Tech Reboot for Wildlife Conservation Projects

With wildlife populations plummeting, conservation teams are fighting back, using cutting-edge technology to track and protect the world’s flora and fauna. We discuss this with:  

Shannon Dubay, director of conservation technology, Panthera, Cape Town, South Africa: Dubay discusses how teams use camera traps, satellite imagery, AI and other technologies in cat conservation, how teams are analyzing and acting on data faster in Zambia, how tech advancements have changed teams’ success metrics in conservation projects, plus how predictive analytics could transform conservation efforts moving forward. 

Dave Thau, global data and technology lead scientist, World Wildlife Fund (WWF), San Francisco: Thau talks about how the role of AI has evolved in conservation and helped boost efficiency. He also discusses WWF’s ManglarIA Project, which uses  AI and other tech to measure the impact of climate change on mangroves in Mexico. Plus, Thau takes listeners through how the team is leaning into innovation and adaptability as well as working with community members to futureproof their efforts. 

Key themes

[02:33] Using camera traps, drones and AI to conserve cats in the wild

[07:41] How teams are analyzing—and acting on—data faster 

[11:10] Tech’s impact on success metrics for conservation projects

[12:02] How predictive analytics could change future conservation projects

[14:58] The evolution of AI in conservation projects

[17:35] How WWF is using AI to measure the impact of climate change on mangroves in Mexico

[20:26] Focusing on innovation, adaptability and community engagement in conservation 

Transcript

Transcript

STEVE HENDERSHOT

Biodiversity has a huge impact on human populations—and when it’s in danger, communities are, too. To protect natural spaces and wildlife, project teams are turning to technology such as drones, sensors, satellite imagery and artificial intelligence. So how is this tech helping teams deliver successful conservation projects? ​

SHANNON DUBAY

Conservation is something I’ve always been really passionate about. I love being in wild places. But conservation has often struggled to keep up with the newest technology and the newest data types—and really harnessing that data. Being able to see how we increase the efficiency in projects, being able to see how our impact is greater and more measurable, and being able to pinpoint that effectiveness has been really powerful for me.

STEVE HENDERSHOT 

In today’s fast-paced and complex business landscape, project professionals lead the way, delivering value while tackling critical challenges and embracing innovative ways of working. On Projectified®, we bring you insights from the project management community to help you thrive in this evolving world of work through real-world stories and strategies, inspiring you to advance your career and make a positive impact.

This is Projectified. I’m Steve Hendershot. 

On this 55th anniversary of Earth Day, we find a planet in peril. According to the World Wildlife Fund, the average size of monitored wildlife populations has plummeted by 73% from 1970 to 2020. Part of this is due to habitat loss, as more than 18 million acres (or 7.3 million hectares) of forest are eliminated every year, according to EarthDay.Org. And ecosystems around the world are taxed by the increasing frequency of extreme weather events along with habitat degradation and loss, overexploitation and disease. 

But project teams are taking action, with many relying on high-tech tools like drones and AI. And today we’re going to speak with a couple of project leaders at the forefront of deploying tech in conservation. 

Before we dive in, if you enjoy Projectified, please leave us a rating or review on Apple Podcasts, Spotify or wherever you listen. Your feedback helps us keep making this show.

Okay, now let’s take a look at efforts designed to conserve and protect wildlife and the spaces they live in. First, we go to Cape Town, South Africa. Shannon Dubay is the director of conservation technology at Panthera, a global organization dedicated to the conservation of wild cats, both big and small, as well as their ecosystems. Shannon talked with Projectified’s Hannah LaBelle about how tech is helping Panthera ensure the health of leopards, lions, lynxes and more.  

MUSICAL TRANSITION 

HANNAH LABELLE 

All right, Shannon. Before we talk about your team’s work, creating and managing all the tech on Panthera’s projects, let’s start here: Why cats? Why is it important for conservationists to be looking at these animals in particular?  

SHANNON DUBAY 

Cats are a really great example of what we call an umbrella species, or in other words, a species that can indicate the health of many other aspects of an ecosystem or a landscape. So, the idea is that by researching cats, we’re able to learn a little bit more about the other animals in the landscape. And also, by protecting cats, a lot of other animals in the landscape will also benefit. 

Cats are very charismatic. They’re very interesting. Lots of people have lots of emotive experiences around cats, and so we often find they’re a really effective way to conserve an area and an ecosystem much bigger than just those cats. 

HANNAH LABELLE 

There’s a lot of technology involved in your work, whether it is the software or these camera traps. But there’s also drones, AI, machine learning. How has that changed in conservation?

SHANNON DUBAY 

Just to set the scene a little bit, initially, camera traps used to be really large. They were big and bulky and heavy. They also used to have a really short battery life, and they were generally quite unreliable. These devices have to be out in nature, so all of the fun conditions that come with that—whether it be weather, extreme temperatures, rain and moisture, floods in certain landscapes. Even animals might come over and sniff around your cameras or even try and take them off and play with them or eat them. So they have to be really durable.  

Camera traps have benefited from enhancements in batteries and other hardware pieces, and they even have some software features now, such as centering algorithms to make sure that the animal that you’re taking an image of is in the middle of the screen. Or some AI components to help filter out unwanted images, such as a blade of grass blowing in the wind that you might not be particularly interested in. Even on the desktop side, when you get back to your computer and you’re going through the data, these images you’ve collected, we now have some smarter software with AI-driven species recognition or even individual animal recognition like unique patterns of certain species, like leopards, jaguars, even tigers. 

So it really helps with the data processing and really decreases those manual processing times. Something that used to maybe take months of effort and time can now be completed in just a few days. 

HANNAH LABELLE

When you talked about the old camera traps, all I could think was, I think a lot of camera traps went down with cats being too curious.

SHANNON DUBAY 

Oh, absolutely. We have lots of images taken from our cameras where it has actually captured a cat walking away with a camera in its mouth. So if we don’t secure them properly, it does happen. Especially in my area of the world, lions are particularly good at that as well as actually elephants and rhinos.

HANNAH LABELLE

So we’ve talked about camera traps and touched on some AI tools associated with them. What other technologies are your teams using? 

SHANNON DUBAY 

There’s been huge advances in satellite imagery and the availability of those images to the general public and to researchers. Government agencies such as NASA and many others have made a lot of their imagery available. And then platforms like Google Earth Engine allow us to access and analyze those images to help inform habitat assessments such as deforestation or other changes that we might be seeing at a landscape level.

One other technology I wanted to highlight, because I just have been learning about it and I think it’s incredible, is we now have AI-powered bioacoustic devices. So there are these little audio devices, basically recorders, that you can put out into a landscape. What’s really incredible is they are now able to put AI alerts onto those devices so that if it hears a particular sound, it will send a real-time alert to a researcher so that they can react in real time to whatever the threat might be. We were mentioning deforestation. They’re able to train AI models to detect chainsaw noises so that they might be able to, in real time, learn if deforestation is happening in an area that it shouldn’t be. 

HANNAH LABELLE

We’ve talked a little bit about some of this general technology. Now let’s talk about the Panthera Integrated Data Systems, or PantheraIDS, project. Tell us about the project itself and how it’s helping your organization’s conservation efforts.

SHANNON DUBAY 

PantheraIDS was born completely out of necessity. If you’ve got a camera that’s indiscriminately taking photos, it can often result in these data sets which are massive. It really primarily helps with storing and standardizing and processing these large data sets. Oftentimes, we’ll put out maybe 80 cameras or 100 cameras, and they’ll be out for multiple weeks to try and get a full survey of the landscape. So now, instead of having to process that all manually, they’re able to put all of that data into PantheraIDS. 

The conservationists on the ground can get insights from their landscapes much more quickly and much more accurately, which is really important. Because these insights, they inform things like policies, they inform species population assessments and species vulnerabilities. It’s really important to be able to have these data insights available as quickly as possible so that we can actually use them in near-real time. 

HANNAH LABELLE

What does the training or upskilling look like when it comes to the teams on the ground that are setting up the technology and then using PantheraIDS in the field 

SHANNON DUBAY 

So those that are more technologically savvy, they’re upskilled through a combination of large training sessions, mostly virtual sessions. We’ve got a lot of documentation, and importantly, we’ve got demos that include some sort of training data set that they can play around with. But those that are less comfortable with technology we find are most effectively trained in person in smaller groups. We prioritize a sort of holistic technology training.

HANNAH LABELLE

Has there been a particular use case when it comes to PantheraIDS that you’ve really enjoyed seeing the growth or kind of like a project that’s like, “Oh, that’s a really good one to talk about? 

SHANNON DUBAY

We have a project in Zambia at the moment, and we do a lot of different conservation activities there, helping to manage a park. We also partake in activities surrounding citizen science. So there are often local communities and some wildlife lodges that are in the area, and we have programs for those people to record any wildlife observations that they see and submit that data for us to be able to better understand the landscape and make some informed conservation decisions while still involving local communities and even tourists, if they’re there for ecotourism.  

In that area, there’s a large focus on lion and leopard protection and conservation. We’re able to see if the different efforts we’re doing with our citizen science program, with our anti-poaching efforts, we can actually see by camera trapping if the lions and the leopards are increasing in population and if we’re having an actual impact that can be measured. There are limited resources for conservation globally, and with the resources that we do have, we want to make sure we’re making the most impact. 

HANNAH LABELLE

How has the increased use of technology impacted your success metrics on projects?  

SHANNON DUBAY

I think what technology has allowed conservation to do is move from reporting and using outputs as our measure of success. If I think of an anti-poaching team that goes out and does patrols, on their patrol, they might see snares, for example, and one of the metrics that historically we might use is how many snares have we collected? It doesn’t really give us an indication on the outcome. What does it mean? And I think with technology, it gets us closer to a space where we can now assess our outcome, which might be how it’s impacting cats and other animals in the landscape. Because at the end of the day, we want to make sure that we are protecting animals.

HANNAH LABELLE  

What do you see as the next big thing and the next big tech advancement in conservation? And how do you anticipate it’s going to impact conservationists and how they execute projects? 

SHANNON DUBAY

Tech advancements in conservation are constant. Many of the really advanced technology tools have been cost-prohibitive for the field for quite a while, so it’s really great to be able to see technology becoming a little bit more accessible to maybe those of us that are not as technologically advanced and also don’t have huge project budgets. 

So, some of the big tech advancements that I see coming in the next few years [are] really going to be real-time, AI-driven monitoring systems, and those that integrate different data types from camera traps, from bioacoustics, from satellite imagery, and really synthesizing that all together. But something that I’m particularly excited about that’s picking up is predictive analytics. 

In conservation, it’s often a struggle for us to use all of the data that we collect in an adaptive way. We really want to be able to harness all of our data and apply it to our actions and adapt accordingly, so that we’re really flexible, really agile, and we can respond to any growing threats that might be emerging. For example, predictive analytics will help park rangers to identify and therefore prioritize different high-risk areas. It’s going to really help synthesize different data streams and also make anti-poaching efforts a lot more proactive and therefore effective. One other thing I have to mention is just having greater reliable internet connectivity is already, and will continue to be, a huge game changer.  

While I’m not a technologist by training—I’m actually an ecologist—but I can see the value in technology and the impact that it can have. And that, for me, is really important. I love being around wildlife, and I want to see some of our most wild places continue to be wild.  

MUSICAL TRANSITION

STEVE HENDERSHOT 

Shannon mentioned using sensors, imagery and other data-collection tools as key components of Panthera’s strategy going forward. Panthera isn’t alone in adopting those technologies. Our next guest is Dave Thau, global data and technology lead scientist at WWF. Dave’s based in San Francisco, and he’s pursuing a similar approach as part of his teams’ efforts to preserve mangrove populations around the world. The plants grow right at the edge of land and water and play a crucial role in the health of the ecosystems where they’re present. 

MUSICAL TRANSITION

STEVE HENDERSHOT 

Okay, Dave. What role is advanced tech playing in conservation? How is that changing? What’s it making possible that didn’t used to be possible, and how did that draw you in?

DAVE THAU

Conservation organizations have been using technology to study the environment for a really long time, and AI has been part of that. Some of the earliest AI applications involved looking at images taken from airplanes and trying to classify different parts of the images as forests or water or crops for natural resource management. And that was work that was done in the ‘70s. 

But what’s happening now is it’s accelerating. It’s available to more people. The systems are much more advanced. Our ability to collect data has increased vastly. And as we have more and more access to more and more data, the importance of artificial intelligence in helping us process the data has grown. What we’ve been seeing is sort of an acceleration in the application of artificial intelligence to deal with the new streams of data. And also, with the newer large language models (LLMs) that have come out, the application of AI has really broadened from collecting data from sensors to sort of all aspects of conservation.

STEVE HENDERSHOT 

And what’s that doing? Sort of writ large, how is this affecting what’s possible in terms of conservation with the upsurge in data available and the LLMs and such to process it?

DAVE THAU

A lot of it is about efficiency—the ability to do a lot more work than we could have done in the past and in broader areas. The ability to process information quickly and to share it widely. That’s just made it so that we can do more work faster, over larger areas and shorter timespans. So it increased the amount of space we could do analyses and also the speed at which those analyses can be done and disseminated. 

Conservation organizations have been around for a really long time. WWF has been around for 60 years. We have an amazing amount of information about the work that we’ve done and the impacts it has had, but that information is usually in large text documents, and it’s so much that no one person can read them all or have them all in their mind. So one of the things that the large language models have been able to do is help us understand all of these documents—not just our own documents, but all the documents from other organizations that are working in the same landscapes that we do. 

STEVE HENDERSHOT 

Let’s talk about the ManglarIA project—the name means AI for mangroves in Spanish. Before we get into specifics, tell me about the main focus of the project: mangroves. What are they, and why do they matter to conservation? 

DAVE THAU

Mangroves are a number of species of trees that live at the nexus of freshwater and saltwater. There are a number of important services that they provide to humans in addition to the ecosystems that they create. So one thing that mangroves do is they store a lot of carbon. They provide coastal protection. They provide fisheries. They also help filter water. They’re difficult to monitor because they’re at this nexus of freshwater and saltwater, and also being on the coast, the access is challenging. Satellite imagery can be difficult. So there are many things that make working in mangroves tough. We really need to understand what’s impacting them.

STEVE HENDERSHOT 

OK, let’s dig into the project’s goals and the tech you’re using to better understand these mangroves. 

DAVE THAU

It’s very focused on the impacts of climate change on mangroves and trying to understand that better. And this is in terms of carbon sequestration, coastal protection and the impact on fisheries. Through a number of discussions with our mangrove experts, we determined a set of sensors that would be useful to answer specific questions. And those sensors, it’s a wide variety. Satellite data is one. We also have drones so we can get more fine-grained information, but we have weather stations that are measuring wind speed and temperature. There’s lidar (light detection and ranging), which is a way of getting a 3D model of the area. We’re looking at currents in the rivers that provide fresh water. We’re also looking at sea level height and sea temperature. There’s also a carbon flux tower. 

There’ve been installations of some of these sensors in the past. What’s new is how we’re integrating them and how they’re all being deployed in the same landscape in a way that lets us do very rapid monitoring. We’re installing motion sensitive cameras to track wildlife, and we’re using artificial intelligence to analyze the images from those camera traps. And then sort of the integration of all of these data streams to address the specific questions that we have. That’s some of the innovation that’s coming out of this project.

STEVE HENDERSHOT 

There’s a project certainty question here, which, I mean, there’s enough new technologies, unfamiliar environment. How have you approached iterating, adjusting, adapting?

DAVE THAU

It’s a pilot project, so this is the first time that we’re doing exactly this methodology. One of the key features of it is all of the information is going into the same, we’ll call it a data lake house, which is then made available to all our partners. The application of artificial intelligence to all of these different streams simultaneously is new, so we’ll see where it goes. And there are kinds of AI that we’d like to apply, which are definitely cutting edge. We’re not sure we’ll be able to do it. And then, there are different kinds of AI models that are developing. And that’s one of the things that makes this project really exciting, is we know we’re going to get some good findings from the technologies that are now kind of standard, and we’re also really excited about seeing where we can go with the technologies that are being developed right now. 

STEVE HENDERSHOT 

How have you approached that from a project structure perspective? Just that combination of uncertainty and certainty in terms of sort of timeline, scope. Obviously this is part and parcel of just making it a pilot project and leaving yourself room to explore, but is that something that you’ve had to address? “We will definitely be able to deliver X. We’ll be exploring the space on Y.”

DAVE THAU

We work with communities. Anything we do has to be in consultation with the local communities. We rely on the relationships that WWF Mexico has built over, many, many, many years. 

Some of these sensors are being deployed with the communities. And so there’s a lot of community engagement that we do to ensure that the projects would be useful for them—and that they can benefit and develop it with us. So there is a lot of management around that. 

Then, the grant is a three-year grant, but we’re installing all of these sensors. We want the sensors to be used after three years. Hopefully we can find continued funding, but no matter what, we have to have people who are there in Mexico who can take over the sensors and continue the data collection and the analysis. Because of that, we have lots of partnerships in Mexico. We’re working with universities and some government agencies so that they understand the technology that’s being developed. They’re developing it with us, and we can turn things over to them depending on what goes on with the grant. 

Google.org, who are the funders, their fundamental desire is to see how we’re applying these technologies and artificial intelligence to make a difference, to find measurable change and be able to make decisions based on that. We know for sure the technology underpinnings, how those will be rolled out. We’re confident of that. And then the insights are going to come largely out of the partnerships that we’re building with the folks in Mexico.

STEVE HENDERSHOT 

How does all that engagement affect your team structure? How are all the teams at different locations working together, especially as you’re looking to future-proof this effort?

DAVE THAU

We have different teams. So I’m leading the technology team. We have a science team, and then we have these project managers in the different sites and also the U.S.-based project manager. 

One thing that’s been great to watch is how communication with these teams has been sort of changing over time and how we’re getting more comfortable with everybody’s cadence of communication. Because we’re working with WWF Mexico and they all already have relationships in place, getting a better understanding of how often people go out into the field and what things can happen quickly and what things are going to take a long time and just getting a better sense for that, has been developing over time. There’s definitely been a lot of back and forth in terms of which technologies we’ll use. People have been very willing to say what their capacities are and what they’re interested in learning more of. So that’s been good to watch. 

STEVE HENDERSHOT 

How did you pick your success measures, and how’s that going now that it’s in practice and not just in grant application form?

DAVE THAU

There’s several different kinds of measures that we’re looking at. The primary one is the number of decisions that can be made with the data that we’re making available, the ability of our partners to decide on what kinds of restoration activities to engage in based on our data. Then there are metrics that are pure technology metrics, like how many different data streams are available, the availability of the data, the uptime. There are a number of features like that as well. And then, how many different data streams we can integrate to make specific observations. And then just partners, right? The strength of the partnership.

We’re hoping that the science-based insights that we have about mangroves in these areas will be transferable to other places, but also it’s the technology. The expectation is that we’re doing both the technology transfer and the knowledge transfer.

STEVE HENDERSHOT 

Thanks so much for talking with us, Dave. 

DAVE THAU

Oh, my pleasure. Thank you for having me on the show.

STEVE HENDERSHOT 

And thank you for listening to Projectified. Like what you heard? Subscribe to the show on your favorite podcast platform and leave us a rating or review. Your feedback matters a lot—it helps us get the support we need to continue making this show. And be sure to visit us online at PMI.org/podcast, where you’ll find the full transcripts for episodes as well as links to related content, like useful tools and templates, the latest research reports and more. Catch you next time!