 Good afternoon and welcome to Sharing Geoscience Online, this year's virtual annual meeting of the European Geosciences Union. This year we have more than 18,000 abstracts and have already had more than 19,000 unique users from around the world participating in our events. I'm Terry Cook, EGU's Head of Media Communications and Outreach, and I'll be hosting this week's press conferences, which include a question and answer period following the presentations by our four speakers. This is the first time we've ever tried completely remote press conferences, and so it's possible that we could experience some technical difficulties. If the platform suddenly quits during the middle of the session, I'll restart it and give you all about five minutes to rejoin the session. I however have been having some internet difficulties, and so if you're not able to log back in, then we will definitely finish recording and still post this on EGU's YouTube video. The transitions can also be a little bit slow, so I ask for your patience while we test this new greener way of holding Geoscience press briefings. Journalists, after the presentations are over and during the Q&A period, please only use the Q&A function in Zoom to ask your questions. Please don't use the chat and don't also use the hand raising function. We're only going to be using the Q&A. The documents and abstracts relating to these press conference are uploaded to the document section of the media website, which is media.egu.eu, and so please check there for more information. And I'm going to go ahead and introduce all four panelists now to make for faster transitions between them. The press conference is titled Epic Journeys, New Insights into Wildlife and Human Migrations. And our speakers today are Oliver Lamb, who's a postdoctoral researcher at the Department of Geological Sciences at the University of North Carolina at Chapel Hill, Philippe Gaspar from Mercator Ocean, Operational Oceanography, Eileen Eckmeyer, a professor in the Department of Geography at Ludwig Maximilians University Tate in Munich, and Lisa Thalheimer, who's a doctoral student at the Environmental Change Institute at the University of Oxford. I'm going to now hand things over to them, and then I will open up the floor for questions after the scientists have finished presenting. Can everyone see my screen? Yes. Great. All right, thank you Terry for the introduction, and thank you for inviting me to take part in this press conference. So today I'll be talking to you about this project that I'm involved with, which really is to try and test a range of geophysical instruments and see how well they work with monitoring of wild African elephants. This project is really led by Michael Shaw, but it also involves a help from Stephen Lee, Jonathan Lees, and Sean Hensman. So just to give you a bit of a background to this project and the motivation behind this project. So some of the motivation for, so we might know that wild elephants are going to do a lot of threat, and that includes some poaching for the ivory habitat loss, as well as human-elephant conflict where elephants might wonder onto agricultural areas and humans might want to drive them away or even kill them. So it's really important that we track wild elephant populations, make sure we know where they are and keep count of how many they are. Elephants are quite well known for their trumpeting noises, which are high frequency, but they also use a lot of low frequency noises, which we also call rumbles, the vocalizations degenerate a lot of acoustic waves. These rumbles are quite important for elephants because they can travel quite far up to three kilometers away from an individual elephant. The elephants are quite large as well, so they might also generate these footsteps that we can record through the ground with seismic waves. So both of these seismic and acoustic waves can offer a really nice and not intrusive solution for monitoring and tracking wild elephant populations. And by not intrusive, what I mean is that we don't have to go up to the elephants and tranquilize them and put tracking collars on these elephants. Therefore, we don't have to introduce any kind of stress to the elephants. So the idea is we just have to leave a sensor in the ground somewhere or in a tree somewhere and just wait for an elephant to come to these places and we can count these elephants and just leave these elephants be. So this is one of the sensors that we've been testing. This is one of the most interesting ones. This is actually a sensor. It's a low cost sensor that was originally designed for recording tectonic earthquakes of a slow acoustic. And I've got an example of what it looks like here. So this is kind of the size of one of these sensors here. This is a Raspberry Shake. This just has the Raspberry, this just has the seismic sensor inside it. What we've been testing is the Raspberry Shake boom, which also has the acoustic sensor. So this particular package here can actually record where it's coming to the ground as well as through the air. And it's really convenient and low cost. As far as I know, this is one of the first documented tests of this particular sensors package in the field for recording elephants. And what prompted us to try and test this was actually there's been a couple of experiments in the past couple of years where people have been using this package in the field. And I had realized that this is actually a really nice low cost sensor using the field that can complement other much higher cost sensors. So we wanted to see how well this would work for elephants. So we took five of these sensors down to an elephant reserve in South Africa, where they have a herd of about seven elephants, two males, three females, and two juveniles. And we were allowed to wander around this 300 hectare area. And what the idea was, we just wanted to record these elephants as they were wandering around this reserve. And we wanted to also record this reunion events. So what about me by reunion events is that the herd might separate into two groups. And when they come back together, they make an awful lot of noise when they come back together. So this might be because they're very happy to see each other again they're just bonding. Noises and also because they're running towards each other. And they make a lot of footsteps that we can record with our sensors. There's an example of what we recorded in the field with our sensors. So the top here I've got acoustic waves that have been recorded over a period of 120 seconds. And at the same time, it's the quarter by the seismic seismic sensor inside the wise be shaking boom. So it's very difficult to actually tell what is elephant and what isn't in these waves. So what we do is we take a frequency spectrogram. So it breaks it down to the various frequency components of the waveform. And what you can actually see here is some nice. Wiggles are coming through the frequency spectrogram. What we think is that these are the elephants as is the rumbling of the elephants individual elephants as they're coming back together. It's not so clear in the seismic waveform, but sometimes you can actually see the rumbling noises, the same rumbling noises appearing in the acoustic as in the seismic sensor. So what we think is that this is the acoustic sound, the acoustics rumblings are so loud, they're actually shaking the ground. And this can be picked up by the geophone sensor. And because we had five sensors, we were able to test how far away the sensors were able to record. And during this particular experiment, we were able to see these rumblings up to 400 meters away from the individual elephants. So this is just a successful recording footsteps. So this example here, each of these peaks here in the seismic sensor is and just over five seconds is what we think is a footstep individual elephants. And you can see these coming up nicely in the frequency spectrogram as well. These lines of frequencies. This is coming from an individual elephant that was within 50 meters of the sensor. This is my last slide. So basically, from what we find is that we think there's a great potential for using device be shaking booms for monitoring ground and populations. The idea is that we can just put one of these sensors in a place that we think elephants would congregate such as a watering hole. We can just let the sensor record data as it's coming in and then just let the sensor automatically count the elephants or just record or send this as an alert when there's an elephant heard in the area. We think we can improve the sensitivity range so we can improve the detection range of the sensors. So we're planning to do more field tests in the near future. We can South Africa and just convert tweak how we put the sensor in the ground to try and improve the sensitivity of the sensor. There are my contact details and you want to know more about this experiment and our plans with the sensor and as an article is detailing experiment is currently under review the biocoustics journal. Thank you very much. I think it's now phoenix turn. Yes, next will be for leap gas bar. This takes a moment to switch over the slides. Yeah. Okay, can you see on my screen. Yes, we can. Okay, so what I will present is the work of the team made of Maxim Lalir, a former colleague at CLS, who's working on sustainable fisheries, development of sustainable fisheries. Tony Candela and myself, who are working at Mercator ocean Mercator ocean is the company that operates the Copernicus marine service for the European Commission. And Tony's work is also supported by up well, US based non governmental, non governmental organization that deals with the protection of sea turtles when they're in the open ocean. Well, the first thing you should know is that we only have seven sea turtle species on hers. All of them are classified as vulnerable and endangered or critically endangered by the International Union for the conservation of nature. And if you only take the biggest and the most emblematic one the letterback turtle. You must know that the Northwest Atlantic population of this turtle has decreased by 60% between 2010 and 2017. If we're speaking of the Eastern and Western Pacific letterback population, they have decreased by 95% in the last three decades. So they're really endangered. And so many lots of efforts have been made to protect and monitor sea turtles on the nesting beaches that that's the place where they are most easily accessible and millions of euros have been used to protect those nesting beach so to protect nesting females and then the when they get out of the nest and go to the sea. But we also need to protect the sea turtles at sea because that's where they spend most of their time to spend over 99% of their time at sea. And they are highly migratory species. For example, turtles that are born letterback turtles that are born in the West Pacific in Papua, New Guinea. They cross the whole Pacific Ocean and come to forage off California. And when they are at sea, they can be threatened by different things that you can see on the slide like oil spills. They can also eat plastic bags and get their guts completely clogged and die from it. They can be entangled in fishing nets or hooked into other fishing gears. So we really need to protect them at sea, but to protect them, we know where they are. Knowing where sea turtles are in the ocean is not really or not too much of a problem for adults because as you see this figure where you see adult letterback equipped with a satellite tag on the back. Adults can be tracked mostly females from their nesting beaches. And then we can have data from satellite tracking as you can see on the figure. And you can see, for example, example of different letterback tracking from different nesting beaches in South America. And so for most or a large number of sea turtle population, we have a reasonably good idea of where adult sea turtles can be found and where we need to protect them. It's not unfortunately the case for hatchlings because hatchlings are much too small when they leave their nesting beach to be equipped with satellite tags. Well, there have been a few attempts, but very limited ones. So basically where the hatchlings go when they get into the water is not known. So hatchlings will get into the water and the next time we see them back is 10 years, 20 years, maybe 30 years later when females come back, adult females come back on the nesting beach. So this juvenile period spend at sea is usually called the lost years because we really don't know where they are. And we only have some ideas and the whole goal of our work has been to try to develop a model to simulate actually, to be able to simulate the dispersal of those hatchlings and then juveniles in the open ocean. Based on the very few knowledge or the very few information we have on their movements. And the thing we know about the hatchlings are really very few. Actually, well, we know that once they get at sea, they are advected by ocean currents. So they move with the ocean currents, they grow, and also they swim. So based on the data we have at Mercator Ocean, so we have ocean current data because it is our task to forecast and to simulate ocean currents. So we have daily forecast or daily simulation of the evolution of the ocean currents. So we know that the hatchlings move with those currents. And we also know that they swim and their swimming speed, we have to model it and that's the difficult part of the job. And so we make reasonable assumption based on also limited knowledge, but we assume that hatchlings will swim at velocity that is proportional to their side. So small hatchlings swim, small, larger hatchlings swim faster and their movements are directed towards what we call favorable habitats. And that's mostly areas where the water temperature is favorable and where food can be found. And if you take my figure here on the screen, you see where you see the Atlantic Ocean. And you see what I call the habitat suitability index that goes from zero that is white to one, which is almost black or dark brown. And you see, for example, here at the very beginning of the simulation, you see that the ocean is all dark brown below 20 north about and all white above it. Simply because what you see is the temperature limit. Basically, hatchlings can survive in water down to 24 degrees C, but when the temperature is below 24 degrees C, they cannot survive in those waters. So basically the limit between the brown and the white is the 24 degree isotherms. And as you will see in the animation, so what we do is that we release simulated sea turtles in the current simulated by the mercator model. And we let them drift with the currents and move according to the law we set up so all our turtles will move in order to remain in acceptable water temperatures and to find food, which is given by primary production, which is also simulated by the model. So let me just show you what the simulation will look like. So you will see little blue dots that are individual sea turtles that are living their nesting beach in French Guiana. And the currents will take them where you will see. And we'll also see the habitat is evolving and mostly we will see the seasonal cycle of the habitat. So you will see temperatures going up and down as the winter comes and then the summer. And the date is on the top of the figure. So you see basically there is one figure every 10 days. So this is typically a simulation of the evolution of the trip of these turtles. The black dots actually the dots turned black with the sea turtle died because they stayed too long in too cold temperatures. So it's called stunning or cold induced mortality. And you see here we already had in year five or six. So you see that after six years, seven years, you start to have turtles arriving along the coast of Europe and northern Africa. And then eventually entering in the Mediterranean Sea. So those simulations show us where those turtles disperse according to what we believe is their movement. Now as a summary, this is the summary of this whole simulation. This shows you all the places where the turtle have dispersed over a period of 18 years. And the color indicates the age of the turtles. So of course on to the west of the basin, you have young turtles and towards the east you have all the turtles. And the model predicts that you have a large density of all the juveniles in the Bay of Biscay. So along the coast of France, along the coast of Galicia in Spain, coast of Portugal. We also see that the turtles enter the Mediterranean Sea, but only the western part of the basin. And they concentrate also mostly in the Gulf of Gabes in Tunisia. And we have turtles also arriving offshore Mauritania. And so this map is the first map of the dispersal of where the back hatchlings from the northwest Atlantic are likely hard. I will not get into the data details, but data from from strandings and bycatch confirmed that actually that actually that the data where juvenile leather bags are found stranded or bycote are those where the model simulates are going. So namely in the Bay of Biscay along the coast of Portugal, in the Mediterranean Sea in particular in Portugal and also in Mauritania. One interesting thing that you can see is that basically what is incredible is that the turtles that are getting first to the eastern part of the Atlantic basin are the Mauritanian one. You can see there is a line of blue dots here. So young turtles going, taking a shortcut basically to Mauritania. And this is confirmed by data actually the smallest leather bag found along the coast of Europe and the Northern Africa has been found in Mauritania. And the model explains for the first time why it is so. So in the future we hope that those data will be used to inform fishermen of the areas where they are most likely found, where they will most likely find juvenile sea turtles. So that they can avoid these areas or take specific care of their fishing gear to avoid interactions with the juvenile turtles. And of course those simulations were made with historical data but we have the capability of running this model in real time. So with nowadays currents or even forecast for the next few days. So we can we can provide fishermen or we could provide fishermen with forecast of the distribution of turtles and hopefully this will help reduce bycatches. Thank you. Thanks so much for the Eileen will be next. And while she's sharing her slides, let me just mention that the photo of the hatchling that fleet showed is available on the documents page of the EU media website. Okay, can I start. Yes, please. Okay. Hi everybody. So firstly I would like to introduce our team. So my name is Eileen Ekmeier. I'm a professor for soil geography at the annual Munich. And my colleagues come from different fields. So Simon Kubler is a geologist, Akida Meier is a soil scientist PhD student, Steven Rossina is a botanist working at the National Museums of Kenya and Nairobi. So we are quite interdisciplinary team. And I hope this will also be presented now by me how these different fields come together to explain one of the biggest migrations which are happening in the world, which is also very famous, the great wildebeest migration. And the seasonal animal movement in the Savannah ecosystem brings together about one million wildebeest species migrating through the region in a regular pattern. So while potential climatic and biological drivers for this large scale migration, including seasonal rainfall patterns and vegetation dynamics and also the seasonal variations in diet and water requirements have been addressed before the role of rock chemistry and also soil diversity as a source and constrained for nutrient provision has not been studied before in greater detail. So our working hypothesis was based on results from the from previous project, which was based in the Rift Valley in Kenya by my colleagues, Simon Kubler et al. He was quite focused on the understanding of early hominin landscape in habitancy and animal migrations related to the distribution of soil nutrients in this region. So Simon and his colleagues found that topography and nutrient availability in the area control the movement of animals. Also in the past, and that the underlying rocks and sediments provided these essential nutrients due to their weathering. So that the presence of already polyolithic sites in this area was already related to the presence of grazing animals, despite of variations in climatic conditions over time. So in general, the Rift Valley is not a completely suitable place for grazing animals, but the nutrient rich hotspots promoted the establishment of grazing areas. The figure shows that the seasonal grazing habitats of wildebeest correlate with specific environmental subregions. They are marked with numbers one to three, which are also our study areas. So for example, in the south, we have higher precipitation, sorry, lower precipitation, where the wildebeest graze when there is more precipitation when there's the raining season. And then they move northwards and to regions which are drier and where we also have different soil conditions. So there are different in general concerning geology precipitation also external factors like soil erosion. In general, we could also see the geochemical variations together with continuous soil formation created a patchwork of soils with different nutrients that is, and that these soil characteristics are subsequently changed by climatic effects, for example, precipitation changes, but also by external factors, for example, over grazing and subsequent erosion of soils. These effects can also reduce the amount of nutrients and vegetation and also have an effect on the amount of plant nutrients and then also biomass available as fodder. So these later factors, the external factors definitely need more attention. So when we come to our results, you could see that the three areas were very different concerning geology. So to be more specific, so the southernmost area was dominated by volcanic rock or ash. So precipitation is rather low and we have lower rates of weathering and therefore also lower releases of nutrients from the rocks. We have higher influence of dust inputs and also a rather short grass vegetation. Here we have deeper organic rich volcanic soils which are high in plant available nutrients, especially due to the impact of the volcanic ashes and the dust. But we also have larger losses due to over grazing and subsequent wind erosion. In area two, the transition zone in the center is characterized by Akian basement rocks, for example, granite. We have here higher precipitation and also higher rate of weathering. This results in slightly less organic rich soils with lower levels in specific nutrients. But here we also have effects of loss due to soil degradation, specifically water erosion or sheet erosion. And in area three, where we have the highest amount of precipitation, we have a patchwork of Akian basement rocks and basaltic lava, but also thick colluvial deposits and other sediments. We have stronger weathering processes and therefore also leaching of nutrients. We have less organic rich soils, lower levels in nutrients, especially on the basement rocks and higher levels on the basaltic rocks. So putting everything together, we could see that the composition and degree of weathering of the different rocks and sediments influence the amounts of plant available nutrients and soils, which therefore control the grazing patterns of the herbivores. And we could also see that the connection between these factors is new. So I was not studied before, but that the interplay in between these factors, geology, climate, and then also grazing patterns is very important to understand also how these patterns could change in the future, especially when we consider climate change. So if you look at the specific areas of our project, what are the negative factors, what is reducing soil fertility is a question, which we will also study further in the future. So thank you very much. Thank you. And now we're going to switch from animal migrations to human migrations with Lisa Thalheimer. Okay, here we go. Hello everyone. Today I would like to talk to you a bit about a bit of our ongoing research on climate displacement, humanitarian needs and forecast based financing. This project is also highly interdisciplinary. We actually work with a team at the Red Cross Climate Center, as well as at IFRC, which is the International Federation of the Red Cross Red Crescent. This research is using very, I would say, unique methods from the realm of climate econometrics, so bringing climate data and methods and tools from econometrics, social science together. And yeah, this is a really great opportunity to show a bit what we have been doing and kind of the thought process of how we can use anticipatory action for climate related displacement. I thought I'd start this talk by a comic which I've seen quite often across Twitter, but which I also thought is very unique to our situation we find ourselves in at the moment. We have climate change going on. We have loads of conflict going on worldwide. The economy is going down as well as the ongoing pandemic COVID-19. And it looks a bit as if the disasters are collaborating better than we are. But we're trying to find a way with anticipatory action to kind of forecast what may or may not happen and act before a disaster happens. So setting the scene. What do we actually want with this talk. First, I think it's really important to dive a bit more into the methodology of what is actually climate displacement. What is forecast based financing or short FBF, as well as the applicability of forecast based financing in the context of climate displacement. Starting with figure one, we thought it's very useful to have to have a definition of what climate displacement actually is. And as you see in number two, the climate displacement is compelled to not voluntary. So this is one very important factor to distinguish migration, which is fairly voluntary from displacement. And you have you don't really have a lot of control over movements over time, but also the direction and mostly climate related displacement is happening in the context of extreme weather events, such as droughts or floods. And if we look on the figure a bit more to the left, we see refugee like events, which is the absolutely forced context essentially and number three migrant like events which I alluded to before, more voluntary, more time, more control over where people go or not. Then we see on the right side figure two. Here we would like to show you what's actually climate related displacement and what is conflict related displacement. Most of the of displacement as a terminology has been put in context of conflict. But we see over the past years from 2008 to 2019 that new displacements are actually much more occurring due to droughts due to floods due to heatwaves and other extreme events. And I think this is a very important distinguishment and also to put this in context of slow onset events such as climate change. But when we look at climate displacement itself. It has been very often put in the context of anthropogenic climate change, but not necessarily with the quantitative results supporting that. What I wanted to show you here on in figure three is that both temperature and precipitation anomalies have been going on over the past years. So here in particular over the month, January 2016 to December 2019. And we see in our data that climate displacement is very dynamic. We see that there are multiple factors coming in, but climate or extreme weather events being only one of these factors. So the question of attribution to climate change very often comes up. We see that increased climate related forced migration so displacement is often portrayed as a key impact of anthropogenic so human induced climate change. However, this kind of causal and quantitative evidence is rather rare or very context specific. So what we're really interested in is what to people in Somalia, which is a region within East Africa, which is fairly poor, but also hit by recurrent recurrent extreme weather events such as floods and droughts in particular. So what do these people actually need when they're displaced when they're arriving at different locations within Somalia. And we plotted here across the time of 2016 to 2019, the top five priority needs, and we see food, livelihood support, protection and water as the most important priority needs over time. And on the on the y-axis you see also the number of newly displaced people which is fairly large over over the years. So how does forecast based action forecast based action come into play here. So, first, I think it's important to kind of define this new terminology. And I try to do that within 20 words, which is, I guess, fairly difficult, but let's give it a go. Anticipation instead of reaction that's kind of the banner on this slide, but forecast based financing enables the access to humanitarian funding. Pre disaster so before a disaster actually hits a certain region. And this is based on both climate and weather forecast and is combined with risk analysis. So we know essentially with these forecasts where an extreme event will hit and what kind of communities could be affected by the type of extreme weather event. And there are obviously differences in the timing and the onset of extreme weather events. Droughts may shape on a much longer scale compared to floods which are very, you have to react very quickly and other priority needs would be provided essentially. And you see on the on the right side. A graph which shows the 18 different regions within Somalia and food as a priority need and here it's it's interesting to see kind of as a takeaway that people get displaced within the countries, but also within urban areas so people flee people get displaced to a certain area and really need food. So kind of going back to the slides, the slide from earlier, we really see that food as a priority need or the biggest priority need. And a couple of other takeaways from our ongoing research that we kind of know the type of priority need within the different regions. And this knowledge then can be used in order to transform the past knowledge into forecasts. And I think that's also very important during during the ongoing pandemic, because we can hopefully provide a couple of very urgent responses and kind of see where which types of priority needs are needed and should be provided across Somalia. And I think I'll leave it at that and thank you for your attention. Thank you very much. I'll give you just a second to absorb that screen. And now I'd like to open it up to questions and we already have a couple. The first ones are for Oliver. And the question is, when did your South African study take place. It took place last October. Just a five day deployment in South Africa last October. And then another question for you. What other novel applications have made use of the raspberry shake and boom, animal related or otherwise. Okay, so there have been a couple of studies already. That I alluded to during my talk. So the first of those was a study where they used not the raspberry shake and boom, but it's a similar device to one that I was holding up the raspberry shake, where they were looking at rock falls in the French Alps with the Swiss Alps. And that was quite successful. And then there was also a study that was done by the United States Geological Survey, the USGS, where they tested these to look at earthquakes in Oklahoma. And actually, right now, I'm running a test on the raspberry shake and boom sensors on a volcano in Chile. I deployed those in January. And I was hoping to pick them up last month, but obviously I couldn't go back down and pick them up because of the pandemic. So hopefully I can go back down and get that data and see how well they work on an active volcano in Chile. That's as far as that there's the only ones that I think know of so far. I think there are a few of the tests that people are looking at around the world. I think there's one in the UK where they're making that small earthquakes in the south of England. That was done by Steve Hicks. And there might be some other ones that are in the works but those are the only ones that I know of. Great. Thank you very much. Question for Philippe now. As higher primary productivity in the ocean is frequently associated with colder waters, less than 24 degrees Celsius. How do you resolve the swimming behavior of little turtles between feeding and swimming toward warmer waters? Yes, thank you. That's an excellent question that will allow me to get into some more details of my habitable. Can I share my screen? Sure, go ahead. I just want to go back to the slide with the animation. Okay, so basically the question is how do we resolve this problem? First of all, I must say that the temperature of 24 degrees I mentioned that that's the minimum temperature in which little turtles and little turtles and little turtles are surviving. Actually, the thermal preferendums or the range of temperatures that these ectothermic animals can sustain goes down as the animal is growing bigger. Actually, small little bugs have a small tolerance to cold temperatures, but as they're growing bigger and older, they become more sturdy. And actually, the adult little bugs are the sea turtles that can sustain the lowest temperatures. Some individuals have been seen, adults have been seen of Newfoundland in nearly freezing waters. So basically in the model, we've factored this in and we have an eat budget of the individual. And based on this, we adapt the minimum temperature. So as the individual grows, the minimum temperature goes down. Also, as the individual grows, the amount of food needed for an individual goes up. So basically that's what you see in the animation. Basically at the beginning, all the warm enough temperature, the warm enough zone is all brown. And as the time will go on, you will see that the limit of the acceptable area goes up. That means towards higher latitudes, so towards colder water. And let me go on with the animation. I will speak as it goes up. First of all, you see the seasonal cycle, but you see also that the center of the gyre is getting less favorable because individuals are growing and they get less food in there. But basically what we see is that they concentrate in, let me see, I will try to stop at the favorable time. Okay, here along this 14-horse area, we have basically the temperature, we have an area, a tongue of favorable area where the temperature is okay and when there is a lot of food. So individuals will tend to concentrate on this area and when you will see when I will let it go is that actually this favorable area goes up and down, inducing a seasonal migration that has been actually observed in other species. But basically the story is that in those cold places, the animals will go up during summer to reach the areas where, colder areas where there is more food, but when the water temperature is warm enough for them. And they will come back towards south during winter to get into warmer temperature, but they will accept less food. And so most of the story when they cross the Atlantic is that they cross the Atlantic doing seasonal migration, always finding the trade-off between temperature and food. So during winter they will go down and accept less food and during summer they will go for food and so they will go up north. Other areas in tropical areas like the off Mauritania, you don't have the seasonal thing because the area is permanently favorable because the water temperature remains acceptable all the time. And there is permanent primary production because of permanent upwelling there. So you can see that sea turtles indeed do a seasonal cycle as a trade-off between temperature and the food. And in areas where the habitat is only good because the food remains there and there is enough temperature, they remain there. So indeed there is a permanent trade-off between looking for food and looking for good temperature. And that's why they're moving. Thank you so much. The next question is for Lisa. And two questions actually. What type of move registers as a displacement and how do you distinguish between climate-related moves and the wider global trend of urbanization? I think this is a very relevant question. So in general, what registers as displacement? I think I'll start with that one first. So anything which is picked up by surveys as forced migration is, for instance, officially registered as displacement. Displacement is also mainly internal. That means within countries, within different regions of a country. So cross-border movements or international migration may be forced or not. We don't see that very often. Mainly people get displaced on a forcefully based and kind of go to the nearest and closest shelter or safe haven that they can go to. We don't really see that they are long-term. However, when we look at internal displacement camps, we see various different dynamic structures of why people cannot leave so-called IDP camps. And also in terms of extreme weather events, the data we have on Somalia actually distinguishes between those who are displaced due to droughts, due to flood and conflict. And we can kind of disentangle these different displacement flows with underlining conflict data or climate and extreme weather events related data to prove kind of the drought or the flood link. And then how do we distinguish between climate-related movements and the kind of global trend of urbanization is also a very relevant question. Because mainly we see that people, especially in the context of drought, we see kind of, first of all, a very limited displacement at the onset of a drought. And people kind of try to adapt in other ways. But the first migration response would be, for instance, young adults who voluntarily search, are in search of new wages, go to urban areas, try to mainly leave their agricultural life and try to have a new life and more money, essentially, in urban areas. And this is often also in nearby urban areas. And as the drought conditions persist, we kind of see outmigration from those areas, which are affected most, but again, mainly internal within the country. So I would say that climate-related displacement actually contributes to the urbanization trend. Thank you very much. The next question is for Oliver. And that is, can you talk a little more about the sorts of things you could conceivably study with these signals? So what aspects of behavior, for example? Yes, absolutely. There's a couple of really, there's two important things that ecologists are quite interested with elephants, that we could study with the signals. The first of those is the movement patterns of the elephants and the herds in particular areas such as national parks. You know, we want to know exactly where exactly the herds are moving, why they're moving from place to place. Are they moving in response to something or are they just moving out of their own free will? So that's one of the most important things. The second important thing is, you want to understand the communication mechanics. So how exactly are the elephants talking to each other? How far away can they talk to each other from each other? So some people think that they can speak to each other using these rumblings as far away as 16 kilometers, but more recent studies may be that they actually may be three kilometers away from each other. So how did the herds talk to each other? So are the herds responding to each other? Are they wanting each other of danger? So it's these communication patterns, it's the other really important thing that we want to understand with these signals. So the movement patterns and communications, those are the two important things that we wanted these signals for. Okay. The next question is also for you Oliver. It seems that the raspberry shaken boom picks up just the rumblings of the elephants. What about footfalls? Is the device able to pick up that? Yes, absolutely. So there was a slide I had in my presentation. I'll just share my screen to show you that slide again. So this particular slide. This particular slide here, they actually showed you that we did record footfalls with the elephants here. So each of these peaks here we think is an individual foot hitting the ground and being picked up by the raspberry shake sensor. The elephant was walking past the sensor during the deployment of South Africa. So yes, we can pick up the elephant rumbling and the footsteps. The question is that we're actually disappointed with the range of the sensor. So this particular sensor only picked it up within 15 meters of the sensor, but there are other studies where they picked it up as far away as a kilometer, maybe a couple of kilometers away. So what we want to do in the future is can we improve the sensitivity of the raspberry shaken boom sensor? Can we change the way we put the sensor in the ground so we can improve the range of that sensor? Thank you so much. The next question is for Philippe. And that is, why don't the turtles spread to the eastern Mediterranean, where water temperatures can be higher than in the western Mediterranean? Yes, good question too. The fact is that, as I explained from previous questions, places where the total go depends on both the temperature and the amount of food available. And I suspect that if they don't go to the eastern Mediterranean sea, it's because they don't find enough food. I mean, it's less productive area than the western red. We haven't looked into the details of that question because we were so happy with the results because the distribution of the turtles that we get in the Mediterranean sea matches almost perfectly data from Italian colleague Paolo Casale, who just actually synthesized all trending data for leatherbacks in the Mediterranean sea. And it is clear that actually the leatherbacks are very rare in the eastern Mediterranean sea. So the model of fits the data, the exact reason we haven't looked into it, but I believe that's because there is less food available in the eastern part of the Mediterranean sea. There are other species, smaller species of turtles present in the eastern Mediterranean sea, a lot of red turtles and green turtles are present there, but I suspect they have a smaller food request. Okay. And then I have another question also for you, Philippe, and that is, how do you deal with the energy consumption of the turtles? So for example, turtles getting tired as they swim? Yeah, that's something that is not yet factored in the model, but that's a very pertinent question. Actually, that's the next improvement of the model that we are planning. Actually, we want to include a model, an equation that will control what we call the fitness of the animal so that it will take into account its energy budget. The energy budget will control not only the fact that the turtle may get tired and swim more slowly, as Michael suggests, but we believe also that the fitness will control the growth of the animal. Basically, we have reptiles, so they have relatively low energy demands, their basal metabolism is low, and so they can adapt to harsh conditions. But anyway, this is something that we need to do and that we will do. The idea is that we have all the elements in the model to take into account and to build an energy budget of the animal that can control not only the swimming speed, but also maybe more importantly, the growth of the animal, and maybe it's mortality too. Okay, and there's one more question for Oliver now. And that is there is some suggestion that we could find in for a sound signals that elephants use as an alarm call and then use those same signals to keep the animals away from populated areas so that they don't come into conflict with humans. Do you have any thoughts on this. Yeah, so I'm really glad you asked this is because that's exactly one of the points of the project we're working on. So what are the idea. So just to start this question, so if the sound that Jonathan is talking about. If the sound is basically the same as the sound that we listen to, but it's below the threshold of human hearing, which is 20 Hertz. So 20 Hertz and below, we can't hear it, but the elephants have been found to generate that kind of noise. So about 17 Hertz is what elephants mostly generate, but sometimes they might be hints of a little bit lower than that. If the sound is quite interesting because you can actually travel quite a lot further away than the sounds that we can hear. So it's quite it's a useful elephants that are trying to communicate to each other for long distances. So Jonathan is quite right that they can use his alarm call. And one of the ideas that we're hoping to do with the vice be shaken boom sensor is develop some kind of system that we can. We can deploy in national parks or in places where elephants might wonder through human through human settlements, and we can develop some kind of alarm system so the vice be shaken boom picks up what they think is an elephant signal. You can send out an alert to the people in that area and say, okay, elephants in the area. You might want to just drive them away peacefully. And for example like that. So yes, this that sort of one of the things we are very interested in that we're hoping to develop in the in the near future with the vice be shaken boom. Thank you. Are there any additional questions. I'll give people just a minute to type in case they have anything else. If there's no more questions then we'll finish here. Thank you all for joining us. And just to let you know that you will present one additional press conference. And that's going to be a journey to the center of the earth, which is loosely based on Jules Verne's classic novel. And that will start in about 20 minutes. And you do need a different link to join that press conference and you can find that on the media website. Thank you all so much. Appreciate your time.