. So number two. Good morning. You said we have to do two things. First, leave half of the earth human free and the second half make sure that our impact at ecosystem is very low. So we cannot use this first half maybe for producing food, is that correct? So my question is, do you think this is possible with nearly at the moment 8 billion people? Think it, I don't know, I really don't know. I'm not that big, I don't know that much about agriculture to really say that. I hear different things, but most people tell me that it should be possible to buy the right mixture of different styles of farming and to buy some people having more intensive farming and some, most of the other people have less intensive farming and so on and so on. I don't know this in detail. Again, I think it must, it needs to be possible. It cannot be the case that we, I mean, what happens in parts of the world, as you know, is we just take a part of a rainforest which is the highest biodiversity of the whole earth, make it, kill everything that's on it and put in one single plant and kill. All the other plants again and again and kill some of the insects, or most of the insects. And this is, I don't have words for this. This cannot go on further and further. We need to find a way to reduce the amount of land we need for farming. Thank you. Also I think that population control cannot be ethically. It's very hard to argue for population control, direct population control ethically. I don't think that this can be directly part of the answer. So the very last question for number two. Hi, when we look at chaos theory where a system swings around different attractors and we add or remove a new attractor, this system will be out of control. We can't predict it with the old data. Will we be able to predict something with these data we collect right now when the whole system already changed? Yeah, let's say we have different samples and we have samples from ecosystems where all the parts are in place. And let's say we have samples from ecosystems where all of the parts are in place except for one. We took out one organism. And let's say we have ecosystems where all the parts are in place but two. And let's hope that we'll find something like this and we can say, okay, it seems like these are all the parts that are in play and we took out one and we took out two and if we put them back, maybe we can go back to the other samples quality. So with that, we need to wrap up the question. Sorry about that. Let's thank the speaker again. Give him a really warm applause for this great talk. Thank you.