- The semifinal testing for a $10 million competition to identify technology that automates the assessment of rainforest biodiversity is underway in Singapore.
- The five-year competition is organized by California-based nonprofit XPRIZE Foundation.
- From robotic dogs to drones and novel methods to gather environmental DNA, 13 teams are competing for a place in the finals next year.
SINGAPORE — It’s been a hectic few weeks for Kevin Marriott.
On a humid May morning in Singapore, he shuttles between two locations at the city’s Windsor Nature Park. At one site, he looks on as a team of scientists and aviation experts get ready to launch a fleet of drones to gather information about biodiversity from tree canopies. A 10-minute walk away, another team attempts to send a robotic dog out into the forest on its own to collect photo and audio data.
The teams are among 13 projects from around the world that qualified for the semifinals of a competition with prize money of $10 million. Organized by California-based nonprofit XPRIZE Foundation, the five-year competition aims to accelerate technological solutions that will automate the monitoring and assessment of biodiversity in rainforests.
As the technical lead for the competition, Marriott has been busy planning logistics and rallying the teams that are taking part. They come from around the world — Brazil, Mexico, U.K., Switzerland and elsewhere — and are given 24 hours to gather data from the forest. They get another 48 hours to analyze the data and draw conclusions about biodiversity from it. While automated collection of environmental DNA (eDNA) — the genetic material left behind by animals — seems to be the primary focus of most of the teams, others are deploying drones and robots to collect visual and audio data.
“We hope the solutions provide new insights into the forest that we haven’t been able to achieve before,” Marriott told Mongabay.
The competition, launched in 2019, is in its penultimate year. The final testing is expected to take place in mid-2024.
Kevin Marriott sat down with Mongabay’s Abhishyant Kidangoor to speak about what he hopes will come out of the competition, why Singapore was chosen as the site for the semifinals testing, and the state of conservation technology today. The following interview has been lightly edited for length and clarity.
Mongabay: To start with, tell me what you’re doing in Singapore.
Kevin Marriott: XPRIZE is here in Singapore to run the semifinals competition of XPRIZE Rainforest, which is a $10 million competition to survey biodiversity in tropical lowland rainforests. It was launched in 2019. It is a five-year competition and we have two years left. Until now, the teams have had time to develop all of their solutions, from conceptualizing ideas through to actual solutions that can be used on the ground. This testing in Singapore is the first time that the teams are actually being assessed and tested by XPRIZE. Prior to this, it was only paper submissions and video submissions. This is the first time that we are actually seeing the teams using the solutions to capture biodiversity.
Mongabay: How do you think it’s going?
Kevin Marriott: The testing process is working well. I think the teams are well-prepared for the competition. They have designed their solutions and they seem to be fit for purpose. There are still a few tweaks and modifications to be made. But after this process, they have still got another 12 months before the finals in which to further develop their solutions to make them as best as they can be.
Mongabay: Out of everything you have seen through this entire process, what excites you the most?
Kevin Marriott: The most innovative solution to me is the deployment method. Some of the teams are using existing and tried-and-tested survey methodologies, and trying to use automated technology to be able to deploy those. For example, getting sensors into tree canopies. As a conservation practitioner, having that in your toolbox is going to be very beneficial to serve biodiversity. Usually, it is manpower-intensive to deploy sensors in rainforest canopies and to rapidly deploy several sensors in such a short period of time. So hopefully, these solutions are going to open up the usability of things like bioacoustics in tree canopies. The tree canopy layer of the forest is one of the most underresearched. So being able to survey that and capture all the data from the tree canopy is, I think, key to making new discoveries within the forest.
Secondly, being able to automate eDNA technology, and using robotic technologies to capture eDNA samples, that is really impressive. The majority of our teams are using eDNA as part of their solutions. So the strongest teams really are the ones that have looked at all the levels of the forest and how they can tackle surveying the biodiversity within 24 hours across all those forest layers.
Mongabay: What is your hope for these projects after the end of the competition next year?
Kevin Marriott: A part of the competition includes a 12-month impact phase at the end. Once we award the final milestone payment, that’s not the end of the competition. That’s only the end of the competing period. But we will still be actively engaged to produce impact for the next 12 months, and help those winning teams get the technologies into the hands of end users and practitioners.
We will, hopefully, enable small local communities to protect their lands and produce the kinds of data that are normally expected by an investor for biodiversity or carbon credits. They expect a certain level of reporting to be provided, and a lot of local communities, in order to provide that, have to partner with some kind of NGO or have biologists on site to be able to do that. I am hoping the technology will be able to empower those people as Indigenous scientists so that they can use the technologies to produce the level of reporting that would be expected of conservation projects.
Surveying biodiversity in areas that are typically dangerous or haven’t really been scientifically researched, or areas that are just too difficult to go to, that’s also another application I would hope for. For instance, there is a large part of the forest in Virunga in the Democratic Republic of Congo that is in rebel territory. We don’t know what’s going on with it. I have worked in Virunga for a number of years and it’s really sad to see what was happening. I really hope that some of the technology that’s going to come out of the competition can give some hope that the biodiversity is still strong within that area, and it’s still surviving the conflict there. So for me personally, that’s what I would like to see: remote sensing of biodiversity in an area that’s just difficult to go to or is just too dangerous.
We also hope that the solutions provide new insights to the forest that we haven’t been able to achieve before. Maybe a new species discovery or the rediscovery of a species that we thought was locally extinct from some areas.
Mongabay: I am curious to know your background and what sparked your interest in this field.
 Marriott, who was with the British Army for 14 years, has previously worked with technical surveillance systems and with intelligence services. Image courtesy of XPRIZE Foundation. |
Kevin Marriott: I was with the British Army for 14 years. Eight years out of that was with the special forces as a radio communicator. I was kind of the tech guy there. So I used to work with technical surveillance systems where we are tracking people as part of counterterrorism efforts globally. I left the military in 2012, and started working as a consultant for the intelligence services around East Africa. It was going in and out of Somalia via Nairobi that got me involved in conservation.
My old sergeant major was working to train rangers in Kenya. His aim was to kind of increase the capabilities of the rangers to effectively carry out law enforcement in northern Kenya. So I went up there, and that’s where I started. And once I started working in conservation, I was hooked. I have always had an affinity for wildlife since I was young. But having been exposed in the frontline pushed me further along. So I started studying toward getting a master’s degree in wildlife biology and conservation. I was lucky enough to work with some very prominent conservationists who helped me and guided me along that academic route.
Mongabay: Why was Singapore selected as the site for the semifinals?
Kevin Marriott: To logistically support so many teams in Congo or the Amazon would have been very difficult. And to throw the teams into a rainforest like that, when some of them have never been in any kind of protected area or outside of their home country, could have been very challenging. It would have induced additional pressures on the team because they are just not used to the environment. So that was one of the reasons why the decision was made to have the semifinals in Singapore, in order to ease the teams into the rainforest environment. So testing them in a smaller rainforest, that still is very similar to the likes of the Amazon or the Congo, but has got additional ease of access so that the teams can trial their ground robotics and not be a complete washout. But the finals will be in a place with a real rainforest where it’s going to be as challenging as it could possibly get. Our aim is to really fully test the solutions to make sure that they are fit for purpose before we roll them out into that impact phase.
Mongabay: What have been the challenges in planning this in an urban environment?
Kevin Marriott: It was difficult in terms of airspace restrictions for flying UAVs. Then there is also the restriction on space since Singapore doesn’t have a large rainforest. It’s just fragmented rainforests in the center of the city.
But the silver lining of doing the competition in a place like Singapore is that the teams’ technologies have been verified by rigorous airworthiness testing and there was a verification process of all of the pilots to make sure the pilots are competent. So that ensured that the UAVs that the teams are operating are safe. So the teams have experienced what it takes to go through those types of airworthiness checks.
We could have chosen somewhere that would have been easier in terms of gaining the permissions to operate in the country. But that’s not really testing. Because then, we would have effectively championed solutions that aren’t actually approved to fly anywhere. Now the teams are faced with having to go through all those verification and airworthiness checks. That’s one of the things within conservation technology that I have seen before. UAV companies or organizations produce these really good UAVs for surveillance or antipoaching efforts. And when they try to actually get it onto the ground and work to assist with missions, they just don’t get approval from the government. So having the teams go through this process means that when they face those types of challenges in the future, they have already done it.
Mongabay: How do you think conservation technology is faring?
Kevin Marriott: There is some really innovative technology that’s coming out to help conservation. I think the gaps are in the deployment systems, which is why I am really excited about this competition. Things like bioacoustics are really coming along and helping us understand species communications and making it easier to detect subaudible-type sounds. But they still require people on the ground to actually implement these solutions. So having the automated side of the technology and the deployment mechanisms like aerial systems or ground robotics is just one of those things that, I think, is really going to help push those solutions even more and get them into the places that could really benefit from them.
Mongabay: How about the analysis part of the collected data? How do you think that is coming along?
Kevin Marriott: One of the things that’s lacking really is the actual libraries of the ecosystems so that AI models can be trained to do the analysis. What I would love to see is big efforts in building and developing those libraries. I think that should be done in conjunction with local communities that probably know all these different species because they live among them. It will be a good way for them to be able to use those libraries, develop them and take ownership of them as well. And potentially monetize them and use the funding to be able to support efforts of their own to protect their ancestral homes.
Mongabay: What are your biggest hope and worst fear for the future of our planet’s biodiversity?
Kevin Marriott: Our rainforests are getting decimated and if we don’t do something about it, it’s going to be too late. The competition will also hopefully raise awareness of the function of the forest and how important it is.
My biggest hope, I think, is that local communities gain a voice and take ownership of a lot of these protected areas. And that governments actually listen and support them. They are not asking for a lot. They are just asking for their homes to be protected.
I also hope that we find some way of being able to compensate countries that are looking after important assets like rainforests, which the whole world benefits from. Ultimately, people have to place a value on biodiversity. I hope that people do realize that we need to conserve biodiversity for our own survival.
Abhishyant Kidangoor is a staff writer at Mongabay. Find him on Twitter @AbhishyantPK.
Bioacoustic analysis made easier: Q&A with Rainforest Connection CEO Bourhan Yassin
