E9 - Beavers and AI

00:12
Kathy

Since we recorded this episode in 2024, a lot has changed. Namely, my guest is now forever connected to Hollywood! 

Pixar – the movie maker - asked Doctor Emily Fairfax to be a science consultant on their animated film Hoppers. It’s all about beavers and was released in March 2026. Her job was to make sure any beavers in the film were portrayed scientifically accurately. On her Instagram account, Emily says “the best part was taking Pixar's art team into the field for a week and showing them how incredible, messy, beautiful, and chaotic beaver ponds are in real life.”

Thanks so much for tuning in. Let’s get you into the original episode:

-------

01:07

Artificial intelligence. It’s what everyone is talking about it these days. At the University of Minnesota, researchers are using AI to study beavers. I’m Kathy Mueller, host of Salthaven’s podcast Wild About Wildlife. In this episode, I speak with one of those researchers, Emily Fairfax. She helps us understand the positive role beavers play in our ecosystem – turns out they are pretty adept at stopping wildfires in their tracks. We explore ways of co-existing with these sometimes destructive mammals, and we debunk some myths about one of Canada’s favourite rodents.

01:47
Kathy
Thanks very much for joining me today Emily. Before we start talking about your research with beavers though, let's get to know a little bit about who you are. I was reading on your bio on LinkedIn that you once told your mom that you wanted to walk on the rings of Saturn. So, like you, you've been interested in science for a long time.

02:08
Emily 
I have yeah ever since I was a little kid I was so fascinated by the world around me and just really trying to find cool ways to be part of it and to understand it and that took a lot of different forms. I don't need to be an astronaut anymore but still very interested in science.

02:25
Kathy
So, how did you end up at the University of Minnesota and studying beavers?

02:32
Emily
So, I have always thought wetlands were super interesting but didn't exactly know that you could do a career out of wetlands. So, I did my undergraduate degree in Minnesota at Carlton College and I studied physics and chemistry and then went to go work as an engineer after undergrad. And in that job as an engineer, I just kept wanting to be outside and to find myself in rivers and in wetlands and I finally was sitting on my couch one day and saw a documentary called Leave it to Beavers that was all about beavers and the ways they engineered the environment, and I was so hooked. And in that documentary they really talked about that there's still a lot we don't understand and there's still questions to answer. And you can be a hydrologist or a geomorphologist or all these different ologies I'd never heard of before and I was like, well, that's going to be me. I can't stop thinking about this so went to grad school to study beavers and really haven't looked back since. I'm still doing it today.

03:26
Kathy
And then okay so let's talk about beavers then and before we get actually into your research, the role that they play in our ecosystem.

03:36
Emily 
Yeah, so beavers are a keystone species and an ecosystem engineer. Keystone species means that a lot of other organisms in the environment depend on beavers to create habitat that they rely on. And ecosystem engineer means that beavers are able to modify the physical environment to better suit their own needs. So, what this all boils down to is that beavers are really powerful agents of change in the environment and they can take a river or a stream and pretty quickly turn it into a wetland. And wetlands are one of the most biodiverse and rare land cover types that we honestly have anymore. So, beavers are out there making this really special home where, in many places, there's nothing else making that home anymore. It's just beavers. So, a lot of folks and animals and plants, they rely on the beavers for wetlands.

04:22
Kathy
And how's the beaver population doing both in the States and in Canada? Are they thriving?

04:30
Emily 
I don't know if I would say the population is thriving. It's better than it was a couple hundred years ago, but it's not what it was in the past. So, historically before the fur trade there were anywhere from 100 to 400 million beavers in North America and that's about one beaver for every kilometre of habitable stream. So, they were literally everywhere. And then, during the fur trade they were really intentionally overharvested and pushed to near extinction. We are probably in the range of 100,000 beavers at the end of the fur trade, so really, really close to losing them. Since then, their population has been able to rebound a bit. It's somewhere between probably 10 and 30 million beavers on the continent today which feels like a lot but that's only 10 percent of their historic population. And there's a lot of places where, you know, in the past it used to be absolutely full of beavers and today you really have to look to find a single beaver colony. So, there's a lot of room for growth still.

05:24
Kathy
And do you think that we need more beavers?

05:28
Emily 
I think so. I mean, beavers were one of the dominant forces on this continent for so long. They learned to build dams and engineer the landscape like 7 million years ago. So, this continent has been shaped by beavers for millions and millions of years. And even in the holocene, the last 10 thousand years where people were really starting to make a bigger impact on the planet, beavers were there in abundance. It's just the very last couple hundred years that we haven't had them and so when we think about what North America should look like and what North America did look like, it has always looked a lot more beavery than it does today.

06:06
Kathy
A lot more beavery. I like that. Now, not everybody likes beavers because of the, you know, that they are building their dams their lodges and that, and that can lead to flooded crops for example. So, can we kind of shift that reputation? Are they getting a bad rap, first of all, and then how can we coexist with beavers and a larger beaver population?

06:31
Emily 
So, beavers are really powerful engineers, and we are also really powerful engineers, and we don't always agree on how to best engineer the landscape and that's where a lot of these conflicts come from. So, if a beaver moves into a valley bottom or a river, it wants to have as much aquatic area as possible. It wants as much aquatic connectivity as possible. So, it’s doing everything it can to increase the amount of water on site which, at a very surface level, it feels like we would agree with that and want to do that as well until we've got our roads in the way or our basements or our houses. Or the beaver needs to cut down some of our favourite trees to create that wetland and that's where we start to really have problems living next to the beavers because we want parts of that area to be dry and beavers want everything to be wet. So, we do come into conflict with them. But there are ways for us to live alongside them more sustainably. You know we are also very smart engineers. I think we sell ourselves a little bit short when it comes to beaver conflict. We're like oh no, a beaver moved in, the only thing I can possibly think of is removing the beaver. There's no other way to do this and it's like um, you know, we've built like rocket ships and things, I think we can figure out how to outsmart a beaver. So, you can install things like pond levelers. It's basically a pipe that goes through the beaver’s dam and you set the water level with that and so you can stop the beaver from flooding the road. As long as you don't take the water level down to nothing, the beaver probably isn't going to notice that you're doing this. It’s just going to not be able to make its pond bigger and then it's up to the beaver to decide if that's acceptable or not. Sometimes it decides not and they'll leave and I guess in that situation like conflict resolved. And sometimes it decides, yeah, this is fine like, good enough pond and they'll stay and you don't have a flooded road but you do have the beaver that is there providing all these other benefits to the landscape. Same with chewing on trees like, we have invented fencing. We can use a fence and if you don't want a big like large chain link fence or something, that's totally fine. Beavers are very easy to redirect with electric fencing which a lot of our farmers and ranchers are very familiar with anyways. Beavers are smart. They learn fast. Once they step on the electric fence line a couple of times they're not going to come walking that way again.

08:40
Kathy
Yeah, like a kid on a hot stove right? And then in terms of the beaver population and predators. So, you mentioned conflict with humans. I would assume that we are their largest predator?

08:57
Emily 
Yeah, so we today are definitely their largest predator. There's a lot of beaver trapping that goes on both for recreation and for food but then also for conflict and nuisance. And so I'm not sure of the numbers in Canada but I know in North America, tens of thousands of beavers are killed every year in a way to do conflict mitigation which is not super sustainable because once a habitat is good beaver habitat, they're going to keep coming back every year so you're on this sort of like never ending treadmill of beaver removal which could be stopped if you know, mitigate in place and work on that site and keep the beavers there but in a modified way. They do have other predators though, so wolves absolutely love to eat beavers. They're really good hunters and because they're pack hunters they can sort of route the beavers round and outsmart them and pick them off more easily mountain lions or cougars, also good hunters of beavers. Grizzly bears can hunt beavers. Once a beaver is an adult, they're going to be anywhere from 40 to 110 pounds which I think, I'm going try to do kilo math in my head, I think that's like 20 to I want to say like 45 kilograms, like pretty big. 

10:05
Emily
And so, when they're adults, like they just have mass on their size. There's not a lot that can put in the right effort to haul and kill a beaver. But, when they are little they have a lot more predators. So, baby beavers are also hunted by things like otters and lynx and bobcat and coyotes and even bald eagles will hunt baby beavers. So, there are natural predators. We're just probably the biggest one.

10:25
Kathy
Okay, and then you talked about, you know, that a beaver will move on if the landscape isn't suitable for them. Do they tend to build a lodge and then stay put or are they moving pretty frequently?

10:41
Emily 
They would prefer to not move all the time. It's a lot of work to build their wetland and, you know, if they move they have to start that process over again. They will move sometimes though if too much mud starts to build up in their ponds and it's hard for them to stay there. They need a certain amount of water depth to feel protected and to make sure the pond isn't going to freeze through in the winter. So, if it starts filling up too much they might leave and wait for that to break and drain some of the sediment off or fill all the way up and then the stream will start cutting across the top of it again and they can rebuild on top of it and sort of build land. Sometimes they will chew on a lot of the trees in their immediate vicinity for food and it'll feel like there's not enough left. Even if there are some trees left it's not enough for them to feel safe and so they'll move on while those trees regrow and then come back. There's this paper, I want to say it's from the 1920s, I don't remember the author off the top my head, but there's this great quote in it that beavers don't mine the landscape, they farm it. And so it's not that they're fully extracting a resource from the landscape but they do have harvests and they like cut it down and then it needs time to regrow. So, that's one of the reasons beavers can move. I would say on average beavers are going to move probably every 7 to 10 years but they are also going to come back and recolonize the sites that they had left 7 to 10 years ago. So once a beaver’s in the system it's pretty much just there on and off ad infinitum.

12:01
Kathy
And how long do beavers live?

12:03
Emily
In the wild beavers are going to live about 10 to 14 years usually, but they can live over 20. I know there have been some wild beavers that have been aged at like 20 and 22 years and I think there is one in captivity that lived to be 24 so, as far as rodents go, they live a long time. And as far as many animals go, they live a pretty long time. So, you know, a beaver can be there occupying a single wetland site for its entire life and that could be like 20 years of the same beaver family.

12:33
Kathy
And then are they having babies every year, and what are the baby beavers called, and is it one baby every year or how does that all work?

12:42
Emily 
So, beavers, they tend to form monogamous pairs and mate for life, and they are going to have one litter of kits which are the baby beavers per year. In a litter of kits, you would expect to see somewhere between one and six babies on average, although some litters that are much larger than that have been reported before. You expect maybe half of those babies to make it to adulthood. So, the mortality in beavers is definitely heavy on the younger beavers. So, if they're having you know 3 or 4 babies per year, about 2 of them are going to actually make it to adulthood. Those babies do stay home with their parents until they are 2 to 3 years old, learning to be better beavers, hanging out with their family, shadowing their parents and their siblings, and then once they reach that 2 to 3 year old mark, that's when they're going to become what's called a dispersing juvenile, basically a teenage beaver getting kicked out of the house for the first time to go set up its own life. So, it's going to go off on its own and try to find some habitat to set up and try to find a mate to start its own family with.

13:38
Kathy
All right I'm talking with Emily Fairfax an assistant professor at the University of Minnesota and Emily, we love beavers up here in Canada, like we have beavers on our nickels right? Let's talk about the research that you're doing I had to write it down because it's quite a mouthful. And you're using something called the Earth Engine Automated Geospatial Elements Recognition or EEAGER, as an eager beaver. First of all, was that acronym chosen specifically so that it would go well with eager beaver?

14:17
Emily
Yeah, it was a bit of a backronym so we knew that we were going to be using earth engines. We had EE and then we knew there was this geospatial, so we had a G already and we were like you know we're pretty close to being able to spell eager. Let's see if we can make this work. It is automatic too. We're getting closer. So yeah, we did want it to be the eager beaver model. So, we a little bit forced that.

14:39
Kathy
Okay, so explain how you're using this system and yeah, tell us about your research.

14:48
Emily 
So, the EEAGER model, it's called the convolutional neural network model. But basically, it's an artificial intelligence or machine learning model that we can give aerial and satellite images and it can find the beaver dams within them. And this was a really helpful model for us to build because finding where beavers are in the landscape is actually pretty time consuming. I can do it in satellite and aerial images. My students can do it after they've been trained, but it takes time where you're sitting in front of a computer and scrolling and your eyes get tired, and it felt like this was really very ripe for automation. And the reason we do this, we spend so much time finding these beaver dams, is because beavers engineer things both at like the individual pond scale. But then when you have a population of beavers, they're changing the landscape at the watershed scale or the basin scale or the regional scale. And to see how beavers interact with larger disturbances like droughts and floods and wildfires which is really the core of my research, we are spending too much time finding the beavers and not enough time seeing what they're actually doing so automate the finding process with EEAGER, and then seeing what they're doing, that's what me and my team can actually focus on, looking at the ways that beaver wetlands provide refugia during climate change.

15:58
Kathy
So, what have you been able to determine so far based on the research that you've done and the dams that you've been able to find?

16:05
Emily 
So far our biggest finding is that beaver wetlands are really uniquely fire resistant. So, even when we have really large and intense fires like megafires, the beaver wetlands within these areas, they are just too wet and too complex to burn. And this is something that is happening really consistently and reliably across different states in the American west and I've recently expanded this research out and we're looking at a whole bunch of other places working in Canada, we're looking in the eastern United States, looking in Alaska and again and again and again, the beaver wetlands are not burning during wildfire. And that matters for the beavers, like they don't want to be caught up in a fire, but it also matters for all of those other organisms that call the beaver wetland home, which is a lot of them, because remember beaver is a keystone species. So, if you're a fish or a frog or an amphibian or a bird or a small mammal or a person even, if you need to find somewhere in the landscape that's not going to burn during a big wildfire today, you don't have a lot of options. Most of the forest is really flammable today. It's drought stressed. It hasn't been managed particularly well for the last couple hundred years. Unless you have like a concrete parking lot or reservoir, it's just a gamble like, I don't know, pick somewhere and hope it doesn't burn. But what we're seeing with the beaver wetlands is that if you have a beaver wetland you basically have a place that you can count on not burning and that if you're one of those animals that can't outrun or outfly the fire, you just have to get to a beaver wetland and you can survive and then all those plants too which are not mobile, when they're in the beaver wetland and they're preserved, that's a mature sort of reproducing population of plants when the rest of the landscape might be burned down to nothing. It might just be seed bank left in the soil, if that. Some of these fires are burning really deep into the soil. So, the beaver wetland is preserving habitat for animals but it's also preserving the vegetation that is the foundation of habitat and can help recolonize that after the fire.

17:58
Kathy
That is absolutely incredible. It's almost acting as a buffer right? And like you're saying, it's a safe place and for creatures of all sizes and ranges. So, especially with climate change and, you know, last year was a horrible year for wildfires here in Canada and here in southwestern Ontario we had some really hazy days because of the smoke that was coming in from wildfires in neighbouring provinces, not even from Ontario. So, how do you see your research being used as climate change becomes more of an issue?

18:38
Emily
I've seen my research be used a few different ways. One of them is as this sort of collaborative effort with fire mitigation. So, there's a lot to do with fire mitigation. I've looked at some of the fires that Ontario had over the last few years and like massive, absolutely massive fires, the scale of work that's needed to protect all of that boreal forest and all of those peatlands is very difficult to like, overstate. Like, it's a lot of work and that needs a lot of people and a lot of time and a lot of money if we do it all ourselves and so, what I'm seeing more folks do is think about all right, well if beavers are already doing this sort of natural fire mitigation in the river corridor or around the borders of our lakes, then what if we focused our efforts more on the places beavers are not working, so the uplands or the hill slopes or the meadows the beavers are not occupying? Instead of being so redundant with our firefighting, more and more organizations that I've talked to and that I've worked with are really starting to think about how can we be strategic and cover as much ground as possible letting the beavers do what they do? And at first that kind of felt like probably a little goofy like, all right, we're firefighters. We're trying to deal with one of the most pressing disasters of our time and I'm going to just pass this off to the beaver, big old rodent. But you know, weirder things have happened. So many places just turn goats loose on the landscape because they want the vegetation thinned and it's like, okay, well we already have trusted other animals with our fire mitigation work, I guess beaver is just part of the crew. 

So, that's happening and then I've also seen folks really try to be more strategic when they think about ways to increase the fire resistance of their landscape, especially if they're a place that does not have beavers right now and they know they have a lot of vulnerable riverscapes or they have like threatened fish populations that can't have ash on their gills. They think all right, where can we encourage beavers to come back? Where can we build what are called beaver dam analogs or fake beaver dams to start simulating what the beavers do when we don't actually have the beavers there to work with right now, and then cross our fingers that beavers show up and we can work with them, because they're definitely more effective at building dams than we are, at least at the beaver dam scale. And you know, sort of all hands, all paws on deck with this. People are trying to be as strategic and as efficient as possible because I think what we learned last year with the wildfire season especially, is you never know when it's going to be an explosively huge year and we don't really have a ton of time to goof around and twiddle our thumbs like, if there's a potential solution, let's get it an action.

21:07
Kathy
Yeah, and then it's great to hear that you're getting some buy in about the idea of using the natural resources that we already have out there. Is there any thought about splitting up beaver populations and relocating some already existing beavers as opposed to encouraging them and hoping that they'll eventually go where we want them to go?

21:31
Emily 
Yeah, so in California they recently established a beaver restoration program within the California Department of Fish and Wildlife and last October they did a relocation. They brought a family of beavers up to Mountain Maidu Consortium land and they introduced them there. And that is a place that has had a really rough history with wildfire and that needs more fire protection and so this was a strategic choice of bringing beavers back to Indigenous land where they belong and where they're culturally significant, of bringing beavers back to a place that has had a really tough time with big wildfires and will probably have a tough time with wildfires again and bringing beavers to a place where they're not going to be in such constant conflict with people. Like, a lot of these beavers that are being utilized in relocation processes, they're being taken from cities or from farmland or from ranch land where people are not tolerating the beaver. So, moving them gives them a chance at life as well as gives us a chance at getting some of those benefits.

22:30
Kathy
You're using AI to find where the beavers are and where their dams and lodges and that are. How much of a game changer is that to be able to use artificial intelligence?

22:45
Emily
It's been such a game changer. It speeds everything up so much. Like, before if we were mapping out all the beaver dams in a given fire scar, let's say it's a 100,000 acre fire, there could be anywhere from 100 to a 1,000 beaver dams within there. My team will literally go through and trace every single creek and look for every single beaver dam and make a mark on it so we know exactly where it is. And then to be super sure of it, we have 2 to 3 people all map it independently of one another and then compare our results. With the EEAGER model with AI, we can either run EEAGER on our fire scars to start with and then that's going to rapidly narrow down where it's likely that there are beavers. Or we can use EEAGER as one of our mappers and then have 2 to 3 mappers, all being human, be replaced by 2 mappers with a computer mapper, or one mapper with a computer mapper. And these are ways for us to save time, save money. I think before it would probably take us, I don't know, three months or so to do a season's worth of wildfires all hand mapped. And now with EEAGER, you know, we can get that first pass mapping done in a week and then our own mapping is sped up by it and so maybe we can do our mapping in a week or two, so a quarter to a third as much time as it used to take.

23:56
Kathy
Are you still sending teams physically out into the wetlands to observe the beaver dams in person? I imagine there's still a lot of info, and a lot of valuable info, that you can get from seeing something first hand as opposed to relying on AI.

24:15
Emily Fairfax
Oh, absolutely. I mean, I don't think the EEAGER model is ever going to fully replace the role of a person doing this.  It's just like, an assistance for us to do it faster and a little bit more efficiently. But we still go to the field. Even after we've hand mapped everything in remote sensing, we want to be 100 percent sure what we're seeing and then also, on ground, like, the level of detail that you can see is not always available in remote sensing or in aerial images or anything like that. I can think of one of the sites that I've been to where like we mapped it out by hand. EEAGER was also able to find this complex. No problem. It's beaver dams. It was pretty green around it even after the wildfire. It looks like it's a good case of fire refugia based on that analysis. When we go there on ground, and what we can see on ground, confirms everything that we saw in the satellites. But then also we can hear the baby beavers groaning and grumbling at their parents. And so, we know that not only did the beavers survive, but they were feeling comfortable enough and healthy enough to reproduce. And then we can see the fish jumping in the pond, and we can see a bunch of birds hanging out, and we can see a black bear which is a little scary, and a lot of information that was not able to be extracted from the satellite imagery. But, we were able to figure out where to go faster because of the AI.

25:30
Kathy
Ok, and so then in terms of artificial intelligence and how you are using it and what it's telling you, what does that look like? When I think of AI, I’m thinking of like, Chat GPT right? And so how does AI actually work in your purposes, if you can describe it in a way that's understandable to a non researcher?

25:54
Emily 
Yeah, totally. So, when we try to find beaver dams by hand, we're looking at like an aerial map basically and trying to find this telltale shape of a beaver dam and the pond behind it and we draw a line on it digitally. What the AI model does is it scans through that image and it puts like a blue box over every pixel that it thinks probably contains a beaver dam. And so we get this map out of it that has these clusters of pixels over likely beaver complexes and that guides our eye to where we would want to look in better detail and do that hand tracing ourself. And it also helps us know where to skip and where to not look. And when we have like western US landscapes where it's kind of obvious you're not going to look out in the middle of the desert for beavers, it's maybe not like, it's good confirmation that it didn't think that was beaver dams,, but I also wouldn't think that was beaver dams. We recently ran it though up in Wood Buffalo National Park, which is super boggy and wetlandy, and if I was going to hand map that I would look at every single part of that image because I have no idea. Beavers could be anywhere. When we ran EEAGER on it we were able to see a lot more clearly, oh, there are these clusters of beaver ponds and then there's other wetlands that actually are not being engineered by the beavers right now and EEAGER did not indicate that any of those were beaver and I confirmed those were not beaver so it was guiding us basically in just like map pictures of bright hotspots where we should pay more attention.

27:17
Kathy
And then, the first time that you used EEAGER, did you trust the results that it gave you or were you sending a team out, you know, like you're saying to verify and make sure that it's giving you accurate information?

27:33
Emily 
I would say we did not trust EEAGER the first time it ran because it was absolutely goofy. And when our first step is for EEAGER to run it and then for someone like me to look at an actual satellite image and hand trace all the beaver dams that I can see and, we do trust my hand mapping pretty solidly; I've been doing it, I've ground truth thousands of these dams in my life. So, we could see pretty fast in the first iteration of EEAGER, it was pretty convinced that cul-de-sacs were beaver ponds, that some specific types of shadows were beaver ponds and that was informative because then we could generate more training data and basically give it a bunch of cul-de-sacs and say these are not beaver ponds, please learn that. And then it stopped doing that. Now I would say that I definitely trust EEAGER a lot more. It's a lot more effective when it does a run. We just did one in California. And I had hand mapped most of that area already, so we were seeing how well it did. And it found this like, cluster in the mountains and I was like, why would a beaver be up there? That doesn't make a lot sense. And I looked and there's like four beaver dams up there. And it turns out it was just this little bitty groundwater spraying. It wasn't connected to the river network, so I wasn't looking there. And it found it and I didn't find it. And that was a really good moment for us because it's like, all right, you know EEAGER still picks up on things that are not beaver dams, like, it's not a perfect model. It probably will never be a perfect model. But it's at least providing us information now that is really complementary to what I could come up with on my own in a fast passive mapping.

29:02
Kathy
And where do you see your research going?

29:07
Emily 
I would love to expand it and run EEAGER and then quality control all those results and actually create like a continental map of beaver activity. We don't really know how many beavers we have today. We definitely don't know how many beaver dams we have today. We don't know if some parts of the country or the nation or the continent have rebounded the same as other parts. We kind of know that they have not based on the kind of research that's come out of different places so like beavers in Ontario doing a lot better than beavers in Alberta. And that's also kind of like a mixed bag with beavers in British Columbia, like, they have not all rebounded the same way and at the same pace and in the same places. So, creating a map of where are there actually beavers doing beaver things is a really big next step for me so that I can say like, all right, at the continental scale we basically have this like, void of no beavers in this area and that's also like high fire risk areas. What if we thought about getting the beavers back there and then maybe we can deprioritize, you know, the boundary waters of Minnesota because it's absolutely full of beavers and we don't, like, they're doing fine. We don't need to focus there. Let's focus where they need our help.

30:14
Kathy
Is there anything that the average person can do to try and support the beaver population? I'm assuming there's not too much the average person can do because it's not like, you know, we have, they're not like squirrels, they’re not running around on our streets or anything like that. But, maybe it's more a question of what can the governments and the different levels of government do, or companies, you mentioned companies and organizations earlier?

30:43
Emily
I think individuals can do something and that's just like, talk about beavers as a normal thing and as part of a healthy ecosystem and a healthy riverscape. We are still overcoming a long narrative of beavers being a pest and a pest only because we have had a lot of conflict with them over how to manage the land. And so, a lot of what we need to do is just shift the perspective on that. Like, we're not denying that beavers are chaotic and difficult neighbours, they absolutely are. But they are a necessary neighbour in many places so just shifting our attitudes a little and that comes from talking to people and like going up to your friends being like, yes, beavers are annoying, but also did you know that they've done all these really great things for us? So, like, let's tolerate them. 

In terms of governments and organizations though, governments I think could be more supportive in terms of finding funding for larger landowners and land managers like ranchers and farmers. If you are asking these folks to give back part of their land to beavers and let the beavers control it, the beaver is not going to be growing hay. It's not going to be growing carrots. Like, it's going to be flooding it and that means you're going to lose money like, you're going to lose that crop and that is not great. And it'd be really nice if the governments had more programs to offset those kinds of losses because while it is a financial loss for the farmer or the rancher, it is a huge gain for the community and society as a whole. And I don't think it's fair to ask these folks to be martyrs and just like give up all of their land for no reason to support everybody. But, if they're getting a little bit of money to offset that, I think it'd be a lot easier argument to make like, yeah I gave back ten acres to the beavers, but I got paid you know a thousand bucks an acre per year and that’s about what I was getting from my crops or whatever. I'm not a farmer. I cannot come up with farm dollars. But the example is just for illustration.

32:31
Kathy
And then is this like, AI, and maybe it already is being used, do you see it being expanded to track other wildlife and see what they're doing?

32:45
Emily 
Yeah, other wildlife would just need to have a big enough impact on the landscape that we can see it from space. That's one of the things that makes beavers real special is they create huge ponds. Some of these ponds are acres and acres big so it's easy to see them. I've seen beaver dams that are 400 metres long and that's the spatial scale that we can see reliably in our aerial and satellite imagery. You can also see them down to like 10 metres, but the 400 metre ones are easy to see. Squirrels don't do that, so tracking a squirrel would be harder. But another researcher has used a similar strategy to find arctic fox dens, for example. They're much greener than the rest of the landscape around them because the fox is doing a lot of nutrient additions within its den. And, you could probably train a model like EEAGER to find the arctic fox dens because that's a very clear signature that would not be explained by anything else and it, basically our rule of thumb is, if you can see it in the image and you can find it reliably, then you can probably train the computer to do it too.

33:43
Kathy
Okay, and you mentioned satellites there and I don't think we've touched on that yet. So that's what you're using the AI technology is communicating with satellites to get you the information that you're looking for?

33:54
Emily
Most of the images that we're looking at are from satellites. So, a lot of the satellites are taking high enough resolution images now that we don't need aerial ones in the same way where you used to like fly a plane to get them all. Ah so yeah, most of what we do is we will have satellite images that were taken at any point in the past really, we can work with. And we feed those images into the model and that's what the model is making its prediction based on. It's that image and then a digital elevation model. So basically, like, how steep is it at any point in the landscape?

34:27
Kathy
Okay, what else should I be asking you that we haven't talked about?

34:33
Emily 
Ooh, a good question. I think my favourite beaver myths that I like to bust whenever I have the opportunity to do so. 

34:40
kats_clicks
Okay, let's do that. What are some of your favourite beaver myths?

34:46
Emily 
So, number one beaver myth is that beavers eat fish. I blame this 100 percent on The Lion, The Witch and The Wardrobe, either the movie or the book because they have the beavers serving up this big platter of fish. And for whatever reason, like, so many people who I've talked to that think beavers eat fish will say like oh yeah, well I saw it in The Lion, The Witch and The Wardrobe. I don't know why. it was like so sticky of a memory for so many people but it was, but beavers don't eat fish. 

35:09
Kathy
Ah, yeah, that one I don't remember that from The Lion, The Witch and The Wardrobe. No.

35:13
Emily
Well, I'm glad. If you had remembered it, you'd probably have this horrible misconception. But I mean, it's understandable. They look a lot like otters and otters definitely do eat fish, but beavers don't eat fish. That's a big one. 

Another one that I hear a lot is that fish cannot get around a beaver dam and that it's just like this absolute obstruction for them and they can't do it. And it's not true for a lot of reasons. If they couldn't do it, fish would be extinct by now because beavers have been building dams for 7.5 million years and fish have also existed during that time period. So, they have ways around. In the field we've seen a number of ways that fish can get past a beaver dam. The most dramatic is when they will jump over the beaver dam. Not all fish can do that but a lot of our migrating salmonids can and so they will just hop the beaver dam, no big deal. Juvenile fish, like up to about maybe 4 or 5 inches long, they can actually swim through the beaver dam. These are really leaky structures and there's a lot of places where there's like underwater breaks in it and the fish just goes straight through. 

And then the one that I really think is underappreciated is beavers dig a ton of canals. Like, they are always digging these like little micro streams. From the downstream to the upstream, from the pond out to the landscape. So, it really does look like a spider web of water pathways and beavers use their canals to get around to their dams. They also don't want to go up and down their dams. Fish love to use their canals. So, there have been plenty of studies now where fish will have tracking tags in them and we will, like, catch them downstream and then we will see them way , but we never get them at the beaver dam and we're like, hmm, how did you get there? Oh surprise. It's just taken the canal, taking the like scenic route around the dam. So, a lot of fish, even the ones that aren't good jumpers, they can get past beaver dams.

36:56
Kathy
And one more beaver myth?

37:00
Emily
One more beaver myth. A beaver myth is that they reproduce like all the other rodents do which is like super fast. And you think about mice and people panic and you're like, oh my god I'm going to get a mice infestation. And so sometimes people worry like if I let beavers back on my land, if I'm going to have a beaver infestation. Are there going to be like 95 beavers here because that would be a lot. But it's not like that. They reproduce much more slowly, just one litter of kits per year and beavers are really, really territorial. So, once you have beavers on your property or in a stream, they are going to maintain and defend a territory about one to two kilometres of stream length. And so you will probably only have one, maybe two families of beavers on your property depending on how big your property is, and they will actually drive away the other beavers. They won't let more settle on your land. And so sometimes the best insurance against having more beavers show up is having some resident beavers in place because they will literally rip each other to pieces to maintain their territory. So, you're not going to get a beaver infestation.

37:58
Kathy
Ok, and then you have made the cutest little beaver video to show this project and the research that you're doing. You’re finding that's a really easy way to explain it to people?

38:16
Emily
Yeah I made that stop motion video when I was going on the job market and really tired of giving my elevator speech because I was always telling people like what do I do, what do I do, what do I do? And I wanted to show them what I did because I'm trying to like, paint this picture which I am 100 percent ready to admit sounds absolutely ridiculous and fantastical like oh yeah, and then the beaver protected the whole area from the wildfire. And just describing it does not always make the mental image I want, so, I decided to just make the image I want and to show people. 

I did not have any real life videos of beavers like doing this because I don't go into wildfires. But, I did have a long history of being a girl scout and doing a lot of crafts. So, I was like all right, I can make this with felt and construction paper. I can do this. So, I did. I made it on my kitchen table, and I was like, this works and it has been like the best tool I've ever had and I've just made it because I was frustrated, just always go for it.

39:13
Kathy
How long did it take you? How long did it take you to pull that together?

39:19
Emily 
I thought about it for a while like, a lot of days I was thinking about like what would this look like if I could do it? And then when I actually sat down to do it, it was maybe 4 hours. It wasn't that long.

39:28
Kathy
Oh, my goodness. Okay, and so we should tell people where can they find it? Like, where's the best place to look for it?

39:33
Emily
So you can find it on my YouTube channel. If you just search Emily Fairfax and then beavers or stop motion it should show up. You can also find it on my Twitter at Emily Fairfax and it should be my pinned tweet on there. So, also if you just look up beaver stop motion I'm pretty sure it's the only one. 

39:51
Kathy
Okay, fantastic. All right was there anything else that you wanted to touch on today. 

39:57
Emily 
I don't think so.

39:58
Kathy
Yeah, okay, that's a really good place to end. Thank you so much for joining me today. Emily.

40:03
Emily 
Yeah, no problem.

40:13
Kathy
Thanks so much for tuning in. 
We’d love to know what you think. Write a review, subscribe to Wild About Wildlife and make sure to tell your friends and family about us. If you have any ideas for future episodes, let us know! Our email is wildaboutwildlife@salthaven.org. 

There’s a lot more information on our website Salthaven.org and you can follow us on Facebook, Twitter and Instagram. 

I’m your host, Kathy Mueller. I’m also a Salthaven volunteer. Together we can keep the wild in wildlife.