‘No end in sight’ for potential of conservation tech: Q&A with Megan Owen

By Abhishyant Kidangoor via MongaBay

Early in her career, Megan Owen recalls, she often found herself wishing for more data to understand certain behavioral patterns in the animals she was studying. However, more often than not, she says, she had to submit to the fact that “there was just no way we’ll ever be able to get that data.”

But that scenario gradually changed as technology witnessed a rapid boom over the years.

“Now, I don’t think we say that anymore,” Owen, a behavioral ecologist and vice president of wildlife conservation science at the nonprofit San Diego Zoo Wildlife Alliance, tells Mongabay.

Owen now leads a team of biologists, ecologists and engineers who run a conservation technology lab at San Diego Zoo Wildlife Alliance with the aim of developing and deploying technologies that could help speed up the implementation of wildlife management and conservation strategies. Apart from developing automated workflows to collect and process vast amounts of data, team members at the lab are also working to develop and mentor the next generation of conservation technology practitioners.

Operating a tech lab in a zoo setting, Owen says, has its perks. “We are in a situation where we can validate a lot of our tools working with the wildlife in our care, and that’s just a tremendous boon,” she says.

Megan Owen spoke with Mongabay’s Abhishyant Kidangoor about the lab’s goals, the challenges in developing new technology, and how she plans to build the next generation of conservation technologists. This interview has been lightly edited for length and clarity.

Mongabay: To start with, could you tell me what the conservation technology lab at SDZWA does?

Megan Owen: The conservation tech lab is part of our conservation science team. Its focus is on the development and implementation of technologies in any form that help advance our work and help us meet our objectives in a more timely and effective way. A big part of the focus is also on developing a pipeline of the next generation of conservation technology practitioners.

Technology is obviously so important to conservation, both in terms of collecting and retrieving data quickly and synthesizing data, and being able to develop management and mitigation strategies based on the information we collect. So it’s just critically important for conservation right now. At the lab, which has been around for about seven years, it has really driven us to coalesce all our activities that have a technological focus.

Mongabay: What gaps are you trying to fill with the lab?

Megan Owen: Our organization has a long history of applying science to conservation to solve conservation challenges and problems. Our team is primarily made up of biologists. But often, as biologists, and I’m a biologist as well, we identify things that we can address but the challenges of getting the data are enormous. I’d say for the past 10-15 years, tech tools have become prominently used by biologists in the field. The tech lab was started to be able to lean into the need to develop and deploy technologies to collect relevant data, and to coalesce all of our programs around that. If we could develop a certain technological solution in one program, there’s a very good chance we could apply similar tech in a different program. So we really wanted to pull all of those different uses of technology together, and then also articulate a need to bring engineers into the conservation science context. Now, we actually have three engineers without biological training working with us and that’s really been a game changer.

Mongabay: Could you walk me through some of the projects at the lab that you are proud of?

Megan Owen: Like many conservation scientists, we work with camera traps, or trail cameras, and we deploy those in many programs all over the world. We generate millions and millions and millions of image files. All of those have to be gone through, classified and turned into usable information. That volume has really prompted the need to develop automated approaches to process those image files. Our team has put a lot of effort into developing those automated workflows for classifying camera trap images, and those efforts have been really invaluable. It has been applicable whether we’re talking about the Amazon or Kenya or here in the U.S. Southwest. We’re able to apply that kind of approach and really speed up the data synthesis process and turn around those findings for management. That same kind of process, we’re also using for acoustics. We’re starting to do a lot more acoustic monitoring and do that same kind of classification. That’s a big part of what we do.

The other piece of what I’m very proud of what we’re doing is an intensive focus on cultivating the next generation of conservation technologists. We have a real strong focus on internships, and we work with a number of universities here in Southern California. Our team puts a tremendous amount of effort into that mentorship.

Mongabay: How would you describe the impact the lab’s work has had on conservation programs?

Megan Owen: I would, again, go back to the processing of trail camera images. The time to process these thousands and thousands of images has been accelerated tremendously. That’s a game changer for what we do. Because often we’re addressing management challenges that are happening now. We don’t have three years or more to wait to get those data back. So that acceleration of the processing of camera traps imagery has been a really important element of our work. We still do work with humans to classify images to some extent. Engaging on a citizen science level with the community has a lot of benefits as well. We had been using citizen scientists to do some of our image classification. We tried to identify those areas in the workflow where having a human do some level of classification is still beneficial because engaging with people on conservation is just as important as ever. So if it doesn’t slow down the process, we want to continue to do that as well.

Another area that I’m proud of the developments we’ve made have been in recognizing autonomous video units for deployment in severely remote and difficult areas. One example would be a collaborative polar bear conservation project with the NGO Polar Bears International. We have a project where we are monitoring emergence behavior of polar bear moms and their cubs in the high Arctic. In order to do that, we work together to develop these camera units that we can readily deploy. They’re able to capture image and video data throughout the winter, and they’re able to withstand all that temperature and wind and wildlife. The data we get back is invaluable. The challenges of working there are significantly greater because of the terrain. The topography is very steep, and so it prompted a whole other level of innovation, both in power sources and configuration. Because you have to basically ski to the point of deployment, it needs to be compact and easy to use.

Mongabay: I’m keen to hear your perspective on how conservation technology has evolved over the years.

Megan Owen: What surprises me the most is the rapid pace of change. There’s no end in sight.

I’m a behavioral ecologist. My thought process has always been that it would be really wonderful if we understood something, but there’s just no way we’ll ever be able to get that data. And now, I don’t think we say that anymore. It’s just this feeling of almost endless possibility, which is phenomenal. The pace at which we’ve been able to automate systems and develop workflows that can crunch through millions of data points, it has just transformed how we approach these challenges.

We’ve also been able to connect so effectively with this broader community of conservation technologists, and that has made a huge difference as well. That community has been out there for a while, but it’s just weaving together more effectively now. The discussion of ideas, the sharing of open-source tools that are developed, that really gives me hope that we’ll continue to develop these tools, and we’ll be able to turn around the data at a pace that actually meets the challenge we’re trying to address.

Mongabay: What have been the challenges in doing this?

Megan Owen: A huge challenge in developing and deploying technology that works is in that calibration and validation phase. How do you interpret the data you’re collecting? What does it mean biologically for the organism of study? Working for the San Diego Zoo Wildlife Alliance, we are in a situation where we can validate a lot of our tools working with the wildlife in our care, and that’s just a tremendous boon. That’s something we’ve worked with other partners in a similar vein where there’s technology being developed by a partner, and they’ve come to us asking if they can validate working with us and the wildlife in our care. So that’s just a tremendous advantage in developing usable technology.

The big challenges are still making sure the tools are low-cost and accessible, fixable, deployable and programmable. I’m not an engineer, and so it’s really important that the tools are readily usable by biologists like me. That continues to be a challenge in some areas. It’s also important to make sure these tools are as low-cost as possible. AudioMoth is a fabulous example. The tools that had been available were out of reach if you wanted to deploy at scale. And then AudioMoth came along, and suddenly you had an incredibly affordable unit that allowed you to scale up. I’m very optimistic that the community as a whole is working towards ensuring that these tools are affordable and readily usable by biologists.

Mongabay: What have been your biggest learning experiences in running a tech lab in a zoo setting? What advice would you give other zoos around the world?

Megan Owen: I think my advice would be to think of it as a community, and to recognize the creativity that’s required to develop innovative solutions. I would also advocate for not overlooking the tools that are available off the shelf. Some of the best solutions that we have ended up deploying are a combination of off-the-shelf tools coupled with some customization. At the end of the day, what that reflects is the focus on the objective you’re trying to meet, and using whatever gets you there fastest.

Mongabay: What does the future look like for the conservation tech lab at SDZWA? What can we expect to hear from you in the next two to five years?

Megan Owen: In conservation technology, that’s a long time frame because of the rapid pace of development. But what I’m hoping is that the machine learning and AI packages that our team have been developing for decoding the data that we’re collecting are going to be readily available and improvable for us and for the conservation community, so that we can continue to put out noninvasive sensors, whether they’re acoustic or visual, and be able to get that data back in real time. Being able to transmit data rather than to have to download data, that would be a huge boon.

I’m also excited about the potential of onboard processing. For example, if we have a targeted project, and we can design sensors that have that onboard processing capacity, doing that data synthesis right there and sending back the information that we’re trying to get would be fantastic. I am very optimistic about what the next two to five years hold.

I’m hoping that we can continue to grow that interest in conservation from the engineering community. I also want to really put a tremendous amount of energy into cultivating the next generation of conservation engineers. I think it’s a field that is going to continue to grow, and we’ll all benefit from it.

Read the original article on Mongabay here.