PETER WREGE: Thanks for coming, everybody. And I'm going to be very informal and just sit here. I don't want to stand up. And my intention really is not to take too much of our time to kind of give you an overview of what the Elephant Project's about and some of the kinds of things that we're doing. And then hopefully, just-- we can talk. Ask questions about what you guys are most interested in, and I can do my best to give you my take on that.
As Syl said, so, Elephant Listening Project is actually part of the bioacoustics research program housed at the Cornell Lab of Ornithology. Financially, we're really on our own. We have a spot to sit at the lab, but I'm responsible for trying to drum up funding for doing all of our research work.
We're surprisingly small. There are only two real staff members, me and one half-time assistant. But we're really fortunate that we've had up to-- well, often, as many as 10 to 15 volunteers, ranging from high school kids in Ithaca, mostly Cornell undergrads, some retired people who just have a huge enthusiasm for elephants and come in and help us with the analysis. And we couldn't possibly get through the kinds of material we get through without those people being involved.
So, the Elephant Listening Project really started in 2000 by Katy Payne. This is the woman who discovered that elephant vocalizations include infrasound, which is frequencies of sound that humans can't hear. And her real insight was that we might be able to use-- basically eavesdrop on the vocalizations of elephants in order to study them.
And specifically, she got interested in how we might use that technology for forest elephants, which live in, for the most part, enclosed canopy rainforest. Very difficult to see. Very difficult to study.
So in essence, the Elephant Listening Project uses recording devices that we put up in lots of different sorts of places in central Africa. They are autonomous recorders. So they just run for long, long periods of time. Often, we have them up for about six months with no one around. And then we go back and collect the data from those to see what we can find out. And we basically use those kinds of recordings both to count how many elephants are in an area, to some degree we can tell what they're doing based on the kinds of vocalizations that we're hearing, and we're beginning to make some progress on determining to some degree the age and sex of the individuals that we're recording.
SPEAKER 1: You can track an individual?
PETER WREGE: No, we can't. From the behavioral observations of Andrea Turkalo, who's-- I'll talk a bit more about her. She weaves in and out. She's the expert who's studied them now for 22 years at one clearing in the Central African Republic.
It's very clear that elephants recognize each other, and from distances where they're not visible. But it's very, very difficult, actually, for researchers to figure out, what is it that's distinctive about a given elephant that would tell you that it's this individual. And gathering the data to do that is also very, very hard. Because you'd need multiple recordings of each individual that you can then statistically figure out, OK, this is what makes this one different than that one. It would be fantastic if we could say, oh, you know, Ralph was here. But not yet.
SPEAKER 1: Can you tell--
SPEAKER 2: Can't you tell--
SPEAKER 1: Oh, sorry.
SPEAKER 2: --male, female?
PETER WREGE: To some degree. We need to do some more work on that. But basically, the fundamental frequencies, the lowest frequency that the vocalization is built on, is dependent to some degree on body size. And males have-- testosterone thickens the vocal cords, which lowers the frequency of their vocalizations.
So there is a range. But they overlap a lot. So, we're getting there. But we're not there yet.
And we've also found, from Andrea's work and our studies at the Dzanga clearing, that males actually don't vocalize as much-- much, much less than females. So a lot of what we're recording, actually, are females talking to their babies and talking to each other. And males are sort of hanging out there, being quiet and whatever.
So, forest elephants are a unique species. Their ancestral range, which is shown in this map in green, included basically almost all the rainforest habitat in Central Africa and West Africa. Probably on the order of a million elephants in that early maximum population or steady state population size. By 1993, which is shown here in yellow, the estimate was that about half that number were still alive, 500,000 or so. And now, restricted mostly just to the Congo River Basin and a couple of other big river drainages in Central Africa.
And the feeling was that about 30% of that 500,000 were in Gabon, on the coast here. And 60% or so were probably in the Democratic Republic of Congo. So then, by 2002, we were basically down to about only half of that range still left, shown in orange here. And the most recent studies suggest that we're now down to about 10% of that original population.
So we think there are only about 100,000 forest elephants left. They've basically been extirpated from Democratic Republic of Congo. They're left only in the Salonga National Park, which is that a little bit here. And this is something that is written through everything that I'm-- you know, everything that we're experiencing is that political instability is not a favorable environment for any of these important species of wildlife for survival. When there is no rule of law, there is also nobody to be stopping any of this poaching going on.
SPEAKER 3: When you said 10%, is that-- you're down to-- there's 50,000 left? Am I doing the math right?
PETER WREGE: 10% of the original million.
SPEAKER 3: Oh.
PETER WREGE: So, we're down around-- we think there are about 100,000.
SPEAKER 3: OK.
PETER WREGE: And for a lot of reasons, we don't actually know how much mortality there is from poaching in forest elephants. Because it's very hard to actually find out where the poaching is occurring. But overall, the guess is that about a hundred elephants are killed a day. And a large proportion of those are forest elephants. Because protection is bad, but it's still way better in East and Southern Africa than it is in Central Africa.
SPEAKER 3: Do some migrate out or is that pretty much where the elephants are?
PETER WREGE: Pretty much where they are. The genetic evidence shows that there's some introgression of savanna genes into the forest elephant population, but not the other way around. And the thought is that forest elephants are smaller, stature-wise, than savanna elephants. And the thought is that basically it's one-way direction because a male forest elephant can't compete with a male savanna elephant for mating opportunities. But the other way it can work, where there's a meeting of the two species.
SPEAKER 4: So where do you have the microphones? Where do you have the listening devices? In which areas?
PETER WREGE: We put these up in trees in the forest or at forest clearings where the elephants come.
SPEAKER 4: Yeah, but is it only in this National Park? Or--
PETER WREGE: No. We're working in forestry concessions and in national parks. And--
SPEAKER 4: So how many locations overall?
PETER WREGE: I think right now, we're monitoring about nine different kind of regions. In some cases-- the project I just came back from, I put 23 of these units up in one area in Republic of Congo where we're trying to estimate the population size and compare it to other methods of estimating the population.
SPEAKER 4: So, nine regions in how many countries?
PETER WREGE: Four, right now. Yep. So one of the key messages that I'd like to get out through you guys, to the degree that this piques your interest, is that first, that there actually is a third species of elephant. Forest elephants are distinct genetically and ecologically from the others. And I think most of the news things that we hear, you know, it's, well, elephants, elephants, elephants. And everything about the conservation and the pressures on savanna elephants versus forest elephants are very different because of the habitats they're in and the political structures in the countries where they are.
SPEAKER 2: This may be a naive question, but if they can interbreed with the savanna elephants, why are they considered a separate species?
PETER WREGE: Well, because they're genetically sufficiently distinct. So there's some hybridization where they meet. But that's a fairly limited area. And again, the introgression is only one way. It's coming from savanna into forest elephants. But there's actually relatively little genetic overlap in that sense. And, you know, the whole species concept-- we shouldn't get into it, but I mean, it's complicated.
And there are many species which do hybridize in certain places and yet we consider them complete species. You know, lions and tigers can breed. But generally, they don't. Because they don't overlap enough. So the evidence now is pretty overwhelming that genetically, they're sufficiently distinct to be considered a different species.
SPEAKER 7: [INAUDIBLE] produce fertile offspring?
PETER WREGE: As far as we know, yeah. I mean, again, this introgression of savanna genes into some parts of the forest elephant population suggests that it is successful.
SPEAKER 4: What are the--
SPEAKER 3: Oh. Sorry. Go ahead.
SPEAKER 4: What are the differences in behavior and vocalization between--
PETER WREGE: The major differences are in their ecology. So one species is in rainforests. The forest elephant has a diet that's very rich in fruits and tree bark. They eat relatively little grassy vegetation, which is typical of the savanna elephants.
We know very little about them, which is the other point here, that basically, forest elephants are surprisingly unknown. We don't really know what their social system is. We think it's similar to savanna elephants, with what we call fission-fusion, where you have kind of extended family groups which sometimes break down into smaller subdivisions. But in fact, we don't really-- haven't pinned it down.
So those are a couple of the main differences. Physically, they're smaller. A little bit different shape of their ears. The tusks tend to be a bit different in how thick they are and how curved they are.
So the second key message really has to do with, although forest elephants seem to be bearing the brunt of the ivory pressure right now, they're also the species that has the best shot at continuing to live kind of a normal evolved elephant lifestyle of any elephant left in the world. Because there still is a lot of rainforest habitat in Central Africa. And they still are basically not-- they're unconstrained. The national parks really are just chunks of land in a matrix of other rainforest.
So forest elephants are still able to really move around as their evolutionary pattern of ecology has-- how they've evolved. More and more savanna elephants are basically in really huge managed parks. More and more of the parks in Africa are being fenced. And that's causing problems in East and Southern-- Southern Africa particularly. Where they're being constrained and so the populations are growing and overpopulating areas.
SPEAKER 3: Is the fence to preserve them by the government? Or do they fence them so they don't destroy things? Like, what's--
PETER WREGE: I think they're mostly fenced so they don't get out and destroy crops. I mean, I'm sure there are lots of different reasons for some of the fencing. But it's also, I think that just the human population growth in proximity to some of the reserves and parks has necessitated some sort of barriers to reduce that human-elephant conflict.
SPEAKER 3: Now, do they have laws there at all, or is it international? For, like, poaching? And the second question I have is--
PETER WREGE: I didn't hear the--
SPEAKER 3: If there are laws. I know in the Arctic, you can't bring Ivory back, or whatever. Number one. But number two, when you go in, then, are there safety problems for--
PETER WREGE: I'll answer the second one first. Where I've been is politically fairly benign. I haven't felt in danger in any of the countries that I've been working in. We rarely run into hunters in the forest. I've never actually run into a real poaching group who might be more aggressive. That has happened, to some degree, in some places.
And really, in answer to your first question, most of these countries, killing elephants is illegal period. But there isn't any enforcement. There is very, very little governmental infrastructure to deal with it.
And part of the problem is that this very, very extensive rainforest is just a really, really hard place to do anything. You can't use planes. You can't use drones to kind of find where the problems are. Because it's this rainforest canopy. And you just can't see what's happening down inside there.
SPEAKER 2: I'm just a little puzzled by your optimism. I mean, given the fact that the forest elephant numbers are going down much quicker than the savanna elephants and also given the enormous political instability in all of those countries, except maybe Gabon, where's the hope?
PETER WREGE: Well, I'm not sure that I would say that I was optimistic. I think that of any of the elephants, they have the best chance, if something happens to stop the poaching, if some of the governments can do something about it and start actually enforcing the laws and catching and putting poachers in jail. As I was talking to Syl coming down here from Ithaca today, it's-- at one level, working on elephants anywhere and working on forest elephants, you would-- if you really look at it purely straight up-- you would stop bothering. The hurdles are huge.
But at the same time, we don't really know what could change in any of these systems in a year, in five years, that might make the difference. And so we kind of keep going and try to learn what we can, and do what we can, and hope that things change. So there's a potential there. I don't know how much hope there really is.
SPEAKER 2: Is the situation in Gabon a lot better than in the Central African Republic and the Congo?
PETER WREGE: Well, it certainly is way better than the DRC. That, basically elephants are gone. That's a hopeless case. There's still good forest and some animals there. But politically, I don't see what's going to happen.
Central African Republic right now is a mess. And it's unclear-- if that remains unstable for long enough, I think we'll lose a lot in the Central African Republic. The area there-- I think I'll show you in another minute or so-- the area of forest there is very much in the very extreme southern part of Central African Republic. And that's the least impacted right at the moment.
Gabon is probably the most stable. Republic of Congo, where I just was, is quite stable politically. Gabon, I think, in the end, will likely be the core of whatever we have for forest elephants. Because the population size is very small-- human population size is small in Gabon compared to any other African country. So the potential is there. But there are lots of hurdles.
SPEAKER 3: Sounds like from your funding, that-- you know, it's so dependent on your funding. So I don't know how much you get, you know, corporations, non-profits, whatever. But anywhere I've lived, the public goes nuts if a new baby elephant's born at a zoo. But they probably are not your donors, right? I mean--
PETER WREGE: No. Well, or I haven't found them yet, anyway.
SPEAKER 3: So I'm just wondering what that dilemma is.
PETER WREGE: Well, the fundraising for us specifically-- it's really difficult. And part of that is, as you say, we have a charismatic species that resonates with most people. And so, if you can reach out and find the right set of people, I think we could bring in money.
Part of the problem is that because we're so small-- it's sort of a catch-22. I'm in Africa a lot because I'm the only person here in our program to go do the field work and analyze the data. And so there isn't enough time to come and do things like this, travel elsewhere around the country trying to find that one or two people who say, this is what I want to help have happen.
SPEAKER 3: Do you compete in, with any, like, the whale program, or birds?
PETER WREGE: Yeah, and-- yeah. And, I mean, that is an issue to some degree at the Lab of Ornithology. Because the people who are attracted to the lab tend to be people interested in birds. And so they end up putting their money into programs in the lab that have to do with birds. And it just has been difficult.
But I see it as, largely, it's my fault. Because I haven't been out getting to the right places. But, anyway, I think you'll see to some degree that the situation is such that I also feel like we've got to do what we can for the conservation stuff, or there won't be any elephants to work with anyway.
SPEAKER 3: I don't mean to dominate the questioning, but you triggered the fact-- when you were talking about voices, I'm thinking of human males. So is there money, even, medically? You know? Is your study ever applicable for them?
PETER WREGE: Probably not very easily. Because we really can't do any kind of experimental stuff. So it's correlational. But I certainly haven't thought of ways that I could tap into NIH or something--
SPEAKER 3: Yeah, yeah.
PETER WREGE: --for the kind of stuff we're doing, where there is big money.
SPEAKER 3: I wondered, you know, like how they do plant science for medicine, or whatever.
PETER WREGE: I mean, one place that would have money, which I just haven't had time to deal with, would be NSF. Because you can get reasonably large grants from NSF. But those grants, I've written many of those when I was a bird guy in previous lifetimes. And they're extremely difficult grants to write. And you need to have lots of time to be thinking about your theoretical-- you know, what's the underpinning, why is this important for science period, not just for-- they don't fund conservation. It's real basic science.
SPEAKER 3: Because there's a Binghamton professor who said, maybe NSF is 3% of the applications.
PETER WREGE: Yes. It's very hard to get the money and a very hard process to write those grants. So it is an avenue that's potential-- to have a post-doc or something. But it's just not one that I've had time for.
SPEAKER 1: Any thoughts about crowdsourcing the audio analysis?
PETER WREGE: We have thought about some of that. And why don't we come back to that, so I can get through the little introduction here and then I don't have to keep looking at my screen here. So, just by way of introduction a little bit, I just have a short series of video clips to kind of show you what these animals look like and where they're active.
This is a particular clearing in Central African Republic. It's the only place you can really directly observe forest elephants. We basically can't work with them very well directly in the rainforest. But they really don't spend a huge amount of their time here. And not all landscapes where forest elephants live have these forest clearings.
I don't know whether we can hear-- we should be able to hear this guy g a bit. This is a baby who's lost his mom.
And I'm going to turn the lights off a little bit because the next pictures are going to be hard for you to see. Whoops. Row one. But this ought to be good fundraising stuff, you know? These baby elephants are great.
SPEAKER 5: Did he just find her there? Is that--
PETER WREGE: Yes. That was the mom. But in fact, 90% of the lives of these animals are spent in the rainforest, walking on trails that they make, like this.
So these are video camera traps that some colleagues from the Max Planck Institute were using in one of my study areas. And they let me use some of their footage. And given how little we know, even this kind of thing has been really important. Because we can see what sort of group sizes are actually walking in the forest; are males ever with a group of females and babies; lots of kinds of things that we can tease apart from this kind of information.
There were actually some rumbles by these elephants that you might not have been able to hear very well because they weren't recorded very loudly. But I'll play some of those in a minute.
SPEAKER 5: They're not nocturnal, but they--
PETER WREGE: They're active at all times of day. Yeah. [LAUGHTER] Stuck on a log. Mom!
So, what makes acoustic recording actually a useful way to study these animals is that they have these low frequency vocalizations. This is a spectrogram or a visual representation of a rumble, which is kind of the primary type of vocalizations elephants use in their communication. And essentially, all this is trying to show is that as the frequency of a sound goes higher, to higher pitch, those sounds run into more interference and attenuation going through any environment. So they bounce off of trees and obstructions much more readily than lower frequencies, which can travel through a habitat, even a forest habitat, with much less attenuation.
So the underpinnings of all of our interest in the infrasound of elephants is that this very low frequency stuff may be allowing elephants to communicate over miles' distance, even in this closed habitat. But it also means that we can record those low frequencies with our acoustic equipment, allowing us to listen to pretty large areas of forest, even with a single recorder. So this is what this rumble sounds like.
So a huge advantage of the acoustic monitoring is that we can basically listen to forest elephants anywhere in an area of up to three square kilometers. And it allows us to listen in areas of closed forests which would be difficult to look at otherwise. This is what a lot of Gabon looks like, and a lot of northern Republic of Congo. Huge, huge, forest.
But you can imagine that it's very difficult both to study an animal, any species, in this kind of habitat. It's equally difficult to find poachers, places where poaching has been happening, when everything's hidden under this canopy of mostly trackless forests. There are very few roads or trails. That's changing. But right now, there still is a lot of this that's like that.
So although poaching is really out of hand, overall, and particularly with forest elephants, this satellite map that I pulled from Google Earth shows that-- this is recent. I just did this a few days ago. This satellite imagery shows that we've got a lot of rainforest still in that Central African region.
Gabon in particular, which has one of the lowest human population densities in Africa, could easily provide a sustainable core for forest elephant populations. But as we were beginning to talk about, there a lot of hurdles in the way for that to actually happen. Gabon, as an example-- I've worked mostly in Gabon-- they recently, in 2002, designated quite a few large areas as national parks. But even 12 years later, the protection in most of these is minimal or nonexistent. Parks were declared but there wasn't even a parks department or ministry in the government. So it's been very, very slow coming up to speed for any kind of real protection.
And a huge challenge is that much of this forest is already designated as forestry concessions or mineral extraction concessions. So these pale things are all lessees or whatever that's called, concessions, that have been sold by the government to various external, mostly international, companies. Most of these are still not active. Some of them are. But many of them have been grabbed up, but the forest is there. The companies haven't actually started working in them.
But in countries like Gabon and all of the Central African countries, natural resources is one of the only sources of income for those countries, for their development. And so this raises a really huge dilemma. These countries are going to exploit these resources. So how can we work with them, with concessions, with forestry companies, somehow, that they can still get some of the income from this but not necessarily destroy this for other animals, and particularly for elephants?
One of the other issues that I'm concerned about, that we've seen here in America, is-- you look at these national parks and you take away all of the forest that's in between them, and you have these islands, which we know now in the US often can't support large animals that range far. So now, in the US, we're spending huge amounts of money trying to make corridors to get gene flow to allow animals to move out of these national parks that are protected through areas that have now been developed. And so I think one of the key things that we're trying to talk about now is, how do we, right now, begin to establish some sorts of corridors, some sorts of areas that would connect these protected areas, so that animals like elephants could move? And more and more, looming in the-- the issue has to do with climate change. And how is that going to change whether some of these areas are going to remain suitable for various really important species?
And I think that to achieve conservation goals, we really face some daunting moral questions, and often, really the problem of high government officials that are really part of the problem, in terms of exploiting ivory, bushmeat, that kind of thing. Most of the people in these countries are desperately poor. And basically, as far as I'm concerned, they have no quality of life. And there is an insistent issue here that these people need to have their standard of living raised so that they have some health care-- reasonable, even minimal health care-- and education, et cetera.
But there are a lot of people-- when you take all of Africa-- take that across the world. Where are the resources going to come from to bring up those people's standard of living? It means more natural resources that have to be exploited. Where do we-- how do we do that?
And the people that I talk to, where they're struggling even to feed their family, conservation isn't even on the radar. They're basically saying, look, don't tell me not to go kill these things. How do I feed my family if I don't do this? So there are huge, huge problems.
I don't know why this is cut off there, but-- so to kind of wrap up here a little bit, how are we using acoustic tools to accomplish something in this area? And we really have two areas of focus-- conservation and basic biology. Of course, a good understanding of basic biology is really critical for good conservation. But the basic biology can also be time consuming to really pin down.
So, the crisis in poaching has really pushed the Elephant Listening Project-- at least my focus-- in the conservation direction. So we've very much been trying to figure out, how can we get information to conservation NGOs, to protected area managers? How can we come up with new tools to help them figure out, where are the problems, where's the poaching happening, where are the resources that elephants need the most, so we can put minimal resources into protecting those sorts of areas.
For example, along with elephant vocalizations, of course, our recorders record gunshots and other kinds of things happening in the forest. This is a small part of a really horrible poaching incident that happened in a clearing that we were monitoring. No humans were monitoring that clearing. It was just one of our recorders.
And I'm going to play this little bit of gunshots and stuff. It sounds a bit strange to you because we record at a very low sampling rate for saving space and stuff on the data cards. And so there are no high frequencies. So it sounds a little bit echoey or kind of funny. But I think you'll get the--
These are elephants screaming after being shot.
So, kind of a horrible thing to have to listen to. But these sorts of data are actually a really powerful wake-up call to protected area managers. If we could bring in data that says, you think you're protecting your area, but we're picking up gunshots here, at this kind of level, and this has changed from last year-- it's now twice as much, that kind of data can be very hard for a manager to ignore. And you say, if we get out reports about our results and that goes to government and to NGOs, it's a way of putting more pressure and saying, you need to be doing something about this.
So, and acoustic recordings also give us an actual measure of what kind of hunting pressure there is. Mostly, the way that it's approached is to look at markets. You know, how much bushmeat is in a market locally, interviewing hunters. With all of the attendant problems with misinformation or whatever that you get from that approach, we feel that having a recorder in there, where we can say, there were this many gunshots in this period of time, is a really good way to measure at least that type of hunting.
The biggest problem in terms of the acoustic approach is that recording a gunshot on our autonomous recorders and finding out that-- for instance, this poaching event, we didn't even find it for two years. Because we didn't happen to analyze this day in our recordings. It doesn't help these elephants. They're dead. And in general, this kind of passive acoustic monitoring, there is this delay. It's important information, but it's not going to solve the problem now.
So an important aspect or an important thing is particularly in bioacoustics, the technology exists now for real-time detection and transmission of this sort of information. But so far, we haven't been able to attract the funding to actually implement it, do a case study to show what could we do if a park manager had a computer screen and it says, we've got poaching. Gunshots are happening right here, right now. Would that be a way--
Again, the gunshots have happened. So something may be dead. But maybe you can get in there and catch the people.
So then, I kind of want to end up here with-- ultimately, the basic biology of the animals is critically important. And as I mentioned at the beginning, it's frightening how little we know about forest elephants. Most of the understanding has come from studies in clearings like the one you see here and that was in the beginning of that video.
But one of the things that those studies have shown, they're almost all-- 99% percent of the observations are daytime. And we know that elephants are very active in these clearings at night. So probably they are more active at night because they've learned to go to avoid poaching and hunting exposure. So what are we missing by not even having any information at night?
So recently, I used a thermal imaging camera to probe what was going on at night. And it was a real eye opener. I'm just going to show you one little example of this. This just gives you an idea of the kind of detail that you can get with a thermal imaging camera.
This is the ending part of a copulation that occurred late at night. It was pitch black. I couldn't even see my hand in front of my face. But I could record all of this and record the sounds that were going on.
Mostly, what I am trying to show here is that there is what we call a mating pandemonium, a vocal display that occurs after many copulations. Lots of individual elephants are involved or participate in that pandemonium. And after having been able to see it linked with the behavior, it gives us another tool where we might be able to search for that vocal signature in order to quantify, where is mating happening, and where are the most important places for mating in forest elephants.
Is it Dzanga Bai? Or is it happening all over the place? Is it happening in the forest? Do elephants only come to clearings for mating? These are all questions we don't really know.
So I want you to notice also that the first real vocalization happens immediately when the male dismounts.
Right now. We don't even know who gave that vocalization. It might be the female.
But then all of this is other animals in the clearing who have heard something. They know that a mating happened. And they're now responding.
So I'll tell you, when I first looked through this camera in the middle of the night, I couldn't believe it. It was like, wow. I can see everything that's going on in this place, where we couldn't see anything before.
So I'm going to end the me sitting here, just talking, so that I can answer questions about any of this, and anything else that is of most interest to you guys.
SPEAKER 5: One of the questions before was about crowdsourcing all of this--
PETER WREGE: Yes.
SPEAKER 5: --mass audio data that you have. And I guess, based on what you said, the technology is there, that we can monitor this in real time, theoretically. So I guess crowdsourcing would kind of just be second-best.
PETER WREGE: Well, it would depend on the question. Something like--
SPEAKER 1: Fund individuals.
PETER WREGE: Yeah. Monitoring gunshots is something that real-time detection would be very useful for. The other sorts of things we're doing, like looking for these sequences of sound, and other kinds of data that we collect from the sound recordings, are very suitable for crowdsourcing.
SPEAKER 1: Would you talk to the whale people, and--
PETER WREGE: Well, they aren't doing it, either. I mean, one of the problems that I've had is-- the technology is all there. I mean, sound files are big. And so there is an issue with where you put them where people can access them-- what repository. But more and more companies like Google and stuff like that-- I'm sure you could get free storage space in the cloud, to put this kind of stuff, for a good reason.
And the real thing holding it back is just developing the interfaces and all that kind of thing to make it available. So it's not that it isn't suitable. It's that I haven't figured out-- haven't found the time and the-- I mean, I can't do that kind of-- I'm not a programmer. I need somebody else to do it. So it's, how do we find somebody who's willing to pay for getting that to happen.
SPEAKER 3: Is it too remote to do, a live video detection?
PETER WREGE: Generally. Most of where I'm working, there's no cell infrastructure. I mean, we're really thinking about trying to have a live video feed from the Dzanga clearing in Central African Republic.
SPEAKER 3: Because when you were talking about reaching the zoo-lover, or whatever, I just wondered if that--
PETER WREGE: Well, it would be fantastic. I mean, the Lab of Ornithology, in the last two years, has been doing these live bird cams. And it has been incredibly popular. People just love seeing-- they can just sit there and watch these behaviors happening.
In a place like Dzanga, there are always elephants there. And you could train some local people to be there some of the time to move the camera to interesting things. I think it would be the most popular thing on the web, I bet.
I mean, people really respond to elephants. They are amazing, amazing animals.
SPEAKER 2: Well, I interviewed Andrea Turkalo. And she mentioned that when you put information on elephants on the web, they get a lot of hits from China.
PETER WREGE: Yep.
SPEAKER 2: So, the poachers and their bosses are also watching.
PETER WREGE: Yep.
SPEAKER 2: So she was very reluctant to get into that. And also, about the acoustical monitoring, I was just in East Africa where I was doing a story on poaching. And they also talk in some of the parks there about acoustical monitoring for gunshots. But the problem is, as you say, once you hear the gunshot, first of all, the incident has already happened. Second of all, unless you have ranger teams really nearby, it's not going to help.
PETER WREGE: No.
SPEAKER 2: And I would think that in West Africa, the situation in terms of rangers seems to be a hell of lot worse than in the big parks.
PETER WREGE: It is. It's much harder, logistically, just in terms of how much time does it take to get a team somewhere that would be effective. It is more difficult.
But, that said, a lot of these protected areas, they're difficult to get in and out of. And as far as I understand-- this isn't where I'm an expert-- but the poachers and the hunters use certain ways in and out. And the park people, in theory, if they're not in on the game, they know where those are.
And so if you had something that said, we know that there's been some poaching happening in this bai, in this place, they might have a couple of days to go get a team in the right place, and catch them coming out. Or, we come up with other sorts of ways where we try and get teams in in a more rapid way. It's not easy. I mean, I completely agree. And ultimately, stopping the incursions in the first place is what we need to do. We need to-- really, ultimately, it's that demand.
SPEAKER 3: Do you have trouble finding your equipment later, or does it get vandalized?
PETER WREGE: We've had a very few things vandalized. And mostly, that's been because of my naivete in where I put them. We had one--
SPEAKER 3: You could set up a camera and not tell the poachers. I mean, that's--
PETER WREGE: Yeah. And the worry that Andrea has expressed, as far as I'm concerned, a place like Dzanga, where-- that's a well-known site for a lot of elephants. It's not like poachers and hunters don't know that that's there. And it seems to me that if you-- a live feed, if that would really, really raise awareness, maybe could really help raise money. So you put a lot of money into protecting Dzanga Bai. You make sure that hunters can't get in there.
SPEAKER 3: And police do it now for crimes, so it's not like you're--
PETER WREGE: So, I don't know. I mean, the other possibility, I guess-- many people, if they saw a video of Dzanga Bai, they'd know it was Dzanga Bai. You don't have to say where it is. You could say, this is in an unknown place in Central Africa. So, I mean, eventually, somebody, if they want to, they'll find out where it is. But--
SPEAKER 2: Well, she said that Dzanga Bai was the best protected. But there was recently a massacre in Dzanga Bai. 40, 50 elephants were killed. So--
PETER WREGE: Yeah. I mean, it's hard to know exactly where it is. There's certainly plenty of circumstantial evidence that best protected is Andrea Turkalo herself. I mean, this happened when she was gone. And she's had almost no poaching in that clearing when she's ever been there. It's always happened when-- everybody locally, they know when she's gone.
And she's a very, very tough woman. And she's willing to go and say, somebody has killed an elephant here, and I know who it is. And she goes to an authority and tries to force them to do something about it.
SPEAKER 2: Right. But the Central African Republic is in chaos now, so--
PETER WREGE: Well, right now, there is-- yeah. And Andrea's here in the States. So it's-- yeah. I mean, I think when you have a breakdown like we've had in Central African Republic, the most we can do is sort of cross our fingers that something really, really disastrous doesn't happen until it gets back enough stable that people can get back in there. But right now, it sounds like that's an area that's just a mess.
SPEAKER 6: How did you decide to use the thermal imaging rather than normal night vision? Is that a new technology?
PETER WREGE: Yeah. I've done normal night vision, which is a lot cheaper. This camera was very expensive to-- I leased it from a company. The difference between what you can see with the thermal imaging, it's like, orders of magnitude more detail.
Normal night vision depends on photo multiplication. So you have to have natural light-- moonlight, starlight-- to see anything. And if the animal you're looking for is, like, under a tree, where there's a shadow, you can't see anything.
The other thing that you can see a bit here is that with the thermal imaging, individuals have different body temperatures. And so if you have a crowd, a clump of elephants, you can often easily distinguish how many are there. And with night vision, you just see these shapes. And it's very, very hard to actually count them.
I was using this a little bit just to count how many elephants were in the clearing-- and this is all at Dzanga. At 10:00 or 11 o'clock at night, there can be 110 or 120 elephants there. With the thermal camera, it was like little blobs of coal glowing in this black landscape. I could go through and go 1, 2, 3, 4, 5, 6, 7, 8, 9, and count very easily.
And you could never do it with night vision. Maybe with the top-of-the-line military equipment. I don't know what they have, because we don't see it. But this is really, really, I think, a valuable tool.
SPEAKER 6: Do a lot of other researchers use it?
PETER WREGE: No one. Certainly, this is the first time it's been used, really, to look at elephants.
SPEAKER 5: So, to the best of your knowledge, what exactly is happening here? So, two elephants were mating. And then there's this pandemonium call, as you called it. And then, all these elephants ran over. And it looks like maybe some of them were even juveniles?
PETER WREGE: Yeah. No. There were some young ones, some older ones. What we think is happening is-- the guess is that maybe the female who was just copulating gives this initial call that somehow is signaling that she just mated. And there are lots of theoretical ideas of why-- why would a female want to advertise this?
Some of them have to do with, well, maybe what it is is to say-- essentially elephants come into estrus only rarely. And they're only fertilizable for a few days in their estrus period. So when you've got a system like this, where males and females aren't together all the time, there's a problem with how do you find a mate when you want to mate.
And for a female, again, theoretically, she may want to incite competition between males. So it could be that it's saying, I just was mating. And if you're another male out there in the forest, you better come in here and check it out. Because maybe you're who should be fathering my babies. So that's one idea.
Another one is that, as I mentioned before, forest elephants spend most of their time distributed in the forests. As far as we can tell, they're very small family groups moving around on their own. So the clearings may be a place for social interaction and where you learn information about the population. So it could also be that that call stimulates other individuals, particularly females, to come and find out something about the quality of the male that just was copulating.
It was a little bit hard to see-- I didn't point it out. I could've. Often, right as the male dismounts, a lot of seminal fluid and female fluids fall to the ground. And that's what these elephants were doing. They were running over and they're smelling these fluids. And elephants are extremely sensitive in their olfaction.
So it may be that there are signals there that tell them about what's the testosterone level of this male, or something that they can use to estimate or learn about this particular male, that they could then use when they come into estrus. They may go, like, nope. I've checked this guy out and he's not worth it. So I'm gonna wait for somebody else.
So basically, we don't know what the purpose of this is. But it seems to be quite a distinctive pattern that happens. It's lots of overlapping calls, lots of roaring going on. And when you look at random sections of the sound from Dzanga with just as many elephants around but no mating, you don't see that kind of pattern or sequence in calling. So it's an area that we need to pursue more. But it's a neat-- it would be a great one to tease apart what's going on.
SPEAKER 7: Can you talk a little bit more about some of the other sounds you've translated? And what you've learned--
PETER WREGE: You mean elephant--
SPEAKER 7: Yeah. Sort of their language.
Yeah. That's a touchy word, language.
SPEAKER 7: Yeah.
PETER WREGE: We basically haven't really interpreted much of anything. We have a lot of calls. Doing that is quite a complicated process. And there certainly is information out there.
The 60 Minutes piece done on forest elephants was "The Secret Language of Forest Elephants," and the dictionary that the Elephant Listening Project is figuring out. Well, we tried to get 60 Minutes away from that, actually. But it's too interesting to people.
So, there's no question that there is some sort of detailed information that elephants are communicating with these calls. But we basically don't really know what that is. And we're trying to do very simple things, in terms of something like this mating pandemonium. You know, that sequencing of calls may send a very clear signal of a very specific meaning. We don't know that for sure.
We think that there may be a very specific kind of call that males give when they're in their musth period, when they're high testosterone and they're looking for matings. Females prefer males in that condition. So there's, again, lots of theoretical reasons you would expect-- if you're in a system where you are not always together, you need to say, hey. I'm a musth male. I'm out here. Are If you're in estrus, let me know.
But again, we have very little conclusive evidence that that call exists. So, I have some people in the lab, some undergraduates, starting to work on some of this by looking at rumble vocalizations that are paired with video of greetings-- you know, when parts of family come together and the females talk back and forth with each other-- and compare those rumbles to rumbles given when the family group is leaving the clearing to go out in the forest. Are there differences? Is there a rumble that means, it's time to go, family, let's gather around where I'm leaving, that's distinct from others? But right now, we're not very far in any of that.
SPEAKER 2: I'm working on a story right now about elephants and bees. Apparently, there was research done on the savanna where they found the warning rumbles that savanna elephants give to bees is different from the warning rumble that they give for people.
PETER WREGE: If they hear people voices.
SPEAKER 2: And then they have warning rumbles for both. And based on this acoustical research, they developed an idea of fences made of beehives. And they work. Because elephants are terrified of bees. They get into the--
PETER WREGE: Into their trunk. Yeah.
SPEAKER 2: --trunk, and they will sting. So, that's one example where acoustical research into the meaning of the calls actually led to some technology that's being used in East Africa now, to build fences keep elephants out of--
PETER WREGE: Out of crops.
SPEAKER 2: --farms, and so on.
PETER WREGE: Yep. Yeah. Yeah. No, that's a very interesting study.
It would be fun to try and see whether we have anything similar in our guys-- not specifically bees. I don't think forest elephants run into quite the same issue. There are honeybees around. But they're not aggressive and as far as I can tell, they're probably not a problem for forest elephants. But whether they have a distinct sort of rumble that would indicate what's the threat that they're afraid of could be a very useful thing for us as well, if we could find out if it's happening with them.
SPEAKER 3: When you did that gunshot, for instance, was there any lead-up? You know--
PETER WREGE: The only thing we could hear in the lead-up was they-- there were elephants in that clearing who knew there was a problem. Because you could hear them running out of the water. So, it was actually a very swampy clearing.
And we heard that, and then the gunshots started. And then there's lots of other stuff that we picked up in there. Afterwards, we could hear the poachers talking. We could hear them chopping the tusks out of the elephants. It's a pretty gruesome bunch of sound.
And in a bigger picture, not even just elephants, these acoustic monitoring, we're recording anything that makes sound in the forest that's within the range of our mic sensitivity-- we have a record of. And that includes chimps, and gorillas, and other kinds of monkeys, and all kinds of birds. It's an incredibly rich bunch of data that are-- and no one has-- we have, now, almost 300,000 hours of sound from Central Africa. And all we're doing is looking for elephant calls in 10% of it in order to get the data that we need. So, it's a--
SPEAKER 8: You mentioned that the elephants are really active at night because they want to avoid poaching or hunters in the day.
PETER WREGE: We think that's maybe why.
SPEAKER 8: OK. Do you think that's, like, a learned behavior? Have they figured out they have a better chance?
PETER WREGE: Well, I think so. I mean, elephants live as long as-- well, live as long as people used to live-- 60, 65 years, if they don't get killed. And so there's a very long time that they could have been exposed to some sort of a poaching attempt. They also may just learn it from, as they're growing up, the behavior of their mom.
It's not black or white, but the evidence seems to show that in areas where poaching has been particularly high, often there are no elephants in these clearings in the day at all. But they do come at night. Dzanga, which again, has been perhaps one of the better protected spots, a lot of elephants do come in in the day, which is why Andrea has been able to get the data she has. But still, the numbers are often at least half, again, larger or double at night, even at Dzanga.
SPEAKER 8: And also, is most of the poaching for bushmeat, or for tusk?
PETER WREGE: I think for elephants, it's mainly ivory poaching. I think in general, for bushmeat, it's a big animal to deal with. And they're dangerous.
And they're not-- most of the hunting, the bushmeat hunting, if they're using guns, it's shotguns. Which isn't the best kind of weapon to kill an elephant with. So it seems as though people aren't really going out expressly to kill an elephant for bushmeat.
SPEAKER 8: But do they take the bushmeat, ever? Or--
PETER WREGE: Well, the poaching incident that was mentioned that happened at Dzanga, local people came in afterwards and took the meat. Which is-- you know, everybody would say, you might as well. Otherwise, what did that animal-- at least it's serving something that helps some local people. So I think that when they can, they'll use it.
And the place I was just now, in Republic of Congo, the town I was in had the most bushmeat activity that I've ever seen. It was absolutely horrible. I did not see anything that looked like it was elephant. And in talking with Wildlife Conservation Society people, they said as far as they know, that market doesn't have any elephant meat.
But there is certainly ivory poaching going on in that forest. So at one level, if you're really going for the money, the tusks are all you want. Bushmeat actually doesn't demand very high price.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact email@example.com if you have any questions about this request.
Peter Wrege, director of the Elephant Listening Project, met with reporters in NYC April 14, 2014 to discuss new discoveries related to forest elephant behaviors and the poaching crisis in Africa.
The Elephant Listening Project is a not-for-profit organization associated with the Bioacoustics Research Program at the Cornell Lab of Ornithology.